From wirges@expert.cc.purdue.edu Thu, 31 Aug 2000 16:21:10 -0500 Date: Thu, 31 Aug 2000 16:21:10 -0500 From: Matt wirges@expert.cc.purdue.edu Subject: [PLUG] More *fun* with port forwarding Can anyone explain to me why I cannot access my local webserver through my firewall. What I mean is that my firewall <192.168.0.255, www.domain.com> port forwards to my webserver <192.168.0.1>. From the outside world you can go to www.mydomain.com and get the webserver. However internally when I goto www.mydomain.com (on any computer, including the webserver) it just sits there. -- ============================================================================== | Matthew Wirges | | Systems Administrator, Student, Programmer, Geek. | | Office Phone: [765]464-1148 | | Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com | | www.sharpwebinnovations.com | ============================================================================== From reh@blacksoul.net Thu, 31 Aug 2000 16:25:02 -0500 (EST) Date: Thu, 31 Aug 2000 16:25:02 -0500 (EST) From: Ryan E. Helfter reh@blacksoul.net Subject: [PLUG] More *fun* with port forwarding Do you have www.domain.com (whichever your domain is) inside your /etc/hosts file? If so, traffic may not be going outside and then back in. And did you set a rule for portforwarding on 192.168.0/24? The /24 rule would be my first guess as to why its not working. ry On Thu, 31 Aug 2000, Matt wrote: > Date: Thu, 31 Aug 2000 16:21:10 -0500 > From: Matt > To: plug@csociety.purdue.edu > Subject: [PLUG] More *fun* with port forwarding > > Can anyone explain to me why I cannot access my local webserver through > my firewall. > What I mean is that my firewall <192.168.0.255, www.domain.com> port > forwards to my webserver <192.168.0.1>. From the outside world you can > go to www.mydomain.com and get the webserver. However internally when I > goto www.mydomain.com (on any computer, including the webserver) it just > sits there. > -- Ryan E. Helfter reh@blacksoul.net Black Soul Networks, LLC -- From porterds@purdue.edu Thu, 31 Aug 2000 16:23:32 -0500 Date: Thu, 31 Aug 2000 16:23:32 -0500 From: Doug Porter porterds@purdue.edu Subject: [PLUG] More *fun* with port forwarding i have run into this problem before too. in the most recent situation it was a very large (2,200 host) network. the firewall was a cisco pix, and it was running nat. i ended up creating internal and external dns servers so that the hostnames would resolve properly for local and external clients. for a small network though i would just edit the host files on local machines to correct the problem. - doug porter Matt wrote: > > Can anyone explain to me why I cannot access my local webserver through > my firewall. > What I mean is that my firewall <192.168.0.255, www.domain.com> port > forwards to my webserver <192.168.0.1>. From the outside world you can > go to www.mydomain.com and get the webserver. However internally when I > goto www.mydomain.com (on any computer, including the webserver) it just > sits there. > -- > ============================================================================== > | Matthew Wirges | > | Systems Administrator, Student, Programmer, Geek. | > | Office Phone: [765]464-1148 | > | Email: wirges@expert.cc.purdue.edu || > sysadmin@sharpwebinnovations.com | > | www.sharpwebinnovations.com | > ============================================================================== > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From cnd@ecn.purdue.edu Fri, 01 Sep 2000 16:04:29 -0500 Date: Fri, 01 Sep 2000 16:04:29 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Computer Organizations Callout Fair What: Computer Organizations Callout Fair When: Thursday, September 7th, 10AM - 2PM Where: MSEE Atrium If you can spare some time to come sit at our table, pass out some flyers, or just come hang out and chat, please stop by. Bring your friends, your family, or your penguin (or little red devil, whichever you please.) -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From andrews@technologist.com Fri, 1 Sep 2000 16:07:31 -0500 Date: Fri, 1 Sep 2000 16:07:31 -0500 From: Will Andrews andrews@technologist.com Subject: [PLUG] Computer Organizations Callout Fair On Fri, Sep 01, 2000 at 04:04:29PM -0500, Christopher N. Deckard wrote: > friends, your family, or your penguin (or little red devil, whichever *smack* I beg to differ. That is called a daemon, not a devil. -- Will Andrews GCS/E/S @d- s+:+ a--- C++ UB++++$ P+ L- E--- W+ N-- !o ?K w--- O- M+ V- PS+ PE++ Y+ PGP+>+++ t++ 5 X+ R+ tv+ b++ DI+++ D+ G++ e>++++ h! r- y? From olenik@purdue.edu Fri, 01 Sep 2000 19:07:36 -0500 Date: Fri, 01 Sep 2000 19:07:36 -0500 From: Hayden Olenik olenik@purdue.edu Subject: [PLUG] Compilation problems I'm trying to compile certain programs involving the playback of DVD's under operating systems used by "hackers" instead of systems sold by monopolies. These progarms that are part of the Linux Video project compiled fine under RH6.2. However, under RH6.9.5, they won't compile, citing errors in pasting or something. Does anyone have binaries for RH6.x? Hayden From wirges@expert.cc.purdue.edu Sat, 02 Sep 2000 16:48:14 -0500 Date: Sat, 02 Sep 2000 16:48:14 -0500 From: matt wirges@expert.cc.purdue.edu Subject: [PLUG] IRC and a IPMASQ Firewall --------------C747214BEBDF4956C8F7245F Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Everytime I try to connect to a dalnet server I get autokilled. I have the ip_masq_irc module installed and loaded. I read somewhere that is because the dalnet server cannot get the ident request from the machine that the client is running on and instead gets the firewall. If you have any suggestions or ideas, let me know. If its relevant: I use BitchX most of the time. I have tried using xchat, and on windows boxes mirc. -- ============================================================================== | Matthew Wirges | | Systems Administrator, Student, Programmer, Geek. | | Office Phone: [765]464-1148 | | Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com | | www.sharpwebinnovations.com | ============================================================================== --------------C747214BEBDF4956C8F7245F Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit Everytime I try to connect to a dalnet server I get autokilled.  I have the ip_masq_irc module installed and loaded.  I read somewhere that is because the dalnet server cannot get the ident request from the machine that the client is running on and instead gets the firewall.  If you have any suggestions or ideas, let me know.
If its relevant:  I use BitchX most of the time. I have tried using xchat, and on windows boxes mirc. 
-- 
==============================================================================
| Matthew Wirges                                                             |
| Systems Administrator, Student, Programmer, Geek.                          |
| Office Phone: [765]464-1148                                                |
| Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com     |
| www.sharpwebinnovations.com                                                |
==============================================================================
  --------------C747214BEBDF4956C8F7245F-- From will@physics.purdue.edu Sat, 2 Sep 2000 19:07:41 -0500 Date: Sat, 2 Sep 2000 19:07:41 -0500 From: Will Andrews will@physics.purdue.edu Subject: [PLUG] IRC and a IPMASQ Firewall On Sat, Sep 02, 2000 at 04:48:14PM -0500, matt wrote: > Everytime I try to connect to a dalnet server I get autokilled. I have Try not connecting to DALnet. ;-) > the ip_masq_irc module installed and loaded. I read somewhere that is > because the dalnet server cannot get the ident request from the machine > that the client is running on and instead gets the firewall. If you > have any suggestions or ideas, let me know. > If its relevant: I use BitchX most of the time. I have tried using > xchat, and on windows boxes mirc. Run an ident server on your NAT box; one that can be configured to provide an ident name when the request doesn't match a client on the machine. Basically it'll return "somename" if a corresponding client username is not found. -- Will Andrews GCS/E/S @d- s+:+ a--- C++ UB++++$ P+ L- E--- W+ N-- !o ?K w--- O- M+ V- PS+ PE++ Y+ PGP+>+++ t++ 5 X+ R+ tv+ b++ DI+++ D+ G++ e>++++ h! r- y? From darth@purdue.edu Sun, 03 Sep 2000 00:14:27 -0500 Date: Sun, 03 Sep 2000 00:14:27 -0500 From: Darth Vader darth@purdue.edu Subject: [PLUG] IRC and a IPMASQ Firewall I have this problem too at the moment. I run Mandrake which comes with pidentd. Most distributions come with some sort of ident server partially setup so see which one you have already. I know pidentd supports masq because I had it working in 6.2 but some idents don't. I lost my setup when I installed Mandrake 7.0 and I can't remember how I set it up before. Has anyone setup pidentd recently and can shed some light on the matter? Jason Will Andrews wrote: > On Sat, Sep 02, 2000 at 04:48:14PM -0500, matt wrote: > > Everytime I try to connect to a dalnet server I get autokilled. I have > > Try not connecting to DALnet. ;-) > > > the ip_masq_irc module installed and loaded. I read somewhere that is > > because the dalnet server cannot get the ident request from the machine > > that the client is running on and instead gets the firewall. If you > > have any suggestions or ideas, let me know. > > If its relevant: I use BitchX most of the time. I have tried using > > xchat, and on windows boxes mirc. > > Run an ident server on your NAT box; one that can be configured to > provide an ident name when the request doesn't match a client on the > machine. Basically it'll return "somename" if a corresponding client > username is not found. > > -- > Will Andrews > GCS/E/S @d- s+:+ a--- C++ UB++++$ P+ L- E--- W+ N-- !o ?K w--- > O- M+ V- PS+ PE++ Y+ PGP+>+++ t++ 5 X+ R+ tv+ b++ DI+++ D+ > G++ e>++++ h! r- y? > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From raj@cerias.purdue.edu Sun, 3 Sep 2000 13:08:09 -0500 (EST) Date: Sun, 3 Sep 2000 13:08:09 -0500 (EST) From: Brian Poole raj@cerias.purdue.edu Subject: [PLUG] IRC and a IPMASQ Firewall Re: [PLUG] IRC and a IPMASQ Firewall I don't know what identd's your distros currently come with, nor how well they support MASQ nor how to set it up even if they do, so I'm not going to try to tell you how to setup pidentd or similar. My personal recommendation would be to grab a daemon that specifically supports identd for MASQed machines. The two most commonly used are oidentd (http://www.numb.org/~odin/), which I have used before and is a very flexible daemon, and midentd (http://p8ur.op.het.net/midentd/), which I have heard good things about. Shouldn't take too long to setup and then you will once again be IRCing in peace. Both of these daemons have a strong aim to support MASQed clients and thus have a good deal of help on it. enjoy, -b (btw, it is still masquerading on Linux, not NAT, NAT comes with the 2.4 kernel :) From cnd@ecn.purdue.edu Sun, 03 Sep 2000 15:53:40 -0500 Date: Sun, 03 Sep 2000 15:53:40 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] reminder - no meeting Reminder that there is NO meeting tomorrow night. Have a great Labor Day. Also, another reminder that Thursday is the Computer Organizations Callout Fair. Come see us or help out. -Chris -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From dresdow@csociety.purdue.edu Mon, 4 Sep 2000 18:33:51 -0500 (EST) Date: Mon, 4 Sep 2000 18:33:51 -0500 (EST) From: weston dresdow@csociety.purdue.edu Subject: [PLUG] Computer Society Callout Come one, Come all! It's the grand event y'all have been waiting for... Purdue Computer Society Fall Callout 2000 when: Thursday, September 7th -- 7:00PM where: EE005 Pizza: Free!!! why: The Csociety is a student organization committed to providing resources aiding computer education outside of the classroom. We accomplish this by sponsoring technical presentations, giving short-courses, and maintaining our own network of computing resources. We currently run an IPv6 development machine, an FTP site containing mirrors of popular open source software, and have several workstations in our office you can use. The Csociety is for anyone interested in Computers. We have several officer positions available including: * Sales * Industrial Relations * Publicity ... and many others. So, come out and join us Thursday. info@csociety.purdue.org http://csociety.purdue.org From heckarsd@purdue.edu Tue, 05 Sep 2000 01:07:23 -0500 Date: Tue, 05 Sep 2000 01:07:23 -0500 From: Seth Heckard heckarsd@purdue.edu Subject: [PLUG] mini-linux I acquired a Cyrix/200 with 32mb of RAM that I would like to put in service as a firewall. It only has a 128mb hard disk and I'm too cheap to buy a larger one. I was looking at the linux router project and while it can do basic firewall needs, I would like a full distribution at my disposal (and I also don't have a floppy drive... too cheap to buy one again). I played with Redhat and Debian tonight -- I got Debian installed however I only have 5MB free and Redhat 6.9 wouldn't even install in less than 400mb. Are there any decent mini-distros? I would like to have all the basic stuff plus a c compiler at least so I could compile new kernels and ssh and the like. Seth From docauerj@purdue.edu Tue, 05 Sep 2000 01:24:27 -0500 Date: Tue, 05 Sep 2000 01:24:27 -0500 From: Alex Docauer docauerj@purdue.edu Subject: [PLUG] mini-linux I would think debian would be the way to go. After reserving about 32 megs for swap, there should be plenty for just a kernel and a shell and maybe some dev tools. Since this is dedicated to a single simple purpose, even some of the packages that are marked as required may not necessarily be so. I'll play with the Debian installer later and see how far I can shave it down. Otherwise, you could just build your own small sytem from scratch and forget the distributions all together, making sure you only have what you absolutely need. Alex Seth Heckard wrote: > I acquired a Cyrix/200 with 32mb of RAM that I would like to put in > service as a firewall. It only has a 128mb hard disk and I'm too cheap > to buy a larger one. I was looking at the linux router project and > while it can do basic firewall needs, I would like a full distribution > at my disposal (and I also don't have a floppy drive... too cheap to buy > one again). I played with Redhat and Debian tonight -- I got Debian > installed however I only have 5MB free and Redhat 6.9 wouldn't even > install in less than 400mb. Are there any decent mini-distros? I would > like to have all the basic stuff plus a c compiler at least so I could > compile new kernels and ssh and the like. > > Seth > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From raj@cerias.purdue.edu Tue, 5 Sep 2000 01:24:26 -0500 (EST) Date: Tue, 5 Sep 2000 01:24:26 -0500 (EST) From: Brian Poole raj@cerias.purdue.edu Subject: [PLUG] mini-linux > I acquired a Cyrix/200 with 32mb of RAM that I would like to put in > service as a firewall. It only has a 128mb hard disk and I'm too cheap > to buy a larger one. I was looking at the linux router project and there are cheap people, then there are _CHEAP_ people. You fit into the latter group. ;) > while it can do basic firewall needs, I would like a full distribution > at my disposal (and I also don't have a floppy drive... too cheap to buy > one again). I played with Redhat and Debian tonight -- I got Debian > installed however I only have 5MB free and Redhat 6.9 wouldn't even > install in less than 400mb. Are there any decent mini-distros? I would > like to have all the basic stuff plus a c compiler at least so I could > compile new kernels and ssh and the like. > Seth I would really recommend you beg a 500+M hard drive off of someone if you wish to actually be able to compile on this machine. You have to be realistic here.. a quick du shows the linux-2.2.14 dir as 90M, the kernel source compressed is at least 13-14M, bzipped. By the time you add in compilers, linking tools, libraries, etc you are way, way over budget, and thats not even accounting for the rest of the OS or swap space, nor space for your logs to fit in, which any firewalling machine will probably generate.. Now if you think about it, you really don't need compilers on the machine, you can compile the kernel on another i386 machine and send it over the wire. You could also do this for ssh and any other programs. However, I do believe it would ultimately be much cheaper for you to get a cheap IDE drive off of someone, almost everyone has at least one extra laying around. Otherwise you get to endure a constant headache from 'that stupid machine without a compiler'. If you would really like to use that 128M drive, god help you, there are a ton of mini distros around. Look in all of the standard places. http://freshmeat.net/appindex/console/mini%20distributions.html comes to mind.. Debian and Slackware should both be able to be installed in that space, as long as you don't try to get all the goodies in. -b From abhinavk@purdue.edu Tue, 5 Sep 2000 01:57:25 -0500 (EST) Date: Tue, 5 Sep 2000 01:57:25 -0500 (EST) From: kumar.abhinav.1 abhinavk@purdue.edu Subject: [PLUG] New Bie Hi I am a newbie to this mailing list and to the world of linux. I installed redhat 6.1 on my p.c. a cleraon 433 128 mb and intel 810 chipset. I have encountered a series of problems( i know some of you will have a good laugh at them but you got to start some where) This is the first one I have a zoom faxmodem which according to linmodes.com is not a winmodem. BUt Linux does not dtect it. I tried configuring my serial port using the setserial but all it showed was I/O error for all ttys. I did move the rc.serial file to the /dev/rc.d/init.d/ and on booting it assigned some address for my serial ports but not irqs. I don't know which addresses were assigned and how do is assign irqs to those devices? Basically I need to get my modem going .. and have no clue how. Abhi From guffin@guffin.org Tue, 5 Sep 2000 06:53:54 -0500 (EST) Date: Tue, 5 Sep 2000 06:53:54 -0500 (EST) From: Josh Guffin guffin@guffin.org Subject: [PLUG] mini-linux On Tue, 5 Sep 2000, Seth Heckard wrote: > I acquired a Cyrix/200 with 32mb of RAM that I would like to put in > service as a firewall. It only has a 128mb hard disk and I'm too cheap > to buy a larger one. I was looking at the linux router project and > while it can do basic firewall needs, I would like a full distribution > at my disposal (and I also don't have a floppy drive... too cheap to buy > one again). I played with Redhat and Debian tonight -- I got Debian > installed however I only have 5MB free and Redhat 6.9 wouldn't even > install in less than 400mb. Are there any decent mini-distros? I would > like to have all the basic stuff plus a c compiler at least so I could > compile new kernels and ssh and the like. > > Seth why use a harddrive at all? ethernet booting works fine, and you can pick up a floppy drive for five bucks. if you're too cheap for five bucks, well sweet baby jesus. =D josh From guffin@guffin.org Tue, 5 Sep 2000 06:57:03 -0500 (EST) Date: Tue, 5 Sep 2000 06:57:03 -0500 (EST) From: Josh Guffin guffin@guffin.org Subject: [PLUG] mini-linux On Tue, 5 Sep 2000, Josh Guffin wrote: > why use a harddrive at all? ethernet booting works fine, and you can pick > up a floppy drive for five bucks. if you're too cheap for five bucks, > well sweet baby jesus. duh what am i talking about. you can keep the kernel in a boot partition on the harddrive and mount /usr, /var, /home, etc. from nfs. easy-cheesy josh From porterds@purdue.edu Tue, 05 Sep 2000 09:41:56 -0500 Date: Tue, 05 Sep 2000 09:41:56 -0500 From: Doug Porter porterds@purdue.edu Subject: [PLUG] mini-linux Seth Heckard wrote: > > I acquired a Cyrix/200 with 32mb of RAM that I would like to put in > service as a firewall. It only has a 128mb hard disk and I'm too cheap > to buy a larger one. I was looking at the linux router project and > while it can do basic firewall needs, I would like a full distribution > at my disposal (and I also don't have a floppy drive... too cheap to buy > one again). i am one of seth's friends, and he is not as cheap as he sounds. he just doesn't want to spend (waste) much money on computer stuff for now. i found something that could be very useful called tiny linux. it is a very small distribution which even provides an x server if you choose to install it. they recommended 8 megs of ram or better, and 50 megs disk minimum, 80 for a more functional system. the only downside is that it must be installed from 14 floppies, but you _could_ borrow the floppy drive from your primary computer. http://tiny.seul.org/en/ - doug porter From heckarsd@purdue.edu Tue, 05 Sep 2000 11:38:20 -0500 Date: Tue, 05 Sep 2000 11:38:20 -0500 From: Seth Heckard heckarsd@purdue.edu Subject: [PLUG] mini-linux Doug Porter wrote: > i am one of seth's friends, and he is not as cheap as he sounds. he just > doesn't want to spend (waste) much money on computer stuff for now. Very true! I spent way too much money over the summer (one of the side effects of buying a car is that you have less money in your savings account afterwards...) and I was trying to make this work as is. But, realistically, 128MB will not be enough for a decent setup, so I will probably make my way out to salvage today and see if I can pick up a ~500MB hard disk for a somewhat decent price. Then it should be much more comfortable. It could be worse, when I first got the computer it only had 8MB of RAM! Also, although a perfect computer-based firewall would have no services running on it, I think that I will have to at least run my mail server and DNS server off of the Cyrix. Add in those two and there's no way I'd be able to cram all that in 128MB. > i found something that could be very useful called tiny linux. it is a > very small distribution which even provides an x server if you choose to > install it. they recommended 8 megs of ram or better, and 50 megs disk > minimum, 80 for a more functional system. the only downside is that it > must be installed from 14 floppies, but you _could_ borrow the floppy > drive from your primary computer. Oh, and a word of wisdom... floppy disks are *not* hot-swappable :-) So don't even try it! Computers have a tendency to reboot when the floppy disk cable comes unseated (and then you have to sit through 15 minutes of fscking fun...) Seth From honeycug@purdue.edu Tue, 5 Sep 2000 13:20:49 -0500 (EST) Date: Tue, 5 Sep 2000 13:20:49 -0500 (EST) From: Ret honeycug@purdue.edu Subject: [PLUG] mini-linux > I acquired a Cyrix/200 with 32mb of RAM that I would like to put in > service as a firewall. It only has a 128mb hard disk and I'm too cheap > to buy a larger one. I was looking at the linux router project and > while it can do basic firewall needs, I would like a full distribution > at my disposal (and I also don't have a floppy drive... too cheap to buy > one again). I played with Redhat and Debian tonight -- I got Debian > installed however I only have 5MB free and Redhat 6.9 wouldn't even > install in less than 400mb. Are there any decent mini-distros? I would > like to have all the basic stuff plus a c compiler at least so I could > compile new kernels and ssh and the like. > > Seth i would either goto kernel.org and compile your own kernel and then goto freshmeat and get only the utils that you need. also if this is a firewall and you dont have much space then check out NetBSD. garrett From cnd@ecn.purdue.edu Tue, 05 Sep 2000 13:27:31 -0500 Date: Tue, 05 Sep 2000 13:27:31 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Events of the Week Here's a brief list of all of the events going on in PLUG and other related organizations. PLUG Callout Monday, September 11th, 2000 6:30pm - 9:30pm Physics 112 CSWN/IBM Resume Workshop Tuesday, September 5th, 2000 7pm - 9pm MTHW 210 Come get helpful pointers on your resume and free food. Computer Organizations Callout Fair Thursday, September 7th, 2000 10am - 2pm MSEE Atrium Come hang out or help out. Computer Society Callout Thursday, September 7th, 2000 7pm - 9pm EE 005 Come join another cool organization. http://csociety.purdue.org -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From cnd@ecn.purdue.edu Tue, 05 Sep 2000 13:40:48 -0500 Date: Tue, 05 Sep 2000 13:40:48 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Can you donate some time? We need help putting up flyers for the Computer Org Callout Fair on Thursday. If you can help put flyers up in buildings and on sidewalks. If you can help out a little (hour between classes or whatever) we'd appreciate it. Send me mail. -Chris -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From ab@eas.purdue.edu Tue, 05 Sep 2000 14:42:33 -0500 Date: Tue, 05 Sep 2000 14:42:33 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] Computer Organizations Callout Fair In message <20000901160731.G73638@radon.gryphonsoft.com>, Will Andrews writes: > > *smack* I beg to differ. That is called a daemon, not a devil. Speaking of which, I asked FreeBSD.org for some free swag for my lab, and they sent me some, which is extremely cool, but I'm about five seats short. Anybody got any extra FreeBSD mousepads or machine logo plates? Anybody want any stickers, bumper stickers, or catalogs? (Seventeen machines running FreeBSD in a lab for our students to use. Used to be eight AIX machines. Meteorologists use UNIX!) ab From will@physics.purdue.edu Tue, 5 Sep 2000 15:10:36 -0500 Date: Tue, 5 Sep 2000 15:10:36 -0500 From: Will Andrews will@physics.purdue.edu Subject: [PLUG] Computer Organizations Callout Fair On Tue, Sep 05, 2000 at 02:42:33PM -0500, A Braunsdorf wrote: > Speaking of which, I asked FreeBSD.org for some free swag for my > lab, and they sent me some, which is extremely cool, but I'm about > five seats short. Anybody got any extra FreeBSD mousepads or > machine logo plates? Anybody want any stickers, bumper stickers, > or catalogs? I got extra FreeBSD sticker sheets. I don't really have much cool stuff with FreeBSD logos on them; just CD sets and sticker sheets. :( Sorry Allen... -- Will Andrews GCS/E/S @d- s+:+ a--- C++ UB++++$ P+ L- E--- W+ N-- !o ?K w--- O- M+ V- PS+ PE++ Y+ PGP+>+++ t++ 5 X+ R+ tv+ b++ DI+++ D+ G++ e>++++ h! r- y? From rjune@ims1.imagestream-is.com Tue, 5 Sep 2000 16:43:07 -0500 (EST) Date: Tue, 5 Sep 2000 16:43:07 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] mini-linux I don't know how far down I can get a *real* distro, however I've gotten a router distro down to about 20MB(With gated and ssh and snmp, etc.) :-) so if You want I'll give you a hand setting up something with a compiler and everything. On Tue, 5 Sep 2000, Seth Heckard wrote: > I acquired a Cyrix/200 with 32mb of RAM that I would like to put in > service as a firewall. It only has a 128mb hard disk and I'm too cheap > to buy a larger one. I was looking at the linux router project and > while it can do basic firewall needs, I would like a full distribution > at my disposal (and I also don't have a floppy drive... too cheap to buy > one again). I played with Redhat and Debian tonight -- I got Debian > installed however I only have 5MB free and Redhat 6.9 wouldn't even > install in less than 400mb. Are there any decent mini-distros? I would > like to have all the basic stuff plus a c compiler at least so I could > compile new kernels and ssh and the like. > > Seth > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From pfitzge1@purdue.edu Tue, 5 Sep 2000 18:21:11 -0500 (EST) Date: Tue, 5 Sep 2000 18:21:11 -0500 (EST) From: patrick.n.fitzgerald.1 pfitzge1@purdue.edu Subject: [PLUG] Computer Organizations Callout Fair On Tue, 5 Sep 2000, A Braunsdorf wrote: > Speaking of which, I asked FreeBSD.org for some free swag for my > lab, and they sent me some, which is extremely cool, but I'm about > five seats short. Anybody got any extra FreeBSD mousepads or > machine logo plates? Anybody want any stickers, bumper stickers, > or catalogs? > I'll take a couple bumper stickers off your hands. I don't think I have any logo plates, but I've probably got some little square stickers about that size with Chuck and "Powered by FreeBSD" if I look around enough. Also, stay tuned for the InstallFest, we usually get tonnes of swag for that. I think my stickers are actually left over from last year's installfest. (I don't have too many machines that can actually claim to be powered by FreeBSD... but once my Linux machine was saved by a FreeBSD boot floppy.) While my car doesn't yet have anything inside it capable of running FreeBSD, it can always serve as a mobile advertisement. I need something to replace the RTFM sticker that got ripped off. Thanks in advance, Patrick N. Fitzgerald Official Plug Videographer and Keeper of the Sacred Flame -- Gravity is a myth, the Earth sucks. --fortune (5) From mpribble@ori.net Tue, 5 Sep 2000 19:28:02 -0500 Date: Tue, 5 Sep 2000 19:28:02 -0500 From: Marv Pribble mpribble@ori.net Subject: [PLUG] Telnet Problem This is a multi-part message in MIME format. ------=_NextPart_000_00B1_01C0176F.6833AE40 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I have a small home network of 3 machines, 1 with Mandrake 7.0. I can = ping all three machines, run Samba on the linux box and access the = drives on my Windows machines, etc. I cannot get telnet or ftp to run = on the linux box. Even from the linux box, when I 'telnet localhost', = the session starts and I receive messages:=20 Trying 127.0.0.1 Connected to localhost.locadomain Escape character is `^]` Then, after a few seconds, 'Connection closed by foreign host.'. =20 I have verified /etc/services contains the info for inetd to start the = telnet daemon. I can run the telnetd daemon from the command line. If anyone knows why, or could suggest some debugging help, I would = appreciate it. TIA Marv ------=_NextPart_000_00B1_01C0176F.6833AE40 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
I have a small home network of 3 = machines, 1 with=20 Mandrake 7.0.  I can ping all three machines, run Samba on the = linux box=20 and access the drives on my Windows machines, etc.  I cannot get = telnet or=20 ftp to run on the linux box.  Even from the linux box, when I = 'telnet=20 localhost', the session starts and I receive messages:
Trying 127.0.0.1
Connected to localhost.locadomain
Escape character is `^]`
 
Then, after a few seconds, 'Connection closed by = foreign=20 host.'. 
 
I have verified /etc/services contains the info for inetd to start = the=20 telnet daemon.  I can run the telnetd daemon from the command = line.
 
If anyone knows why, or could suggest some = debugging help,=20 I would appreciate it.
 
TIA
 
Marv
------=_NextPart_000_00B1_01C0176F.6833AE40-- From Biggs@fuzzydice.net Tue, 5 Sep 2000 19:23:53 +0000 (/etc/localtime) Date: Tue, 5 Sep 2000 19:23:53 +0000 (/etc/localtime) From: Matt Vleaminck Biggs@fuzzydice.net Subject: [PLUG] Mandrake boot disk I need a Mandrake boot disk pronto. Thanks. Matt Vleaminck a.k.a. Biggs VR Web Design Co-Owner Primary E-mail: biggs@fuzzydice.net Secondary E-mail: biggs@vrwebdesign.org From Biggs@fuzzydice.net Tue, 5 Sep 2000 19:24:37 +0000 (/etc/localtime) Date: Tue, 5 Sep 2000 19:24:37 +0000 (/etc/localtime) From: Matt Vleaminck Biggs@fuzzydice.net Subject: [PLUG] boot disk 2 that is Mandrake 7.0 by the way :) Matt Vleaminck a.k.a. Biggs VR Web Design Co-Owner Primary E-mail: biggs@fuzzydice.net Secondary E-mail: biggs@vrwebdesign.org From pweber1@purdue.edu Tue, 5 Sep 2000 19:54:01 -0500 (EST) Date: Tue, 5 Sep 2000 19:54:01 -0500 (EST) From: peter pweber1@purdue.edu Subject: [PLUG] streaming audio & a suggestion i'm looking for a way to save streaming audio (most notably RealAudio) to file(s). does such a beast exist? my suggestion is that i think it would be very useful to be able to search the PLUG list archives. this seems like it would be a fairly straightforward thing to get set up, and i for one would find it to be invaluable. peter ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From docauerj@purdue.edu Tue, 05 Sep 2000 19:55:46 -0500 Date: Tue, 05 Sep 2000 19:55:46 -0500 From: Alex Docauer docauerj@purdue.edu Subject: [PLUG] Telnet Problem Did you check your tcp wrappers? Look at /etc/hosts.allow and /etc/hosts.deny hosts.deny should look like: ALL:ALL and hosts.allow should looke something like: ALL:LOCAL ALL: in.telnetd: in.ftpd: Alex Marv Pribble wrote: > I have a small home network of 3 machines, 1 with Mandrake 7.0. I can > ping all three machines, run Samba on the linux box and access the > drives on my Windows machines, etc. I cannot get telnet or ftp to run > on the linux box. Even from the linux box, when I 'telnet localhost', > the session starts and I receive messages:Trying 127.0.0.1Connected to > localhost.locadomainEscape character is `^]` Then, after a few > seconds, 'Connection closed by foreign host.'. I have verified > /etc/services contains the info for inetd to start the telnet daemon. > I can run the telnetd daemon from the command line. If anyone knows > why, or could suggest some debugging help, I would appreciate > it. TIA Marv From mpribble@ori.net Tue, 5 Sep 2000 20:53:49 -0500 Date: Tue, 5 Sep 2000 20:53:49 -0500 From: Marv Pribble mpribble@ori.net Subject: [PLUG] Telnet Problem Alex, I made the changes you suggested and stopped and restarted inetd. Still the same result. Is there a log anywhere or some logging I can turn on to help determine the problem? Marv ----- Original Message ----- From: Alex Docauer To: Marv Pribble Cc: Sent: Tuesday, September 05, 2000 7:55 PM Subject: Re: [PLUG] Telnet Problem > Did you check your tcp wrappers? Look at /etc/hosts.allow and > /etc/hosts.deny > > hosts.deny should look like: > ALL:ALL > > and hosts.allow should looke something like: > ALL:LOCAL > ALL: > in.telnetd: > in.ftpd: > > Alex > > Marv Pribble wrote: > > > I have a small home network of 3 machines, 1 with Mandrake 7.0. I can > > ping all three machines, run Samba on the linux box and access the > > drives on my Windows machines, etc. I cannot get telnet or ftp to run > > on the linux box. Even from the linux box, when I 'telnet localhost', > > the session starts and I receive messages:Trying 127.0.0.1Connected to > > localhost.locadomainEscape character is `^]` Then, after a few > > seconds, 'Connection closed by foreign host.'. I have verified > > /etc/services contains the info for inetd to start the telnet daemon. > > I can run the telnetd daemon from the command line. If anyone knows > > why, or could suggest some debugging help, I would appreciate > > it. TIA Marv > > From docauerj@purdue.edu Tue, 05 Sep 2000 21:02:04 -0500 Date: Tue, 05 Sep 2000 21:02:04 -0500 From: Alex Docauer docauerj@purdue.edu Subject: [PLUG] Telnet Problem One of the logs, I think either messages or security, should say something if the connection attempt is refused. Also, make sure that your dns informatin is set up correctly. TCP wrappers uses dns to verify host information and can get really screwed up if it can't perform a lookup. You may also want to try setting hosts.allow to: ALL:ALL just to see if that's the problem, although I wouldn't reccomend leaving it like that for long. Put a # sign in front of all the other entries. Alex Marv Pribble wrote: > Alex, > > I made the changes you suggested and stopped and restarted inetd. Still the > same result. Is there a log anywhere or some logging I can turn on to help > determine the problem? > > Marv > > ----- Original Message ----- > From: Alex Docauer > To: Marv Pribble > Cc: > Sent: Tuesday, September 05, 2000 7:55 PM > Subject: Re: [PLUG] Telnet Problem > > > Did you check your tcp wrappers? Look at /etc/hosts.allow and > > /etc/hosts.deny > > > > hosts.deny should look like: > > ALL:ALL > > > > and hosts.allow should looke something like: > > ALL:LOCAL > > ALL: > > in.telnetd: > > in.ftpd: > > > > Alex > > > > Marv Pribble wrote: > > > > > I have a small home network of 3 machines, 1 with Mandrake 7.0. I can > > > ping all three machines, run Samba on the linux box and access the > > > drives on my Windows machines, etc. I cannot get telnet or ftp to run > > > on the linux box. Even from the linux box, when I 'telnet localhost', > > > the session starts and I receive messages:Trying 127.0.0.1Connected to > > > localhost.locadomainEscape character is `^]` Then, after a few > > > seconds, 'Connection closed by foreign host.'. I have verified > > > /etc/services contains the info for inetd to start the telnet daemon. > > > I can run the telnetd daemon from the command line. If anyone knows > > > why, or could suggest some debugging help, I would appreciate > > > it. TIA Marv > > > > From cnd@ecn.purdue.edu Tue, 05 Sep 2000 22:39:33 -0500 Date: Tue, 05 Sep 2000 22:39:33 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] streaming audio & a suggestion > i'm looking for a way to save streaming audio (most notably RealAudio) to Try lynx maybe? lynx -s source http://blah > file.rm > my suggestion is that i think it would be very useful to be able to search > the PLUG list archives. this seems like it would be a fairly > straightforward thing to get set up, and i for one would find it to be > invaluable. Are you willing to put in the time??? -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From cnd@ecn.purdue.edu Tue, 05 Sep 2000 22:40:51 -0500 Date: Tue, 05 Sep 2000 22:40:51 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Computer Organizations Callout Fair Computer Organizations Callout Fair MSEE Atrium Thursday, September 7th 10:00-2:00 Come check out all of Purdue's computer related student organizations. Representatives from the Purdue Linux User's Group (PLUG), the Purdue Macintosh User's Group (PUMUG), Purdue SIGgraph, Purdue Computer Society, Purdue Web Development Org., Purdue Perl Mongers, and the Purdue chapter of ACM will all be there to give information and answer questions. -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From wirges@expert.cc.purdue.edu Wed, 06 Sep 2000 00:10:24 -0500 Date: Wed, 06 Sep 2000 00:10:24 -0500 From: Matt wirges@expert.cc.purdue.edu Subject: [PLUG] Telnet Problem Marv, Make sure that your inetd.conf file has these lines in it: ftp stream tcp nowait root /usr/sbin/tcpd in.ftp -l -a telnet stream tcp nowait root /usr/sbin/tcpd in.telnetd If u don't have those lines there, add them. If they are commented out, uncomment them. Make sure you restart inetd! -Matt -- ============================================================================== | Matthew Wirges | | Systems Administrator, Student, Programmer, Geek. | | Office Phone: [765]464-1148 | | Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com | | www.sharpwebinnovations.com | ============================================================================== > Marv Pribble wrote: > > I have a small home network of 3 machines, 1 with Mandrake 7.0. I can > ping all three machines, run Samba on the linux box and access the > drives on my Windows machines, etc. I cannot get telnet or ftp to run > on the linux box. Even from the linux box, when I 'telnet localhost', > the session starts and I receive messages: > Trying 127.0.0.1 > Connected to localhost.locadomain > Escape character is `^]` > > Then, after a few seconds, 'Connection closed by foreign host.'. > > I have verified /etc/services contains the info for inetd to start the > telnet daemon. I can run the telnetd daemon from the command line. > > If anyone knows why, or could suggest some debugging help, I would > appreciate it. > > TIA > > Marv From sergent@csociety.purdue.edu Tue, 05 Sep 2000 22:44:37 -0700 Date: Tue, 05 Sep 2000 22:44:37 -0700 From: Jonathan Sergent sergent@csociety.purdue.edu Subject: [PLUG] streaming audio & a suggestion > i'm looking for a way to save streaming audio (most notably RealAudio) to > file(s). does such a beast exist? On the client side or on the sender side? --jss. From bhlewis@pythia.cc.purdue.edu 06 Sep 2000 06:40:55 -0500 Date: 06 Sep 2000 06:40:55 -0500 From: Benjamin Lewis bhlewis@pythia.cc.purdue.edu Subject: [PLUG] Computer Organizations Callout Fair A Braunsdorf writes: > Speaking of which, I asked FreeBSD.org for some free swag for my > lab, and they sent me some, which is extremely cool, but I'm about > five seats short. Anybody got any extra FreeBSD mousepads or > machine logo plates? Anybody want any stickers, bumper stickers, > or catalogs? I have at least five of the 1" square FreeBSD logo case plates that I'd be more than happy to donate to your lab. If you let me know where to send them, I can drop them in campus mail tomorrow. -Ben -- Benjamin Lewis Thank goodness modern convenience is a Database Analyst/Programmer thing of the remote future. Purdue University Computing Center -- Pogo, by Walt Kelly bhlewis@purdue.edu From cnd@ecn.purdue.edu Wed, 06 Sep 2000 11:36:01 -0500 Date: Wed, 06 Sep 2000 11:36:01 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] P&G and Hughes P&G and Hughes are in the MSEE Atrium talking with people and taking resumes. It's also P&G Fun Day. I think there are little 'events' going on all over campus. I know there are some out here in the Engineering Mall. Come on over, talk with some people, get a job, or just get some cool free stuff. -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From rjune@ims1.imagestream-is.com Wed, 6 Sep 2000 12:18:24 -0500 (EST) Date: Wed, 6 Sep 2000 12:18:24 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] P&G and Hughes I know people that work for P&G, I've not heard good things about it. On Wed, 6 Sep 2000, Christopher N. Deckard wrote: > P&G and Hughes are in the MSEE Atrium talking with people and taking > resumes. > > It's also P&G Fun Day. I think there are little 'events' going on all > over campus. I know there are some out here in the Engineering Mall. > > Come on over, talk with some people, get a job, or just get some cool > free stuff. > > -Chris > PLUG President > > -- > ---------------------------------------------------------------------- > Christopher N. Deckard | Lead Web Technician > cnd@ecn.purdue.edu | Engineering Computer Network > http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ > ---------------------------------------------------------------------- > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From doobie@doobie.org Wed, 6 Sep 2000 23:21:53 -0500 (EST) Date: Wed, 6 Sep 2000 23:21:53 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] || Execution of perl scripts? Threading? Okay, I've got a question to pose to those perl mongers out there. I need to do multiple things at one time (aka 500+). The problem is that fork() is not very memory conservative, my database starts dying from lack of memory. Anyone know of a good solution, can you do threads in perl? Thanks, Jas From heckarsd@purdue.edu Wed, 6 Sep 2000 23:42:29 -0500 (EST) Date: Wed, 6 Sep 2000 23:42:29 -0500 (EST) From: Seth Heckard heckarsd@purdue.edu Subject: [PLUG] || Execution of perl scripts? Threading? On Wed, 6 Sep 2000, Jason Dubrow wrote: > > Okay, I've got a question to pose to those perl mongers out there. I need > to do multiple things at one time (aka 500+). The problem is that fork() > is not very memory conservative, my database starts dying from lack of > memory. Anyone know of a good solution, can you do threads in perl? > > Thanks, > Jas there are threads in perl, but i don't know if they work reliably yet... i've never used them. if you can get ahold of 'programming perl' 3rd ed. (just came out a month ago or so), chapter 17 has about 20 pages on threads. it's not much, but it might help... seth From mpribble@ori.net Thu, 7 Sep 2000 00:54:08 -0500 Date: Thu, 7 Sep 2000 00:54:08 -0500 From: Marv Pribble mpribble@ori.net Subject: [PLUG] Telnet Problem Alex, Thanks for your help. When you mentioned TCP wrappers, I found the README in /usr/doc/tcp_wrappers and read it. It mentioned 2 helpful programs tcpdchk and tcpdmatch which verify your inetd.conf, hosts.allow, and hosts.deny files. It pointed out several problems and I fixed those. Everything (telnet and ftp) is working now from my local Linux box. I should have known it would work today. My hub died yesterday. Marv From zamboni@cerias.purdue.edu Thu, 07 Sep 2000 08:19:43 -0500 Date: Thu, 07 Sep 2000 08:19:43 -0500 From: Diego Zamboni zamboni@cerias.purdue.edu Subject: [PLUG] || Execution of perl scripts? Threading? doobie@doobie.org said: > Anyone know of a good solution, can you do threads in perl? There is thread support in Perl, but it has to be compiled in, and it is still considered experimental. There is some documentation about them installed with the newer installations of Perl: perldoc Thread perldoc perlthrtut --Diego From cnd@ecn.purdue.edu Thu, 07 Sep 2000 08:41:06 -0500 Date: Thu, 07 Sep 2000 08:41:06 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Computer Organizations Callout Fair TODAY TODAY TODAY TODAY TODAY Computer Organizations Callout Fair MSEE Atrium Thursday, September 7th 10:00-2:00 Come check out all of Purdue's computer related student organizations. Representatives from the Purdue Linux User's Group (PLUG), the Purdue Macintosh User's Group (PUMUG), Purdue SIGgraph, Purdue Computer Society, Purdue Web Development Org., Purdue Perl Mongers, and the Purdue chapter of ACM will all be there to give information and answer questions. -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From heckarsd@purdue.edu Thu, 7 Sep 2000 11:29:21 -0500 (EST) Date: Thu, 7 Sep 2000 11:29:21 -0500 (EST) From: Seth Heckard heckarsd@purdue.edu Subject: [PLUG] x and redhat 6.9.5 i'm running redhat 6.9.5 (pinstripe) and i can't do any remote x sessions with pcxware. gdm is set up properly i think and i have a feeling that it is a problem with my font server (pcxware complains that it can't find all the fonts). other than that, i have no idea :-) anyone run into this before? seth From pweber1@purdue.edu Thu, 7 Sep 2000 11:47:01 -0500 Date: Thu, 7 Sep 2000 11:47:01 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] help help! i just found out that someone traced a network attack of some kind to my ip address. this was definitely NOT me, so my best guess is that my machine got cracked and someone used it to launch an attack. how can i figure out what happened? i'm running RH6.1 (i was going to be upgrading to 6.2 as we speak). services include(d): httpd (dumb, i know) ftp ssh afpd (quite insecure, i know) papd (ditto) sendmail and some other stuff i'm not thinking of right now. i have this machine running as a router/firewall, and i guess i *really* need to improve how that's set up.. :) thanks, peter p.s. if one of you filed the complain (the assistant dean of students told me the attack started about 2pm yesterday) please contact me, i really want to figure out what happened! -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From porterds@purdue.edu Thu, 07 Sep 2000 12:03:22 -0500 Date: Thu, 07 Sep 2000 12:03:22 -0500 From: Doug Porter porterds@purdue.edu Subject: [PLUG] help most likely it was ftp... if you were running the wu-ftpd from redhat 6.1 it has a huge root exploit. there are many scripts floating around the net script kiddies use which require no intelligence to use. i was once in a similar situation to you... the assistant dean of students called me and told me they were getting complaints that i was "attempting to break into some computer." of course the lady that calls knows nothing about computers so i told her to get more details. she called me back a week later and read me some lines from a log file. turns out that i was fingering some box, hardly a break in attempt, but that is kind of hard to explain to the lady i was talking to that knew nothing about computers. i haven't heard anything about it since... - doug porter peter wrote: > > help! i just found out that someone traced a network attack of some > kind to my ip address. this was definitely NOT me, so my best guess > is that my machine got cracked and someone used it to launch an > attack. how can i figure out what happened? i'm running RH6.1 (i was > going to be upgrading to 6.2 as we speak). > > services include(d): > > httpd (dumb, i know) > ftp > ssh > afpd (quite insecure, i know) > papd (ditto) > sendmail > > and some other stuff i'm not thinking of right now. i have this > machine running as a router/firewall, and i guess i *really* need to > improve how that's set up.. :) > > thanks, > peter > > p.s. if one of you filed the complain (the assistant dean of students > told me the attack started about 2pm yesterday) please contact me, i > really want to figure out what happened! > -- > ------------------------------------------------------------------ > peter weber "To know that we know what we know, > N9AZ and that we do not know what we do > pweber1@purdue.edu not know, that is true knowledge." > -Confucius > ------------------------------------------------------------------ > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From rjune@ims1.imagestream-is.com Thu, 7 Sep 2000 12:05:57 -0500 (EST) Date: Thu, 7 Sep 2000 12:05:57 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] help You can use rpm to verify that those haven't changed at all rpm --verify use a man rpm to find out what the output means. That would be a possible start. > From pweber1@purdue.edu Thu, 7 Sep 2000 12:07:38 -0500 Date: Thu, 7 Sep 2000 12:07:38 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] help i already axed ftp. (i have my brother trying to access from a way off-campus ip.) how can i figure out what outgoing connections were happening during a specific time frame? thanks, peter >most likely it was ftp... if you were running the wu-ftpd from redhat >6.1 it has a huge root exploit. there are many scripts floating around >the net script kiddies use which require no intelligence to use. > >i was once in a similar situation to you... the assistant dean of >students called me and told me they were getting complaints that i was >"attempting to break into some computer." of course the lady that calls >knows nothing about computers so i told her to get more details. she >called me back a week later and read me some lines from a log file. >turns out that i was fingering some box, hardly a break in attempt, but >that is kind of hard to explain to the lady i was talking to that knew >nothing about computers. i haven't heard anything about it since... > >- doug porter > >peter wrote: >> >> help! i just found out that someone traced a network attack of some >> kind to my ip address. this was definitely NOT me, so my best guess >> is that my machine got cracked and someone used it to launch an >> attack. how can i figure out what happened? i'm running RH6.1 (i was >> going to be upgrading to 6.2 as we speak). >> >> services include(d): >> >> httpd (dumb, i know) >> ftp >> ssh >> afpd (quite insecure, i know) >> papd (ditto) >> sendmail >> >> and some other stuff i'm not thinking of right now. i have this >> machine running as a router/firewall, and i guess i *really* need to >> improve how that's set up.. :) >> >> thanks, >> peter >> >> p.s. if one of you filed the complain (the assistant dean of students >> told me the attack started about 2pm yesterday) please contact me, i >> really want to figure out what happened! >> -- >> ------------------------------------------------------------------ >> peter weber "To know that we know what we know, >> N9AZ and that we do not know what we do >> pweber1@purdue.edu not know, that is true knowledge." >> -Confucius >> ------------------------------------------------------------------ >> >> ____________________________________________________ >> The Purdue Linux Users' Group (PLUG) mailing list. >> plug mailing list - plug@csociety.purdue.edu >> http://csociety.ecn.purdue.edu/mailman/listinfo/plug -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From porterds@purdue.edu Thu, 07 Sep 2000 12:15:22 -0500 Date: Thu, 07 Sep 2000 12:15:22 -0500 From: Doug Porter porterds@purdue.edu Subject: [PLUG] help you can try the command "last | grep ftpd". that will show you who logged on to your ftp server since the beginning of this month. you can also check out the /etc/log/xferlog which shows all ftp transfers (up or down). two other log files you should look at in are /var/log/messages and /var/log/secure. but remember, if this person knew what they were doing they could have covered their tracks. i would also check /etc/passwd and look for peculiar accounts, specifically any account with a user id of zero other than root. - doug peter wrote: > > i already axed ftp. (i have my brother trying to access from a way > off-campus ip.) how can i figure out what outgoing connections were > happening during a specific time frame? > > thanks, > peter > > >most likely it was ftp... if you were running the wu-ftpd from redhat > >6.1 it has a huge root exploit. there are many scripts floating around > >the net script kiddies use which require no intelligence to use. > > > >i was once in a similar situation to you... the assistant dean of > >students called me and told me they were getting complaints that i was > >"attempting to break into some computer." of course the lady that calls > >knows nothing about computers so i told her to get more details. she > >called me back a week later and read me some lines from a log file. > >turns out that i was fingering some box, hardly a break in attempt, but > >that is kind of hard to explain to the lady i was talking to that knew > >nothing about computers. i haven't heard anything about it since... > > > >- doug porter > > > >peter wrote: > >> > >> help! i just found out that someone traced a network attack of some > >> kind to my ip address. this was definitely NOT me, so my best guess > >> is that my machine got cracked and someone used it to launch an > >> attack. how can i figure out what happened? i'm running RH6.1 (i was > >> going to be upgrading to 6.2 as we speak). > >> > >> services include(d): > >> > >> httpd (dumb, i know) > >> ftp > >> ssh > >> afpd (quite insecure, i know) > >> papd (ditto) > >> sendmail > >> > >> and some other stuff i'm not thinking of right now. i have this > >> machine running as a router/firewall, and i guess i *really* need to > >> improve how that's set up.. :) > >> > >> thanks, > >> peter > >> > >> p.s. if one of you filed the complain (the assistant dean of students > >> told me the attack started about 2pm yesterday) please contact me, i > >> really want to figure out what happened! > >> -- > >> ------------------------------------------------------------------ > >> peter weber "To know that we know what we know, > >> N9AZ and that we do not know what we do > >> pweber1@purdue.edu not know, that is true knowledge." > >> -Confucius > >> ------------------------------------------------------------------ > >> > >> ____________________________________________________ > >> The Purdue Linux Users' Group (PLUG) mailing list. > >> plug mailing list - plug@csociety.purdue.edu > >> http://csociety.ecn.purdue.edu/mailman/listinfo/plug > > -- > ------------------------------------------------------------------ > peter weber "To know that we know what we know, > N9AZ and that we do not know what we do > pweber1@purdue.edu not know, that is true knowledge." > -Confucius > ------------------------------------------------------------------ From jclewis@purdue.edu Thu, 07 Sep 2000 12:34:15 -0500 Date: Thu, 07 Sep 2000 12:34:15 -0500 From: James Lewis jclewis@purdue.edu Subject: [PLUG] help You should at the very least be running tcplogd and icmplogd so you can tell when outside connection attempts are being made to your computer. peter wrote: > i already axed ftp. (i have my brother trying to access from a way > off-campus ip.) how can i figure out what outgoing connections were > happening during a specific time frame? > > thanks, > peter > > >most likely it was ftp... if you were running the wu-ftpd from redhat > >6.1 it has a huge root exploit. there are many scripts floating around > >the net script kiddies use which require no intelligence to use. > > > >i was once in a similar situation to you... the assistant dean of > >students called me and told me they were getting complaints that i was > >"attempting to break into some computer." of course the lady that calls > >knows nothing about computers so i told her to get more details. she > >called me back a week later and read me some lines from a log file. > >turns out that i was fingering some box, hardly a break in attempt, but > >that is kind of hard to explain to the lady i was talking to that knew > >nothing about computers. i haven't heard anything about it since... > > > >- doug porter > > > >peter wrote: > >> > >> help! i just found out that someone traced a network attack of some > >> kind to my ip address. this was definitely NOT me, so my best guess > >> is that my machine got cracked and someone used it to launch an > >> attack. how can i figure out what happened? i'm running RH6.1 (i was > >> going to be upgrading to 6.2 as we speak). > >> > >> services include(d): > >> > >> httpd (dumb, i know) > >> ftp > >> ssh > >> afpd (quite insecure, i know) > >> papd (ditto) > >> sendmail > >> > >> and some other stuff i'm not thinking of right now. i have this > >> machine running as a router/firewall, and i guess i *really* need to > >> improve how that's set up.. :) > >> > >> thanks, > >> peter > >> > >> p.s. if one of you filed the complain (the assistant dean of students > >> told me the attack started about 2pm yesterday) please contact me, i > >> really want to figure out what happened! > >> -- > >> ------------------------------------------------------------------ > >> peter weber "To know that we know what we know, > >> N9AZ and that we do not know what we do > >> pweber1@purdue.edu not know, that is true knowledge." > >> -Confucius > >> ------------------------------------------------------------------ > >> > >> ____________________________________________________ > >> The Purdue Linux Users' Group (PLUG) mailing list. > >> plug mailing list - plug@csociety.purdue.edu > >> http://csociety.ecn.purdue.edu/mailman/listinfo/plug > > -- > ------------------------------------------------------------------ > peter weber "To know that we know what we know, > N9AZ and that we do not know what we do > pweber1@purdue.edu not know, that is true knowledge." > -Confucius > ------------------------------------------------------------------ > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From calx@purdue.edu Thu, 7 Sep 2000 12:37:46 -0500 (EST) Date: Thu, 7 Sep 2000 12:37:46 -0500 (EST) From: David calx@purdue.edu Subject: [PLUG] help Caution: rant. Don't like rants? Don't read. Extra caution: Chock full of opinion. > p.s. if one of you filed the complain (the assistant dean of students > told me the attack started about 2pm yesterday) please contact me, i > really want to figure out what happened! Yeah, I had someone "file a complaint" last year. The assistant dean, or the representative called me and accused me of ha><0RiN6 some slob's box. After constant questioning, I finally got it out of them that all I did was attempt to ftp into the dork's box. The weenie was running FreeBSD, so I don't think they were completely clueless. They could have looked at those magic logging files and perhaps attempted to contact me. Or they could have used that ipfw thingie that does stuff with "INternet". If anyone here gets "a77a><0r3D", why not check your logs, and write a friendly email to that person. It's rather stupid to go running to the arms of ResNet/PUCC/FBI etc, when chances are it's most likely not a vicious attack. They didn't shut down my damn connection that instant, so what's the point? When has attempting to utilize a bureaucracy worked better than personal interaction? Perhaps those people have a problem with confrontation? Perhaps they truly believe that the "offenders" service will cease the instant they report the "devious" behavior. Sorry about the rant, but I would be really pissed off if I got my service disconnected because I fingered a box (ewww) or ftp'd or www'd or telnet'd etc. It's not a damn port scan, and I didn't try to break in, it was a singular instance. I would be even more pissed off if an attack was launched from my box and no one informed me. Looking at the mail added to the list while I was writing this, I see that I am not the only one who had this happen. I believe that *NIX users are (for the most part) a step above the average computer user, and we all have an inherent curiosity. I think that we can police/protect ourselves much better than the school can. Hmm. Gratuitous use of sarcastic quotations... David From pweber1@purdue.edu Thu, 7 Sep 2000 13:30:00 -0500 Date: Thu, 7 Sep 2000 13:30:00 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] help to everyone who has answered: thanks for your help, i'm still working on it, and everything you've given me has been very helpful. (sorry i haven't answered each of you personally, i've gotten a LOT of emails in the last couple of hours!) i've got what i think are some good leads, and i'll post a followup once i get this figured out. (and more questions in the meantime, if i have them! :) thanks, peter -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From docauerj@purdue.edu Thu, 07 Sep 2000 13:43:38 -0500 Date: Thu, 07 Sep 2000 13:43:38 -0500 From: Alex Docauer docauerj@purdue.edu Subject: [PLUG] help Doug Porter wrote: > i was once in a similar situation to you... the assistant dean of > students called me and told me they were getting complaints that i was > "attempting to break into some computer." of course the lady that calls > knows nothing about computers so i told her to get more details. she > called me back a week later and read me some lines from a log file. > turns out that i was fingering some box, hardly a break in attempt, but > that is kind of hard to explain to the lady i was talking to that knew > nothing about computers. i haven't heard anything about it since... Almost the same exact thing happened to me. One day I got a call from the office of the dean of students and they accused me of flooding other people out of irc channels. Apparently a guy broke into my RedHat 6.1 box and was running an irc egg-drop program on my system. I deleted the program, (and the crond scripts that kept reinstalling it), and used rpm -Va to figure out if he changed anything else. As it turned out, he had also modified /bin/login, but I couldn't delete it as the system gave me some error about not being able to unlink it. I couldn't rename it either; all I could do was rename /bin to /bin.old and copy the contents (except login) from bin.old back to bin. It's still a mystery to me why I couldn't remove it, and /bin.old/login still sits there to this day. Anyone have any ideas on why I can't remove it? Alex From defouwj@purdue.edu Thu, 7 Sep 2000 13:55:55 -0500 (EST) Date: Thu, 7 Sep 2000 13:55:55 -0500 (EST) From: Jeff DeFouw defouwj@purdue.edu Subject: [PLUG] help On Thu, 7 Sep 2000, Alex Docauer wrote: > It's still a mystery to me why I couldn't remove it, and > /bin.old/login still sits there to this day. Anyone have any ideas on > why I can't remove it? It probably had the ext2 "immutable" flag set. Do an lsattr /bin.old/login and look for an "i". You can use chattr -i /bin.old/login to remove the flag and then do whatever with it. -- Jeff DeFouw From pweber1@purdue.edu Thu, 7 Sep 2000 14:24:36 -0500 Date: Thu, 7 Sep 2000 14:24:36 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] help i've found a few suspicious things. in /var/log/messages: Sep 5 16:46:43 hawk-c-019 sshd[713]: connection from "193.89.248.8" Sep 6 01:26:02 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: Unknown opcode from host 128.211.202.78 Sep 6 02:34:30 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: Unknown opcode from host 128.211.202.78 Sep 6 03:34:31 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: Unknown opcode from host 128.211.202.78 the first one (193.89.248.8) is www.baa.dk and also shows an entry in /var/log/secure (see below). the next 3 entries are also interesting. these are approximately 12 hours before the time they say the 'attack' started. i don't know exactly that this is about other than it has something to do with gdm. i find it interesting that the 3 messages are about an hour apart. in /var/log/secure: Sep 5 16:46:43 hawk-c-019 in.ftpd[11058]: connect from 193.89.248.8 Sep 6 10:19:17 hawk-c-019 in.ftpd[15739]: connect from 168.234.190.4 Sep 6 18:18:48 hawk-c-019 in.ftpd[18885]: connect from 24.168.41.251 the first address is the aforementioned www.baa.dk. the second address won't resolve: [pweber1@mentor pweber1]$ nslookup 168.234.190.4 Server: mentor.cc.purdue.edu Address: 128.210.10.12 *** mentor.cc.purdue.edu can't find 168.234.190.4: Non-existent host/domain and the third is from NYC(?) [pweber1@mentor pweber1]$ nslookup 24.168.41.251 Server: mentor.cc.purdue.edu Address: 128.210.10.12 Name: 24-168-41-251.nyc.rr.com Address: 24.168.41.251 the first two are the ones that really concerns me, since they happened before the time that the 'attack' reportedly started (2pm or so yesterday). so i guess what i have is: 1) 128.211.202.78 tried connecting to gdm about 12 hours before the attack 2) 193.89.248.8 (in Denmark?) connecting almost simultaneously via ssh and ftp before the attack 3) an unresolvable address (168.234.190.4) that connected via ftp before the attack 4) 24.168.41.251 connected via ftp about 4 hours after the attack started does this shed any light on anything? (i'm trying to get the name of the contact that informed the Dean of Students office, so that i get my hands on some useful info. i should be able to get that name in about 2 hours.) thanks again, peter -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From ab@eas.purdue.edu Thu, 07 Sep 2000 14:27:22 -0500 Date: Thu, 07 Sep 2000 14:27:22 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] help Sounds like you guys need to run something like Tripwire. ab From pweber1@purdue.edu Thu, 7 Sep 2000 14:27:53 -0500 Date: Thu, 7 Sep 2000 14:27:53 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] help oops, forgot one piece. there are two entries in /var/log/secure from about 2.5 hours ago. someone in hilltop tried to ftp in, and the connections were refused (ala /etc/hosts.deny) (maybe that was someone on this list who was trying to check my services for me? [he hopes optimistically]) peter -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From psmith@lib.purdue.edu Thu, 7 Sep 2000 14:30:27 -0500 Date: Thu, 7 Sep 2000 14:30:27 -0500 From: Preston Smith psmith@lib.purdue.edu Subject: [PLUG] help On Thu, Sep 07, 2000 at 02:24:36PM -0500, peter (pweber1@purdue.edu) wrote: > i've found a few suspicious things. > > Sep 5 16:46:43 hawk-c-019 sshd[713]: connection from "193.89.248.8" > Sep 6 01:26:02 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > Unknown opcode from host 128.211.202.78 > Sep 6 02:34:30 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > Unknown opcode from host 128.211.202.78 > Sep 6 03:34:31 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > Unknown opcode from host 128.211.202.78 > > the first one (193.89.248.8) is www.baa.dk and also shows an entry in > /var/log/secure (see below). The entire network was portscanned by www.baa.dk the other day, I believe. (All of my boxes at the Libraries were probed) Probably scanned *.purdue.edu... -- Preston M. Smith http://www.lib.purdue.edu/~psmith "History repeats itself only if one does not listen the first time." From sharpa@ecn.purdue.edu Thu, 7 Sep 2000 14:33:43 -0500 (EST) Date: Thu, 7 Sep 2000 14:33:43 -0500 (EST) From: Andrew Sharp sharpa@ecn.purdue.edu Subject: [PLUG] help > in /var/log/messages: > > Sep 5 16:46:43 hawk-c-019 sshd[713]: connection from "193.89.248.8" > Sep 6 01:26:02 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > Unknown opcode from host 128.211.202.78 > Sep 6 02:34:30 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > Unknown opcode from host 128.211.202.78 > Sep 6 03:34:31 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > Unknown opcode from host 128.211.202.78 Hey! Thats the same guy that was bothering me the other day. From my logs:Sep 5 16:23:23 goatsonparade portsentry[66506]: attackalert: Connect from host: www.baa.dk/193.89.248.8 to TCP port: 143 Sep 5 16:23:23 goatsonparade portsentry[66506]: attackalert: Host 193.89.248.8 has been blocked via wrappers with string:"ALL: 193.89.248.8" Andrew From notz@purdue.edu Thu, 7 Sep 2000 14:40:19 -0500 (EST) Date: Thu, 7 Sep 2000 14:40:19 -0500 (EST) From: Patrick K Notz notz@purdue.edu Subject: [PLUG] help Call me crazy, but ("You're a Crazy Butt!") perhaps all of this suggests that a PLUG supported lecture on securing linux boxen would be a valuable event. I know I'd attend. I've been thinking about getting a ResNet ASDL connection at home but I think I'll wait until I learn how to secure my machine a little better. Pat From meuserj@purdue.edu Thu, 7 Sep 2000 14:45:36 -0500 (EST) Date: Thu, 7 Sep 2000 14:45:36 -0500 (EST) From: John C Meuser meuserj@purdue.edu Subject: [PLUG] help > > in /var/log/messages: > > > > Sep 5 16:46:43 hawk-c-019 sshd[713]: connection from "193.89.248.8" > > Sep 6 01:26:02 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > > Unknown opcode from host 128.211.202.78 > > Sep 6 02:34:30 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > > Unknown opcode from host 128.211.202.78 > > Sep 6 03:34:31 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > > Unknown opcode from host 128.211.202.78 > > > Hey! Thats the same guy that was bothering me the other day. From my > logs:Sep 5 16:23:23 goatsonparade portsentry[66506]: attackalert: Connect > from host: www.baa.dk/193.89.248.8 to TCP port: 143 > Sep 5 16:23:23 goatsonparade portsentry[66506]: attackalert: Host > 193.89.248.8 has been blocked via wrappers with string:"ALL: 193.89.248.8" > > Andrew > He tried to get me too: Sep 5 18:08:31 wily-c-071 sshd[24761]: fatal: Timeout before authentication for 193.89.248.8. Should somebody report this or what? John From calx@purdue.edu Thu, 7 Sep 2000 14:49:14 -0500 (EST) Date: Thu, 7 Sep 2000 14:49:14 -0500 (EST) From: David calx@purdue.edu Subject: [PLUG] help > Call me crazy, but ("You're a Crazy Butt!") perhaps all of this > suggests that a PLUG supported lecture on securing linux boxen would > be a valuable event. I know I'd attend. I've been thinking about > getting a ResNet ASDL connection at home but I think I'll wait until I > learn how to secure my machine a little better. Agreed. Just as long as the lecture doesn't turn into a "in order to have true security, you need to use OpenBSD and not change anything" sort of lecture. :) (Nothing against OpenBSD, mind you, I know it's very secure, etc, just that it would be a boring lecture. Sort of like a lecture on getting Windows to crash. :) Could PLUG get some sysadmins from Purdue to speak? It would probably be easier to get some of the people securing our network to talk than anyone from the outside. Then again, I don't know what sort of connections PLUG has. David From calx@purdue.edu Thu, 7 Sep 2000 14:57:01 -0500 (EST) Date: Thu, 7 Sep 2000 14:57:01 -0500 (EST) From: David calx@purdue.edu Subject: [PLUG] help > Should somebody report this or what? Whenever someone from the outside attacks the bio network, the sysadmin can only send email to whoever appears to be responsible and then simply deny access from that host. If it turns out to be a network wide problem (@home), then the entire network is denied. There is a GIANT list of denied IPs for bio, and unfortunately, the response rate to complaints is very, very low. I think that what has been going on in this thread is a really great thing. It helps keep us aware of what others are doing. Perhaps there should be a mailing list for security issues like this? (Is there? I thought there was, but not as specific as this.) BTW, that same person attempted to login to my machines as well. Grrr. David > > > Sep 5 16:46:43 hawk-c-019 sshd[713]: connection from "193.89.248.8" > > > > Hey! Thats the same guy that was bothering me the other day. From my > > logs:Sep 5 16:23:23 goatsonparade portsentry[66506]: attackalert: Connect > > from host: www.baa.dk/193.89.248.8 to TCP port: 143 > > Sep 5 16:23:23 goatsonparade portsentry[66506]: attackalert: Host > > 193.89.248.8 has been blocked via wrappers with string:"ALL: 193.89.248.8" > He tried to get me too: > > Sep 5 18:08:31 wily-c-071 sshd[24761]: fatal: Timeout before > authentication for 193.89.248.8. > > Should somebody report this or what? From lev@purdue.edu Thu, 7 Sep 2000 15:02:44 -0500 Date: Thu, 7 Sep 2000 15:02:44 -0500 From: Lev Gorenstein lev@purdue.edu Subject: [PLUG] help On Thu, 7 Sep 2000, Patrick K Notz wrote: PKN> Call me crazy, but ("You're a Crazy Butt!") perhaps all of this PKN> suggests that a PLUG supported lecture on securing linux boxen would PKN> be a valuable event. I know I'd attend. I certainly second(3F) that. We could also make(1) it in sync(1M) with InstallFest... say, a week after InstallFest. When people get(1) their boxes installed, have a week to play with them and somewhat get familiar - then it'd be a good time to make them secure. Granted, for those who would only do occasional dial-ups it's less of an issue compared to the *DSL'ed people, but still it is an issue. Kind of important one ;-)). Lev -- To hell with HalloWIN! He have HalloUNIX every SUNday! From rajak@purdue.edu Thu, 7 Sep 2000 15:15:23 -0500 (EST) Date: Thu, 7 Sep 2000 15:15:23 -0500 (EST) From: Brian Poole rajak@purdue.edu Subject: [PLUG] help On Thu, 7 Sep 2000, David wrote: > Date: Thu, 7 Sep 2000 14:49:14 -0500 (EST) > From: David > To: Patrick K Notz > Cc: plug@csociety.purdue.edu > Subject: Re: [PLUG] help > > > Call me crazy, but ("You're a Crazy Butt!") perhaps all of this > > suggests that a PLUG supported lecture on securing linux boxen would > > be a valuable event. I know I'd attend. I've been thinking about > > getting a ResNet ASDL connection at home but I think I'll wait until I > > learn how to secure my machine a little better. > > Agreed. Just as long as the lecture doesn't turn into a "in order to have > true security, you need to use OpenBSD and not change anything" sort of > lecture. :) (Nothing against OpenBSD, mind you, I know it's very secure, > etc, just that it would be a boring lecture. Sort of like a lecture on > getting Windows to crash. :) > > Could PLUG get some sysadmins from Purdue to speak? It would probably be > easier to get some of the people securing our network to talk than anyone > from the outside. Then again, I don't know what sort of connections PLUG > has. Truth be told most of the security tips and configuration are pretty basic. It doesn't take an extremely experienced sys admin to know how to keep a machine relatively safe (especially from remote users, local users are another story, and requires a book or two ;) In Linux (and most *nixes for that matter) the common steps are.. 1. Edit /etc/inetd.conf and remove unnecessary services by commenting them out with a '#' then killall -HUP inetd to restart the daemon. 2. Edit your rc.d scripts to stop unnecessary services (rpc.*, sendmail, etc), this is distro specific, there are a couple of different styles, wouldn't take long to show the two main. 3. Setup TCP wrappers if you still have any services running out of inetd (if you don't you can remove it from your rc.d scripts as well, its pointless running it). The interface is simple and easy to block and allow people through with. 4. If you feel like it you can also setup a firewall which is extremely simple these days, there are a hundred firewall scripts for ipchains, some with GUIs included. Specify what services you need on the firewall, what machines that are masqed need specific ports fwded, etc. This is needed to protect other services not run out of inetd, remember TCP wrappers only works by default with inetd services, other services _may_ support it, but normally not by default (eg sshd, httpd, etc) 5. A little lecture on good policy. Don't pick bad passwords, don't let root login remotely (use su), don't give out shells, etc. 6. Install SSH over telnet if possible, it uses encryption which keeps usernames & passwords from being sniffed. There are free Window SSH clients as well, so its not much of a hassle. 7. Try to keep up to date on errata, if you use Redhat sign up for the Redhat security list, if you use Slackware the Slackware's. These are generally pretty low traffic (unlike, say BUGTRAQ) but are still pretty complete. 8. Install file integrity tools such as Tripwire, ViperDB, whatever works for you. It is after the fact security, you don't get notified until something has been changed already, but at least you know and can stop the machine and figure out what happened and keep the attacker from using the machine to springboard to others. If you actually follow these steps you are a very hard nut to crack and most people will just skip over you. Why attack a person with only a ssh port open that is also filtered at the firewall, when you have a person 20 IPs down who is running rpc.statd? After you've done this for a while it doesn't take all that long to secure a machine pretty well, even if it does seem a lot of work right off. I think a full lecture on this would be nice though, as long as the lecturer remembers not to keep it to distro-specific. -b > > David > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From rrussel5@purdue.edu Thu, 7 Sep 2000 15:59:38 -0500 Date: Thu, 7 Sep 2000 15:59:38 -0500 From: Alex Russell rrussel5@purdue.edu Subject: [PLUG] help Perchance someone (Chris) could query the CERIAS folks and see if any of them wanted to give a talk like the one suggested earlier? Just a thought, who knows, perhaps a Linux-only admin might give a more relevant talk. Alex +------------------------------------------------+ | Thomson Consumer Electronics www.rca.com | +------------------------------------------------+ | Alex Russell | Desk: (765) 495-8084 | | Web Developer | Cell: (317) 514-8455 | | www.netwindows.org | russella@tce.com | | russella.csoft.net | rrussel5@purdue.edu | | AIM: russcoon | russella@smoke.csoft.net | +------------------------------------------------+ -----Original Message----- From: plug-admin@csociety.purdue.edu [mailto:plug-admin@csociety.purdue.edu]On Behalf Of Lev Gorenstein Sent: Thursday, September 07, 2000 3:03 PM Cc: plug@csociety.purdue.edu Subject: Re: [PLUG] help On Thu, 7 Sep 2000, Patrick K Notz wrote: PKN> Call me crazy, but ("You're a Crazy Butt!") perhaps all of this PKN> suggests that a PLUG supported lecture on securing linux boxen would PKN> be a valuable event. I know I'd attend. I certainly second(3F) that. We could also make(1) it in sync(1M) with InstallFest... say, a week after InstallFest. When people get(1) their boxes installed, have a week to play with them and somewhat get familiar - then it'd be a good time to make them secure. Granted, for those who would only do occasional dial-ups it's less of an issue compared to the *DSL'ed people, but still it is an issue. Kind of important one ;-)). Lev -- To hell with HalloWIN! He have HalloUNIX every SUNday! ____________________________________________________ The Purdue Linux Users' Group (PLUG) mailing list. plug mailing list - plug@csociety.purdue.edu http://csociety.ecn.purdue.edu/mailman/listinfo/plug From zamboni@cerias.purdue.edu Thu, 07 Sep 2000 16:21:28 -0500 Date: Thu, 07 Sep 2000 16:21:28 -0500 From: Diego Zamboni zamboni@cerias.purdue.edu Subject: [PLUG] help >>>>> On Thu, 7 Sep 2000, "Brian" == Brian Poole wrote: Brian> I think a full lecture on this would be nice though, as long as the Brian> lecturer remembers not to keep it to distro-specific. Or better yet, not to keep it OS-specific. Most of the steps you outlined are not Linux-specific, and can very well be applied to *BSD, Solaris, or anything else. The only exception being item 4, since ipchains is not available everywhere, although there are other options (such as ipf, which comes with OpenBSD). --Diego From cnd@ecn.purdue.edu Thu, 07 Sep 2000 16:25:43 -0500 Date: Thu, 07 Sep 2000 16:25:43 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] help David wrote: > Could PLUG get some sysadmins from Purdue to speak? It would probably be > easier to get some of the people securing our network to talk than anyone > from the outside. Then again, I don't know what sort of connections PLUG > has. Oh. I know some people. :-) What kinds of things do you guys want covered and I'll try and get some lightning talks on security. I think that that would be a good way of doing it. Have maybe 5 or 6 people talk on differnet security concepts and present different things on what to do or not do. -Chris PLUG President From raj@cerias.purdue.edu Thu, 7 Sep 2000 16:28:16 -0500 (EST) Date: Thu, 7 Sep 2000 16:28:16 -0500 (EST) From: Brian Poole raj@cerias.purdue.edu Subject: [PLUG] help Agreed. ;) Might make it a bit harder to display visually the differences on BSD, Solaris, and Linux, but most of the steps are indeed common. I'd just be a bit worried about time, after all I'm sure most of us would get a bit carried away and talk about security and its risks for hours :) -b On Thu, 7 Sep 2000, Diego Zamboni wrote: > >>>>> On Thu, 7 Sep 2000, "Brian" == Brian Poole wrote: > > Brian> I think a full lecture on this would be nice though, as long as the > Brian> lecturer remembers not to keep it to distro-specific. > > Or better yet, not to keep it OS-specific. Most of the steps you outlined are > not Linux-specific, and can very well be applied to *BSD, Solaris, or anything > else. The only exception being item 4, since ipchains is not available > everywhere, although there are other options (such as ipf, which comes with > OpenBSD). > > --Diego > From sharpa@ecn.purdue.edu Thu, 7 Sep 2000 16:36:07 -0500 (EST) Date: Thu, 7 Sep 2000 16:36:07 -0500 (EST) From: Andrew Sharp sharpa@ecn.purdue.edu Subject: [PLUG] help > What kinds of things do you guys want covered and I'll try and get some > lightning talks on security. I think that that would be a good way of > doing it. Have maybe 5 or 6 people talk on differnet security concepts > and present different things on what to do or not do. I think a bit about what programs like Tripwire (and others) can do for you. I'm a little hazy in that area. Also a discussion of good firewalling rules would be helpful I'm sure. Then of course, the basic stuff w/inetd, ssh etc. Andrew From pweber1@purdue.edu Thu, 7 Sep 2000 17:21:39 -0500 Date: Thu, 7 Sep 2000 17:21:39 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] Bastille Linux anyone have any experience with this? according to the website it has been updated to deal with RH6.2 and can be run on "non-virgin" systems. ---- from http://www.securityfocus.com/focus/linux/articles/linux-install.html BASTILLE LINUX It is also apt here to consider installing Bastille-Linux (http://www.bastille-linux.org). From the Bastille Linux home page: "The Bastille Hardening script attempts to provide the most secure, yet functional, Redhat 6.0 system available." Bastille Linux is a script that attempts to appropriately harden a Redhat 6.0 system. It acts fairly similar to tools available for other operating systems, such as Titan for Sun Solaris. Presently, Bastille-Linux is only available (and has been thoroughly tested) for Redhat 6.0. However, it should shortly be availble for other versions of Redhat and hopefully other distributions of Linux. If it is available for the version of Linux you are running, it would be wise to download and install this script. Just remember that if you do use Bastille-Linux, it should be performed immediately after Redhat has been installed (ie even before patches are added). -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From olenik@purdue.edu Thu, 07 Sep 2000 17:24:34 -0500 Date: Thu, 07 Sep 2000 17:24:34 -0500 From: Hayden Olenik olenik@purdue.edu Subject: [PLUG] Bastille Linux I'm using Bastille now. It certainly is adaptable, as it seems to be working great, and I'm running RedHat 6.9.5 (Pinstripe). Hayden On Thu, 7 Sep 2000, peter wrote: > Date: Thu, 7 Sep 2000 17:21:39 -0500 > To: plug@csociety.purdue.edu > From: peter > Sender: plug-admin@csociety.purdue.edu > Subject: [PLUG] Bastille Linux > > anyone have any experience with this? according to the website it has > been updated to deal with RH6.2 and can be run on "non-virgin" > systems. > > ---- > > from http://www.securityfocus.com/focus/linux/articles/linux-install.html > > BASTILLE LINUX > > It is also apt here to consider installing Bastille-Linux > (http://www.bastille-linux.org). From the Bastille Linux home page: > "The Bastille > Hardening script attempts to provide the most secure, yet > functional, Redhat 6.0 system available." Bastille Linux is a script > that > attempts to appropriately harden a Redhat 6.0 system. It acts fairly > similar to tools available for other operating systems, such as Titan > for Sun Solaris. Presently, Bastille-Linux is only available (and > has been thoroughly tested) for Redhat 6.0. However, it should shortly > be availble for other versions of Redhat and hopefully other > distributions of Linux. > > If it is available for the version of Linux you are running, it > would be wise to download and install this script. Just remember that > if you > do use Bastille-Linux, it should be performed immediately after > Redhat has been installed (ie even before patches are added). > -- > ------------------------------------------------------------------ > peter weber "To know that we know what we know, > N9AZ and that we do not know what we do > pweber1@purdue.edu not know, that is true knowledge." > -Confucius > ------------------------------------------------------------------ > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From vids@mentor.cc.purdue.edu Thu, 7 Sep 2000 18:08:36 -0500 (EST) Date: Thu, 7 Sep 2000 18:08:36 -0500 (EST) From: Vidyut Samanta (aka vids) vids@mentor.cc.purdue.edu Subject: [PLUG] help On Thu, 7 Sep 2000, Andrew Sharp wrote: > > What kinds of things do you guys want covered and I'll try and get some > > lightning talks on security. I think that that would be a good way of > > doing it. Have maybe 5 or 6 people talk on differnet security concepts > > and present different things on what to do or not do. > > > I think a bit about what programs like Tripwire (and others) can do > for you. I'm a little hazy in that area. Also a discussion of good > firewalling rules would be helpful I'm sure. > > Then of course, the basic stuff w/inetd, ssh etc. > I'd also be interested in Portscan Detectors like Dragon IDS, or Shadow. And also to go a step further on setting up Honeypots. maybe FakeBO or some thing similar. -vids > Andrew > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From cnd@ecn.purdue.edu Fri, 08 Sep 2000 09:56:01 -0500 Date: Fri, 08 Sep 2000 09:56:01 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Re: Mandrake 7.1 and assorted issues > I've been tooling around a bit with mandrake 7.1 and I need a few Go Mandrake!! > 1. is there a script that automatically mounts modules at startup? i > need to insmod pegsus.o so I can get the ol' ethernet card running. I think you should be able to put in /etc/conf.modules: alias eth0 pegsus If not, you can put in /etc/rc.d/rc.local: insmode pegsus > 2. I am having troubles with with soundcard. it is a PCI live! value and > all the configuration utils refuse to set it up. what do I do to get > that running? I think you have to get a special kernel from Creative. Not sure on that one. I don't have a live card. I know it works, just never set one up. > 3. I have a lexmark 3200 printer. it is listed as unsupported by the > kernel. ould I get it to run in a genric mode or something like that? Welcome to the wonderful world of Lexmark printers... Only some are supported. If it says it isn't, it's not. Try running, as root, 'printtool'. This will allow you to look through a list of supported printers. If it isn't there, and your on resnet, you can set up your dorm's printer. Most of the dorms have an HP LaserJet 4 MX I think. Hope that helps... -Chris -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From reh@blacksoul.net Fri, 8 Sep 2000 10:06:51 -0500 (EST) Date: Fri, 8 Sep 2000 10:06:51 -0500 (EST) From: Ryan E. Helfter reh@blacksoul.net Subject: [PLUG] Re: Mandrake 7.1 and assorted issues On Fri, 8 Sep 2000, Christopher N. Deckard wrote: > > 3. I have a lexmark 3200 printer. it is listed as unsupported by the > > kernel. ould I get it to run in a genric mode or something like that? HAHA, Chris, I remember that Lexmark Printer... You still have that? man that thing was beautiful when it printed, but never worked under Linux :) Anyway, there is a beta driver for that printer... Go here: http://www.geocities.com/dgordini/ > > Welcome to the wonderful world of Lexmark printers... Only some are > supported. If it says it isn't, it's not. Try running, as root, > 'printtool'. This will allow you to look through a list of supported > printers. If it isn't there, and your on resnet, you can set up your > dorm's printer. Most of the dorms have an HP LaserJet 4 MX I think. > > Hope that helps... > > -Chris > > -- Ryan E. Helfter reh@blacksoul.net Black Soul Networks, LLC -- From olenik@purdue.edu Fri, 08 Sep 2000 10:27:24 -0500 Date: Fri, 08 Sep 2000 10:27:24 -0500 From: Hayden Olenik olenik@purdue.edu Subject: [PLUG] Compilation Problems Hey, I know others here are using RH6.9.5 (Pinstripe), so maybe you can help me with a problem. It seems that every program I try to compile, all of which compiled under RH6.2, have problems compiling. This is usually with an error saying "pasting would not give a valid preprocessing token". Searching google, this seems to be a common problem with gcc 2.95/96 (I have gcc 2.96 as well as all c++ libs v2.96). Has anyone else had this trouble and/or does anyone know how to fix it? Hayden From pweber1@purdue.edu Fri, 8 Sep 2000 10:37:54 -0500 (EST) Date: Fri, 8 Sep 2000 10:37:54 -0500 (EST) From: peter pweber1@purdue.edu Subject: [PLUG] help Those of you who have had complaints filed against you and got more detailed info, how did you go about getting it? I was referred by the ODOS to Jeff Schwab at PDN who could only tell me that it was "port scans against remote computers starting around 2PM Wednesday." peter ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From porterds@purdue.edu Fri, 08 Sep 2000 10:59:17 -0500 Date: Fri, 08 Sep 2000 10:59:17 -0500 From: Doug Porter porterds@purdue.edu Subject: [PLUG] help peter wrote: > > Those of you who have had complaints filed against you and got more > detailed info, how did you go about getting it? I was referred by the ODOS > to Jeff Schwab at PDN who could only tell me that it was "port scans > against remote computers starting around 2PM Wednesday." > > peter well, when i ran into trouble once i didn't get much more information than that. i just asked the original lady that called me to get more info. she told me she would try to get more details and call me back in a few days. when she called back later all she could tell me was the time of the supposed attack and tcp port 79. all i can figure is that i must have been doing some malicious fingering :-) it was probably someone in a dorm who didn't like the fact that i fingered their box once or twice so they childishly reported it to pucc as break in attempt. she told me to call jeff schwab to find out more, and i did, but he never returned my call. i haven't heard anything about it since... i can't believe they even act on port scanning. i get port scanned once or twice a day usually, and i have port scanned several of my friends computers before. hehe, i have even port scanned my own computer from expert.cc before :-) - doug From calx@purdue.edu Fri, 8 Sep 2000 11:49:22 -0500 (EST) Date: Fri, 8 Sep 2000 11:49:22 -0500 (EST) From: David calx@purdue.edu Subject: [PLUG] help > Those of you who have had complaints filed against you and got more > detailed info, how did you go about getting it? I was referred by the ODOS > to Jeff Schwab at PDN who could only tell me that it was "port scans > against remote computers starting around 2PM Wednesday." When the assistant dean, or whomever called, I did most of the talking, asking questions over and over. Asst. Dean: You have been doing bad things. We noticed and "flagged" your file. Me: What bad things? Asst. Dean: Things that are against the computer network usage guidelines. Me: Ok. What did I do? Asst Dean: You attempted to break in to a student's computer. Me: That's strange, since I don't remember trying to do such a thing. Is there any way you could be a bit more specific. I believe that if I am accused of "hacking" I should at least have the right to hear what things I have done. Asst. Dean: (Now getting pissed.) I will look at the report. .... It appears that you were trying to "FTP" into a FreeBSD computer. Me: Are you serious? Asst. Dean: Yes. Me: How is that against network usage policy? Asst. Dean: It irritated the owner of the computer, and they filed a report. Me: Ok. Did I do any damage? Asst. Dean: Not they I can read from the report. Me: So I apparently irritated somebody by attempting unsuccessfully to log in to their ftp server? How is this manevolent? Asst. Dean: (Now getting really impatient) You just shouldn't be messing with things you don't know anything about. Me: (Realizing the futility of the discussion) OK. I will watch myself. At least I knew what I was accused of. I would have been really mad if they didn't tell me. They probably get tons of these and just don't want to deal with explaining what each is about. (At least I hope that's the reason.) Just pester them. :) I believe that you have a right to know considering that it wasn't you intentionally attacking a box. David From rjune@ims1.imagestream-is.com Fri, 8 Sep 2000 12:26:02 -0500 (EST) Date: Fri, 8 Sep 2000 12:26:02 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] Re: Mandrake 7.1 and assorted issues On Fri, 8 Sep 2000, Christopher N. Deckard wrote: > > 1. is there a script that automatically mounts modules at startup? i > > need to insmod pegsus.o so I can get the ol' ethernet card running. > > I think you should be able to put in /etc/conf.modules: > > alias eth0 pegsus That should do it. > > 2. I am having troubles with with soundcard. it is a PCI live! value and > > all the configuration utils refuse to set it up. what do I do to get > > that running? All you should have to do is insmod the emu10k1 module, I've got a Live and that's I'll I had to do. > I think you have to get a special kernel from Creative. Not sure on > that one. I don't have a live card. I know it works, just never set > one up. Chris, no offense intended but what crack were you on when you thought that up? a special kernel to use a sound card? a pox on any company that tries to give out special kernels instead of drivers. From rjune@ims1.imagestream-is.com Fri, 8 Sep 2000 12:26:34 -0500 (EST) Date: Fri, 8 Sep 2000 12:26:34 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] :Cue:Cat Anyone gotten the CueCat to work under Linux/Alpha? From pweber1@purdue.edu Fri, 8 Sep 2000 12:37:15 -0500 Date: Fri, 8 Sep 2000 12:37:15 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] help she wasn't even that helpful with me. said she couldn't get any more info and she actually berated me for spending time worrying about this because i "should be spending that time studying." her logic apparently was that i didn't need to know anything else, all i needed to do was "secure my machine". feh. i don't even know *how* to do a port scan!! >peter wrote: >> >> Those of you who have had complaints filed against you and got more >> detailed info, how did you go about getting it? I was referred by the ODOS >> to Jeff Schwab at PDN who could only tell me that it was "port scans >> against remote computers starting around 2PM Wednesday." >> >> peter > >well, when i ran into trouble once i didn't get much more information >than that. i just asked the original lady that called me to get more >info. she told me she would try to get more details and call me back in >a few days. when she called back later all she could tell me was the >time of the supposed attack and tcp port 79. all i can figure is that i >must have been doing some malicious fingering :-) it was probably >someone in a dorm who didn't like the fact that i fingered their box >once or twice so they childishly reported it to pucc as break in >attempt. she told me to call jeff schwab to find out more, and i did, >but he never returned my call. i haven't heard anything about it >since... > >i can't believe they even act on port scanning. i get port scanned once >or twice a day usually, and i have port scanned several of my friends >computers before. hehe, i have even port scanned my own computer from >expert.cc before :-) > >- doug > >____________________________________________________ >The Purdue Linux Users' Group (PLUG) mailing list. >plug mailing list - plug@csociety.purdue.edu >http://csociety.ecn.purdue.edu/mailman/listinfo/plug -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From pollackc@purdue.edu Fri, 08 Sep 2000 12:37:45 -0500 Date: Fri, 08 Sep 2000 12:37:45 -0500 From: Christopher Alan Pollack pollackc@purdue.edu Subject: [PLUG] :Cue:Cat I have gotten the hardware to work fine. When you scan a barcode, it outputs a string of characters with a very distinct pattern. However I have not messed around with any software stuff yet. Chris (evolution) Pollack Richard June wrote: > > Anyone gotten the CueCat to work under Linux/Alpha? > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From porterds@purdue.edu Fri, 08 Sep 2000 12:44:31 -0500 Date: Fri, 08 Sep 2000 12:44:31 -0500 From: Doug Porter porterds@purdue.edu Subject: [PLUG] :Cue:Cat Christopher Alan Pollack wrote: > > I have gotten the hardware to work fine. When you scan a barcode, it > outputs a string of characters with a very distinct pattern. However I > have not messed around with any software stuff yet. > > Chris (evolution) Pollack did you recently get an alpha chris and not tell me about it? richard was asking whether anyone could get it to work with an alpha, not a dual celeron. - doug > Richard June wrote: > > > > Anyone gotten the CueCat to work under Linux/Alpha? > > > > ____________________________________________________ > > The Purdue Linux Users' Group (PLUG) mailing list. > > plug mailing list - plug@csociety.purdue.edu > > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From scott@purdue.edu Fri, 8 Sep 2000 13:09:29 -0500 Date: Fri, 8 Sep 2000 13:09:29 -0500 From: Scott Stembaugh scott@purdue.edu Subject: [PLUG] Security Considering the recent dicussions on security has any looked at Securing and Optimizing Linux: Red Hat Edition (http://www.linuxdoc.org/guides.html#securing_linux)? Opinions? --scott From rajak@purdue.edu Fri, 8 Sep 2000 13:15:30 -0500 (EST) Date: Fri, 8 Sep 2000 13:15:30 -0500 (EST) From: Brian Poole rajak@purdue.edu Subject: [PLUG] Security I haven't read it myself, just have to put in the obligatory 'Special editions for specific distros aren't worth reading' bit. Learn to secure a machine or learn to secure Linux, don't learn to secure Redhat. If you are interested in security you should read books that talk about security issues in general, then go and apply it to your specific distro/OS, otherwise you are limiting your knowledge and flexibility. Redhat is undoubtedly one of the most popular Linux distributions, but there are a lot of others and you will not always be using Redhat, so why limit yourself? -b On Fri, 8 Sep 2000, Scott Stembaugh wrote: > Date: Fri, 8 Sep 2000 13:09:29 -0500 > From: Scott Stembaugh > To: plug@csociety.purdue.edu > Subject: [PLUG] Security > > Considering the recent dicussions on security has any looked at Securing and > Optimizing Linux: Red Hat Edition > (http://www.linuxdoc.org/guides.html#securing_linux)? Opinions? > > > --scott From pfitzge1@purdue.edu Fri, 8 Sep 2000 13:22:04 -0500 (EST) Date: Fri, 8 Sep 2000 13:22:04 -0500 (EST) From: patrick.n.fitzgerald.1 pfitzge1@purdue.edu Subject: [PLUG] help On Thu, 7 Sep 2000, John C Meuser wrote: > > > in /var/log/messages: > > > > > > Sep 5 16:46:43 hawk-c-019 sshd[713]: connection from "193.89.248.8" > > > Sep 6 01:26:02 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > > > Unknown opcode from host 128.211.202.78 > > > Sep 6 02:34:30 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > > > Unknown opcode from host 128.211.202.78 > > > Sep 6 03:34:31 hawk-c-019 gdm[27172]: gdm_xdmcp_decode_packet: > > > Unknown opcode from host 128.211.202.78 > > > > > > Hey! Thats the same guy that was bothering me the other day. From my > > logs:Sep 5 16:23:23 goatsonparade portsentry[66506]: attackalert: Connect > > from host: www.baa.dk/193.89.248.8 to TCP port: 143 > > Sep 5 16:23:23 goatsonparade portsentry[66506]: attackalert: Host > > 193.89.248.8 has been blocked via wrappers with string:"ALL: 193.89.248.8" > > > > Andrew > > > He tried to get me too: > > Sep 5 18:08:31 wily-c-071 sshd[24761]: fatal: Timeout before > authentication for 193.89.248.8. Ooh! Ooh! Me too! Sep 5 16:46:12 phor04 sshd[22907]: log: Connection from 193.89.248.8 port 4435 Sep 5 16:46:12 phor04 sshd[22907]: fatal: Did not receive ident string. Sep 5 16:46:12 phor04 ftpd-BSD[22906]: connect from www.baa.dk earlier in the week... Sep 3 13:32:16 phor04 sshd[21327]: log: Connection from 24.94.206.211 port 2269 Sep 3 13:32:16 phor04 sshd[21327]: fatal: Did not receive ident string. Sep 3 13:32:17 phor04 ftpd-BSD[21328]: connect from wks-94-206-211.kscable.com It's a fact of life. I get portscanned all the time, and there ain't much I can do about it. Keep your daemons few, and your traffic encrypted. OpenBSD looks better and better every day. Patrick F. -- Gravity is a myth, the Earth sucks. --fortune (5) From rajak@purdue.edu Fri, 8 Sep 2000 13:29:07 -0500 (EST) Date: Fri, 8 Sep 2000 13:29:07 -0500 (EST) From: Brian Poole rajak@purdue.edu Subject: [PLUG] Re: Mandrake 7.1 and assorted issues On Fri, 8 Sep 2000, Richard June wrote: > Date: Fri, 8 Sep 2000 12:26:02 -0500 (EST) > From: Richard June > To: Plug > Subject: Re: [PLUG] Re: Mandrake 7.1 and assorted issues > > On Fri, 8 Sep 2000, Christopher N. Deckard wrote: > > > > 1. is there a script that automatically mounts modules at startup? i > > > need to insmod pegsus.o so I can get the ol' ethernet card running. > > > > I think you should be able to put in /etc/conf.modules: /etc/conf.modules is being deprecated in newer modutils, so be forewarned, you should be using /etc/modules.conf, the new modutils will warn you if it detects something it doesnt like, as illustrated below.. modprobe version 2.3.15 Warning: modutils is reading from /etc/modules.conf and ignoring /etc/conf.modules. The use of /etc/conf.modules is deprecated, please remove /etc/conf.modules as soon as possible. Command rm /etc/conf.modules or.. Warning: You do not need a link from /etc/conf.modules to /etc/modules.conf. The use of /etc/conf.modules is deprecated, please remove /etc/conf.modules as soon as possible. Command rm /etc/conf.modules or.. Warning: modutils is reading from /etc/conf.modules because /etc/modules.conf does not exist. The use of /etc/conf.modules is deprecated, please rename /etc/conf.modules to /etc/modules.conf as soon as possible. Command mv /etc/conf.modules /etc/modules.conf > > > > alias eth0 pegsus > That should do it. > > > > 2. I am having troubles with with soundcard. it is a PCI live! value and > > > all the configuration utils refuse to set it up. what do I do to get > > > that running? > All you should have to do is insmod the emu10k1 module, I've got a > Live and that's I'll I had to do. > > > I think you have to get a special kernel from Creative. Not sure on > > that one. I don't have a live card. I know it works, just never set > > one up. > Chris, no offense intended but what crack were you on when you thought > that up? a special kernel to use a sound card? a pox on any company that > tries to give out special kernels instead of drivers. Soundblaster did indeed release a special kernel, although not a complete one. Rather it was a kernel module, which as we all know is part of the kernel. Bad wording agreed, but still essentially correct. At the beginning it was a closed source (binary) module, however they have opensourced the module last I checked. Check out http://opensource.creative.com/ and http://developer.creative.com/linux/ for more information on SB's sound card support in Linux. -b From rjune@ims1.imagestream-is.com Fri, 8 Sep 2000 13:29:35 -0500 (EST) Date: Fri, 8 Sep 2000 13:29:35 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] :Cue:Cat Thanks for paying attention Doug :-) Here's the error. drivers/char/char.a(cuecat.o): In function `cuecat_init': cuecat.c(.text+0xb0): undefined reference to `kernel_thread' cuecat.c(.text+0xb4): undefined reference to `kernel_thread' I've added #include to cuecat.c but kernel_thread is still undefined. any ideas? On Fri, 8 Sep 2000, Doug Porter wrote: > Christopher Alan Pollack wrote: > > > > I have gotten the hardware to work fine. When you scan a barcode, it > > outputs a string of characters with a very distinct pattern. However I > > have not messed around with any software stuff yet. > > > > Chris (evolution) Pollack > > did you recently get an alpha chris and not tell me about it? richard > was asking whether anyone could get it to work with an alpha, not a dual > celeron. > > - doug > > > Richard June wrote: > > > > > > Anyone gotten the CueCat to work under Linux/Alpha? > > > > > > ____________________________________________________ > > > The Purdue Linux Users' Group (PLUG) mailing list. > > > plug mailing list - plug@csociety.purdue.edu > > > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > > > > ____________________________________________________ > > The Purdue Linux Users' Group (PLUG) mailing list. > > plug mailing list - plug@csociety.purdue.edu > > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From cnd@ecn.purdue.edu Fri, 08 Sep 2000 13:40:13 -0500 Date: Fri, 08 Sep 2000 13:40:13 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Re: Mandrake 7.1 and assorted issues > > I think you have to get a special kernel from Creative. Not sure on > > that one. I don't have a live card. I know it works, just never set > > one up. > Chris, no offense intended but what crack were you on when you thought > that up? a special kernel to use a sound card? a pox on any company that > tries to give out special kernels instead of drivers. I probably worded that wrong. They used to have binary only drivers that would only work with specific kernels. That is what I was referring to. This was about a year ago, before they opened the code to the drivers (I think they've opened it.) -Chris -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From reh@blacksoul.net Fri, 8 Sep 2000 14:16:27 -0500 (EST) Date: Fri, 8 Sep 2000 14:16:27 -0500 (EST) From: Ryan E. Helfter reh@blacksoul.net Subject: [PLUG] Sun Keyboard Question Does anyone know specifically how to enable the on/off key for the internal speaker on an UltraSparc? I know this sounds like a trivial question, but its just annoying to have MP3's playing through the HK speakers only to be ruined by the internal speaker... Trust me, I have thought about just opening the case and disconnecting it rather than messing with the xmodmap sequence. I have looked all through docs.sun.com and sunsolve.sun.com to no avail... This is not a critical thing that needs to be immediatly resolved... But would appreciated any helpful info. thanks -- Ryan E. Helfter reh@blacksoul.net Black Soul Networks, LLC -- From reh@blacksoul.net Fri, 8 Sep 2000 14:22:27 -0500 (EST) Date: Fri, 8 Sep 2000 14:22:27 -0500 (EST) From: Ryan E. Helfter reh@blacksoul.net Subject: [PLUG] Sun Keyboard Question Nevermind, the /usr/openwin/bin/audiocontrol binary did the trick... sorry to bother you... I emailed the list prematurely... ry On Fri, 8 Sep 2000, Ryan E. Helfter wrote: > Date: Fri, 8 Sep 2000 14:16:27 -0500 (EST) > From: Ryan E. Helfter > To: Purdue Linux Users Group > Subject: [PLUG] Sun Keyboard Question > > Does anyone know specifically how to enable the on/off key for the > internal speaker on an UltraSparc? > > I know this sounds like a trivial question, but its just annoying to have > MP3's playing through the HK speakers only to be ruined by the internal > speaker... Trust me, I have thought about just opening the case and > disconnecting it rather than messing with the xmodmap sequence. > > I have looked all through docs.sun.com and sunsolve.sun.com to no avail... > > This is not a critical thing that needs to be immediatly resolved... But > would appreciated any helpful info. > > thanks > > -- Ryan E. Helfter reh@blacksoul.net Black Soul Networks, LLC -- From malex@purdue.edu Fri, 08 Sep 2000 14:23:11 -0500 Date: Fri, 08 Sep 2000 14:23:11 -0500 From: Oleksandr Moskalenko malex@purdue.edu Subject: [PLUG] OpenBSD woes Hello, I remember that some people mentioned at PLUG meetings that they ran OpenBSD on their firewall boxen. Please if any of you read this help me sort this one out. I FTP installed OpenBSD 2.7 onto a box that ran Mandrake 7.0 for several months. I was going to run ipf/nat/ftpd on this box. However, since the hour 1 troubles started rolling in. Every few hours OBSD segfaults seemingly at random. Last one was this morning when I was working on the configuration of ssh, previous one happened when I was trying to obtain source tree through anonCVS. I can't see a pattern. Last one that I saw gave me a message: uvm_fault (0xe3l8dedf, 0x0, 0, 3) -> 1 kernel: page fault trap, code = 0 stopped at -crfree+0xld decw 0(%edx) I already tried searching OBSD mailing list archives and only found one message with "uvm_fault" in it that wasn't really answered. There was one answer that said that it be hardware related. I am totally lost as I ran linux on this box for more than a half a year and only had some troubles with kernel compilation when I had mixed FPM/EDO memory in it that I rid of. Dmesg output of my machine is listed below. Please give me an advice on what I can do to sort this out. I already dusted the box and reseated all the cards. Then I ran DOS based config utility to set ISA 3Com Etherlink III card to non-plug n' pray and put it onto a fixed IRQ 11. What else can I do. It's working now, but it's been less than a half an hour since I booted it. Thanks, Alex. Begin dmesg ======================================== OpenBSD 2.7 (GENERIC) #25: Sat May 13 18:04:26 MDT 2000 deraadt@i386.openbsd.org:/usr/src/sys/arch/i386/compile/GENERIC cpu0: F00F bug workaround installed cpu0: Intel Pentium (P54C) ("GenuineIntel" 586-class) 133 MHz cpu0: FPU,V86,DE,PSE,TSC,MSR,MCE,CX8 real mem = 16363520 (15980K) avail mem = 10575872 (10328K) using 225 buffers containing 921600 bytes (900K) of memory mainbus0 (root) bios0 at mainbus0: AT/286+(64) BIOS, date 03/07/96 pci0 at mainbus0 bus 0: configuration mode 1 (bios) pchb0 at pci0 dev 0 function 0 "Intel 82437FX (Triton) PCI/Cache/DRAM" rev 0x01 pcib0 at pci0 dev 7 function 0 "Intel 82371FB (Triton) PCI-ISA" rev 0x02 pciide0 at pci0 dev 7 function 1 "Intel 82371FB (Triton) IDE" rev 0x02: DMA, channel 0 wired to compatibility, channel 1 wired to compatibility wd0 at pciide0 channel 0 drive 0: wd0: can use 32-bit, PIO mode 4, DMA mode 2 wd0: 16-sector PIO, LBA, 3093MB, 6704 cyl, 15 head, 63 sec, 6335280 sectors pciide0: channel 0 interrupting at irq 14 wd0(pciide0:0:0): using PIO mode 4, DMA mode 2 (using DMA data transfers) atapiscsi0 at pciide0 channel 1 scsibus0 at atapiscsi0: 2 targets cd0 at scsibus0 targ 0 lun 0: SCSI0 5/cdrom removable cd0: can use 32-bit, PIO mode 4, DMA mode 2 pciide0: channel 1 interrupting at irq 15 cd0(pciide0:1:0): using PIO mode 4, DMA mode 2 (using DMA data transfers) "S3 ViRGE" rev 0x06 at pci0 dev 18 function 0 not configured isa0 at pcib0 isadma0 at isa0 we0: changing IRQ 9 to 3 we0 at isa0 port 0x280/32 iomem 0xd0000/16384 irq 3: SMC8216/SMC8216C (16-bit) we0: address 00:00:c0:eb:79:7f ep0 at isa0 port 0x300/16 irq 11: address 00:20:af:3f:c2:c7, utp/aui (default utp) pcppi0 at isa0 port 0x61 midi0 at pcppi0: sysbeep0 at pcppi0 lpt0 at isa0 port 0x378/4 irq 7 npx0 at isa0 port 0xf0/16: using exception 16 pccom0 at isa0 port 0x3f8/8 irq 4: ns16550a, 16 byte fifo pccom1: irq 3 already in use vt0 at isa0 port 0x60/16 irq 1: vga 80 col, color, 8 scr, mf2-kbd pms0 at vt0 irq 12 fdc0 at isa0 port 0x3f0/6 irq 6 drq 2 fd0 at fdc0 drive 0: 1.44MB 80 cyl, 2 head, 18 sec biomask c040 netmask c848 ttymask d8ca pctr: 586-class performance counters and user-level cycle counter enabled dkcsum: wd0 matched BIOS disk 80 root on wd0a rootdev=0x0 rrootdev=0x300 rawdev=0x302 ====================================================== END From reh@blacksoul.net Fri, 8 Sep 2000 15:19:29 -0500 (EST) Date: Fri, 8 Sep 2000 15:19:29 -0500 (EST) From: Ryan E. Helfter reh@blacksoul.net Subject: [PLUG] OpenBSD woes After it crashes... If you do a that should bring the machine to debug mode. Then if you type "panic" this should make the machine core out... You can debug from there... Or send it off to the experts... Other than that, I am fresh out of ideas... I would look / read the OpenBSD handbook, as it is full of info. ry On Fri, 8 Sep 2000, Oleksandr Moskalenko wrote: > Date: Fri, 08 Sep 2000 14:23:11 -0500 > From: Oleksandr Moskalenko > To: plug@csociety.purdue.edu > Subject: [PLUG] OpenBSD woes > > > Hello, > > I remember that some people mentioned at PLUG > meetings that they ran OpenBSD on their firewall > boxen. Please if any of you read this help me sort > this one out. > I FTP installed OpenBSD 2.7 onto a box that ran > Mandrake 7.0 for several months. I was going to > run ipf/nat/ftpd on this box. However, since the > hour 1 troubles started rolling in. Every few hours > OBSD segfaults seemingly at random. > Last one was this morning when I was working on the > configuration of ssh, previous one happened when I > was trying to obtain source tree through anonCVS. I > can't see a pattern. Last one that I saw gave me a > message: > > uvm_fault (0xe3l8dedf, 0x0, 0, 3) -> 1 > kernel: page fault trap, code = 0 > stopped at -crfree+0xld decw 0(%edx) > > I already tried searching OBSD mailing list archives > and only found one message with "uvm_fault" in it that > wasn't really answered. There was one answer that said > that it be hardware related. I am totally > lost as I ran linux on this box for more than a half > a year and only had some troubles with kernel compilation > when I had mixed FPM/EDO memory in it that I rid of. > > Dmesg output of my machine is listed below. Please give me > an advice on what I can do to sort this out. I already > dusted the box and reseated all the cards. Then I ran DOS > based config utility to set ISA 3Com Etherlink III card to > non-plug n' pray and put it onto a fixed IRQ 11. What else > can I do. It's working now, but it's been less than a half > an hour since I booted it. > > Thanks, > > Alex. > > Begin dmesg > ======================================== > OpenBSD 2.7 (GENERIC) #25: Sat May 13 18:04:26 MDT 2000 > deraadt@i386.openbsd.org:/usr/src/sys/arch/i386/compile/GENERIC > cpu0: F00F bug workaround installed > cpu0: Intel Pentium (P54C) ("GenuineIntel" 586-class) 133 MHz > cpu0: FPU,V86,DE,PSE,TSC,MSR,MCE,CX8 > real mem = 16363520 (15980K) > avail mem = 10575872 (10328K) > using 225 buffers containing 921600 bytes (900K) of memory > mainbus0 (root) > bios0 at mainbus0: AT/286+(64) BIOS, date 03/07/96 > pci0 at mainbus0 bus 0: configuration mode 1 (bios) > pchb0 at pci0 dev 0 function 0 "Intel 82437FX (Triton) PCI/Cache/DRAM" rev 0x01 > pcib0 at pci0 dev 7 function 0 "Intel 82371FB (Triton) PCI-ISA" rev 0x02 > pciide0 at pci0 dev 7 function 1 "Intel 82371FB (Triton) IDE" rev 0x02: DMA, > channel 0 wired to compatibility, channel 1 wired to compatibility > wd0 at pciide0 channel 0 drive 0: > wd0: can use 32-bit, PIO mode 4, DMA mode 2 > wd0: 16-sector PIO, LBA, 3093MB, 6704 cyl, 15 head, 63 sec, 6335280 sectors > pciide0: channel 0 interrupting at irq 14 > wd0(pciide0:0:0): using PIO mode 4, DMA mode 2 (using DMA data transfers) > atapiscsi0 at pciide0 channel 1 > scsibus0 at atapiscsi0: 2 targets > cd0 at scsibus0 targ 0 lun 0: SCSI0 5/cdrom removable > cd0: can use 32-bit, PIO mode 4, DMA mode 2 > pciide0: channel 1 interrupting at irq 15 > cd0(pciide0:1:0): using PIO mode 4, DMA mode 2 (using DMA data transfers) > "S3 ViRGE" rev 0x06 at pci0 dev 18 function 0 not configured > isa0 at pcib0 > isadma0 at isa0 > we0: changing IRQ 9 to 3 > we0 at isa0 port 0x280/32 iomem 0xd0000/16384 irq 3: SMC8216/SMC8216C (16-bit) > we0: address 00:00:c0:eb:79:7f > ep0 at isa0 port 0x300/16 irq 11: address 00:20:af:3f:c2:c7, utp/aui (default > utp) > pcppi0 at isa0 port 0x61 > midi0 at pcppi0: > sysbeep0 at pcppi0 > lpt0 at isa0 port 0x378/4 irq 7 > npx0 at isa0 port 0xf0/16: using exception 16 > pccom0 at isa0 port 0x3f8/8 irq 4: ns16550a, 16 byte fifo > pccom1: irq 3 already in use > vt0 at isa0 port 0x60/16 irq 1: vga 80 col, color, 8 scr, mf2-kbd > pms0 at vt0 irq 12 > fdc0 at isa0 port 0x3f0/6 irq 6 drq 2 > fd0 at fdc0 drive 0: 1.44MB 80 cyl, 2 head, 18 sec > biomask c040 netmask c848 ttymask d8ca > pctr: 586-class performance counters and user-level cycle counter enabled > dkcsum: wd0 matched BIOS disk 80 > root on wd0a > rootdev=0x0 rrootdev=0x300 rawdev=0x302 > ====================================================== > END > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > -- Ryan E. Helfter reh@blacksoul.net Black Soul Networks, LLC -- From sergent@csociety.purdue.edu Fri, 08 Sep 2000 15:34:41 -0500 Date: Fri, 08 Sep 2000 15:34:41 -0500 From: Jonathan Sergent sergent@csociety.purdue.edu Subject: OT: Re: [PLUG] Sun Keyboard Question > Nevermind, the /usr/openwin/bin/audiocontrol binary did the trick... > > sorry to bother you... I emailed the list prematurely... BTW: /usr/dt/bin/sdtaudiocontrol is better... especially on Solaris 8. You can even adjust the mixer volume for each program that has the device open, and view all kinds of statistics about the audio device. From pweber1@purdue.edu Fri, 8 Sep 2000 15:51:46 -0500 Date: Fri, 8 Sep 2000 15:51:46 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] follow-up i said that i'd post a follow-up once i got as much info as possible, so here goes... since i found out that the offending connections that were traced to my IP were "port scans of remote computers", it's a sure thing that i was hacked. i'm still working on seeing if i can figure out who did it, but at least i know that it was indeed a break-in. so what i've decided to do in response to this is start from scratch. (i'd been thinking about it anyway.) i'm going to back everything up, reformat and reinstall. then i'm going to spend as much time as i need to get the machine as secure as possible BEFORE it goes back online. (i almost guarantee i'll be asking questions about this!) with this in mind, i'm curious about perhaps trying out a different distro of linux or a *BSD. (from what i've heard OpenBSD would be best in terms of security.) how much of a paradigm shift would i be looking at? how is hardware support for *BSD (Open or Free)? (last time i did a fresh install {RH6.0} i tried to install a version FreeBSD, and it didn't work. i have a new motherboard now.) i run almost no linux-specific software (i like being able to use WordPerfect 8, for instance, but it's by no means necessary), but IIRC you can install the linux libraries under *BSD, is this correct? would that have any effect on security? i don't mind making a switch in how to administrate a box (in fact, it would be neat to learn a different system), as long as i can do (basically) all of the same things. i apologize in advance if i inadvertantly start a *BSD vs. Linux battle. :) thanks, peter -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From will@physics.purdue.edu Fri, 8 Sep 2000 16:05:25 -0500 Date: Fri, 8 Sep 2000 16:05:25 -0500 From: Will Andrews will@physics.purdue.edu Subject: [PLUG] follow-up On Fri, Sep 08, 2000 at 03:51:46PM -0500, peter wrote: > since i found out that the offending connections that were traced to > my IP were "port scans of remote computers", it's a sure thing that i > was hacked. i'm still working on seeing if i can figure out who did > it, but at least i know that it was indeed a break-in. Just for the record, that would be "cracked", not "hacked". > so what i've decided to do in response to this is start from scratch. > (i'd been thinking about it anyway.) i'm going to back everything up, > reformat and reinstall. then i'm going to spend as much time as i > need to get the machine as secure as possible BEFORE it goes back > online. (i almost guarantee i'll be asking questions about this!) There's not really much to do in terms of blocking off 99% of intruders. Close off all your unnecessary services, don't use telnet (ssh only), firewall everything out except services you know about, block certain protocols, and so forth.. It may also help to break TCP/IP protocol by dropping SYN+FIN packets, because that makes it extremely difficult for nmap et al to figure out what OS you're running. > with this in mind, i'm curious about perhaps trying out a different > distro of linux or a *BSD. (from what i've heard OpenBSD would be > best in terms of security.) how much of a paradigm shift would i be > looking at? how is hardware support for *BSD (Open or Free)? (last > time i did a fresh install {RH6.0} i tried to install a version > FreeBSD, and it didn't work. i have a new motherboard now.) There's not really much difference between FreeBSD and OpenBSD as far as security goes. I'd have a hard time believing that you couldn't secure a FreeBSD machine as well as OpenBSD is by default, and then some. Thus, I think the "security by obscurity" feature of OpenBSD's is overhyped. There are some advantages that OpenBSD has over FreeBSD (size, non-i386/alpha ports, others?), but not much in my opinion. If you need help installing FreeBSD, I'll be glad to help (I have used FreeBSD for more than three and a half years, and am part of the FreeBSD Project's developer team). I can also bring my FreeBSD 4.0 CD set, and show you how to upgrade it to the newest versions. > i run almost no linux-specific software (i like being able to use > WordPerfect 8, for instance, but it's by no means necessary), but > IIRC you can install the linux libraries under *BSD, is this correct? > would that have any effect on security? Yes, you can install Linux emulation on FreeBSD (don't know about NetBSD or OpenBSD). If you want a word processor, try KWord (from the koffice suite, which is part of the FreeBSD Ports Collection). Or AbiWord (but I hear that one sucks worse). AFAIK, the only security issues involving using Linux binaries is if the binaries themselves have holes. There's not much you can do about that unless the binaries themselves are fixed. But if you properly firewall your machines, there's not much chance an attacker could take advantage of any security holes. Additionally, if you don't run the Linux bins as root (where possible), damage can be limited. > i apologize in advance if i inadvertantly start a *BSD vs. Linux battle. :) Heheh. -- Will Andrews GCS/E/S @d- s+:+ a--- C++ UB++++$ P+ L- E--- W+ N-- !o ?K w--- O- M+ V- PS+ PE++ Y+ PGP+>+++ t++ 5 X+ R+ tv+ b++ DI+++ D+ G++ e>++++ h! r- y? From honeycug@purdue.edu Fri, 8 Sep 2000 16:18:44 -0500 (EST) Date: Fri, 8 Sep 2000 16:18:44 -0500 (EST) From: garrett honeycug@purdue.edu Subject: [PLUG] follow-up On Fri, 8 Sep 2000, peter wrote: > so what i've decided to do in response to this is start from scratch. > (i'd been thinking about it anyway.) i'm going to back everything up, > reformat and reinstall. then i'm going to spend as much time as i > need to get the machine as secure as possible BEFORE it goes back > online. (i almost guarantee i'll be asking questions about this!) OpenBSD is a very secure OS as it was built with this in mind; however, you can secure an OS [even an ugly red hat distro]. limit what services you run, have services run as their own user [NOT root], keep updated on whats going on in security, what kernels / programs you run have been exploited, et cetera. setting up a firewall would also be ideal if you have another machine. -garrett From reh@blacksoul.net Fri, 8 Sep 2000 16:26:00 -0500 (EST) Date: Fri, 8 Sep 2000 16:26:00 -0500 (EST) From: Ryan E. Helfter reh@blacksoul.net Subject: [PLUG] follow-up If you really want to secure your box a bit more, I suggest you read up on some DOC and HOWTO's on TCPWRAPPERS. (i.e. hosts.allow and hosts.deny) If not, portsentry and the such programs watch for probing and with the use of IPCHAINS, they block traffic of the source and email you what has been done, i.e. which port, what time, etc... To be really honest, any unix is as secure as it weakest link. You could have the most secure operation system in the world, if you run "untrusted services" then your box is untrusted. Firewalls and the such are useless in my book. (just another hop in the TCP chain). Please visit sites like www.securityfocus.com and get on the bugtraq mailing lists... They are mostly postings on known bugs that are just found, but at least you get on the inside scoop. Trust me, I used to be on the malicious side. And then I found ways to use the skills to help fight against people like my old self. Security will always be an issue in this world, learn it, use it, market yourself on it... BIG $$$ in this business, day, and age for it. I would suggest using the unix flavor that you are most comfortable with, and the only way to do that is to experiment with as many flavors as possible... Here in Chicago, the flavors of business choice here are HP/UX, (HPUX sucks btw, but people use it for HP-OPENVIEW, even though it runs on Solaris with less functionality) Solaris, and Linux. If you are into the ISP biz, FreeBSD is the most popular. IMO: If you want a server, use FreeBSD/OpenBSD. If you want a workstation, follow my advise and kill inetd or a good use of TCPWRAPPERS, and use Linux. just my 2 cents On Fri, 8 Sep 2000, peter wrote: > Date: Fri, 8 Sep 2000 15:51:46 -0500 > From: peter > To: plug@csociety.purdue.edu > Subject: [PLUG] follow-up > > i said that i'd post a follow-up once i got as much info as possible, > so here goes... > > since i found out that the offending connections that were traced to > my IP were "port scans of remote computers", it's a sure thing that i > was hacked. i'm still working on seeing if i can figure out who did > it, but at least i know that it was indeed a break-in. > > so what i've decided to do in response to this is start from scratch. > (i'd been thinking about it anyway.) i'm going to back everything up, > reformat and reinstall. then i'm going to spend as much time as i > need to get the machine as secure as possible BEFORE it goes back > online. (i almost guarantee i'll be asking questions about this!) > > with this in mind, i'm curious about perhaps trying out a different > distro of linux or a *BSD. (from what i've heard OpenBSD would be > best in terms of security.) how much of a paradigm shift would i be > looking at? how is hardware support for *BSD (Open or Free)? (last > time i did a fresh install {RH6.0} i tried to install a version > FreeBSD, and it didn't work. i have a new motherboard now.) > > i run almost no linux-specific software (i like being able to use > WordPerfect 8, for instance, but it's by no means necessary), but > IIRC you can install the linux libraries under *BSD, is this correct? > would that have any effect on security? > > i don't mind making a switch in how to administrate a box (in fact, > it would be neat to learn a different system), as long as i can do > (basically) all of the same things. > > i apologize in advance if i inadvertantly start a *BSD vs. Linux battle. :) > > thanks, > peter > -- Ryan E. Helfter reh@blacksoul.net Black Soul Networks, LLC -- From doobie@doobie.org Fri, 8 Sep 2000 16:37:21 -0500 (EST) Date: Fri, 8 Sep 2000 16:37:21 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] :Cue:Cat CueCat has been sending legal cease letters to anyone who posts copies of software to decode/do whatever. Here is a snip of code that I found however: No clue how well it works, it is a cgi web page.... #!/usr/bin/perl -w use CGI; my ($raw); $cgi= new CGI; $cgi->header; $s=$cgi->param('raw'); @fields = split(/\./,$s); @data = map(decode($_), @fields[1..$#fields]); print "$data[2]"; #print "
$data[1]"; print <Actions:
EOF $cgi->end_html; sub decode { my ($seq) = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789+-'; ($encoded) = @_; @s = map { index($seq,$_); } split(//,$encoded); $l = ($#s+1) % 4; if ($l) { if ($l == 1) { print "Error!"; return; } $l = 4-$l; $#s += $l; } $r = ''; while ($#s >= 0) { $n = (($s[0] << 6 | $s[1]) << 6 | $s[2]) << 6 | $s[3]; $r .=chr(($n >> 16) ^ 67) . chr(($n >> 8 & 255) ^ 67) . chr(($n & 255) ^ 67); @s = @s[4..$#s]; } $r = substr($r,0,length($r)-$l); return $r; } On Fri, 8 Sep 2000, Richard June wrote: > Anyone gotten the CueCat to work under Linux/Alpha? > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From doobie@doobie.org Fri, 8 Sep 2000 16:38:35 -0500 (EST) Date: Fri, 8 Sep 2000 16:38:35 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] :Cue:Cat I should probably read more clearly......Yes I can get the bar code crap coming up on my screen with both Alpha's I have tried it on. What model of Alpha do you have? Jason On Fri, 8 Sep 2000, Richard June wrote: > Anyone gotten the CueCat to work under Linux/Alpha? > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From rjune@ims1.imagestream-is.com Fri, 8 Sep 2000 16:41:57 -0500 (EST) Date: Fri, 8 Sep 2000 16:41:57 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] :Cue:Cat I've got an XL366 I am trying to get the kernel driver to compile. don't worry about not actually reading the question, a few people have had that problem already :-) I posted the error I was getting already From wallacms@cs.purdue.edu Fri, 8 Sep 2000 16:47:43 -0500 (EST) Date: Fri, 8 Sep 2000 16:47:43 -0500 (EST) From: Matt Wallace wallacms@cs.purdue.edu Subject: [PLUG] help > Call me crazy, but ("You're a Crazy Butt!") perhaps all of this > suggests that a PLUG supported lecture on securing linux boxen would > be a valuable event. I know I'd attend. I've been thinking about > getting a ResNet ASDL connection at home but I think I'll wait until I > learn how to secure my machine a little better. > > Pat I'm just curious. Why is anyone here running anything other than SSH on their boxen anyway (excluding people running boxen for business purposes)? Why do you need HTTP, FTP, Telnet, ??? If it's just for the coolness factor of "Wow, look at me I've got a web server", I think the security concerns outwiegh the coolness.. -Matt From pweber1@purdue.edu Fri, 8 Sep 2000 16:53:37 -0500 Date: Fri, 8 Sep 2000 16:53:37 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] help well, i was running ftp for connectivity with my non-unix-running brother, httpd so that i could easily spread info to my family and for other personal use, and a couple other things for connectivity with other non-unix machines i have running. also, i was allowing remote connections to gdm so that i could use pc-xware in PUCC labs. obviously this all led to some fairly serious problems, so i'm changing how i do things, but the reason for me was basically was convenience and functionality. (i wasn't running telnet, fwiw.) peter > > Call me crazy, but ("You're a Crazy Butt!") perhaps all of this >> suggests that a PLUG supported lecture on securing linux boxen would >> be a valuable event. I know I'd attend. I've been thinking about >> getting a ResNet ASDL connection at home but I think I'll wait until I >> learn how to secure my machine a little better. >> >> Pat > >I'm just curious. Why is anyone here running anything other than SSH on >their boxen anyway (excluding people running boxen for business purposes)? >Why do you need HTTP, FTP, Telnet, ??? If it's just for >the coolness factor of "Wow, look at me I've got a web server", I think >the security concerns outwiegh the coolness.. > >-Matt > > >____________________________________________________ >The Purdue Linux Users' Group (PLUG) mailing list. >plug mailing list - plug@csociety.purdue.edu >http://csociety.ecn.purdue.edu/mailman/listinfo/plug -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From wli@holomorphy.com Fri, 8 Sep 2000 15:23:47 -0700 Date: Fri, 8 Sep 2000 15:23:47 -0700 From: William Lee Irwin III wli@holomorphy.com Subject: [PLUG] help On Fri, Sep 08, 2000 at 04:47:43PM -0500, Matt Wallace wrote: > I'm just curious. Why is anyone here running anything other than SSH > on their boxen anyway (excluding people running boxen for business > purposes)? Why do you need HTTP, FTP, Telnet, ??? If > it's just for the coolness factor of "Wow, look at me I've got a web > server", I think the security concerns outwiegh the coolness.. Well, it's not as if there is anything intrinsically dangerous about these services, but that's I suppose a different attitude toward system configuration than what's taken here. HTTP and FTP are good for providing services to people you don't necessarily want to provide accounts to. Anonymous FTP (especially for incoming files) and HTTP for various snippets I write up now and then (for instance, my "Integration by Example" document). Furthermore, I much prefer to have my email delivered directly to my machine so that I might make use of (a) the email address for my domain and (b) receive my mail more or less instantly with the least amount of implementation effort. Security is not the first consideration, it's functionality. Security is very important, but the question (at least to me) is "I want to do this. How do I do this securely?" not "How do I make things secure regardless of convenience?" As far as telnet and r* services go, well, I don't have to deal with terribly many systems that don't have the ssh equivalents, and they're not otherwise useful, so they can go. It's not coolness, it's what I use the machines for. Cheers, Bill -- "I'm sure that nobody here would dream of misusing the Arpanet. It's as unthinkable as fornication, or smoking pot." -- Richard M. Stallman From heckarsd@purdue.edu Fri, 08 Sep 2000 17:42:34 -0500 Date: Fri, 08 Sep 2000 17:42:34 -0500 From: Seth Heckard heckarsd@purdue.edu Subject: [PLUG] help William Lee Irwin III wrote: > Well, it's not as if there is anything intrinsically dangerous about > these services, but that's I suppose a different attitude toward > system configuration than what's taken here. > > HTTP and FTP are good for providing services to people you don't > necessarily want to provide accounts to. Anonymous FTP (especially for > incoming files) and HTTP for various snippets I write up now and then > (for instance, my "Integration by Example" document). Furthermore, I > much prefer to have my email delivered directly to my machine so that > I might make use of (a) the email address for my domain and (b) receive > my mail more or less instantly with the least amount of implementation > effort. Security is not the first consideration, it's functionality. > Security is very important, but the question (at least to me) is > "I want to do this. How do I do this securely?" not "How do I make > things secure regardless of convenience?" As far as telnet and r* > services go, well, I don't have to deal with terribly many systems that > don't have the ssh equivalents, and they're not otherwise useful, so > they can go. > > It's not coolness, it's what I use the machines for. > > Cheers, > Bill i agree. here's what i run (don't crack me please :-) and why i need it. ssh: duh! dns: i have two domains. one points to my computer at purdue, the other points to my sparc serving up my home network from cable. both boxes run primary dns for their domain and secondary dns for their sister domain. http: if you go to my web page, you will get nothing. but... i need apache running for a couple reasons, one being that i made a custom status report webpage to save me time for my job. but i'm the only one that sees that. ftp: there have been several times where i needed to get a file from my box. of course i could use scp, but what if i'm on a windows box and don't have/can't get/don't want to mess with ssh? it's justified for me. mail: it's cool to have the email address seth@heckard.com even though i don't use it much :-) but i do use stuff like ebay@... or buycom@... whenever i buy stuff to minimize spam. nfs: only for my friends. it's convenient. probably insecure too... gdm: would *like* to use this, however it seems broken currently. ntp: i have a thing for atomic clocks. and yes, i have tripwire/portsentry/logcheck/etc. installed. and a minimal firewall, but nothing to lock down ports, just stupid stuff like the reserved class a, b, c addrs, etc. seth From rrussel5@purdue.edu Fri, 8 Sep 2000 18:37:14 -0500 Date: Fri, 8 Sep 2000 18:37:14 -0500 From: Alex Russell rrussel5@purdue.edu Subject: [PLUG] help I can think of a couple of hundred reasons to run some of these services, the main one probably being the lack of SCP support in a lot of SSH clients, not to mention client-server SSH incompatibilities (ssh2 is a moving target). Should people run these? Probably not, and at least not until the know how to chroot them and the like...but skewering those who need to because their personal needs don't fit yours might not be the best way to promote security consciousness. Alex -----Original Message----- From: plug-admin@csociety.purdue.edu [mailto:plug-admin@csociety.purdue.edu]On Behalf Of Matt Wallace Sent: Friday, September 08, 2000 4:48 PM To: Patrick K Notz Cc: plug@csociety.purdue.edu Subject: Re: [PLUG] help > Call me crazy, but ("You're a Crazy Butt!") perhaps all of this > suggests that a PLUG supported lecture on securing linux boxen would > be a valuable event. I know I'd attend. I've been thinking about > getting a ResNet ASDL connection at home but I think I'll wait until I > learn how to secure my machine a little better. > > Pat I'm just curious. Why is anyone here running anything other than SSH on their boxen anyway (excluding people running boxen for business purposes)? Why do you need HTTP, FTP, Telnet, ??? If it's just for the coolness factor of "Wow, look at me I've got a web server", I think the security concerns outwiegh the coolness.. -Matt ____________________________________________________ The Purdue Linux Users' Group (PLUG) mailing list. plug mailing list - plug@csociety.purdue.edu http://csociety.ecn.purdue.edu/mailman/listinfo/plug From wallacms@cs.purdue.edu Fri, 8 Sep 2000 19:15:40 -0500 (EST) Date: Fri, 8 Sep 2000 19:15:40 -0500 (EST) From: Matt Wallace wallacms@cs.purdue.edu Subject: [PLUG] help All valid answers. I wasn't intending to skewer anyone, just asking an honest question. And also trying to raise the point that unneccesary services are in my opinion the biggest security hole. All of the people that responded had legitemit reasons for running the services they did. Though I would argue that HTTP doesn't really qualify since 90% of the people on this list can serve a web page of off one of various university machines. But again, not trying to hurt anyone's feelings, just adding to the discussion. -Matt On Fri, 8 Sep 2000, Alex Russell wrote: > I can think of a couple of hundred reasons to run some of these services, > the main one probably being the lack of SCP support in a lot of SSH clients, > not to mention client-server SSH incompatibilities (ssh2 is a moving > target). Should people run these? Probably not, and at least not until the > know how to chroot them and the like...but skewering those who need to > because their personal needs don't fit yours might not be the best way to > promote security consciousness. > > Alex > > -----Original Message----- > From: plug-admin@csociety.purdue.edu > [mailto:plug-admin@csociety.purdue.edu]On Behalf Of Matt Wallace > Sent: Friday, September 08, 2000 4:48 PM > To: Patrick K Notz > Cc: plug@csociety.purdue.edu > Subject: Re: [PLUG] help > > > Call me crazy, but ("You're a Crazy Butt!") perhaps all of this > > suggests that a PLUG supported lecture on securing linux boxen would > > be a valuable event. I know I'd attend. I've been thinking about > > getting a ResNet ASDL connection at home but I think I'll wait until I > > learn how to secure my machine a little better. > > > > Pat > > I'm just curious. Why is anyone here running anything other than SSH on > their boxen anyway (excluding people running boxen for business purposes)? > Why do you need HTTP, FTP, Telnet, ??? If it's just for > the coolness factor of "Wow, look at me I've got a web server", I think > the security concerns outwiegh the coolness.. > > -Matt > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > > From notz@purdue.edu Fri, 8 Sep 2000 23:12:36 -0500 (EST) Date: Fri, 8 Sep 2000 23:12:36 -0500 (EST) From: Patrick K Notz notz@purdue.edu Subject: [PLUG] Linux/BSD Security [was: help] I second some of the responses. I've been using Linux since 1992 but I don't have bumper stickers and t-shirts. I do research in a unix environment and on any given day I have to work on AIX, Solaris, IRIX and (ugh) sometimes even HP-UX. I use things like cc/CC/f77 (yes, f77), cvs, make, bash, perl, grep, sed, etc. I stick with free software so that my makefiles, source files etc are usable on all these other OSes and so that I don't have to worry about the different usages for the proprietary variants of these tools. My linux/GNU/free cool factor comes this unity and from having all of this available at home. Nothing personal, but I don't want to be a sysadmin; I'm a proud user. While linux is wonderful for folks like me, I don't have the time (and not too much interest) to spend my free hour (singular) reading a bazillion scattered FAQs, HOWTOs, articles, books, man pages, info pages, the SAG and the NAG just to find out what lines like: iso-tsap 102/tcp tsap # part of ISODE. in my /etc/services mean. I would love a 1-1.5 hour intro to *nix/*bsd security to learn (a) what kind of threats are our there, (b) how might my machine be vulnerable, and (c) what are some good starting points for documentation and information. The problem, also, with many linux distros is that a lot of these services are enabled by default. Installers don't ask and don't warn. Isn't it cool to have a "full blown, decked out distro with 4400 packages installed"? Ahh, no. Then there's always the "don't fix it if it ain't broke" ideology. I'm not supporting disillusionment but I've mucked up my share of systems by tinkering. It really is a shame that distro makers don't pay more attention to this. It's tempting to try one of the *BSD variants. Incidently, if anyone knows of a good document which describes the /etc/services, /etc/inetd.conf, tcpwrappers, etc. I'd be greatful. I've been looking around but everything I find is in bits-n-pieces. Anyway, I feel like I've already learned a lot just from reading this thread... Cheers ~ pat From mahdawia@hotmail.com Sun, 10 Sep 2000 01:55:37 GMT Date: Sun, 10 Sep 2000 01:55:37 GMT From: King Sultan mahdawia@hotmail.com Subject: [PLUG] pppd question I use Red hat 6.2. When I connect to ISP and run pppd -d -detach noipdefault 38400, I get 'LCP: timeoute sending Config-Requests, connection is terminated' message. Can anyone help? Thanks, sheirak ===== _________________________________________________________________________ Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com. Share information about yourself, create your own public profile at http://profiles.msn.com. From doobie@doobie.org Sat, 9 Sep 2000 22:23:40 -0500 (EST) Date: Sat, 9 Sep 2000 22:23:40 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] :Cue:Cat Simple fix: :) alpha:/usr/src/linux/drivers/char> diff cuecat.c cuecat.c~ 116c116 < __kernel_thread(barcode_decoder_timeout,NULL,0); --- > kernel_thread(barcode_decoder_timeout,NULL,0); Jason On Fri, 8 Sep 2000, Richard June wrote: > I've got an XL366 I am trying to get the kernel driver to compile. > don't worry about not actually reading the question, a few people have had > that problem already :-) I posted the error I was getting already > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From pweber1@purdue.edu Sat, 9 Sep 2000 22:38:13 -0500 Date: Sat, 9 Sep 2000 22:38:13 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] server side includes does expert support server side includes? i moved my entire httpd directory to expert, and after changing all of the necessary things to make my pages load correctly and have my scripts work, i can't seem to get server side includes to work. all i get is: [an error occurred while processing this directive] i'm just using the exec directive. i'm also experiencing erratic behavior using .htaccess. it works for some directories and not for others. the .htaccess files are exactly the same for each directory. any ideas? thanks, peter -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From deepak@purdue.edu Sat, 9 Sep 2000 23:43:35 -0500 (EST) Date: Sat, 9 Sep 2000 23:43:35 -0500 (EST) From: Deepak Dinesh deepak@purdue.edu Subject: [PLUG] Mandrake 7.0 Hi, After I updated glibc for Mandrake 7.0 I get the following error when I compile any programs: perl: warning: Setting locale failed. perl: warning: Please check that your locale settings: LANGUAGE = "en", LC_ALL = "en", LANG = "en" are supported and installed on your system. Also many of my programs fail with "locale setting failed". Thanks. From wli@holomorphy.com Sat, 9 Sep 2000 22:42:40 -0700 Date: Sat, 9 Sep 2000 22:42:40 -0700 From: William Lee Irwin III wli@holomorphy.com Subject: [PLUG] Mandrake 7.0 On Sat, Sep 09, 2000 at 11:43:35PM -0500, Deepak Dinesh wrote: > Hi, > After I updated glibc for Mandrake 7.0 I get the following > error when I compile any programs: > > perl: warning: Setting locale failed. > perl: warning: Please check that your locale settings: > LANGUAGE = "en", > LC_ALL = "en", > LANG = "en" > are supported and installed on your system. > > Also many of my programs fail with "locale setting failed". > Thanks. Try "en_US" instead of "en". Cheers, Bill -- "Microsoft shouldn't be broken up. It should be shut down." -- Security expert Bruce Schneier on the ILOVEYOU virus. From lewis1@purdue.edu Sun, 10 Sep 2000 04:02:59 -0500 (EST) Date: Sun, 10 Sep 2000 04:02:59 -0500 (EST) From: James C. Lewis lewis1@purdue.edu Subject: [PLUG] Cue::Cat Hey... I went to radio shack today and picked up one of those cue::cat thingys. Since I don't have windows... anyone have or know where I can get the drivers for gnome/linux. I kind of like these and I really like cmdr taco's idea about indexing DVD's or somethign with them. BTW, if you say where to get it... i promise not to tell digital convergance ;) hehehehe Thanks in advance, -James C. Lewis _ /_ _ _ _ / ) / _ ' _ (_/(///)(-_) (__ . (__(-((//_) PUCC Information Center Consultant jclewis@purdue.edu "Only those who dare to fail greatly can ever achieve greatly." -- Robert F. Kennedy Please place all complaints in this box --> [] From cnd@ecn.purdue.edu Sun, 10 Sep 2000 10:09:30 -0500 Date: Sun, 10 Sep 2000 10:09:30 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] X-Box - MS Technical Presentation I thought that you all might be interested in this. Console gaming is definitely becoming bleeding edge when it comes to graphics. It also seems as though these console machines are more and more like real computers each generation. -Chris PLUG President >From the CS dept: Microsoft Corp. is entering the video game world with the introduction of a future-generation dedicated video game console, currently code-named X-Box. We invite you to attend this technical discussion and demo of this product. Date: 9/12/00 Time: 6:30P-8:30P Location: Loeb Playhouse Food: Pizza...first come, first serve. When it's gone it's gone. X-Box is designed to deliver intense, action-packed games, which will deliver the most realistic, videogame experiences available. It will deliver the richest, most compelling graphics and will play better and faster than any other games console on the market. X-Box will be a superior game console that enables gamers to play videogames alone or side-by-side on their TV screen. X-Box will also be able to play DVD movies and is wired for broadband, which will allow gamers to play fast-action. * Delivers more than 300+million polygons/second * 600MHz AMD CPU * 64MB RAM (unified memory architecture) * 8GB Hard Drive * 4X DVD Drive * 4 Game Controller Ports * 100 Mbps Ethernet X-Box, the next level! Jill Albrecht Development Assistant Computer Sciences Purdue University 765-496-3525 -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From heckarsd@purdue.edu Sun, 10 Sep 2000 14:05:16 -0500 Date: Sun, 10 Sep 2000 14:05:16 -0500 From: Seth Heckard heckarsd@purdue.edu Subject: [PLUG] Cue::Cat "James C. Lewis" wrote: > > Hey... > > I went to radio shack today and picked up one of those cue::cat thingys. > > Since I don't have windows... anyone have or know where I can get the > drivers for gnome/linux. > > I kind of like these and I really like cmdr taco's idea about indexing > DVD's or somethign with them. > > BTW, if you say where to get it... i promise not to tell digital > convergance ;) hehehehe > > Thanks in advance, > > -James C. Lewis Check out http://freshmeat.net/search/?q=cuecat. Most of the results are just little perl scripts that decode the data and do nothing more. I think there was a fancy gtk program, but it was the first to go offline with the whole DC brouhaha. So, if you want to do anything fancy with it, you're probably going to have to write it yourself. My problem is that I have one, but have no idea what to do with it! I have more computers in my room than DVDs :-) Seth From doobie@doobie.org Sun, 10 Sep 2000 18:24:14 -0500 (EST) Date: Sun, 10 Sep 2000 18:24:14 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] uncompressing sea? Is there a unix tool to uncompress sea archives? From wirges@expert.cc.purdue.edu Sun, 10 Sep 2000 21:45:09 -0500 Date: Sun, 10 Sep 2000 21:45:09 -0500 From: matt wirges@expert.cc.purdue.edu Subject: [PLUG] uncompressing sea? --------------2CA28D1F43CA6D7938D231E0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I believe that .sea's are Self Extracting Archives and they can only be executed on a mac. However, alladin systems has a version of Expander (avail at www.alladinsys.com) for linux. If you can't extract it with that, you can't extract it unless you are using a mac. -Matt -- ============================================================================== | Matthew Wirges | | Systems Administrator, Student, Programmer, Geek. | | Office Phone: [765]464-1148 | | Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com | | www.sharpwebinnovations.com | ============================================================================== Jason Dubrow wrote: > Is there a unix tool to uncompress sea archives? > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug --------------2CA28D1F43CA6D7938D231E0 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit I believe that .sea's are Self Extracting Archives and they can only be executed on a mac.
However, alladin systems has a version of Expander (avail at www.alladinsys.com) for linux.
If you can't extract it with that, you can't extract it unless you are using a mac.

-Matt 

-- 
==============================================================================
| Matthew Wirges                                                             |
| Systems Administrator, Student, Programmer, Geek.                          |
| Office Phone: [765]464-1148                                                |
| Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com     |
| www.sharpwebinnovations.com                                                |
==============================================================================


Jason Dubrow wrote:

Is there a unix tool to uncompress sea archives?

____________________________________________________
The Purdue Linux Users' Group (PLUG) mailing list.
plug mailing list  -  plug@csociety.purdue.edu
http://csociety.ecn.purdue.edu/mailman/listinfo/plug


  --------------2CA28D1F43CA6D7938D231E0-- From wirges@expert.cc.purdue.edu Sun, 10 Sep 2000 23:12:43 -0500 Date: Sun, 10 Sep 2000 23:12:43 -0500 From: matt wirges@expert.cc.purdue.edu Subject: [PLUG] uncompressing sea? --------------8868D8D81C258B6358386C6C Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I believe that .sea's are Self Extracting Archives and they can only be executed on a mac. However, alladin systems has a version of Expander (avail at www.alladinsys.com) for linux. If you can't extract it with that, you can't extract it unless you are using a mac. -Matt -- ============================================================================== | Matthew Wirges | | Systems Administrator, Student, Programmer, Geek. | | Office Phone: [765]464-1148 | | Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com | | www.sharpwebinnovations.com | ============================================================================== --------------8868D8D81C258B6358386C6C Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit  
I believe that .sea's are Self Extracting Archives and they can only be executed on a
mac.
However, alladin systems has a version of Expander (avail at www.alladinsys.com) for
linux.
If you can't extract it with that, you can't extract it unless you are using a mac.

-Matt 

-- 
==============================================================================
| Matthew Wirges                                                             |
| Systems Administrator, Student, Programmer, Geek.                          |
| Office Phone: [765]464-1148                                                |
| Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com     |
| www.sharpwebinnovations.com                                                |
==============================================================================
  --------------8868D8D81C258B6358386C6C-- From cnd@ecn.purdue.edu Mon, 11 Sep 2000 00:04:46 -0500 Date: Mon, 11 Sep 2000 00:04:46 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] CALLOUT - Monday Night Almost too late is better than never right? PLUG (the Purdue Linux Users Group) will be having its fall semester callout Monday, September 11, in Physics 112 from 6:30 - 9:30 pm. You do not have to come for the whole time, but you are welcome to hang out with us for as long as you can/want. I am going to give a brief presentation about PLUG, some events we will be having this semester, and talking about being a member in PLUG. I will also be giving out prizes. Prizes include: -books -posters -distro media -Star Office -some O'Reilly cool stuff -VA Linux T-shirts -A full Deluxe Boxed Version of Corel Linux (with Civilization Call to Power and a penguin) Be prepared for some whacky ways of getting prizes. As some of the winners may not be around when we draw names and count scores and stuff, prizes will be given out at the next PLUG meeting. On display will be some books from our library and some other stuff that I haven't quite figured out yet. We are also going to have some members (or proxies to those members as they have class) talk about PLUG Special Interest Groups that they are in charge of. We may also have some cool hardware displays. I think Andy Sydelko from Computer Society (and PLUG, he's just famous over there...) will demo his diskless dual Xeon machine. As we've had a recent EXPLOSION in mailing list activity, maybe I can get some of you security interested folks to talk about security issues when running Linux/Unix on Resnet, @Home, or in some other environment. We will also have some punch and pie available. I think you all know the definition of that one... PLUG is going to be cool this year. Some of you were probably jealous of what we did this summer. DO NOT FEAR, I will do my best to have things like that this fall and coming spring. Some notes on membership information. Please see our Membership info at http://www.ecn.purdue.edu/PLUG/Membership/. Dues are $10 for the year. I will also have computer account applications available for those who are paying dues and want a PLUG computer account. That's about it. Bring your roomate(s) with you Monday night. Also, the CS dept has decided to hold a job fair on the same night as our callout and ruin our fun (just kidding, just kidding.) If you are planning on going to the callout, you can come see us afterwards as we will be there until 9:30ish. See you all there!!! -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From cnd@ecn.purdue.edu Mon, 11 Sep 2000 09:51:34 -0500 Date: Mon, 11 Sep 2000 09:51:34 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Re: Linux for a newbie > I have a 20GB HDD with a single NTFS partition with Windows 2000 pro > installed. I plan to resize the partition to 10GB using partitionmagic and > I want to install Linux on the other half. > > Should I use a primary or extended partition for the Linux file system? > Will LILO boot Win2000? And if not, will the NT boot menu boot Linux? Primary is fine. As I hate dealing with paritions, I always use primaries. Make sure that you create a swap partition too. > Will Linux work with my USB keyboard and mouse? (if not, it's a laptop so I > can use the built-ins). Mandrake 7.1 has USB support. I think RedHat 6.2 does, but I don't use RedHat so don't take my word on that. Suse 6.4 may. > Will my combo NIC/modem work? I know that the NIC part works in most > Linuxes (Linuces?) I heard that the modem will probably not. It's a 3com > 3CCFEM656B. Look at hardware compatibility lists: http://www.linux-mandrake.com/en/fhard.php3 > How would I go about setting up the boot manager? I already know how to > install Linux but I've never been serious about it before. You can use LILO, NTLDR, or GRUB. Grub looks cool, but make sure you pay attention to what's going on. I've been told you can severely hose the system. I'm attaching the NTLDR scheme that Paul talked about since it only went to the newsgroup. LILO is simple and the installer will set everything up for you. I have a dual boot machine and everything works fine. > Finally, what's the best distro for a newbie such as myself? I asked in > alt.os.linux and heard Mandrake 7.1 and Redhat 6.2. You need one that supports USB. Most current distros should, but look at them closely. I use Mandrake. It's package compatible with RedHat and it seems to run faster (compiled with pentium optimizations). > Oh yeah, one other thing. And I hope this doesn't start a flame war. What > is the "best" shell for X Free 86? I'm hearing GNOME and KDE. Remember > that I know almost nothing about this. Flame away. GO GNOME!!! Also note that with Mandrake 7.1 you can choose between XFree 3.whatever and XFree 4. XFree 4 is very nice, but I don't think it's stable enough to use for prime time. -Chris > Will LILO boot Win2000? And if not, will the NT boot menu boot Linux? I won't try to help with the other stuff (please don't ask me to tell you why it took me three days to install Win98, Win2K, and RH6.1 on the same machine), but I *can* tell you how to get the NT Loader to give you an option to boot Linux. First, when you're installing Linux via whatever installer you have, install LILO on the *boot partiton*, NOT the MBR. NT Loader will stay in the MBR. Got that? This is important. Also, make a note of which partition is your /boot partition. OK, that done, after you've installed Linux, if you can, mount your Windows C: drive partition into the filesystem. (Note: I think you need a special driver or something to access NTFS partitions. If you don't have it, I'll tell you the workaround after this....) Now, at the command line, type this: dd if=/dev/hda1 of=/mnt/kefka/linux.dat bs=446 count=1 Replace /dev/hda1 with your /boot partition. Replace /mnt/kefka with wherever your C: drive is mounted. (If you can't mount your NTFS C: drive, change that path to somewhere else in your filesystem, then copy it to a floppy. Then, after booting into Win2K later, copy it to C:) A file called linux.dat will be created in that directory, containing a copy of the first 446 bytes of your boot sector. OK, once you have that file, reboot your system into Windows 2000. Once that's done, open up the C:\boot.ini file. This is where NT Loader gets its menu from. Un-Read-Only the file and open it in Notepad. Add the following line to the bottom of the file: C:\linux.dat="Red Hat Linux 6.2" /win95dos If you decided to name your file something other than linux.dat, change that there. In this example, "Red Hat Linux 6.2" is the text that will appear in the NT Loader menu. Change it to whatever best represents your distro and install. /win95dos stays no matter what. Save the file and reboot. You should see your new option to boot Linux added to the bottom of the NT Loader boot menu. If you select it, LILO will appear, and from there, you can boot into Linux. Since LILO is now relegated to a tiny layer between NT Loader and your Linux boot system, before you do any of this (i.e., before creating your linux.dat file), you might want to change your LILO timeout period to 1 or 2 seconds, and remove anything but "linux" from its menu. After all, you aren't using LILO as your boot menu anyway. Anyway, this paragraph's just a suggestion. Voila! You've just dual-booted Win2K and Linux on your system, and are using NT Loader to boot! Wow...being a newbie myself, I never thought I'd actually be able to answer a support-type question about Linux. Of course, I did only answer one of your questions. But since I installed my OSes from a clean HD, I don't know what'll be all involved in repartitioning your Windows 2000 drive etc. I'll leave that for someone who knows what they're talking about. (BTW: Don't try to install Red Hat 6.1. I speak from experience. There's so many gotchas in the installer it isn't funny. *That's* why it took me three days to get my OSes set up. If you use Red Hat, download 6.2 and use that.) From olenik@purdue.edu Mon, 11 Sep 2000 09:58:34 -0500 Date: Mon, 11 Sep 2000 09:58:34 -0500 From: Hayden Olenik olenik@purdue.edu Subject: [PLUG] Re: Linux for a newbie On Mon, 11 Sep 2000, Christopher N. Deckard wrote: > Date: Mon, 11 Sep 2000 09:51:34 -0500 > To: Eric Rongo , Plug > From: "Christopher N. Deckard" > Sender: plug-admin@csociety.purdue.edu > Subject: [PLUG] Re: Linux for a newbie > > > I have a 20GB HDD with a single NTFS partition with Windows 2000 pro > > installed. I plan to resize the partition to 10GB using partitionmagic > and > > I want to install Linux on the other half. > > > > Should I use a primary or extended partition for the Linux file system? > > Will LILO boot Win2000? And if not, will the NT boot menu boot Linux? > > Primary is fine. As I hate dealing with paritions, I always use > primaries. Make sure that you create a swap partition too. > > > Will Linux work with my USB keyboard and mouse? (if not, it's a laptop > so I > > can use the built-ins). > > Mandrake 7.1 has USB support. I think RedHat 6.2 does, but I don't use > RedHat so don't take my word on that. Suse 6.4 may. RedHat 6.2 has USB support also. As for XFree86 4.0, you might want to see what video card your laptop has. My VAIO has a Neomagic 256AV+, which requires tweaking under XFree86 3.3.6, because they only have a server up to the 256AV. XFree86 4.0.1 has a server for the Neomagic 256AV+. Hayden From will@physics.purdue.edu Mon, 11 Sep 2000 10:34:51 -0500 Date: Mon, 11 Sep 2000 10:34:51 -0500 From: Will Andrews will@physics.purdue.edu Subject: [PLUG] Re: Linux for a newbie On Mon, Sep 11, 2000 at 09:51:34AM -0500, Christopher N. Deckard wrote: > > Oh yeah, one other thing. And I hope this doesn't start a flame war. What > > is the "best" shell for X Free 86? I'm hearing GNOME and KDE. Remember > > that I know almost nothing about this. > > Flame away. GO GNOME!!! Bah. GNOME is a project led by people who are "free as in speech" fanatics. Who cares if QT isn't under the GPL (well, it is now)? It still kicks GTK's ass. ;-> Hmm... /me goes to find that code to get data from an FTP site in only two or three lines of code. Does GNOME have a similar API? ;) -- Will Andrews GCS/E/S @d- s+:+ a--- C++ UB++++$ P+ L- E--- W+ N-- !o ?K w--- O- M+ V- PS+ PE++ Y+ PGP+>+++ t++ 5 X+ R+ tv+ b++ DI+++ D+ G++ e>++++ h! r- y? From ab@eas.purdue.edu Mon, 11 Sep 2000 10:50:23 -0500 Date: Mon, 11 Sep 2000 10:50:23 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] uncompressing sea? In message , Jason Dubr ow writes: > > Is there a unix tool to uncompress sea archives? Uh yeah, you just doubleclick on them. Wait, I'm not on my NeXT... I think one of unmac's friends will do it. I haven't been doing that lately, but I used to import a lot of PostScript fonts from Macland, and I'm pretty sure my NeXT did it. I'll check when I get it on the DSL. (I doubt I have any on its hard drive.) ab From ab@eas.purdue.edu Mon, 11 Sep 2000 11:09:44 -0500 Date: Mon, 11 Sep 2000 11:09:44 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] Re: Linux for a newbie [ It may be unthinkable, because I use UNIX pretty much exclusively. I don't use X. Really. I'll put more in the UNIX flavors thread when I get back to it. -ab ] > Oh yeah, one other thing. And I hope this doesn't start a flame war. What > is the "best" shell for X Free 86? I'm hearing GNOME and KDE. Remember > that I know almost nothing about this. I now have some hope for the GNUstep/Window Maker/Xws.app combination, but it's really not ready for primetime yet. Depends what you mean by "best". My definition of best includes "easiest to write applications that comply to a consistent interface". That's why I'm still using NeXTs most places. (The only place I use X is at home on my FreeBSD machine, and I'm GNUstepping that as fast as I can.) GNUstep might get us there if development keeps picking up. I've had no fun at all trying to use the older X toolkits, but X and I parted ways about ten years ago. How is programming under this newer stuff like GNOME or KDE? Has anyone written any non-trivial applications? Has anyone written anything more complicated than then stuff I do in my OPENSTEP demos? But you're not writing apps so who cares how hard it is? The people writing the apps you use, dummy. Unless it's easy (OPENSTEP has a disproportionate number of classy apps for its userbase) or profitable (MS Windows), you won't get anything of note. You certainly won't get anything that integrates well with other programs, which undercuts the whole UNIX thing, as well as the program-to-program consistency that users expect (and, really, are entitled to). Even if you've got a reasonable interface, if it's not easier to write conforming programs than non-conforming ones, people won't, you won't get a consistent interface, and users are better off using MacOS or maybe even MS Windows if your interface is really yucky. ab From mpribble@ori.net Mon, 11 Sep 2000 12:42:47 -0500 Date: Mon, 11 Sep 2000 12:42:47 -0500 From: Marv Pribble mpribble@ori.net Subject: [PLUG] Computer Case This is a multi-part message in MIME format. ------=_NextPart_000_005A_01C01BED.C9AAFFC0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I have a small home network with 2 windows PC's (for work) and a Linux = box for fun and doing real computer stuff. I have been thinking of = putting together another box to run a firewall. However, with 3 PC's = in this room, the last thing I need is more heat, noise, and electrical = use. Plus, I want to experiment with a small, floppy-only distro for = just firewalls. For the firewall, I was thinking I can probably find a motherboard with = a network connection builtin. Then, if I had a floppy for it, I could = run one of the small, firewall-only versions of Linux with no harddrive, = no monitor, no keyboard, no mouse, etc. I have an old case, but it = would be much cooler (both heat-wise and tech-wise) to find a small = case, no fan(?), and a small power supply. With the floppy controller = built-in and the ethernet connection built-in, I actually would not even = need any ISA or PCI slots. So it really would not even have to be thick = enough to hold those cards. It would take up less space, less = electricity, and most of all, generate less heat. Do you know of any place where I could get such a case and power-supply? = I tried the search engiines, but I am having no luck narrowing down the = choices and finding what I want. One other thought I just had was to find an old laptop. That would not = be as neat, but it seems like it might work also. TIA, Marv ------=_NextPart_000_005A_01C01BED.C9AAFFC0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
I have a small home network with 2 windows PC's = (for work)=20 and a Linux box for fun and doing real computer stuff.  I have been = thinking of putting together another box to run a firewall.  = However, =20 with 3 PC's in this room, the last thing I need is more heat, noise, and = electrical use.  Plus, I want to experiment with a small, = floppy-only=20 distro for just firewalls.
 
For the firewall, I was thinking I can probably = find a=20 motherboard with a network connection builtin.  Then, if I had a = floppy for=20 it, I could run one of the small, firewall-only versions of Linux with = no=20 harddrive, no monitor, no keyboard, no mouse, etc.  I have an old = case, but=20 it would be much cooler (both heat-wise and tech-wise) to find a small = case, no=20 fan(?), and a small power supply.  With the floppy controller = built-in and=20 the ethernet connection built-in, I actually would not even need any ISA = or PCI=20 slots.  So it really would not even have to be thick enough to hold = those=20 cards.  It would take up less space, less electricity, and most of = all,=20 generate less heat.
 
Do you know of any place where I could get such = a case and=20 power-supply?  I tried the search engiines, but I am having no luck = narrowing down the choices and finding what I want.
 
One other thought I just had was to find an old=20 laptop.  That would not be as neat, but it seems like it might work = also.
 
TIA,
Marv
------=_NextPart_000_005A_01C01BED.C9AAFFC0-- From wallacms@cs.purdue.edu Mon, 11 Sep 2000 13:26:46 -0500 (EST) Date: Mon, 11 Sep 2000 13:26:46 -0500 (EST) From: Matt Wallace wallacms@cs.purdue.edu Subject: [PLUG] Re: Linux for a newbie On Mon, 11 Sep 2000, Will Andrews wrote: > On Mon, Sep 11, 2000 at 09:51:34AM -0500, Christopher N. Deckard wrote: > > > Oh yeah, one other thing. And I hope this doesn't start a flame war. What > > > is the "best" shell for X Free 86? I'm hearing GNOME and KDE. Remember > > > that I know almost nothing about this. > > > > Flame away. GO GNOME!!! I'll spare the flame, but I have to say for a newbie, I think KDE is a much easier system to use. Now I haven't tried GNOME in a year or so, but I imagine the philosphy and therefore the implementation remains essentially the same. First, KDE come default with a much more Windows-like environment. I'm not saying this is good or bad, just easier for a newbie. Second, KDE is a much self-contained thing in my opinion. When I used GNOME it seemed to be a bunch of applications that were loosely connected by a framework. KDE has a much more integrated feel. Also, I personally think that kwm is a much better window manager than WindowMaker (is that still the preferred GNOME window maneger? All the NeXT-ish wm's run together for me..) And finally I think that KOffice will be much better than anything the GNOME is still patching together (If only becuase it has a significant head start). Matt From wallacms@cs.purdue.edu Mon, 11 Sep 2000 13:30:49 -0500 (EST) Date: Mon, 11 Sep 2000 13:30:49 -0500 (EST) From: Matt Wallace wallacms@cs.purdue.edu Subject: [PLUG] Re: Linux for a newbie One more thing to add. It is very easy (and desireable, IMO) to run both GNOME and KDE on the same box. Well, that is, you can run both Qt and GTK+ apps on the same box, I'm fairly sure you can only get the integration/desktop environment stuff of one or the other. So if it's application support that your worried about, don't be. Matt From honeycug@purdue.edu Mon, 11 Sep 2000 14:28:29 -0500 (EST) Date: Mon, 11 Sep 2000 14:28:29 -0500 (EST) From: garrett honeycug@purdue.edu Subject: [PLUG] NeXT software Does anyone know if there are any firewall / proxy software available for NeXT's? --garrett From will@physics.purdue.edu Mon, 11 Sep 2000 14:36:34 -0500 Date: Mon, 11 Sep 2000 14:36:34 -0500 From: Will Andrews will@physics.purdue.edu Subject: [PLUG] Re: Linux for a newbie On Mon, Sep 11, 2000 at 11:09:44AM -0500, A Braunsdorf wrote: > [ It may be unthinkable, because I use UNIX pretty much exclusively. > I don't use X. Really. I'll put more in the UNIX flavors thread > when I get back to it. -ab ] There's nothing wrong with not using X. There are times when I prefer the console to X, and I could live without X. > I now have some hope for the GNUstep/Window Maker/Xws.app combination, > but it's really not ready for primetime yet. I use Window Maker personally, although I've looked to KDE more lately because of its powerful APIs and its growing quality in terms of applications (which can be important, for example, if i want to use a word processor or a spreadsheet for class and *DON'T* want to learn LaTeX or anything like that ;). > as fast as I can.) GNUstep might get us there if development keeps It's still rather slow compared to KDE/GNOME. :-( > and I parted ways about ten years ago. How is programming under > this newer stuff like GNOME or KDE? Has anyone written any I haven't written any non-trivial apps in KDE yet, but I'm thinking about doing it once the openpackages/ports project reaches its goals. FreeBSD ports are just becoming a cliche for me these days.. I've heard of people writing network code in just a couple lines (ftp, http, whatever).. it sure beats having to understand sockets. -- Will Andrews GCS/E/S @d- s+:+ a--- C++ UB++++$ P+ L- E--- W+ N-- !o ?K w--- O- M+ V- PS+ PE++ Y+ PGP+>+++ t++ 5 X+ R+ tv+ b++ DI+++ D+ G++ e>++++ h! r- y? From cnd@ecn.purdue.edu Mon, 11 Sep 2000 15:23:48 -0500 Date: Mon, 11 Sep 2000 15:23:48 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Re: Linux for a newbie > I'll spare the flame, but I have to say for a newbie, I think KDE is a > much easier system to use. Easier to use probably. I like Gnome because of the ability to configure its look however you want. I like little panel apps to tell me system status and I like to have other weird things (Wanda the Fish sometimes). Gnome offers all these and they all integrate into the panel very nicely. KDE seems to lack that, but it is my understanding that all of that will change when the new KDE (2.0?) is released. Who knows... Maybe when KDE 2 comes out I'll switch. I've changed WMs yearly it seems. > Also, I personally think that kwm is a much better window manager than > WindowMaker (is that still the preferred GNOME window maneger? Sawfish (was sawmill) is the current distributed WM, but Enlightenment KWM, WindowMaker and others work with Gnome. -Chris From marcom@purdue.edu Mon, 11 Sep 2000 17:18:30 -0500 Date: Mon, 11 Sep 2000 17:18:30 -0500 From: Marcom, Greg marcom@purdue.edu Subject: [PLUG] FTP server I am trying to FTP Mandrake 7.1 from the csociety server. For the server I am putting csociety-ftp.ecn.purdue.edu For the path I am putting /pub/mandrake/7.1/Mandrake/. It keeps giving me a servre error. Greg From heckarsd@purdue.edu Mon, 11 Sep 2000 17:16:37 -0500 Date: Mon, 11 Sep 2000 17:16:37 -0500 From: Seth Heckard heckarsd@purdue.edu Subject: [PLUG] NeXT software garrett wrote: > > Does anyone know if there are any firewall / proxy software available for > NeXT's? > > --garrett as for firewall, i don't believe it's possible because you can only have one ethernet interface. although the next cube has expansion slots, there was never an ethernet card manufactured. i don't even think you can do packet forwarding due to some limitation on the bsd it's based on. seth From ab@eas.purdue.edu Mon, 11 Sep 2000 17:22:05 -0500 Date: Mon, 11 Sep 2000 17:22:05 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] NeXT software In message <39BD59C5.53A32294@purdue.edu>, Seth Heckard writes: > > as for firewall, i don't believe it's possible because you can only have > one ethernet interface. although the next cube has expansion slots, > there was never an ethernet card manufactured. i don't even think you > can do packet forwarding due to some limitation on the bsd it's based > on. They (ISC, I think, someone who does DHCP) says bpf is available. You can load kernel modules, so if you're not afraid to do that on a kernel you don't have sources to, you could do most anything. There aren't any NeXTs with multiple ethernet adaptors, but if you had a SCSI one (do they make such a thing?), you might be able to get it working. That'd be a big "why?", though. :-) The cube has slots (I wouldn't call them expansion- they're all the same as the one the motherboard's in, I think), but only a couple of cards were commercially produced (NeXTdimension, the >video< card being the most famous), and a couple of others were built for internal use (like the Barco adaptor in the loaner I used to have). ab From will@physics.purdue.edu Mon, 11 Sep 2000 19:22:25 -0500 Date: Mon, 11 Sep 2000 19:22:25 -0500 From: Will Andrews will@physics.purdue.edu Subject: [PLUG] Re: Linux for a newbie On Mon, Sep 11, 2000 at 03:23:48PM -0500, Christopher N. Deckard wrote: > Easier to use probably. I like Gnome because of the ability to > configure its look however you want. I like little panel apps to tell > me system status and I like to have other weird things (Wanda the Fish > sometimes). Gnome offers all these and they all integrate into the > panel very nicely. KDE seems to lack that, but it is my understanding > that all of that will change when the new KDE (2.0?) is released. Yes, KDE2 offers these things. > Who knows... Maybe when KDE 2 comes out I'll switch. I've changed WMs > yearly it seems. That happened to me too until I decided Window Maker is the WM of gods. :P Seriously though, I don't consider GNOME or KDE "window managers", they are more like full-blown "environments". IMO, the result is a window manager (these "environments" come with "window managers" which interact with X, etc.) that is more bloated and can't get rid of extra baggage when you need to get the job done. So I set my machine up to use Window Maker about two years ago and use GNOME/KDE apps as slaves, depending on whether I needed a word processor or a game or something like that, and that way I can fall back to the simpler, less bloated interface to do REAL work (i.e. system administration and such). However, for user interfaces, nothing beats GNOME or KDE. CDE, OpenWindows, and all those other lame environments should just die. Die die die. Yes, including the ones on *.ecn.purdue.edu Sparc machines. ;-) -- Will Andrews GCS/E/S @d- s+:+ a--- C++ UB++++$ P+ L- E--- W+ N-- !o ?K w--- O- M+ V- PS+ PE++ Y+ PGP+>+++ t++ 5 X+ R+ tv+ b++ DI+++ D+ G++ e>++++ h! r- y? From ab@eas.purdue.edu Tue, 12 Sep 2000 01:45:46 -0500 Date: Tue, 12 Sep 2000 01:45:46 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] That :Cue:C.A.T. Thing Wired subscribers get a :Cue:C.A.T. with this month's issue. ab From ab@eas.purdue.edu Tue, 12 Sep 2000 09:27:28 -0500 Date: Tue, 12 Sep 2000 09:27:28 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] Re: Linux for a newbie In message <20000911143634.P1309@radon.gryphonsoft.com>, Will Andrews writes: > > There's nothing wrong with not using X. There are times when I prefer > the console to X, and I could live without X. I log in from a NeXT (or my Mac which is running MacOS X Server- same thing but uglier) to do most of my work. My work is systems administration, and X-based tools don't do me much good there. I process my email with mh. I do my graphics work in PostScript on my NeXT, which is the right place to do it. X just hasn't made my life better, and my NeXTs have immensely. In the interest of improving things for my X users here, and to help out with the GNUstep project (now that I believe it's actually getting somewhere), I am about to put an X-based system on my desk here at work. I'm keeping my Mac too, just in case, but I'm taking the plunge. I hope it doesn't make me go postal or anything... > I use Window Maker personally, although I've looked to KDE more lately > because of its powerful APIs and its growing quality in terms of > applications (which can be important, for example, if i want to use a > word processor or a spreadsheet for class and *DON'T* want to learn > LaTeX or anything like that ;). > > > as fast as I can.) GNUstep might get us there if development keeps > > It's still rather slow compared to KDE/GNOME. :-( You're missing the point. GNUstep is being built from the ground up in OPENSTEP's image. One of these days (and it looks like they're close now), it'll hit critical mass and all us NeXTpeople can get back to work. There'll be more cool applications than you can imagine so fast your head'll spin. And they'll run on MacOS X and MS Windows too. Have you seen the applications available on a NeXT from Lighthouse Design and some of those guys? Do you realize that with minor recoding (they weren't all OPENSTEP compliant and GNUstep might not be either) and a recompile you could run them under GNUstep if it worked? Now imagine if comparable open source projects were rolling. The future is cool. Too bad it took ten years too long to get here. :-) OPENSTEP is a well thought out and engineered system. It's not just a style of buttons and scrollbars to dress up crappy old programs. It's more than you think. ("It's a religion?" I wouldn't go that far, but it's at least a lifestyle. :-) ) > I haven't written any non-trivial apps in KDE yet, but I'm thinking > about doing it once the openpackages/ports project reaches its goals. > FreeBSD ports are just becoming a cliche for me these days.. > > I've heard of people writing network code in just a couple lines (ftp, > http, whatever).. it sure beats having to understand sockets. So what, I can do that stuff in one line of sh too. Doesn't mean I'm going to save the world. (And you probably should understand sockets too if you want to be an educated dude.) Yawn. We've got lots of networking code that works. How about stuff users care about? Can you write something like MS Word without losing your mind? Can the components you write be reused in other programs so no one ever has to write a whole text editor again? ab From pratte@ecn.purdue.edu Tue, 12 Sep 2000 10:02:48 -0500 (EST) Date: Tue, 12 Sep 2000 10:02:48 -0500 (EST) From: Eric Pratt pratte@ecn.purdue.edu Subject: [PLUG] syslogd failing On booting my Linux machine (redhat 6.2) syslogd fails. I read some HOW-TOs and whatever I can find on syslogd, but I cannot find enough info to troubleshoot it. /etc/services shows UDP port 514 assigned to syslog. Wierd thing is, even though it shows as failing on boot, it appears to continue to write messages to /var/log/messages. I guess I have to say that on boot, I see that system logging services fails and I assumed this is syslogd. I see in /var/log/messages that syslogd failed. It is the first thing that gets put into the log after a reboot. Here is the line that appears in the log: Sep 11 19:39:47 eightbits syslog: syslogd startup failed Anyone have any ideas on what exactly is failing or if this is just an erroneous mesage? Thanks. Eric From cnd@ecn.purdue.edu Tue, 12 Sep 2000 12:49:07 -0500 Date: Tue, 12 Sep 2000 12:49:07 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Free Software Company What do you all think of this? -Chris Begin forwarded message: From: Radhakrishnan C V To: pdftex@tug.org Subject: Free Software Company Message-ID: MIME-Version: 1.0 Sender: owner-pdftex@tug.org Precedence: bulk Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 5984 What I am giving below is off topic but relevant to everyone associated with free software. I hope most of you might have heard about Richard M. Stallman (RMS) and the Free Software Foundation (FSF), especially those aligned with TeX are no doubt allies of the free software movement ignited by RMS. An intense discussion is going on at freedevelopers@topica.com for the formation of a Free Software Company entirely owned by the free developers world wide (100,000 developers expected) to safeguard the moral, social and ethical foundations on which FSF was built up, to provide appropriate monetary compensation and quality life to a free software developer he deserves in all the fairness and to fight the merciless commerce of the proprietary software corporates. These being the primary objectives, Tony Stanco (an associate of RMS, who moderates the list) describe it in the following way: 1. Proprietary code is the enemy. It must be destroyed for developers and the world to be free. Open source is an ally. 2. Developers can be paid salaries and/or stock options to work on free code without violating the core principles of free code. 3. Mergers and acquisitions of proprietary companies are not objectionable in defeating proprietary. 4. A company of free developers, by free developers, for free developers is an acceptable vehicle to achieve the ends of free code. 5. A requirement in the certificate of incorporation that all code owned by the company is licensed under GPL or other tying to FSF is appropriate to ensure that the core principles of free software are observed going forward and to protect from slipping back to proprietary. 6. A democratic, free developer run corporation does not require special safeguards to protect ordinary world citizens. As one of the developing nations, in India or in any developing nation, where automation has just started, Free Software Company and FSF have plenty of implications. 1. In the first place, our poor economy cant afford to the fancy prices of proprietary software (maybe due to the foriegn exchange conversion magic). 2. Free Software can meet any objective, functionality that is claimed by the proprietary. 3. The huge man power resources generated in each and every place of higher academic learning in this country get a chance to contribute to the free software movement, while he earns a substantially increased income comparable to his counterpart in any proprietary corporate, as an employee of the proposed company. The discussions tend to provide the standard wages in India as in any part of the world. That will surely solve the disturbing problems of migration in many an Indian family. 4. There are plenty of requirement for software in this country for meeting its target of total automation for which each and every government or other agencies stand for today. Free software can meet their objectives on sound moral, ethical and social foundations than any other proprietary corporate. A Case Study: - ------------ The Govt. of Kerala has formed an IT mission to automate the 1000 and odd Gram Panchayats (the lowest unit of elected body at village level) of this state. It is a massive and aggressive project to bring details of all the citizens of this state into a huge database, each Panchayat becoming a resource center for the government and at the same time act as the information exchange medium between the public and the government. This is an ideal project for the free software movement. The government have earmarked around Rs. 800,000 per Panchayat for using proprietary software and related development as initial investment and Rs. 150,000 as annual recurring expenses. While this came as a proposal, the Linux Users and TeX Users Groups came forward to negotiate with the government to do the project at a cost of Rs. 150,000 per Panchyat as initial expenses and rs. 20,000 as recurring expenses. But we were turned down, just because, we were considered to be a group of free thinkers, whom no responsible government can rely upon. Had there been a corporate entity with FSF objectives to compete with the proprietary agencies, the public exchecquer would have saved millions of rupees, the developers in this or neighbouring states would have got employment. Still the project is not finalized due to the media stir we raised and the seminars of users groups wherein the government nominees were special invitees. To make matters difficult for the government, the Cochin Linux Users Group came up with a viable, stable, functional software model which the political bureaucracy cannot overlook or deny. The final decision was postponed and still lingering. That means we have not lost the race, the global Free Software Company can still fight it out once it is formally incorporated. Tens of thousands of projects are in the offing which we all can undertake. This is the case in many other countries world wide. There are a quite a number of posts in the freedevelopers list similar to this from West European countries. Therefore, if you believe in the non-exploitative future of your country and the world at large and if you believe in the objectives of free software movement, you are always welcome to make the Free Software Company a reality. It requires the developer in you, not for free, but for a standard compensation as in any other proprietary corporate. Why don't you subscribe to the discussions by sending a mail to: freedevelopers-subscribe@topica.com with a word `subscribe' in the subject line? You are not alone, but you are going to meet/associate with some of the greatest minds of this era in this movement. And remember, Free Software Company is going to be the greatest event in the history of Internet and every country should have its fare representation and share. Comments are welcome. - -- Radhakrishnan ------- End of Forwarded Message From dave@sr71.net Tue, 12 Sep 2000 13:03:17 -0500 Date: Tue, 12 Sep 2000 13:03:17 -0500 From: David C. Hansen dave@sr71.net Subject: [PLUG] Local Mirrors Besides the csociety-ftp.ecn mirror, what local mirrors for Linux stuff do we have? I am trying to get an ISO of one of the PPC distros and it is taking forever. -- David C. Hansen dave@sr71.net From rajak@purdue.edu Tue, 12 Sep 2000 13:05:12 -0500 (EST) Date: Tue, 12 Sep 2000 13:05:12 -0500 (EST) From: Brian Poole rajak@purdue.edu Subject: [PLUG] Local Mirrors We have a bit of stuff mirrored at CERIAS, but no PPC stuff, sorry. Currently have full mirrors of OpenBSD, slackware, and debian here at ftp.cerias.purdue.edu. A full mirror of FreeBSD is in the works but is waiting until I get more RAM for the FTP server. -b On Tue, 12 Sep 2000, David C. Hansen wrote: > Date: Tue, 12 Sep 2000 13:03:17 -0500 > From: David C. Hansen > To: plug@csociety.purdue.edu > Subject: [PLUG] Local Mirrors > > Besides the csociety-ftp.ecn mirror, what local mirrors for Linux stuff do we have? > > I am trying to get an ISO of one of the PPC distros and it is taking forever. > > -- > David C. Hansen > dave@sr71.net > > > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From rrussel5@purdue.edu Tue, 12 Sep 2000 13:29:00 -0500 Date: Tue, 12 Sep 2000 13:29:00 -0500 From: Alex Russell rrussel5@purdue.edu Subject: [PLUG] Free Software Company This is a multi-part message in MIME format. ------=_NextPart_000_0002_01C01CBD.69069690 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 "1. Proprietary code is the enemy. It must be destroyed for developers and the world to be free. Open source is an ally." PHULEEZE!!!! Me, I'm as much of an advocate of free (as in speech) software as any other nerd on this mailing list, but let's all be rational about this. The idea that we are locked in a mortal struggle with the evil forces of the world is a little silly. Is greed evil? Yes, but then so is vanity, and isn't that part of why we do free software? ...hmm... People need food an clothes a heck of a lot more than they need the GPL attached to their software. I am all for working for any company that will pay me to do free software development, and in fact free software is changing how and why development is done _everywhere_. However, the spirit of free software, to me, does not include vindictive needs to "show the world what we can do" because we are being oppressed. I do it for the love of what I'm doing, beyond that, I'm not sure I want any part of a jihad. The battle for open minds and open software will be waged company-by-company, and if it's the right tool, I have no doubt that it will win out over other solutions. And indeed I hope that it will. But until such time as it is a panacea for everything, I am not going to label it as such. Just my $0.02. Alex -----BEGIN PGP SIGNATURE----- Version: PGPfreeware 6.5.8 for non-commercial use iQA/AwUBOb517J9fSMV9QwcyEQKGJgCg71rZ6iZ+t74H1ehXzf5i73ZCeCMAn3zR 9TWDYGkUsf2uuu/VgK6FFdlX =cxEE -----END PGP SIGNATURE----- ------=_NextPart_000_0002_01C01CBD.69069690 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable ------=_NextPart_000_0002_01C01CBD.69069690-- From jgennis@mindspring.com Tue, 12 Sep 2000 11:58:03 GMT Date: Tue, 12 Sep 2000 11:58:03 GMT From: Jamie Gennis jgennis@mindspring.com Subject: [PLUG] Diskless booting information Where can I get more info on the diskless booting project that was demonstrated at the callout yesterday? -jamie From rjune@ims1.imagestream-is.com Tue, 12 Sep 2000 23:27:18 -0500 (EST) Date: Tue, 12 Sep 2000 23:27:18 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] Diskless booting information I wasn't there, this is the etherboot homepage, it may be what you want. http://etherboot.sourceforge.net/ On Tue, 12 Sep 2000, Jamie Gennis wrote: > > Where can I get more info on the diskless booting project that was demonstrated > at the callout yesterday? > > -jamie > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From gennis@purdue.edu Wed, 13 Sep 2000 07:33:51 -0500 Date: Wed, 13 Sep 2000 07:33:51 -0500 From: Jamie Moline Gennis gennis@purdue.edu Subject: [PLUG] Diskless booting information Actually, there is a project going on here at Purdue where they are trying to set up a bunch of machines to do diskless booting & are going to create a cluster out of them... That's the project I was looking for. Thanks though :) Jamie Gennis Richard June wrote: > > I wasn't there, this is the etherboot homepage, it may be what you want. > http://etherboot.sourceforge.net/ > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From will@physics.purdue.edu Wed, 13 Sep 2000 22:29:39 -0500 Date: Wed, 13 Sep 2000 22:29:39 -0500 From: Will Andrews will@physics.purdue.edu Subject: [PLUG] Security docco Here's an interesting article which covers a bunch of common security tips etc. not only for Linux but for any UNIX machine. http://www.rootprompt.org/article.php3?article=922 -- Will Andrews GCS/E/S @d- s+:+ a--- C++ UB++++$ P+ L- E--- W+ N-- !o ?K w--- O- M+ V- PS+ PE++ Y+ PGP+>+++ t++ 5 X+ R+ tv+ b++ DI+++ D+ G++ e>++++ h! r- y? From deepak@purdue.edu Wed, 13 Sep 2000 22:34:33 -0500 (EST) Date: Wed, 13 Sep 2000 22:34:33 -0500 (EST) From: Deepak Dinesh deepak@purdue.edu Subject: [PLUG] Mandrake 7.0 I tried that and it didn't work. I have the locales-en rpm installed if that makes any difference. I am unable to use most gnome programs (GDK ?!). Thanks. On Sat, 9 Sep 2000, William Lee Irwin III wrote: > On Sat, Sep 09, 2000 at 11:43:35PM -0500, Deepak Dinesh wrote: > > Hi, > > After I updated glibc for Mandrake 7.0 I get the following > > error when I compile any programs: > > > > perl: warning: Setting locale failed. > > perl: warning: Please check that your locale settings: > > LANGUAGE = "en", > > LC_ALL = "en", > > LANG = "en" > > are supported and installed on your system. > > > > Also many of my programs fail with "locale setting failed". > > Thanks. > > Try "en_US" instead of "en". > > Cheers, > Bill > -- > "Microsoft shouldn't be broken up. It should be shut down." > -- Security expert Bruce Schneier on the ILOVEYOU virus. > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From deepak@purdue.edu Wed, 13 Sep 2000 22:37:58 -0500 (EST) Date: Wed, 13 Sep 2000 22:37:58 -0500 (EST) From: Deepak Dinesh deepak@purdue.edu Subject: [PLUG] Tripwire Hi, There has been some discussion about security in this list. I think that though tripwire is the only real solution to this problem there is no point running it if the kernel is compromised. Does anyone have some instructions on building a floppy with just the kernel, tripwire program and the required libraries ? Thanks. From pweber1@purdue.edu Wed, 13 Sep 2000 23:40:13 -0500 Date: Wed, 13 Sep 2000 23:40:13 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] security and firewalls i was talking with someone about firewalls and his assertion is that "when you run anything else on a firewall, it ceases to be a firewall". (or something to that effect.) is this literally true? how far does this go? is there any way to run services (for instance, nfs, samba, netatalk or even httpd) and make them available to ONLY the internal network? is it really worth it to have a whole seperate machine to do the firewalling? i have absolutely no security issues on the inside of my network, (i'm the only one using it and i doubt the service worker for this floor is a malicious cracker) so i have no problem running less-than-completely-secure services internally as long as they're not causing any external security problems. i have a number of machines, but they're all running different OS's for different purposes, and i'd rather not have to get another entire machine set up. (i already have 4 in a small dorm room as it is.) thanks, peter -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From meling@myweb.com.my Thu, 14 Sep 2000 12:55:08 +0800 Date: Thu, 14 Sep 2000 12:55:08 +0800 From: meling meling@myweb.com.my Subject: [PLUG] Tripwire > There has been some discussion about security in this list. I > think that though tripwire is the only real solution to this > problem Not entirely true. The job of a security manager does not end by installing Tripwire or some other baseline security software. One needs to keep up with software patches and upgrades (i missed a wu-ftpd security patch couple months ago, and one of my system was instantly rooted), one needs to harden the system before rolling it into production (running the neccessary services, configuring tcpwrappers, running nmap scans, installing ssh, etc), and yup, one needs to have a backup/recovery plan handy, as well as what-to-do-when-compromised-HOWTO. Also benificial is to have an intrusion detection available, redundant logging and active system monitoring (i've developed some handy scripts for active system monitoring). After all this is set up, you have to be proactive, meaning that, once in a while, try to break into your own system. Try crashing it. Another important point to keep in mind if you are a security manager is that you also need to educate your web developers on secure programming... --mel From heckarsd@purdue.edu Wed, 13 Sep 2000 23:57:43 -0500 Date: Wed, 13 Sep 2000 23:57:43 -0500 From: Seth Heckard heckarsd@purdue.edu Subject: [PLUG] security and firewalls peter wrote: > > i was talking with someone about firewalls and his assertion is that > "when you run anything else on a firewall, it ceases to be a > firewall". (or something to that effect.) > > is this literally true? how far does this go? is there any way to run > services (for instance, nfs, samba, netatalk or even httpd) and make > them available to ONLY the internal network? is it really worth it to > have a whole seperate machine to do the firewalling? i have > absolutely no security issues on the inside of my network, (i'm the > only one using it and i doubt the service worker for this floor is a > malicious cracker) so i have no problem running > less-than-completely-secure services internally as long as they're > not causing any external security problems. i have a number of > machines, but they're all running different OS's for different > purposes, and i'd rather not have to get another entire machine set > up. (i already have 4 in a small dorm room as it is.) > > thanks, > peter it is a good idea to run as few services on your firewall box as possible. but, realistically speaking, that isn't going to happen. of course, you could pick up a 486 to do it, but that requires more money and like you said, is just one more system. nfs and samba are real easy to set up to only work on the internal network. just look at the 'hosts allow' line in your smb.conf for samba, and for nfs, only put 192.168. (or 172.16. or 10. if you prefer) hosts in your exports file. and if you're still worried, you can always firewall the smb and nfs ports to the outside. that's probably a good idea anyway, unless you're addicted to your floor's network shares. just keep up to date with the security updates -- that's the most important thing. check your logs too, install a package like logcheck if you prefer. a properly configured linux/*bsd system that is *up to date* should be almost uncrackable. firewalls aren't really an excuse to have insecure systems inside, they're just an extra line of defense on top of the security measures you have inside. seth From deepak@purdue.edu Thu, 14 Sep 2000 00:18:21 -0500 (EST) Date: Thu, 14 Sep 2000 00:18:21 -0500 (EST) From: Deepak Dinesh deepak@purdue.edu Subject: [PLUG] Tripwire On Thu, 14 Sep 2000, meling wrote: > > > There has been some discussion about security in this list. I > > think that though tripwire is the only real solution to this > > problem > > Not entirely true. The job of a security manager does not end by > installing > Tripwire or some other baseline security software. One needs to keep up > with software patches and upgrades (i missed a wu-ftpd security patch > couple > months ago, and one of my system was instantly rooted), one needs to > harden > the system before rolling it into production (running the neccessary > services, > configuring tcpwrappers, running nmap scans, installing ssh, etc), and I have done all of this, but I am sure that crackers are always one step ahead. I would like to shutdown my machine everyday and reboot with the tripwire floppy and be assured that I have not been cracked. > yup, > one needs to have a backup/recovery plan handy, as well as > what-to-do-when-compromised-HOWTO. Also benificial is to have an > intrusion > detection available, redundant logging and active system monitoring > (i've > developed some handy scripts for active system monitoring). After all I guess you mean a separate machine to do the logging. I don't have a home network, so this is not an option, though it would be nice to have a script send an email when an attack is detected. > this is > set up, you have to be proactive, meaning that, once in a while, try to > break > into your own system. Try crashing it. How do you go about doing this ? From vids@mentor.cc.purdue.edu Thu, 14 Sep 2000 01:36:01 -0500 (EST) Date: Thu, 14 Sep 2000 01:36:01 -0500 (EST) From: Vidyut Samanta (aka vids) vids@mentor.cc.purdue.edu Subject: [PLUG] uudecode Anyone know of a windows program to uudecode a mail attachment ? Also is there any other way to send a binary attachment using mailx ? -vids From wirges@expert.cc.purdue.edu Thu, 14 Sep 2000 01:39:22 -0500 Date: Thu, 14 Sep 2000 01:39:22 -0500 From: matt wirges@expert.cc.purdue.edu Subject: [PLUG] Format an ext2 partition --------------0F1D20A217B10BCD3C487619 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I just created a new partition with fdisk called hdb3 when I go to format the disk (mkfs.ext2 -c /dev/hdb3) it says "Device not configured while trying to determing the filesystem size. I also tried giving it the number of blocks, 'mkfs.ext -c /dev/hdb3 3341520', and I got the same error. The device /dev/hdb3 does exist. Any suggestions /solutions? Thanks -- ============================================================================== | Matthew Wirges | | Systems Administrator, Student, Programmer, Geek. | | Office Phone: [765]464-1148 | | Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com | | www.sharpwebinnovations.com | ============================================================================== --------------0F1D20A217B10BCD3C487619 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit I just created a new partition with fdisk called hdb3 when I go to format the disk (mkfs.ext2 -c /dev/hdb3) it says "Device not configured while trying to determing the filesystem size.  I also tried giving it the number of blocks, 'mkfs.ext -c /dev/hdb3 3341520', and I got the same error.  The device /dev/hdb3 does exist.
Any suggestions /solutions?
Thanks 
-- 
==============================================================================
| Matthew Wirges                                                             |
| Systems Administrator, Student, Programmer, Geek.                          |
| Office Phone: [765]464-1148                                                |
| Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com     |
| www.sharpwebinnovations.com                                                |
==============================================================================
  --------------0F1D20A217B10BCD3C487619-- From sergent@csociety.purdue.edu Wed, 13 Sep 2000 23:50:25 -0700 Date: Wed, 13 Sep 2000 23:50:25 -0700 From: Jonathan Sergent sergent@csociety.purdue.edu Subject: [PLUG] uudecode > Anyone know of a windows program to uudecode a mail attachment ? Use the web, Luke! I'm sure you can find lots. Many mail programs do it internally, even... > Also is there any other way to send a binary attachment using mailx ? Not using mailx. If you just need to do it from the command line or from a script, you can use MH. (mhmail is useful from scripts.) --jss. From sergent@csociety.purdue.edu Wed, 13 Sep 2000 23:55:15 -0700 Date: Wed, 13 Sep 2000 23:55:15 -0700 From: Jonathan Sergent sergent@csociety.purdue.edu Subject: [PLUG] Format an ext2 partition > I just created a new partition with fdisk called hdb3 when I go to > format the disk (mkfs.ext2 -c /dev/hdb3) it says "Device not configured > while trying to determing the filesystem size. I also tried giving it > the number of blocks, 'mkfs.ext -c /dev/hdb3 3341520', and I got the > same error. The device /dev/hdb3 does exist. Did you reboot after making the partition? --jss. From olenik@purdue.edu Thu, 14 Sep 2000 08:58:52 -0500 Date: Thu, 14 Sep 2000 08:58:52 -0500 From: Hayden Olenik olenik@purdue.edu Subject: [PLUG] Samba and PUCC Has anyone successfully used Samba or something else to mount their career account drive under Linux? It's getting to be a pain that I have to go use my neighbor's Wintel pc to get the matlab files I saved on my h:\. Any suggestions about how to do this would be appreciated. Thanks. Hayden From lewis1@purdue.edu Thu, 14 Sep 2000 09:02:08 -0500 (EST) Date: Thu, 14 Sep 2000 09:02:08 -0500 (EST) From: James C. Lewis lewis1@purdue.edu Subject: [PLUG] Samba and PUCC On Thu, 14 Sep 2000, Hayden Olenik wrote: > my neighbor's Wintel pc to get the matlab files I saved on my h:\. Any > suggestions about how to do this would be appreciated. Thanks. Is there something wrong with ftp? -James C. Lewis _ /_ _ _ _ / ) / _ ' _ (_/(///)(-_) (__ . (__(-((//_) PUCC Information Center Consultant jclewis@purdue.edu "Only those who dare to fail greatly can ever achieve greatly." -- Robert F. Kennedy Please place all complaints in this box --> [] From cnd@ecn.purdue.edu Thu, 14 Sep 2000 09:08:23 -0500 Date: Thu, 14 Sep 2000 09:08:23 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Samba and PUCC Check the PLUG archives from March: http://csociety/pipermail/plug/2000-March/thread.html I set up autofs to automount PUCC samba shares. Granted, it is insecure for other purposes as your password resides in the autofs files, but hey, it works. :-) -Chris Hayden Olenik wrote: > > Has anyone successfully used Samba or something else to mount their career > account drive under Linux? It's getting to be a pain that I have to go use > my neighbor's Wintel pc to get the matlab files I saved on my h:\. Any > suggestions about how to do this would be appreciated. Thanks. > > Hayden > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From josh@guffin.org Thu, 14 Sep 2000 09:33:24 -0500 (EST) Date: Thu, 14 Sep 2000 09:33:24 -0500 (EST) From: Josh Guffin josh@guffin.org Subject: [PLUG] Mandrake 7.0 -----BEGIN PGP SIGNED MESSAGE----- Hash: RIPEMD160 On Wed, 13 Sep 2000, Deepak Dinesh wrote: > I tried that and it didn't work. I have the locales-en rpm installed > if that makes any difference. I am unable to use most gnome programs > (GDK ?!). > Thanks. i had this problem when i setup my terminal to use cyrillic characters... i found that if you just set the variables to null, nothing complains. for (t)csh: setenv LANGUAGE setenv LANG josh -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.2 (GNU/Linux) Comment: For info see http://www.gnupg.org iQEXAwUBOcDhuNmFfuPAuKo0FAMcEAP+J/P98v95bI2muae0RdFG+0OJ/oU0CVXL 9Gm6pVGpBX7AV0FNk9U4peVkQGGRnQioVwHpvc+ukM4M0BlfBgrBZm3XoZTVHD1U 0jSopjll0yQEIOk4MDeaEQolHI/oPPfmpXqIWKaYuq4LHjLqV0Gjf8NVctpNXY6f ZX53zaBEN60D/A7KqE9rnHUL8MANAwqJ0iP2eKTXtuAgNqqfytc5w22izYI4HSaL HWZHff532acZuj69B/K3F2fj2NU8mhcX8pl19Qlihnx59xM7UDXY7tqPhTaDx9Cl TLR5JrNB1CA2/t+kmNtO/4p/1VF5VsHJDiYTA3RQ1u0f9P+b85x7WBwK =TnPk -----END PGP SIGNATURE----- From deepak@purdue.edu Thu, 14 Sep 2000 10:44:22 -0500 (EST) Date: Thu, 14 Sep 2000 10:44:22 -0500 (EST) From: Deepak Dinesh deepak@purdue.edu Subject: [PLUG] Mandrake 7.0 That shuts up the compiler but the GNOME programs still don't work: Gdk-WARNING **: locale not supported by C library Thanks. On Thu, 14 Sep 2000, Josh Guffin wrote: From deepak@purdue.edu Thu, 14 Sep 2000 11:30:58 -0500 (EST) Date: Thu, 14 Sep 2000 11:30:58 -0500 (EST) From: Deepak Dinesh deepak@purdue.edu Subject: [PLUG] OOM Hi, Has anyone tried the kernel patches to kill processes in OOM condition ? I found a bunch of these in my /var/log/messages: VM: do_try_to_free_pages failed for syslogd... At the same time there was a lot of disk activity (so much that I had to reboot to fix it (no response over the network or keyboard/mouse)). Thanks. From josh@guffin.org Thu, 14 Sep 2000 11:39:03 -0500 (EST) Date: Thu, 14 Sep 2000 11:39:03 -0500 (EST) From: Josh Guffin josh@guffin.org Subject: [PLUG] Mandrake 7.0 -----BEGIN PGP SIGNED MESSAGE----- Hash: RIPEMD160 On Thu, 14 Sep 2000, Deepak Dinesh wrote: > That shuts up the compiler but the GNOME programs still don't work: > > Gdk-WARNING **: locale not supported by C library > > Thanks. yah that happens to mine too. i usually just start programs in the background and then close the eterm i started them from, so i dont see the error messages.... does anyone know how to stop this though? thanks josh -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.2 (GNU/Linux) Comment: For info see http://www.gnupg.org iQEXAwUBOcD/K9mFfuPAuKo0FAPXmgP/Sp2ENEFqYEpudyeUCsOweUyAstTzQxNF hXb+83xihmh8axbCLqlIgHO9kC9AgauXbr85t/VryaiRrts6qiU5z3yZ16PaWRVI cRponb2U/Dl6JUiamdeWyfoFRRyaUMeeL4t4RVM97eIr6aUL8UrXpuzfJPbdHcc8 LRLRQQ6egtID/0IelSNkAx2Udfs2nxtc+iIIsVfPTOM6I7tA3fhPyUlKxz0wnZEG sE9jrPdoJVarvs6XHOq/bSz5odwMQGm3qmQrOfzcmz35CDn38HNvW8qKDbYhMLTK 9Nay0r6wIeC1VFlca0WfK0GzUp7gXSapCopLb+u5aGTOKNPg+Gg3IaPh =5wxo -----END PGP SIGNATURE----- From sergent@csociety.purdue.edu Thu, 14 Sep 2000 10:33:43 -0700 Date: Thu, 14 Sep 2000 10:33:43 -0700 From: Jonathan Sergent sergent@csociety.purdue.edu Subject: [PLUG] Mandrake 7.0 > On Thu, 14 Sep 2000, Deepak Dinesh wrote: > > Gdk-WARNING **: locale not supported by C library > yah that happens to mine too. i usually just start programs in the > background and then close the eterm i started them from, so i dont see the > error messages.... does anyone know how to stop this though? Try env | grep LC_ Unset any that show up, and then set LC_ALL="C". --jss. From deepak@purdue.edu Thu, 14 Sep 2000 12:51:26 -0500 (EST) Date: Thu, 14 Sep 2000 12:51:26 -0500 (EST) From: Deepak Dinesh deepak@purdue.edu Subject: [PLUG] Mandrake 7.0 That worked. But I still don't know what went wrong :-(. No problems with Mandrake 7.1. Cheers. On Thu, 14 Sep 2000, Jonathan Sergent wrote: > > On Thu, 14 Sep 2000, Deepak Dinesh wrote: > > > Gdk-WARNING **: locale not supported by C library > > yah that happens to mine too. i usually just start programs in the > > background and then close the eterm i started them from, so i dont see the > > error messages.... does anyone know how to stop this though? > > Try > env | grep LC_ > Unset any that show up, and then set LC_ALL="C". > > > --jss. > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From lekrigbaum@freh-01.freh.purdue.edu Thu, 14 Sep 2000 13:08:11 EST Date: Thu, 14 Sep 2000 13:08:11 EST From: Larry Krigbaum lekrigbaum@freh-01.freh.purdue.edu Subject: [PLUG] Paid help wanted I'm running RH 6.2 and I'm having what seem to be the usual problems getting the modem, an external 3Com USR 56K Faxmodem, to work. I've recruited a fair amount of advice from various ineternet resouces, but it's still not working, so as much as it frosts me to do it, I'd like to hire someone to get the serial port / modem problem ironed out. A couple of caveats... I live in Crawfordsville so there's a bit of transportation involved, though I wouldn't mind running someone there and back if they don't have a car. Also, I don't want to pay someone to futz around for a while, scratch their head, and say "duh, I dunno." (I've been quite sucessful doing that myself.), so if your experience is limited to your curent installation, you're really not who I'm looking for. On the otherhand, I don't mind paying what it's worth for someone who knows how to get the job done. I suspect it's probably a fairly simple, straightforward omission on my part that I should be able to fix myself if I wasn't so ^*^&#^ clueless. If you're interested, give me a call at 494-0514 or email me at lekrigbaum@purdue.edu and I can give you more details. Sorry if I sound kind of grumpy. I'm just a bit upset with myself for not being able to figure it out. Larry Krigbaum, DBA Purdue University (765) 494-0514 lekrigbaum@adpc.purdue.edu From reh@blacksoul.net Thu, 14 Sep 2000 13:57:12 -0500 (EST) Date: Thu, 14 Sep 2000 13:57:12 -0500 (EST) From: Ryan E. Helfter reh@blacksoul.net Subject: [PLUG] Paid help wanted Wait wait wait... there is a utility called kudzu that autodetects hardware... I have an external US Robotics 56K modem. It found it, set it all up for me. (I know, I know... ICK! I hate when an OS thinks it knows better than me...) Give that a whirl. If its not installed, look under the cdrom under RedHat/RPM for the kudzu rpm. Install it, run /etc/rc.d/init.d/kudzu start have fun :) ry On Thu, 14 Sep 2000, Larry Krigbaum wrote: > Date: Thu, 14 Sep 2000 13:08:11 EST > From: Larry Krigbaum > To: plug@csociety.purdue.edu > Subject: [PLUG] Paid help wanted > > I'm running RH 6.2 and I'm having what seem to be the usual > problems getting the modem, an external 3Com USR 56K > Faxmodem, to work. I've recruited a fair amount of advice from > various ineternet resouces, but it's still not working, so as much as > it frosts me to do it, I'd like to hire someone to get the serial port / > modem problem ironed out. A couple of caveats... I live in > Crawfordsville so there's a bit of transportation involved, though I > wouldn't mind running someone there and back if they don't have a > car. Also, I don't want to pay someone to futz around for a while, > scratch their head, and say "duh, I dunno." (I've been quite > sucessful doing that myself.), so if your experience is limited to > your curent installation, you're really not who I'm looking for. On > the otherhand, I don't mind paying what it's worth for someone who > knows how to get the job done. I suspect it's probably a fairly > simple, straightforward omission on my part that I should be able to > fix myself if I wasn't so ^*^&#^ clueless. If you're interested, give > me a call at 494-0514 or email me at lekrigbaum@purdue.edu and I > can give you more details. Sorry if I sound kind of grumpy. I'm just > a bit upset with myself for not being able to figure it out. > > > > > > Larry Krigbaum, DBA > Purdue University > (765) 494-0514 > lekrigbaum@adpc.purdue.edu > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > -- Ryan E. Helfter reh@blacksoul.net Black Soul Networks, LLC -- From notz@purdue.edu Thu, 14 Sep 2000 15:00:19 -0500 (EST) Date: Thu, 14 Sep 2000 15:00:19 -0500 (EST) From: Patrick K Notz notz@purdue.edu Subject: [PLUG] Free Linux CDs from Pogo For those looking to try some recent linux distros (and don't have a CD burner), Pogo is giving away free CDs (even the shipping is free). So far they have (a) RedHat 6.2 w/ "pogo enhancments" (b) RedHat 7.0Beta and (c) Mandrake 7.1. http://www.pogolinux.com/systems/accessories/linux/linuxcd.html From cnd@ecn.purdue.edu Thu, 14 Sep 2000 19:38:17 -0500 Date: Thu, 14 Sep 2000 19:38:17 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] ACM Programming Contest Let's get some people from PLUG over there to put those ACMers where they belong. (Ok, so I'm an ACM member too (once I pay my dues, doh!!)) Anyway. Check it out, and if you are from PLUG and you go send me an email. If you win, definitely send me email so we can all bow down to your glory. Here's the announcement: SUNDAY! SUNDAY! SUNDAY! (Imagine for yourself the monster truck guy saying this) Purdue's best programmers go head to head in a fierce battle to decide who will represent Purdue in the ACM Collegiate Programming Contest Regional Finals. Prizes for the winners supplied by Microsoft. Judging the event, none other than Lord Brylow.. Seriously though, here are the details. Sunday, September 24th in CS G040 from 12-6pm we'll be having our local programming contest. Food will be provided. Teams of three will compete to solve several math/puzzle type programming problems in the shortest period of time. All undergraduate students are welcome to participate (Freshman, don't be afraid, this is a great learning experience). 1st year grad students are also welcome, but only 1 per team. Also, you are not required to have 3 people to compete. This is a really fun event and a great way to hone your programming skills. Also there's a chance to win some great prizes from Microsoft. If you have any questions please mail wallacms@cs.purdue.edu. -Matt Wallace ACM President -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From zamboni@cerias.purdue.edu Thu, 14 Sep 2000 21:33:32 -0500 Date: Thu, 14 Sep 2000 21:33:32 -0500 From: Diego Zamboni zamboni@cerias.purdue.edu Subject: [PLUG] ACM Programming Contest On Thu, Sep 14, 2000 at 07:38:17PM -0500, Christopher N. Deckard wrote: > Let's get some people from PLUG over there to put those ACMers where > they belong. (Ok, so I'm an ACM member too (once I pay my dues, doh!!)) I don't know what the specific rules are for this contest, but read this story if you know Perl and are thinking of participating: http://chicago.pm.org/perlstory.html --Diego From wallacms@cs.purdue.edu Thu, 14 Sep 2000 23:04:29 -0500 (EST) Date: Thu, 14 Sep 2000 23:04:29 -0500 (EST) From: Matt Wallace wallacms@cs.purdue.edu Subject: [PLUG] ACM Programming Contest > On Thu, Sep 14, 2000 at 07:38:17PM -0500, Christopher N. Deckard wrote: > > Let's get some people from PLUG over there to put those ACMers where > > they belong. (Ok, so I'm an ACM member too (once I pay my dues, doh!!)) > > I don't know what the specific rules are for this contest, but read > this story if you know Perl and are thinking of participating: > http://chicago.pm.org/perlstory.html Sorry, no Perl for you! (Not our rules, national's) Matt From cnd@ecn.purdue.edu Fri, 15 Sep 2000 10:44:22 -0500 Date: Fri, 15 Sep 2000 10:44:22 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Presentation - DHTML - Monday Sept 18, 7pm, Physics 112 DHTML Monday - September 18, 2000 Physics 112 7-9pm Alex Russell is going to talk about next-gen DHTML and the development of his netWindows project. He will discuss problems in developing it, solutions, and other outstanding issues that need to be resolved for the interface. Alex will also talk a little on web standards, why they matter, and the possibilities available to standards-based browsers. For a quick peek at netWindows, point Mozilla over to http://www.netwindows.org. (For those who were here this summer, you saw an early demo of this project which Alex had sent me to show off for him.) -Chris PLUG President P.S. For those of you who were unable to make it to the callout, I will have membership applications at the meeting. Dues are $10 a year. Membership info is available at http://www.ecn.purdue.edu/PLUG/Membership/ -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From cnd@ecn.purdue.edu Fri, 15 Sep 2000 11:17:07 -0500 Date: Fri, 15 Sep 2000 11:17:07 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] installfest meeting Those of you interested in helping with installfest we will have a meeting following the normal PLUG meeting on Monday night. It'll only be 15 minutes or so. We need to decide who is going to chair this event and who is going to help do planning for it. I don't have time to organize it, but I'll help out. We will also select the Grand Poobah (aka The Chair). If you aren't on the installfest list go to: http://csociety.ecn.purdue.edu/mailman/listinfo/installfest/ to sign up. -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From cnd@ecn.purdue.edu Fri, 15 Sep 2000 11:24:51 -0500 Date: Fri, 15 Sep 2000 11:24:51 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] new books for the library Ok, so I'm waiting on getting space to put all of our books somewhere, but I thought that you all might be interested in knowing that we got a bunch of books sent to us. Thanks goes to Garrett Honeycutt for getting the donation. Everyone give him a high five or pat on the back. Here's the list: Open Source - The Unauthorized White Papers Samba for Dummies Corel Linux for Dummies (2 copies) RedHat Linux for Dummies (4 copies) Art for Dummies (2 copies) RedHat Linux Secrets KDE Bible RedHat Linux Bible TurboLinux Bible Gimp Bible for Linux PHP4 Bible This is all in addition to our existing collection of O'Reilly books (a list of which I forget) including: Linux Network Administration Programming Perl System Administration with Perl CVS Pocket Reference Programming Python Python Pocket Reference Mastering Regular Expressions with Perl Open Source Bibliography and a few others that I forget the names of. Now if we just had space.... -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From lewis1@purdue.edu Fri, 15 Sep 2000 14:46:33 -0500 (EST) Date: Fri, 15 Sep 2000 14:46:33 -0500 (EST) From: James C. Lewis lewis1@purdue.edu Subject: [PLUG] ISBN and DVDs Anyone know of a ISBN search that returns DVD titles? I'm using my cute little CueCat to catalog my DVDs. I thought Amazon or B&N would work, but it appears they only have books in their database. I figure someone somewhere must be tracking DVD/Movie titled with ISBNs, or why else would they be on there? -James C. Lewis _ /_ _ _ _ / ) / _ ' _ (_/(///)(-_) (__ . (__(-((//_) PUCC Information Center Consultant jclewis@purdue.edu "Only those who dare to fail greatly can ever achieve greatly." -- Robert F. Kennedy Please place all complaints in this box --> [] From bhlewis@pythia.cc.purdue.edu 15 Sep 2000 15:39:23 -0500 Date: 15 Sep 2000 15:39:23 -0500 From: Benjamin Lewis bhlewis@pythia.cc.purdue.edu Subject: [PLUG] ISBN and DVDs "James C. Lewis" writes: > Anyone know of a ISBN search that returns DVD titles? The URL below has a pretty good UPC code lookup. I found the few DVDs and CDs that I tried. http://www.barpoint.com/ -Ben From lev@purdue.edu Fri, 15 Sep 2000 16:19:26 -0500 Date: Fri, 15 Sep 2000 16:19:26 -0500 From: Lev Gorenstein lev@purdue.edu Subject: [PLUG] Security docco On Thu, 14 Sep 2000, Paul Kuliniewicz wrote: PK> Another page you might want to take a look at is PK> http://www.enteract.com/~lspitz/linux.html, which also talks about some PK> things you can do to lock down your Linux box. PK> PK> Related to that, I've got a small problem with RH6.2. I wanted to change PK> the /etc/issue file, which is what's displayed when you log in. I was going PK> to change it into a warning message for trespassers, or at least get rid of PK> the system identification information that's there by default. (Yeah yeah, PK> obscurity != security, but why help out an intruder?) So, I edit PK> /etc/issue, log out, and it works. But if I do a reboot of the system, Red PK> Hat replaces /etc/issue with the default banner! PK> PK> How do I get RH to stop overwriting the /etc/issue file? It appears to do PK> this either on reboot/shutdown or startup. Note that I don't know where the PK> appropriate scripts are located (yet), so please be specific in your reply. PK> I figure all I have to do is comment out a line in some script somewhere, PK> but the question is *which* line.... I just can't help but ROTFL. If you had read the same article that you suggest, you would've found the answer right there ;-)). About 2/3 from the top, it says: ------- Begin quote ------ Last, create the file /etc/issue. This file is an ASCII text banner that appears for all telnet logins (example B). This legal warning will appear whenever someone attempts to login to your system. If you want to continue using the same /etc/issue file, you will have to modify /etc/rc.d/init.d/S99local. By default, Linux creates a new /etc/issue file on every reboot. -------- End quote ------- Lev -- Karma? Let me guess... It's the Toyota Factory! From kuliniew@purdue.edu Fri, 15 Sep 2000 16:22:43 -0500 Date: Fri, 15 Sep 2000 16:22:43 -0500 From: Paul Kuliniewicz kuliniew@purdue.edu Subject: [PLUG] Security docco >I just can't help but ROTFL. > >If you had read the same article that you suggest, you would've >found the answer right there ;-)). About 2/3 from the top, it >says: *snip* Um...thanks. I guess that's what I get for skipping around the page following different suggestions. Still seems like goofy behavior to me, but oh well. Thanks for pointing that out. From rjune@ims1.imagestream-is.com Fri, 15 Sep 2000 17:07:40 -0500 (EST) Date: Fri, 15 Sep 2000 17:07:40 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] installfest meeting I won't be able to make the meeting, however like last year, I can bring computers, and hubs, and a burner. On Fri, 15 Sep 2000, Christopher N. Deckard wrote: > Those of you interested in helping with installfest we will have a > meeting following the normal PLUG meeting on Monday night. It'll only > be 15 minutes or so. > > We need to decide who is going to chair this event and who is going to > help do planning for it. I don't have time to organize it, but I'll > help out. We will also select the Grand Poobah (aka The Chair). > > If you aren't on the installfest list go to: > http://csociety.ecn.purdue.edu/mailman/listinfo/installfest/ to sign up. > > -Chris > PLUG President > > -- > ---------------------------------------------------------------------- > Christopher N. Deckard | Lead Web Technician > cnd@ecn.purdue.edu | Engineering Computer Network > http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ > ---------------------------------------------------------------------- > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From rajak@purdue.edu Fri, 15 Sep 2000 17:36:26 -0500 (EST) Date: Fri, 15 Sep 2000 17:36:26 -0500 (EST) From: Brian Poole rajak@purdue.edu Subject: [PLUG] Security docco On Fri, 15 Sep 2000, Paul Kuliniewicz wrote: > following different suggestions. Still seems like goofy behavior to me, > but oh well. Thanks for pointing that out. As the issue contains 2 variables, that of the current version of the distro and that of the current kernel running, this behavior really isn't goofy but necessary. Otherwise if you either upgrade your distro or kernel the issue will not give up to date information thus defeating the whole purpose of having it there (not to say that I know what that purpose is, but since they do have it, they should also keep it correct). -b From cnd@ecn.purdue.edu Mon, 18 Sep 2000 08:43:08 -0500 Date: Mon, 18 Sep 2000 08:43:08 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] TONIGHT - DHTML - Monday Sept 18, 7pm, Physics 112 TONIGHT TONIGHT TONIGHT TONIGHT DHTML Monday - September 18, 2000 Physics 112 7-9pm Alex Russell is going to talk about next-gen DHTML and the development of his netWindows project. He will discuss problems in developing it, solutions, and other outstanding issues that need to be resolved for the interface. Alex will also talk a little on web standards, why they matter, and the possibilities available to standards-based browsers. For a quick peek at netWindows, point Mozilla over to http://www.netwindows.org. (For those who were here this summer, you saw an early demo of this project which Alex had sent me to show off for him.) -Chris PLUG President P.S. For those of you who were unable to make it to the callout, I will have membership applications at the meeting. Dues are $10 a year. Membership info is available at http://www.ecn.purdue.edu/PLUG/Membership/ -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From wirges@purdue.edu Mon, 18 Sep 2000 11:21:33 -0500 Date: Mon, 18 Sep 2000 11:21:33 -0500 From: Matthew Laurence Wirges wirges@purdue.edu Subject: [PLUG] Debian, Networking and lost sleep.... Okay, got a problem that kept me up late last night and bugged me all morning.. I have just installed Debian Slink (btw yes, I know potato exists and I can upgrade) on my Toshiba Tecra 510CDT. Well, when I was installing it last night I moronically gave my laptop my desktop computer's IP address.. Needless to say one of my roommates was pissed when he kept losing his connection to xdm, heh.. So the first thing I did was I ran ifconfig for the new ip as follows: ifconfig eth0 192.168.0.54 netmask 255.255.255.0 up well, that worked and I was up and running (i.e. I could connect to my lan and the internet and vice versa) but I knew that once I restarted it would just go back to the old ip.. After an hour of searching on the web and /etc I finally found the file in /etc/pcmcia/network.opts (I think it was network.opts, I don't remeber) I changed the IP in there and rebooted.. Sure enough when the laptop booted back up it had its correct ip, I also made sure to change my entry in my hosts file... Now, here is my real problem.. Since I changed my laptop's ip to the correct one I cannot connect to any other computer in my network or the internet.. Nor can any other machine connect to me. I checked the hosts.deny and removed the line ALL:Paranoia, my hosts.allow was empty. I rebooted it a couple times... I have checked the inetd.conf file to make sure that me telnet and ftp servers were running and they were.. My laptop cannot ping any other computers but itself.. When I try to ping other computers I get some sort of -1 error (I would tell you the error message but I don't have my laptop with me right now).. So if anyone understands any of that incoherent crap I just wrote, and can offer any helpful solutions, suggestions or magic to fix this problem.. Let me know so I can get some sleep. -Matt -- =================================================== | Matt Wirges | | Student, SysAdmin, Programmer, Geek | | Office: [765]464-1148 | | Email: wirges@expert.cc.purdue.edu || | | sysadmin@sharpwebinnovations.com | | www.sharpwebinnovations.com | =================================================== From redmanr@minga.com Mon, 18 Sep 2000 11:15:49 -0500 (EST) Date: Mon, 18 Sep 2000 11:15:49 -0500 (EST) From: Rick Redman redmanr@minga.com Subject: [PLUG] Debian, Networking and lost sleep.... I have a simular problem on my debian machines (PCMCIA). Check your routes. Remove your default route, this MIGHT make it so you can talk to the local LAN (255.255.255.0) without any problem. Thats what I have to do sometimes when my router isnt working/plugged in. Minga From he@expert.cc.purdue.edu Mon, 18 Sep 2000 15:25:08 -0500 (EST) Date: Mon, 18 Sep 2000 15:25:08 -0500 (EST) From: he@expert.cc.purdue.edu he@expert.cc.purdue.edu Subject: [PLUG] Debian, Networking and lost sleep.... On Mon, 18 Sep 2000, Matthew Laurence Wirges wrote: > Okay, got a problem that kept me up late last night and bugged me all > morning.. > ifconfig eth0 192.168.0.54 netmask 255.255.255.0 up ^^^^^^^^^^^^^ This looks like an internel IP. Are you going to internet though NAT? > > Now, here is my real problem.. Since I changed my laptop's ip to the > correct one I cannot connect to any other computer in my network or the > internet.. Nor can any other machine connect to me. I checked the You'd better always make sure to connect to your local lan first. You can do "ifconfig -a" for several times after several boot, check that whether you have the same nonzero MAC address. If it doesn't your driver has problem. Then you can do an "arp", you should be able to see all the MACs of your LAN IPs listed there. And when you do an "arp" in other machine in the same LAN, you should see your MAC there. If driver and arp works and your IP/MASK is set OK and you still cant ping within your LAN, probably check the routing table. kevin > hosts.deny and removed the line ALL:Paranoia, my hosts.allow was empty. > I rebooted it a couple times... I have checked the inetd.conf file to > make sure that me telnet and ftp servers were running and they were.. My > laptop cannot ping any other computers but itself.. When I try to ping > other computers I get some sort of -1 error (I would tell you the error > message but I don't have my laptop with me right now).. > > So if anyone understands any of that incoherent crap I just wrote, and > can offer any helpful solutions, suggestions or magic to fix this > problem.. Let me know so I can get some sleep. > -Matt > -- > =================================================== > | Matt Wirges | > | Student, SysAdmin, Programmer, Geek | > | Office: [765]464-1148 | > | Email: wirges@expert.cc.purdue.edu || | > | sysadmin@sharpwebinnovations.com | > | www.sharpwebinnovations.com | > =================================================== > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From ab@eas.purdue.edu Mon, 18 Sep 2000 16:13:00 -0500 Date: Mon, 18 Sep 2000 16:13:00 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] ACM Programming Contest These things are bogus. I've been thinking for some time that we need a real programming contest of some sort. My first thought is a game-player tournament. Set up a server to control the game state and give people stubs of code and the API to write a player. Play a lot of rounds in some sort of elimination tournament (or scoring, depending on the game), last/best one standing wins. To make it fun, the game shouldn't be totally deterministic or should have some hidden information (in other words, not chess for pete's sake). The trendy one now is backgammon. I was thinking a card game like Uno or poker, myself. Uno's got a scoring system, a poker player could wager (we'd have to stake them all equally, obviously). Ought to be easy to figure out who the winner is. I'll think about this some more. Anyone interested in a contest like that? ab From zamboni@cerias.purdue.edu Mon, 18 Sep 2000 16:36:01 -0500 Date: Mon, 18 Sep 2000 16:36:01 -0500 From: Diego Zamboni zamboni@cerias.purdue.edu Subject: [PLUG] ACM Programming Contest On Mon, Sep 18, 2000 at 04:13:00PM -0500, A Braunsdorf wrote: > I was thinking a card game like Uno or poker, myself. Uno's got > a scoring system, a poker player could wager (we'd have to stake > them all equally, obviously). Ought to be easy to figure out who > the winner is. > > I'll think about this some more. Anyone interested in a contest > like that? Another possibility would be something like RealTimeBattle (http://realtimebattle.sourceforge.net/), in which robots battle and try to destroy each other. You program your robot, and the server pits it against all the others. It certainly can be fun to watch :-) --Diego From wallacms@cs.purdue.edu Mon, 18 Sep 2000 18:35:34 -0500 (EST) Date: Mon, 18 Sep 2000 18:35:34 -0500 (EST) From: Matt Wallace wallacms@cs.purdue.edu Subject: [PLUG] ACM Programming Contest > These things are bogus. Why? I organized the programming contest this year and if you have input as to how we can make the progamming contest better, I'd love to hear it. Probably the format for this particular contest won't ever change, becuase it's a qualifier for the national contest, so we have to play by their rules, but we're definitely looking to do another contest in the spring. I have to say that in my opinion these types of contests are very cool.. It really tests your ability to think on your feet, becuase you don't have weeks to plan out your program. Most of the problems we use have simple solutions once you "see" them right, but look quite daunting at first. I definitely agree that they exercise a different skill then the more long term type of contest, but I would consider them far from bogus.. > > My first thought is a game-player tournament. Set up a server to > control the game state and give people stubs of code and the API > to write a player. Play a lot of rounds in some sort of elimination > tournament (or scoring, depending on the game), last/best one > standing wins. If you're interested in this sort of thing, UIUC ACM is doing that exact thing at Reflections/Projections in October. You write an AI to control a MechWarrior player. Also Crowe Chizek sponsors a type of programming contest like this in the spring. In the past three years they have done Stratego, Pente, and a TankWars game where you move a tank around a gameboard and try to kill the other player. > I'll think about this some more. Anyone interested in a contest > like that? > If people are, ACM is definitely interested in sponosoring a contest like this. -Matt From naval@purdue.edu Mon, 18 Sep 2000 18:42:28 -0500 Date: Mon, 18 Sep 2000 18:42:28 -0500 From: Emmanuel Magana Naval naval@purdue.edu Subject: [PLUG] ACM Programming Contest AI programming Contests seem more fun. One possible environment can also be a simple MUD. There's a whole lot of things you can do with AI on a MUD or MUD type game. Ah...That brings back memories of food fight and trade wars. I most definitely would be interested...and not to mention, include PERL as an acceptable language! Manny A Braunsdorf wrote: > > These things are bogus. I've been thinking for some time that we > need a real programming contest of some sort. > > My first thought is a game-player tournament. Set up a server to > control the game state and give people stubs of code and the API > to write a player. Play a lot of rounds in some sort of elimination > tournament (or scoring, depending on the game), last/best one > standing wins. > > To make it fun, the game shouldn't be totally deterministic or > should have some hidden information (in other words, not chess for > pete's sake). The trendy one now is backgammon. > > I was thinking a card game like Uno or poker, myself. Uno's got > a scoring system, a poker player could wager (we'd have to stake > them all equally, obviously). Ought to be easy to figure out who > the winner is. > > I'll think about this some more. Anyone interested in a contest > like that? > > ab > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From wli@holomorphy.com Mon, 18 Sep 2000 17:04:03 -0700 Date: Mon, 18 Sep 2000 17:04:03 -0700 From: William Lee Irwin III wli@holomorphy.com Subject: [PLUG] ACM Programming Contest On Mon, Sep 18, 2000 at 06:42:28PM -0500, Emmanuel Magana Naval wrote: > AI programming Contests seem more fun. > One possible environment can also be a simple > MUD. There's a whole lot of things you can do > with AI on a MUD or MUD type game. > Ah...That brings back memories of food fight > and trade wars. > I most definitely would be interested...and not > to mention, include PERL as an acceptable language! Other programming contests have formulated similar ideas. The International Contest in Functional Programming last year basically used the idea of creating a "bot" or whatever on a MUD as its contest problem. (well, it was a little more involved than that). This year it used a raytracer. I won't try bringing in too many of my own ideas, because I know from past experience that they bore the rest of the world to tears. =) There are many other subfields of AI to explore, for instance, natural language processing, image and sound recognition, and theorem proving. Cheers, Bill P.S.: Don't give matrix logic puzzles as problems and allow Prolog or other logic languages. Solving those types of problems in logic languages consists of little more than writing down the problem in a slightly weird notation. -- "I'd crawl over an acre of 'Visual This++' and 'Integrated Development That' to get to gcc, Emacs, and gdb. Thank you." -- Vance Petree, Virginia Power From ab@eas.purdue.edu Mon, 18 Sep 2000 22:34:39 -0500 Date: Mon, 18 Sep 2000 22:34:39 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] ACM Programming Contest In message , Matt Wallace writes: > > These things are bogus. > > Why? In the literal sense. "Fake." "Fraudulent." "Misleading." Certainly reflective of something, but is it something we care about? Maybe. Maybe >not<. I honestly don't know the particulars of this contest, but I worry when I hear things like contestants being barred from using Perl because it works too well. Oh yeah, that's Duty Now For The Future. That's exactly the message >I< want to send the youth of America. My calculus teacher in high school wouldn't let us use calculators. Why not? Because, he said, we wouldn't always have one. (The obvious retort is "We won't always be doing calculus", but...) For years I carried one everywhere I went, hoping I'd run into that prick. Now I carry a computer. We're talking about >computer< >programming< here- we're committed to technology already. Setting up artificial barriers is trying to unring the bell. > I have to say that in my opinion these types of contests are very cool.. > It really tests your ability to think on your feet, becuase you don't > have weeks to plan out your program. Most of the problems we use have > simple solutions once you "see" them right, but look quite daunting at > first. I definitely agree that they exercise a different skill then the > more long term type of contest, but I would consider them far from bogus.. Oh, I agree that thinking on your feet is important. But so's using the right tool and reusing code. I have a bigger problem with students not using existing tools and code than I do with them not being able to think on their feet. Why? "We weren't taught that." "That's cheating." No, it's the whole open source thing, writ large. Do you believe or don't you?! Not letting people use Perl isn't the problem, it's a symptom of the problem. I understand going into battle underequipped for the fun of it. I understand tests and games as a rite of passage. Heck, I've lost Fastest Finger under actual playing conditions- I know from game shows. But what're you trying to prove? What are you trying to test? That two guys with their hands tied behind their backs can butt heads? I just think a little more big picture thinking might make this more interesting. I don't go >anywhere< without a computer in my pocket. The idea of not being allowed to use it at a serious task is ludicrous, yet that's what's expected in a lot of testing and contest situations. Why's that? Because it makes my life too easy? Compared to what? You don't cut wood without a saw. You don't pound nails without a hammer. You don't bring a knife to a gunfight. Do you really want to program without the right tools? Especially when, thanks to open source, there's no such what-if as "what if you don't have them"? Which is easiest to scrounge up when you need it? A saw, hammer, gun, or gcc? In context, the compiler is much easier than the miracle of conjuring a solid object from thin air. We're talking about programming computers here- the "roughing it" train left the station a long time ago. Now, as the saying goes, we're just haggling over the price. If the idea is really a hi-tech version of "Survivor"- getting by with stone knives and bearskins- that's OK, but is that really what you want to test? I mean if you blow Michael Jordan's kneecaps off, I might be able to beat him at basketball too. It doesn't mean I'll ever make a living at it. > If you're interested in this sort of thing, UIUC ACM is doing that exact > thing at Reflections/Projections in October. You write an AI to control a > MechWarrior player. MechWarrior. Jeez, is blowing things up in 3D all anyone cares about anymore? > Also Crowe Chizek sponsors a type of programming contest like this in the > spring. In the past three years they have done Stratego, Pente, and a > TankWars game where you move a tank around a gameboard and try to kill the > other player. Stratego and Pente both yield to look-ahead, right? (I don't know Stratego.) And I already weighed in on the blowing things up issue. :-) You want to do a game that doesn't yield to brute force, either because it's computationally infeasible or (much more fun) because it depends on a random or unknown element. That makes it more fun for human players too. > > I'll think about this some more. Anyone interested in a contest > > like that? > If people are, ACM is definitely interested in sponosoring a contest like > this. That's the spirit! ab From brylow@cs.purdue.edu Tue, 19 Sep 2000 02:07:17 -0500 Date: Tue, 19 Sep 2000 02:07:17 -0500 From: Dennis Brylow brylow@cs.purdue.edu Subject: [PLUG] ACM Programming Contest Quoting A Braunsdorf (ab@eas.purdue.edu): > In the literal sense. "Fake." "Fraudulent." "Misleading." > Certainly reflective of something, but is it something we care > about? Maybe. Maybe >not<. I guess that all depends on who you are defining as "we", and what you allege it is that we care about. > My calculus teacher in high school wouldn't let us use calculators. > Why not? Because, he said, we wouldn't always have one. (The > obvious retort is "We won't always be doing calculus", but...) > For years I carried one everywhere I went, hoping I'd run into that > prick. Now I carry a computer. We're talking about >computer< > >programming< here- we're committed to technology already. Setting > up artificial barriers is trying to unring the bell. I'm sorry your high school employed Luddites as instructors, but there is an equal and opposite peril on the far end of that spectrum. One of the common misconceptions about Computer Science is the notion that as technology has progressed and advanced programming languages have proliferated, all of the more basic competencies have become less important. Witness introductory programming courses across the nation that start with object-oriented programming, rather than more traditional instructional languages. Java and its ilk are "in" and "hot", but the end results are sophomores who can recite from memory the definition of inheritance, but couldn't write a loop to compare two strings to save their lives. See graduating CS seniors, diplomas proudly in hand, who cannot explain simple two's complement arithmetic, the benefits of tail recursion, or basic loop optimizations. These are, of course, gross generalizations. The bottom line, however, is that we have an epidemic in this field of "programmers" who are very good at putting together building blocks created by *other* people, but who can neither understand these pieces, nor build their own. I guess that's OK, if you're only interested in training code monkeys to churn out millions of lines of braindead bloatware. Is this something we care about? I would assert that yes, it is. > I have a bigger problem > with students not using existing tools and code than I do with them > not being able to think on their feet. Why? "We weren't taught > that." "That's cheating." I agree; in all things there must be balance. We do not ask incoming mechanical engineering students to first re-invent the wheel, nor should we. But there is an implicit assumption that all ME's know how a wheel works, and appreciate its elegance as a solution, before moving on to Wankel rotary engines. Not so in Computer Science, where it's not safe to assume everybody knows the basics, yet we hand them the power tools on the first day. "Why not? It's only software..." This competition is for undergraduates, and first-year grad students. It is not meant to challenge or entertain seasoned professionals. > But what're you trying to prove? What are you trying to test? That > two guys with their hands tied behind their backs can butt heads? I > just think a little more big picture thinking might make this more > interesting. It's ironic that you should make this point during the Olympics. What are we trying to prove by swimming relay races, when motor boats have been the obvious way to go for decades? Are marathon runners just poor, misguided dolts who can't see the "big picture"? They really could get there a lot faster, after all, using the latest tools and technology. > You don't cut wood without a saw. You don't pound nails without > a hammer. You don't bring a knife to a gunfight. You shouldn't fell a tree with a power saw, if you have yet to master steak knives. There's a lot more to building a house than simply pounding in the nails. And getting drawn into unexpected gunfights is often a sign of poor planning. > Do you really > want to program without the right tools? Especially when, thanks > to open source, there's no such what-if as "what if you don't have > them"? In industry? No, never. In academia? Yes, if using the "right" tools would serve only to obscure the underlying concepts, and foster reliance on someone else's pre-packaged skill. There will be plenty of time for that in industry. They can hold their own programming competitions. Ironically, I think the founders of the open-source movement would concede that their incredible depth of skill would not have been possible without having been forced, at some point, to understand the innermost workings of their primitive tools. It remains to be seen whether or not the next generation can acquire this same skill level without having to endure the same trial by fire. > We're talking about programming computers here- the "roughing it" > train left the station a long time ago. An arguable point. See Isaac Asimov's "The Feeling of Power". :) > If the idea is really a hi-tech version of "Survivor"- getting by > with stone knives and bearskins- that's OK, but is that really what > you want to test? Yes. I believe that the programming equivalent of "stone knives and bearskins" remains a valuable skillset that is just as important as learning the intricacies of more complex and powerful tools. The difficult thing about Computer Science is that the "basics" are still among mankind's most abstract and complex creations. There are more degrees of freedom in a programming problem than in virtually any analogous problem in the physical world. The story that set off this entire discussion is an amusing, but questionable bit of Perl propoganda. That UCLA's non-ACM-related programming contest has inconsistent rules is fairly irrelevant to the question at hand. Official ACM programming contests have not "banned" Perl, because they have never allowed Perl in the first place. ACM Competitions use Pascal, C, and C++; they do not see the need to jump on the bandwagon for every new-fangled language on the horizon with a rapidly changing feature set. Over the past decade, I have heard several variations on this theme, describing how ACM has "banned" one language or another because it "makes the competitions too easy." Such languages have included, but are not limited to, Scheme, ADA, Objective C, and yes, even FORTRAN. (I'm pretty sure that the FORTRAN proponent was joking, but one can never tell...) I'm sure there are plenty of people on this list who are willing to expound at great length on which of these is the most preposterous. Any kind of competition can suffer from poor design. I suspect that it is a sign of poor problem writing if competitors are loosing time generating from scratch code that is so standardized as to have been incorporated into a language like Perl. Problems should be algorithmically challenging -- thought problems -- not busy work centered around minute details of implementing I/O or string processing. > > If people are, ACM is definitely interested in sponosoring a contest like > > this. > > That's the spirit! There we are in agreement. Powertool users should have challenging contests, as well. But don't bash the basics just because you're bored with them. They are no less important now, than when you first learned them back in the Dark Ages. But that's all just my opinion. -D From pratte@ecn.purdue.edu Tue, 19 Sep 2000 02:44:21 -0500 (EST) Date: Tue, 19 Sep 2000 02:44:21 -0500 (EST) From: Eric Pratt pratte@ecn.purdue.edu Subject: [PLUG] ACM Programming Contest Damn.... you guys are deep! On Tue, 19 Sep 2000, Dennis Brylow wrote: > Quoting A Braunsdorf (ab@eas.purdue.edu): > > In the literal sense. "Fake." "Fraudulent." "Misleading." > > Certainly reflective of something, but is it something we care > > about? Maybe. Maybe >not<. > > I guess that all depends on who you are defining as "we", and what > you allege it is that we care about. > > > My calculus teacher in high school wouldn't let us use calculators. > > Why not? Because, he said, we wouldn't always have one. (The > > obvious retort is "We won't always be doing calculus", but...) > > For years I carried one everywhere I went, hoping I'd run into that > > prick. Now I carry a computer. We're talking about >computer< > > >programming< here- we're committed to technology already. Setting > > up artificial barriers is trying to unring the bell. > > I'm sorry your high school employed Luddites as instructors, but > there is an equal and opposite peril on the far end of that spectrum. > One of the common misconceptions about Computer Science is the notion > that as technology has progressed and advanced programming languages > have proliferated, all of the more basic competencies have become less > important. > Witness introductory programming courses across the nation that > start with object-oriented programming, rather than more traditional > instructional languages. Java and its ilk are "in" and "hot", but the > end results are sophomores who can recite from memory the definition > of inheritance, but couldn't write a loop to compare two strings to > save their lives. > See graduating CS seniors, diplomas proudly in hand, who cannot > explain simple two's complement arithmetic, the benefits of tail > recursion, or basic loop optimizations. > These are, of course, gross generalizations. The bottom line, > however, is that we have an epidemic in this field of "programmers" > who are very good at putting together building blocks created by > *other* people, but who can neither understand these pieces, nor build > their own. > I guess that's OK, if you're only interested in training code > monkeys to churn out millions of lines of braindead bloatware. > Is this something we care about? I would assert that yes, it is. > > > I have a bigger problem > > with students not using existing tools and code than I do with them > > not being able to think on their feet. Why? "We weren't taught > > that." "That's cheating." > > I agree; in all things there must be balance. We do not ask > incoming mechanical engineering students to first re-invent the wheel, > nor should we. But there is an implicit assumption that all ME's know > how a wheel works, and appreciate its elegance as a solution, before > moving on to Wankel rotary engines. > Not so in Computer Science, where it's not safe to assume everybody > knows the basics, yet we hand them the power tools on the first day. > "Why not? It's only software..." > This competition is for undergraduates, and first-year grad students. > It is not meant to challenge or entertain seasoned professionals. > > > But what're you trying to prove? What are you trying to test? That > > two guys with their hands tied behind their backs can butt heads? I > > just think a little more big picture thinking might make this more > > interesting. > > It's ironic that you should make this point during the Olympics. > What are we trying to prove by swimming relay races, when motor boats > have been the obvious way to go for decades? Are marathon runners > just poor, misguided dolts who can't see the "big picture"? They > really could get there a lot faster, after all, using the latest tools > and technology. > > > You don't cut wood without a saw. You don't pound nails without > > a hammer. You don't bring a knife to a gunfight. > > You shouldn't fell a tree with a power saw, if you have yet to > master steak knives. There's a lot more to building a house than > simply pounding in the nails. And getting drawn into unexpected > gunfights is often a sign of poor planning. > > > Do you really > > want to program without the right tools? Especially when, thanks > > to open source, there's no such what-if as "what if you don't have > > them"? > > In industry? No, never. In academia? Yes, if using the "right" > tools would serve only to obscure the underlying concepts, and foster > reliance on someone else's pre-packaged skill. There will be plenty > of time for that in industry. They can hold their own programming > competitions. > Ironically, I think the founders of the open-source movement would > concede that their incredible depth of skill would not have been > possible without having been forced, at some point, to understand the > innermost workings of their primitive tools. It remains to be seen > whether or not the next generation can acquire this same skill level > without having to endure the same trial by fire. > > > We're talking about programming computers here- the "roughing it" > > train left the station a long time ago. > > An arguable point. See Isaac Asimov's "The Feeling of Power". :) > > > If the idea is really a hi-tech version of "Survivor"- getting by > > with stone knives and bearskins- that's OK, but is that really what > > you want to test? > > Yes. I believe that the programming equivalent of "stone knives > and bearskins" remains a valuable skillset that is just as important > as learning the intricacies of more complex and powerful tools. The > difficult thing about Computer Science is that the "basics" are still > among mankind's most abstract and complex creations. There are more > degrees of freedom in a programming problem than in virtually any > analogous problem in the physical world. > > The story that set off this entire discussion is an amusing, but > questionable bit of Perl propoganda. That UCLA's non-ACM-related > programming contest has inconsistent rules is fairly irrelevant to the > question at hand. Official ACM programming contests have not "banned" > Perl, because they have never allowed Perl in the first place. ACM > Competitions use Pascal, C, and C++; they do not see the need to jump > on the bandwagon for every new-fangled language on the horizon with a > rapidly changing feature set. Over the past decade, I have heard > several variations on this theme, describing how ACM has "banned" one > language or another because it "makes the competitions too easy." > Such languages have included, but are not limited to, Scheme, ADA, > Objective C, and yes, even FORTRAN. (I'm pretty sure that the > FORTRAN proponent was joking, but one can never tell...) I'm sure > there are plenty of people on this list who are willing to expound > at great length on which of these is the most preposterous. > > Any kind of competition can suffer from poor design. I suspect > that it is a sign of poor problem writing if competitors are loosing > time generating from scratch code that is so standardized as to have > been incorporated into a language like Perl. Problems should be > algorithmically challenging -- thought problems -- not busy work > centered around minute details of implementing I/O or string > processing. > > > > If people are, ACM is definitely interested in sponosoring a contest like > > > this. > > > > That's the spirit! > > There we are in agreement. Powertool users should have challenging > contests, as well. But don't bash the basics just because you're > bored with them. They are no less important now, than when you first > learned them back in the Dark Ages. > > > But that's all just my opinion. > > -D > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From ab@eas.purdue.edu Tue, 19 Sep 2000 03:18:46 -0500 Date: Tue, 19 Sep 2000 03:18:46 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] ACM Programming Contest In message <20000919020717.A8970@cs.purdue.edu>, Dennis Brylow writes: > > I'm sorry your high school employed Luddites as instructors, but > there is an equal and opposite peril on the far end of that spectrum. First, that was a common attitude when I was in high school- times have changed, and that's part of my point. Fifteen years from now, I hope the idea of drilling "basics" at the University level is as fantastically quaint as a spelling bee would be. Second, it's not an equal peril. If we can advance the state of the art instead of reinventing the wheel, that'd be better, but you're right that... > One of the common misconceptions about Computer Science is the notion > that as technology has progressed and advanced programming languages > have proliferated, all of the more basic competencies have become less > important. You're talking about something different now. I'm more about knowing the basics than anyone. I'm an atomist by creed. I agree fully that most students these days really don't get enough of that, but that's not something I'd neither marginalize nor elevate to competition for University students. I'd like to see students really learn the basics. I'd like to see them learn them really early in basic math classes, and a lot of them long before they write a line of code. > I guess that's OK, if you're only interested in training code > monkeys to churn out millions of lines of braindead bloatware. I'm not. A lot of people are, though. Isn't that what CPTs are for? > Not so in Computer Science, where it's not safe to assume everybody > knows the basics, yet we hand them the power tools on the first day. > "Why not? It's only software..." Why is it not safe? I'll agree they don't, in a lot of cases, but computers are far from inscrutable. Unlike >anything< in nature, people created them from the ground up in microscopic detail. If there's anything people can grasp readily, you'd think this'd be it. The problem is that computers are still used by the masses as TV sets for watching the web. Grade schools consider being able to use a web browser "computer literacy". Those kids need a better math curriculum that covers the fundamentals of CS. These are fundamental skills to this era as much as arithmetic was a century ago. One weekend in college is not the time to pick that up. My point is, we don't see English majors at Purdue competing in spelling bees. They're supposed to know that stuff already (yeah, some of them don't). Ditto for math majors and their multiplication tables. Are CS students just dumber? I hope not. Is the field that much more complicated? I'll argue not, but even if it is, it means people need to be even more prepared. Yeah, I think a lot more people should flunk out too. Not everyone can do this, apparently. > This competition is for undergraduates, and first-year grad students. > It is not meant to challenge or entertain seasoned professionals. I understand that, and I'm over-seasoned. :-) I just think sixth graders ought to be doing this and supposedly winning adventurers should be doing bigger and better stuff. > It's ironic that you should make this point during the Olympics. > What are we trying to prove by swimming relay races, when motor boats > have been the obvious way to go for decades? Are marathon runners > just poor, misguided dolts who can't see the "big picture"? They > really could get there a lot faster, after all, using the latest tools > and technology. If your goal is to get there, hell yes the boat is the right answer. If the goal is the journey, or the exertion, swimming is great. In the big picture, yeah, I do sort of think they're dolts, but I understand why they do what they do. They're trying to make themselves into something better and do things people haven't done before. You know what? That's selfish. That's OK, and I'm not going to stop them, but here in the land of computers, adding to the code base of the world helps the world. We're hoping that code lives on long after we're all dust. Yes, I'm serious. That's not to say I don't want to be the best I can be, nor that I don't get a lot of satisfaction from that, but I really am in this to save the world, indirectly by being a curmudgeon, if not by action. :-) > You shouldn't fell a tree with a power saw, if you have yet to > master steak knives. There's a lot more to building a house than > simply pounding in the nails. And getting drawn into unexpected > gunfights is often a sign of poor planning. Not every programmer is going to be a designer, just as not every construction worker is an architect. The really cool architects don't use hammers at all. The really cool software designers needn't write code either. > In industry? No, never. In academia? Yes, if using the "right" > tools would serve only to obscure the underlying concepts, and foster > reliance on someone else's pre-packaged skill. Use of a tool does not necessarily obscure the underlying concepts. It can, certainly, but even if it does, does everyone need to implement everything? Where do you draw the line? Can you only use a compiler after you've written one? OK. Can you only use a microprocessor after you've fabbed one? I think everyone should understand everything (not just in their field of expertise- everything everything), but I don't think they need to do everything (even in their field). I believe strongly in students learning the basics of computer science. I think it should be an integral part of their math education from birth. Does that mean they need to keep drilling their entire lives? Maybe, if they feel they need it. Probably not, though. > Ironically, I think the founders of the open-source movement would > concede that their incredible depth of skill would not have been > possible without having been forced, at some point, to understand the > innermost workings of their primitive tools. It remains to be seen > whether or not the next generation can acquire this same skill level > without having to endure the same trial by fire. I only half agree with it. Believe me, I got in microcomputers at the ground level as a kid. I endured because I had to, and had no idea how bad off we really were. And yes, I learned a lot that people don't anymore- a lot of which is actually useful. Could I have learned it a lot faster and then got on to bigger things sooner with better tools and support. Ya damn skippy. > Yes. I believe that the programming equivalent of "stone knives > and bearskins" remains a valuable skillset that is just as important > as learning the intricacies of more complex and powerful tools. The > difficult thing about Computer Science is that the "basics" are still > among mankind's most abstract and complex creations. There are more > degrees of freedom in a programming problem than in virtually any > analogous problem in the physical world. Which is why we have to educate earlier. If you can't do arithmetic or read, you shouldn't get into college. Ditto for basic computer skills. > The story that set off this entire discussion is an amusing, but > questionable bit of Perl propoganda. Surely, but I can give examples that aren't. People who've written several parsers, but never graduated to using lex and yacc so that they could work faster and, more importantly, other people could fix the bugs caused by their stubbornness. (And, in this case, stunted development of a whole class of applications.) A certain former videogame powerhouse that refused to support developers writing code in C even after being shown that the compiler emitted the exact same code as their hotshot programmer. The billions of crappy FTP clients being written while lots of more useful, important, or even more entertaining programs aren't. > Any kind of competition can suffer from poor design. I suspect > that it is a sign of poor problem writing if competitors are loosing > time generating from scratch code that is so standardized as to have > been incorporated into a language like Perl. Problems should be > algorithmically challenging -- thought problems -- not busy work > centered around minute details of implementing I/O or string > processing. Bingo! If you lose a competition to someone using Perl, the response shouldn't be "That jerk!", it should be "Perl? I should learn that!" > There we are in agreement. Powertool users should have challenging > contests, as well. But don't bash the basics just because you're > bored with them. They are no less important now, than when you first > learned them back in the Dark Ages. I'm not bored with the basics. Some nights I think about nothing but. I just feel a >lot< of unqualified people start CS here. And many of them are little better when they leave. ab From pratte@ecn.purdue.edu Tue, 19 Sep 2000 03:35:44 -0500 (EST) Date: Tue, 19 Sep 2000 03:35:44 -0500 (EST) From: Eric Pratt pratte@ecn.purdue.edu Subject: [PLUG] ACM Programming Contest more depth.... On Tue, 19 Sep 2000, A Braunsdorf wrote: > > In message <20000919020717.A8970@cs.purdue.edu>, Dennis Brylow writes: > > > > I'm sorry your high school employed Luddites as instructors, but > > there is an equal and opposite peril on the far end of that spectrum. > > First, that was a common attitude when I was in high school- times > have changed, and that's part of my point. Fifteen years from now, > I hope the idea of drilling "basics" at the University level is as > fantastically quaint as a spelling bee would be. > > Second, it's not an equal peril. If we can advance the state of > the art instead of reinventing the wheel, that'd be better, but > you're right that... > > > One of the common misconceptions about Computer Science is the notion > > that as technology has progressed and advanced programming languages > > have proliferated, all of the more basic competencies have become less > > important. > > You're talking about something different now. I'm more about > knowing the basics than anyone. I'm an atomist by creed. I agree > fully that most students these days really don't get enough of > that, but that's not something I'd neither marginalize nor elevate > to competition for University students. > > I'd like to see students really learn the basics. I'd like to see > them learn them really early in basic math classes, and a lot of > them long before they write a line of code. > > > I guess that's OK, if you're only interested in training code > > monkeys to churn out millions of lines of braindead bloatware. > > I'm not. A lot of people are, though. Isn't that what CPTs are > for? > > > Not so in Computer Science, where it's not safe to assume everybody > > knows the basics, yet we hand them the power tools on the first day. > > "Why not? It's only software..." > > Why is it not safe? I'll agree they don't, in a lot of cases, but > computers are far from inscrutable. Unlike >anything< in nature, > people created them from the ground up in microscopic detail. If > there's anything people can grasp readily, you'd think this'd be > it. > > The problem is that computers are still used by the masses as TV > sets for watching the web. Grade schools consider being able to > use a web browser "computer literacy". > > Those kids need a better math curriculum that covers the fundamentals > of CS. These are fundamental skills to this era as much as arithmetic > was a century ago. One weekend in college is not the time to pick > that up. > > My point is, we don't see English majors at Purdue competing in > spelling bees. They're supposed to know that stuff already (yeah, > some of them don't). Ditto for math majors and their multiplication > tables. > > Are CS students just dumber? I hope not. Is the field that much > more complicated? I'll argue not, but even if it is, it means > people need to be even more prepared. Yeah, I think a lot more > people should flunk out too. Not everyone can do this, apparently. > > > This competition is for undergraduates, and first-year grad students. > > It is not meant to challenge or entertain seasoned professionals. > > I understand that, and I'm over-seasoned. :-) I just think sixth > graders ought to be doing this and supposedly winning adventurers > should be doing bigger and better stuff. > > > It's ironic that you should make this point during the Olympics. > > What are we trying to prove by swimming relay races, when motor boats > > have been the obvious way to go for decades? Are marathon runners > > just poor, misguided dolts who can't see the "big picture"? They > > really could get there a lot faster, after all, using the latest tools > > and technology. > > If your goal is to get there, hell yes the boat is the right answer. > If the goal is the journey, or the exertion, swimming is great. > > In the big picture, yeah, I do sort of think they're dolts, but I > understand why they do what they do. They're trying to make > themselves into something better and do things people haven't done > before. > > You know what? That's selfish. That's OK, and I'm not going to > stop them, but here in the land of computers, adding to the code > base of the world helps the world. We're hoping that code lives > on long after we're all dust. Yes, I'm serious. > > That's not to say I don't want to be the best I can be, nor that > I don't get a lot of satisfaction from that, but I really am in > this to save the world, indirectly by being a curmudgeon, if not > by action. :-) > > > You shouldn't fell a tree with a power saw, if you have yet to > > master steak knives. There's a lot more to building a house than > > simply pounding in the nails. And getting drawn into unexpected > > gunfights is often a sign of poor planning. > > Not every programmer is going to be a designer, just as not every > construction worker is an architect. The really cool architects > don't use hammers at all. The really cool software designers > needn't write code either. > > > In industry? No, never. In academia? Yes, if using the "right" > > tools would serve only to obscure the underlying concepts, and foster > > reliance on someone else's pre-packaged skill. > > Use of a tool does not necessarily obscure the underlying concepts. > It can, certainly, but even if it does, does everyone need to > implement everything? > > Where do you draw the line? Can you only use a compiler after > you've written one? OK. Can you only use a microprocessor after > you've fabbed one? > > I think everyone should understand everything (not just in their > field of expertise- everything everything), but I don't think they > need to do everything (even in their field). > > I believe strongly in students learning the basics of computer > science. I think it should be an integral part of their math > education from birth. > > Does that mean they need to keep drilling their entire lives? > Maybe, if they feel they need it. Probably not, though. > > > Ironically, I think the founders of the open-source movement would > > concede that their incredible depth of skill would not have been > > possible without having been forced, at some point, to understand the > > innermost workings of their primitive tools. It remains to be seen > > whether or not the next generation can acquire this same skill level > > without having to endure the same trial by fire. > > I only half agree with it. Believe me, I got in microcomputers at > the ground level as a kid. I endured because I had to, and had no > idea how bad off we really were. And yes, I learned a lot that > people don't anymore- a lot of which is actually useful. Could I > have learned it a lot faster and then got on to bigger things sooner > with better tools and support. Ya damn skippy. > > > Yes. I believe that the programming equivalent of "stone knives > > and bearskins" remains a valuable skillset that is just as important > > as learning the intricacies of more complex and powerful tools. The > > difficult thing about Computer Science is that the "basics" are still > > among mankind's most abstract and complex creations. There are more > > degrees of freedom in a programming problem than in virtually any > > analogous problem in the physical world. > > Which is why we have to educate earlier. If you can't do arithmetic > or read, you shouldn't get into college. Ditto for basic computer > skills. > > > The story that set off this entire discussion is an amusing, but > > questionable bit of Perl propoganda. > > Surely, but I can give examples that aren't. > > People who've written several parsers, but never graduated to using > lex and yacc so that they could work faster and, more importantly, > other people could fix the bugs caused by their stubbornness. > (And, in this case, stunted development of a whole class of > applications.) > > A certain former videogame powerhouse that refused to support > developers writing code in C even after being shown that the compiler > emitted the exact same code as their hotshot programmer. > > The billions of crappy FTP clients being written while lots of more > useful, important, or even more entertaining programs aren't. > > > Any kind of competition can suffer from poor design. I suspect > > that it is a sign of poor problem writing if competitors are loosing > > time generating from scratch code that is so standardized as to have > > been incorporated into a language like Perl. Problems should be > > algorithmically challenging -- thought problems -- not busy work > > centered around minute details of implementing I/O or string > > processing. > > Bingo! If you lose a competition to someone using Perl, the response > shouldn't be "That jerk!", it should be "Perl? I should learn > that!" > > > There we are in agreement. Powertool users should have challenging > > contests, as well. But don't bash the basics just because you're > > bored with them. They are no less important now, than when you first > > learned them back in the Dark Ages. > > I'm not bored with the basics. Some nights I think about nothing > but. I just feel a >lot< of unqualified people start CS here. > And many of them are little better when they leave. > > ab > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From will@physics.purdue.edu Tue, 19 Sep 2000 04:33:35 -0500 Date: Tue, 19 Sep 2000 04:33:35 -0500 From: Will Andrews will@physics.purdue.edu Subject: [PLUG] ACM Programming Contest On Tue, Sep 19, 2000 at 03:35:44AM -0500, Eric Pratt wrote: > more depth.... Aw, stuff it. CmpE/CS majors are *always* deep. :-) That's why girls like us. -- Will Andrews GCS/E/S @d- s+:+ a--- C++ UB++++$ P+ L- E--- W+ N-- !o ?K w--- O- M+ V- PS+ PE++ Y+ PGP+>+++ t++ 5 X+ R+ tv+ b++ DI+++ D+ G++ e>++++ h! r- y? From pfitzge1@purdue.edu Tue, 19 Sep 2000 07:12:03 -0500 (EST) Date: Tue, 19 Sep 2000 07:12:03 -0500 (EST) From: patrick.n.fitzgerald.1 pfitzge1@purdue.edu Subject: [PLUG] ACM Programming Contest Okay, gents. Let's keep it clean. No head-butting, no blows below the belt, and no semiautomatic weapons on the mailing list please. Shake hands, go back to your corners and come out fighting! (/me pops some popcorn.. love to watch a good flamewar!) Patrick Fitzgerald Official Plug Videographer and Keeper of the Sacred Flame -- Gravity is a myth, the Earth sucks. --fortune (5) From ab@eas.purdue.edu Tue, 19 Sep 2000 09:02:55 -0500 Date: Tue, 19 Sep 2000 09:02:55 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] ACM Programming Contest In message , "pa trick.n.fitzgerald.1" writes: > > (/me pops some popcorn.. love to watch a good flamewar!) Flamewar? Heavens no. If this were a flamewar, you'd know. I've won awards. :-) ab From ab@eas.purdue.edu Tue, 19 Sep 2000 09:06:32 -0500 Date: Tue, 19 Sep 2000 09:06:32 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] ACM Programming Contest In message , Eric Pratt writes: > Damn.... you guys are deep! Damn... you need to trim your quotes! ab From cnd@ecn.purdue.edu Tue, 19 Sep 2000 09:36:30 -0500 Date: Tue, 19 Sep 2000 09:36:30 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Want to test a Magneto Optical Drive? Here's the deal. I have one of these to give away to someone who is willing to review the drive. Olympus/Verbatim Magneto Optical Drive 640MB External Ultra-SCSI Bootable What you need to have: A computer A scsi card that can do Ultra-SCSI external What you need to do is send me a detailed description of what you will do to test the drive. We want to know OS compatibilty with Linux and other Unices and a bunch of other stuff. Benchmark the drive with bonnie. Make it bootable and boot it with different OSes. Partition the drive and see what it's capable of from that stand point. And anything else you can think of. The lucky winner will get to test the drive and when all is said and done, be able to keep it. The drive retails for $350 and 640MB MO media is $20. Not too bad. The drive is also backwards compatible with older MO media from 128MB to 640MB. Olympus/Verbatim will have you fill out a short questionaire to complete the agreement of the test, but we have also added on this other stuff because we want to start publishing reviews about cool stuff. The more reviews we have, the more hardware people will give us to test and keep. You need to mail me TODAY. No late comers, sorry. My mail goes out to Verbatim this afternoon. -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From pweber1@purdue.edu Tue, 19 Sep 2000 11:23:13 -0500 Date: Tue, 19 Sep 2000 11:23:13 -0500 From: peter pweber1@purdue.edu Subject: [PLUG] Want to test a Magneto Optical Drive? i would love to do this, though offhand i'm not sure if my hardware is up to spec. i have serveral machines that i could try it on, a x86 *n?x box that has an AHA1542CF (i can probably borrow a newer card), a slightly older Mac compatible that's dual-booting LinuxPPC and MacOS, and a NeXTstation. sorry i don't have great compatibility info, i'm not a SCSI afficianado, just an occasional user. :) let me know if you're interested in having me do some testing. peter At 9:36 AM -0500 9/19/00, Christopher N. Deckard wrote: >Here's the deal. I have one of these to give away to someone who is >willing to review the drive. > >Olympus/Verbatim Magneto Optical Drive >640MB >External Ultra-SCSI >Bootable > >What you need to have: >A computer >A scsi card that can do Ultra-SCSI external > >What you need to do is send me a detailed description of what you will >do to test the drive. We want to know OS compatibilty with Linux and >other Unices and a bunch of other stuff. Benchmark the drive with >bonnie. Make it bootable and boot it with different OSes. Partition >the drive and see what it's capable of from that stand point. And >anything else you can think of. > >The lucky winner will get to test the drive and when all is said and >done, be able to keep it. > >The drive retails for $350 and 640MB MO media is $20. Not too bad. The >drive is also backwards compatible with older MO media from 128MB to >640MB. > >Olympus/Verbatim will have you fill out a short questionaire to complete >the agreement of the test, but we have also added on this other stuff >because we want to start publishing reviews about cool stuff. The more >reviews we have, the more hardware people will give us to test and >keep. > >You need to mail me TODAY. No late comers, sorry. My mail goes out to >Verbatim this afternoon. > >-Chris >PLUG President > >-- >---------------------------------------------------------------------- > Christopher N. Deckard | Lead Web Technician > cnd@ecn.purdue.edu | Engineering Computer Network > http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ >---------------------------------------------------------------------- > >____________________________________________________ >The Purdue Linux Users' Group (PLUG) mailing list. >plug mailing list - plug@csociety.purdue.edu >http://csociety.ecn.purdue.edu/mailman/listinfo/plug -- ------------------------------------------------------------------ peter weber "To know that we know what we know, N9AZ and that we do not know what we do pweber1@purdue.edu not know, that is true knowledge." -Confucius ------------------------------------------------------------------ From cnd@ecn.purdue.edu Tue, 19 Sep 2000 13:24:37 -0500 Date: Tue, 19 Sep 2000 13:24:37 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] (updated) Want to test a Magneto Optical Drive? Here's the deal. I have one of these to give away to someone who is willing to review the drive. Olympus/Verbatim Magneto Optical Drive 640MB External Ultra-SCSI Bootable What you need to have: A computer A scsi card that can do Ultra-SCSI external What you need to do is send me a detailed description of what you will do to test the drive. We want to know OS compatibilty with Linux and other Unices and a bunch of other stuff. Benchmark the drive with bonnie. Make it bootable and boot it with different OSes. Partition the drive and see what it's capable of from that stand point. And anything else you can think of. The lucky winner will get to test the drive and when all is said and done, be able to keep it. The drive retails for $350 and 640MB MO media is $20. Not too bad. The drive is also backwards compatible with older MO media from 128MB to 640MB. Olympus/Verbatim will have you fill out a short questionaire to complete the agreement of the test, but we have also added on this other stuff because we want to start publishing reviews about cool stuff. The more reviews we have, the more hardware people will give us to test and keep. You need to mail me TODAY. No late comers, sorry. My mail goes out to Verbatim this afternoon. One last thing. You also need to be a dues paying member. Like I said, opportunity for free stuff if you're a member. You can however pay your dues now and get in on this special deal. :-) -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From wirges@expert.cc.purdue.edu Tue, 19 Sep 2000 17:59:31 -0500 Date: Tue, 19 Sep 2000 17:59:31 -0500 From: Matt wirges@expert.cc.purdue.edu Subject: [PLUG] More Debian probs OKay, how about this one.. Sometimes when I type xinit and it goes to start the x-server the whole machine just resets.. Sometimes I can go into X first, then after a bit it will just reset... I would say hardware problem, but none of the other distros I have used on it have any problems remotely similiar to any of the ones I have from debian -- ============================================================================== | Matthew Wirges | | Systems Administrator, Student, Programmer, Geek. | | Office Phone: [765]464-1148 | | Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com | | www.sharpwebinnovations.com | ============================================================================== From meuserj@purdue.edu Tue, 19 Sep 2000 17:56:01 -0500 Date: Tue, 19 Sep 2000 17:56:01 -0500 From: John C Meuser meuserj@purdue.edu Subject: [PLUG] More Debian probs Matt wrote: > > OKay, how about this one.. > Sometimes when I type xinit and it goes to start the x-server the whole > machine just resets.. Sometimes I can go into X first, then after a bit > it will just reset... I would say hardware problem, but none of the > other distros I have used on it have any problems remotely similiar to > any of the ones I have from debian > I had a similar problem for a while with Windows. For no appearant reason what-so-ever, the machine would reset. It wasn't purely a hardware problem, because Linux ran for days with no problems. I never did figure it out, it didn't incovenience me much, because I very rarely ran Windows. Since then I've gotten a new motherboard and haven't had any problems. John From sdh@purdue.edu Tue, 19 Sep 2000 17:58:27 -0500 Date: Tue, 19 Sep 2000 17:58:27 -0500 From: Seth Heckard sdh@purdue.edu Subject: [PLUG] More Debian probs John C Meuser wrote: > I had a similar problem for a while with Windows. For no appearant > reason what-so-ever, the machine would reset. It wasn't purely a > hardware problem, because Linux ran for days with no problems. Welcome to the wonderful world of Windows :-) Seth From lev@purdue.edu Tue, 19 Sep 2000 22:01:08 -0500 Date: Tue, 19 Sep 2000 22:01:08 -0500 From: Lev Gorenstein lev@purdue.edu Subject: [PLUG] ACM Programming Contest On Tue, 19 Sep 2000, Default wrote: D> By relying on the calculator, you deprive yourself of the chance to learn the D> things that will help you solve problems. I don't care how smart you are, you D> can't think with an empty head. Drilling the basics is vital. If you doubt that, D> just watch the checkout clerks when the power goes out or the cash registers go D> down. The young ones can't count change; there're no buttons! I should certainly agree with this. I've seen some (well... a lot) of freshmen General Chemistry students, who would (no, I'm not kidding) reach for the calculator in order to multiply a number by 10. Because they didn't know the elementary basics. Well, actually, deep inside they did know. After getting the result off the screen, some of them would pause and say "DUH!". But the point is that long search for calculator in the bag was the first thing they did. NOT evaluating the problem, but rather grabbing a calc immediately. And that habit of "calculations always mean calculator" had made them soft ;-). This is a basic stuff. I personally wouldn't remember a rule of division by 11 ;-), but multiplication by 10's got to be instantaneous. The art of "estimate the number of" is also almost entirely gone among younger people. The ability to do a quick and dirty ball-park estimate is in my opinion one of the very important skills for engineers and scientists. You know, the estimate that you do without calculator, substituting 3 for Pi, etc. I've given this type of problems to my gen. chem students (e.g. "estimate the number of atoms in the average human body") - and man, this was the toughest job! After which they would bring me the answers with 10 significant figures ('cause their displays showed that much...). Regards, Lev -- If you are going to try cross-country skiing, start with a very small country. From ab@eas.purdue.edu Wed, 20 Sep 2000 09:02:16 -0500 Date: Wed, 20 Sep 2000 09:02:16 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] Want to test a Magneto Optical Drive? In message , peter writes: > i would love to do this, though offhand i'm not sure if my hardware > is up to spec. i have serveral machines that i could try it on, a x86 > *n?x box that has an AHA1542CF (i can probably borrow a newer card), > a slightly older Mac compatible that's dual-booting LinuxPPC and > MacOS, and a NeXTstation. Probably no ultra SCSI there, but NeXTs have nice tools to display all the parameters of SCSI drives. I often use mine to troubleshoot SCSI problems. Between that and the Adaptec stuff, you ought to be able to make sure it's doing SCSI stuff right. (So it'll work as well as any SCSI device does on your system.) Depending on what version of MacOS you're running, it can be picky about SCSI devices. They've lightened up a lot, but it's useful for some people to know whether devices work right with older Mac stuff. It didn't help that some older SCSI devices really >were< messed up, so it's not all MacOS's fault. :-) ab From ab@eas.purdue.edu Wed, 20 Sep 2000 10:18:33 -0500 Date: Wed, 20 Sep 2000 10:18:33 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] ACM Programming Contest In message , Le v Gorenstein writes: > > I should certainly agree with this. I've seen some (well... a lot) > of freshmen General Chemistry students, who would (no, I'm not > kidding) reach for the calculator in order to multiply a number by > 10. Because they didn't know the elementary basics.... > > The art of "estimate the number of" is also almost entirely gone > among younger people. The ability to do a quick and dirty > ball-park estimate is in my opinion one of the very important > skills for engineers and scientists. And they'll lose versus people who do. There's a competitive advantage to having these skills. They'll win out. If they're really needed, that is. If they aren't, maybe their importance is exaggerated. I think the students are soft because the schoolwork is soft. If being able to estimate would make things easier and faster to a degree that it affected their test scores or course grades, the cream would rise to the top. Make your tests multiple choice (so the estimators don't need to really work them all) and triple the number of questions. See what that does. When I was a kid, calculators were fairly exotic. Only accountants and other people who got paid to do arithmetic bothered with the expense. My father had one (a Director II, which although paperless, you plugged into the wall) because he's a science teacher and tax preparer. So did my grandfather the bookie (who handicapped, in the mathematical not mechanical manner) horses. The widespread use of calculators probably has undermined basic skills, but it's also made arithmetic commonplace. People who wouldn't have bothered to do it at all thirty years ago now will. That's a good thing. Ditto with computers. Computers were first used by people to do what they did better. Now they're used by people to do things they wouldn't be doing without them. The Internet is the same story. Are people doing things they don't really understand. Oh yeah. Big time. Do we want them to stop? Some of them. The answer academically is to keep pushing forward. Consider basic skills a "gimme" and build on top of them mercilessly. The foundation will crack if it doesn't hold. If it holds, maybe it matters less then you think. ab From notz@purdue.edu Wed, 20 Sep 2000 10:38:04 -0500 (EST) Date: Wed, 20 Sep 2000 10:38:04 -0500 (EST) From: Patrick K Notz notz@purdue.edu Subject: [PLUG] ACM Programming Contest >> >> The art of "estimate the number of" is also almost entirely >> gone among younger people. The ability to do a quick and dirty >> ball-park estimate is in my opinion one of the very important >> skills for engineers and scientists. I can personally attest to the value of 'ball-park' figuring. I'm fortunate enough to have an advisor who takes me (and my groupmates) along when he visits the companies we collaborate with. Typically, we start by reporting on our reseach progress and then the conversation usually takes off from there. We have a lot of 'what-if' / 'would it be possible to...' type of conversations where quick and dirty estimates can lead to research programs forming or dying. We're usually looking at estimating force balances and that sort of thing. When you do this sort of thing, gravity becomes 10 (not 9.8... m/s^2), Pi becomes 10^1/2, etc. If these talks happen over dinner and the check comes before you've got a clue, then you're probably looking elswhere for funding. You don't know what will come up so you have to be able to do any kind of figuring (force balances, economics, macroscopic balances, even rigorous derivations) on command. No books, no Maple, and no computers. The other side of this is perception. Wouldn't you rather hire someone who actually KNOWS and UNDERSTANDS at a deep level the things you're working with? Sure, you'll always have your books and you'll always have access to computers, but EVERYONE does. Thinking and figuring on your toes will set you appart from the 200 other people with the same degree and the same inflated GPA. Your skills should become your second nature, not a set of books and software titles. When you limit yourself to ideas which have already made it to publications (software, paper, or whatever) then you're limited by existing ideas. $0.02 From wallacms@cs.purdue.edu Wed, 20 Sep 2000 11:31:54 -0500 (EST) Date: Wed, 20 Sep 2000 11:31:54 -0500 (EST) From: Matt Wallace wallacms@cs.purdue.edu Subject: [PLUG] ACM Programming Contest > The answer academically is to keep pushing forward. Consider basic > skills a "gimme" and build on top of them mercilessly. The foundation > will crack if it doesn't hold. If it holds, maybe it matters less > then you think. I realize that this thread has gone way beyond just the programming contest, but I did want to point out that the main purpose of the programming contest is fun. So whether or not it has redeeming social value is not real important to me. -Matt From ab@eas.purdue.edu Wed, 20 Sep 2000 13:03:43 -0500 Date: Wed, 20 Sep 2000 13:03:43 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] ACM Programming Contest In message , Matt Wallace writes: > > I realize that this thread has gone way beyond just the programming > contest, but I did want to point out that the main purpose of the > programming contest is fun. Oh, I agree with that too. Contests are supposed to be fun, and they aren't all for everyone. Lots of different kinds of contests are needed. I'd just like to see the big programming contest (and this is the big one here, right?) have some more social value. That's the kind of guy I am. :-) I couldn't find any programmers I know who had much good to say about the contest. True, most of them (us, is probably fair) are snobs and crusty old boogers like me, but that's discouraging. I'd like to see something where I and my colleagues would be really pumped about sending a team into battle. > So whether or not it has redeeming social > value is not real important to me. In the particular, it's not to me either. It just smells like a symptom of greater problems which are really important to me. ab From rjune@ims1.imagestream-is.com Thu, 21 Sep 2000 11:21:45 -0500 (EST) Date: Thu, 21 Sep 2000 11:21:45 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] Printers. I'm looking to buy a printer, can anyone recommend one that works with linux(color isn't required, but would be nice)? if any of these are known to work specificly, that would be cool. Lexmark Z52/Z42/Z31 HP 648C/842C/895Cse/932C Epson SC900/SC1160/SC760 From calx@purdue.edu Thu, 21 Sep 2000 11:26:01 -0500 (EST) Date: Thu, 21 Sep 2000 11:26:01 -0500 (EST) From: David calx@purdue.edu Subject: [PLUG] Printers. I recently bought an Epson Stylus 740. Works great and can use USB as well. It was also only $99 at BestBuy. David From rocky@purdue.edu Thu, 21 Sep 2000 11:39:50 -0500 Date: Thu, 21 Sep 2000 11:39:50 -0500 From: Rocky Scaletta rocky@purdue.edu Subject: [PLUG] Printers The best resource for this is sort of thing is          http://www.linuxprinting.org/database.html They've got every printer ever made (it seems) with detailed compatibility information.  In any case, every printer I've ever looked up can be _made_ to work with linux, eventually. On Thu, 21 Sep 2000, you wrote: > I'm looking to buy a printer, can anyone recommend one that works with > linux(color isn't required, but would be nice)? if any of these are known > to work specificly, that would be cool. > > Lexmark Z52/Z42/Z31 > HP 648C/842C/895Cse/932C > Epson SC900/SC1160/SC760 > > > _____________________________________________ -- THE INFINITE AND BEYOND "My God, it's full of stars..." -Dave Bowman, as he approached the Jovian Monolith 2001. . . it awaits us all -Rocky. From sdh@purdue.edu Thu, 21 Sep 2000 11:36:25 -0500 Date: Thu, 21 Sep 2000 11:36:25 -0500 From: Seth Heckard sdh@purdue.edu Subject: [PLUG] Printers. Richard June wrote: > > I'm looking to buy a printer, can anyone recommend one that works with > linux(color isn't required, but would be nice)? if any of these are known > to work specificly, that would be cool. > > Lexmark Z52/Z42/Z31 > HP 648C/842C/895Cse/932C > Epson SC900/SC1160/SC760 > If you can find one, go for the Lexmark Optra 40. It groks postscript and it's an inkjet. Fully compatible with linux of course (nothing better than going cat blah.ps > /dev/lp0). However, it has a couple of drawbacks. One is that it only comes with 4MB RAM, which is fine for text but is severly lacking for anything else. I just went to salvage and picked up an 8MB SIMM and all is well now (it has 1 72-pin SIMM socket in the back). Also, color printing really, really sucks. But black & white text, on the other hand, is beautiful... the best I've ever seen on an inkjet printer. Of course, the best part is the price: I only paid $99 about 6 months ago. But the lack of photo-quality color bothers me sometimes. Seth From cnd@ecn.purdue.edu Thu, 21 Sep 2000 11:48:37 -0500 Date: Thu, 21 Sep 2000 11:48:37 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Printers. My personal experience would be to stay away from Lexmark printers. They are great printers, but some don't work under Linux. Some don't even work under Windows NT. I'd go with HP. They seem to be supported most of the time in Linux. -Chris Richard June wrote: > > I'm looking to buy a printer, can anyone recommend one that works with > linux(color isn't required, but would be nice)? if any of these are known > to work specificly, that would be cool. > > Lexmark Z52/Z42/Z31 > HP 648C/842C/895Cse/932C > Epson SC900/SC1160/SC760 > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From ab@eas.purdue.edu Thu, 21 Sep 2000 11:58:48 -0500 Date: Thu, 21 Sep 2000 11:58:48 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] Printers. In message <39CA3BE5.B84510CC@ecn.purdue.edu>, "Christopher N. Deckard" writes: > My personal experience would be to stay away from Lexmark printers. > They are great printers, but some don't work under Linux. Some don't > even work under Windows NT. I'd go with HP. They seem to be supported > most of the time in Linux. Linux, Linux, Linux. You guys, geez. This isn't a Linux problem, is it? It better not be. Linux doesn't support any printers, right? It's gs, or lpd, or ifhp, or the device drivers for the ports you need to worry about. If the printer speaks a "standard" language of some sort, and you can get a communication channel to it, you should be OK. If it uses some kind of proprietary protocol, you shouldn't be using it. The guys who make those should all be roasting in hell. Some of the MS Windowscentric printers these days I guess are like that, but, thankfully, that's out of style these days. I'd stick to a PostScript printer myself, but I'm the PostScript guy. :-) The more recent PCL printers, at least the HP-made ones are consistent enough, and you can feed them from GhostScript to make them play ball with the world of PostScript. For non-PostScript printers, I'd check out what gs supports these days and pick one of those that has the features you want and the right kind of interface for your computer (don't get a USB printer if you don't have a USB port, for example). ab From cnd@ecn.purdue.edu Thu, 21 Sep 2000 12:09:08 -0500 Date: Thu, 21 Sep 2000 12:09:08 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Printers. A Braunsdorf wrote: > > In message <39CA3BE5.B84510CC@ecn.purdue.edu>, "Christopher N. Deckard" writes: > > My personal experience would be to stay away from Lexmark printers. > > They are great printers, but some don't work under Linux. Some don't > > even work under Windows NT. I'd go with HP. They seem to be supported > > most of the time in Linux. > > Linux, Linux, Linux. You guys, geez. This isn't a Linux problem, > is it? It better not be. No. The problem is that companies are lazy and don't make REAL printers. So the printers that they do make, the less expensive ones HAVE to have special drivers to make them work. Pretty much any printer that works in Win NT will work with a real OS as well, but it is not the same with the other Windows OSes. The entire printer systems are different between NT and 9x. In 9x you get nifty little programs that use all of the CPU and don't let you do any other work while printing. > Linux doesn't support any printers, right? It's gs, or lpd, or > ifhp, or the device drivers for the ports you need to worry about. > > If the printer speaks a "standard" language of some sort, and you > can get a communication channel to it, you should be OK. That's the problem. SOme don't speak a standard language. I think it's pig greek. -Chris From cnd@ecn.purdue.edu Thu, 21 Sep 2000 12:16:50 -0500 Date: Thu, 21 Sep 2000 12:16:50 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] PHP - Richard Lynch, Zend Technologies Monday, September 25th, 2000 7pm - 9pm Physics 112 Richard Lynch from Zend Technologies, the authors of the PHP Zend Optimizer, will be speaking about PHP and writing web applications using PHP on Monday night. He is going to develop a simple web application during the presentation. PHP is the PHP Hypertext Preprocessor. It is used for writing dynamically generated and database driven websites. Here's the description from the website: PHP Version 3.0 is an HTML-embedded scripting language. Much of its syntax is borrowed from C, Java and Perl with a couple of unique PHP-specific features thrown in. The goal of the language is to allow web developers to write dynamically generated pages quickly. Related Links: http://www.zend.com/ - Zend Technologies http://www.php.net/ - PHP Main Website Punch and pie will be available for refreshments. -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From zamboni@cerias.purdue.edu Thu, 21 Sep 2000 13:13:44 -0500 Date: Thu, 21 Sep 2000 13:13:44 -0500 From: Diego Zamboni zamboni@cerias.purdue.edu Subject: [PLUG] Printers Speaking of printers... I'd be very interested in hearing about experiences with network-capable inkjet printers, like the Epson 850N or 900N. I have a fairly heterogeneous network at home (including Sparcs, a Mac and a NeXT, but no PCs so far), so I have been thinking that a network printer might be the best thing to serve all of them, since it doesn't depend on any of the machines. But they are quite more expensive than non-network printers. Any thoughts? --Diego From meuserj@purdue.edu Thu, 21 Sep 2000 13:22:17 -0500 (EST) Date: Thu, 21 Sep 2000 13:22:17 -0500 (EST) From: John C Meuser meuserj@purdue.edu Subject: [PLUG] Printers. I heard about something but haven't tried it. There is SOME support for Epson Stylus, but it isn't complete. But I have heard (someplace, not sure where) that there is a Gimp plugin that can talk to gs that will fully support many inkjet printers that had only so-so support in *nix before. Just a thought. John From deepak@purdue.edu Sat, 23 Sep 2000 11:03:34 -0500 (EST) Date: Sat, 23 Sep 2000 11:03:34 -0500 (EST) From: Deepak Dinesh deepak@purdue.edu Subject: [PLUG] Locale Problem with 7.0 Fixed Hi , The fix for the locale problem in Mandrake 7.0 is here: http://www.linux-mandrake.com/en/updates/MDKA-2000-007.php3?dis=7.0 Cheers. From deepak@purdue.edu Sat, 23 Sep 2000 11:13:43 -0500 (EST) Date: Sat, 23 Sep 2000 11:13:43 -0500 (EST) From: Deepak Dinesh deepak@purdue.edu Subject: [PLUG] User Agent with Junkbuster Hi, The junkbuster proxy (http://www.junkbusters.com/) claims to hide the UserAgent value of a browser. It appears that it doesn't work for https. Anyone have any experience with this ? Is it possible to hide user agent when connecting to a secure server ? Cheers. From doobie@doobie.org Sat, 23 Sep 2000 13:47:45 -0500 (EST) Date: Sat, 23 Sep 2000 13:47:45 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] User Agent with Junkbuster If you want it to be secure, no. Everything including the headers is encrypted inside the ssl protocol, so you'de need to do SSL to the juncklbuster server, then have junk buster send out a new request....not 100% sure how junkbusters works so it may somehow be possible. Just hack the binary for your web browser and set it to whatever you want it to be. jason On Sat, 23 Sep 2000, Deepak Dinesh wrote: > Hi, > The junkbuster proxy (http://www.junkbusters.com/) claims to > hide the UserAgent value of a browser. It appears that it > doesn't work for https. Anyone have any experience with this ? > Is it possible to hide user agent when connecting to a secure > server ? > Cheers. > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From deepak@purdue.edu Sat, 23 Sep 2000 15:10:32 -0500 (EST) Date: Sat, 23 Sep 2000 15:10:32 -0500 (EST) From: Deepak Dinesh deepak@purdue.edu Subject: [PLUG] User Agent with Junkbuster Hi , I should have RTFM before I wrote. From the faq: If you enter a ``Secure Document Area,'' cookies and other header information such as User Agent and Referer are sent encrypted, so they cannot be filtered. Cheers, Deepak Dinesh. On Sat, 23 Sep 2000, Jason Dubrow wrote: > >If you want it to be secure, no. Everything including the headers is >encrypted inside the ssl protocol, so you'de need to do SSL to the >juncklbuster server, then have junk buster send out a new request....not >100% sure how junkbusters works so it may somehow be possible. Just hack >the binary for your web browser and set it to whatever you want it to be. > >jason > >On Sat, 23 Sep 2000, Deepak Dinesh wrote: > >> Hi, >> The junkbuster proxy (http://www.junkbusters.com/) claims to >> hide the UserAgent value of a browser. It appears that it >> doesn't work for https. Anyone have any experience with this ? >> Is it possible to hide user agent when connecting to a secure >> server ? >> Cheers. >> >> >> ____________________________________________________ >> The Purdue Linux Users' Group (PLUG) mailing list. >> plug mailing list - plug@csociety.purdue.edu >> http://csociety.ecn.purdue.edu/mailman/listinfo/plug >> > From david_hansen@ibm.net Sun, 24 Sep 2000 13:57:00 -0500 Date: Sun, 24 Sep 2000 13:57:00 -0500 From: David Christopher Hansen david_hansen@ibm.net Subject: [PLUG] Web caching I wanted to know everyone feels about the web caching server that has been installed. For those of you who don't know, there is a machine on campus that caches all of your web requests. It, of course, only affects those who use the university internet connection. Its purpose is to alleviate some of the network load. If 50 people request www.yahoo.com/ , it will only be fetched once, the next 49 people will get a copy. What are the consequences of having this thing in the way? I remember last year when they were installing it, we got lots of strange errors because of it. Pages would fail to load, and site a cryptic error message that the caching server generated. Has anyone noticed any problems? BTW, look at these two pages: one is on campus, the other is not. Look at the REMOTE_ADDR field. http://icdweb.cc.purdue.edu/~hansendc/ct.cgi http://hoohoo.ncsa.uiuc.edu/cgi-bin/test2.cgi ------------------------ | David C. Hansen | dave@sr71.net ------------------------ From doobie@doobie.org Sun, 24 Sep 2000 14:23:14 -0500 (EST) Date: Sun, 24 Sep 2000 14:23:14 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] Web caching Good side: less bandwidth from the web cache to the rest of the world. you can big a better 'big brother' (good if you are the univ, bad for students) Bad: * won't work on non-cacheable pages (aka dynamic, aka my.yahoo.com, most large sites), however it'll work on their images, which is a big plus * adds an extra layer of indirection, slower * ssl probably won't work * for a university, you'll prolly need a bunch of servers to do it (aka load would be high) * are theer multiple points from the users to the pipe out of purdue? * you probably aren't going to get students to use it if they don't have to (see good side about being big brother) thats what I thought of.... On Sun, 24 Sep 2000, David Christopher Hansen wrote: > I wanted to know everyone feels about the web caching server that has > been installed. For those of you who don't know, there is a machine on > campus that caches all of your web requests. It, of course, only > affects those who use the university internet connection. Its purpose > is to alleviate some of the network load. If 50 people request > www.yahoo.com/ , it will only be fetched once, the next 49 people will > get a copy. > What are the consequences of having this thing in the way? I remember > last year when they were installing it, we got lots of strange errors > because of it. Pages would fail to load, and site a cryptic error > message that the caching server generated. Has anyone noticed any > problems? > > BTW, look at these two pages: one is on campus, the other is not. Look > at the REMOTE_ADDR field. > http://icdweb.cc.purdue.edu/~hansendc/ct.cgi > http://hoohoo.ncsa.uiuc.edu/cgi-bin/test2.cgi > ------------------------ > | David C. Hansen > | dave@sr71.net > ------------------------ > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From mds@ecn.purdue.edu Sun, 24 Sep 2000 14:39:21 -0500 (EST) Date: Sun, 24 Sep 2000 14:39:21 -0500 (EST) From: Mark Senn mds@ecn.purdue.edu Subject: [PLUG] 28.8k modem $10 PLUG, 28.8K modem for sale $10. Guaranteed. No documentation included but will give it to you later if I come across it. Standard U.S. Robotics modem. First email takes it. Mark Senn, staff advisor From rajak@purdue.edu Sun, 24 Sep 2000 15:39:55 -0500 (EST) Date: Sun, 24 Sep 2000 15:39:55 -0500 (EST) From: Brian Poole rajak@purdue.edu Subject: [PLUG] Web caching Alrightly, lets tear this apart ;) On Sun, 24 Sep 2000, Jason Dubrow wrote: > Good side: > less bandwidth from the web cache to the rest of the world. I don't think this point can be overemphasized. Purdue's bandwidth to the real world for ResNet is _constantly_ peaked this semester even though they have dedicated a T3 line to it. They _must_ conserve b/w in all forms if they want anything to get through. Go take a look at PDN's MRTG if you haven't already, you might appreciate the situation a bit more (http://www.purdue.edu/PUCC/PDN/stats/summary/offcampus.html). Albeit there are other steps they should perhaps be taking to reduce the bandwidth load *cough*mp3s*cough*warez*cough*, but that is another subject we won't get into. The point is they are trying to reduce the b/w used, which is an excellent idea and one I am very glad for. > you can big a better 'big brother' (good if you are the univ, bad for > students) You can big a better? erg, that one doesn't quite make sense.. Maybe 'you can build a better' big brother, but I don't put much faith in that one, all they can do is a bit of generalization considering all the traffic that is going through them and I'm sure they could already do that if they so desired. If they want to spy on you they have always been fully capable given that they control every point of access. > > Bad: > * won't work on non-cacheable pages (aka dynamic, aka my.yahoo.com, most > large sites), however it'll work on their images, which is a big plus It works fine, it just passes the traffic through. What it won't do is be able to cache the page, but thats fine, it is not any worse than what Netscape does. It just adds the bit of indirection that you speak of below. This isn't worthy of its own bullet in the bad section.. > * adds an extra layer of indirection, slower Yes, there is certainly a bit of processing it has to wait for, the question is, does this layer outweigh the speed increase you just gained from caching pages & images? In my experience web caches _generally_ make browsing faster after the cache builds. There are limited cases where like dynamic pages that it doesn't make the page faster, but it still does cache the images, which is probably the largest part anyways. > * ssl probably won't work SSL works fine in my experience, I would imagine it just doesn't cache it. > * for a university, you'll prolly need a bunch of servers to do it (aka > load would be high) Shrug, buying new computers has never seemed to be a problem. I'm sure they can handle spending a bit of money on new hardware. I've seen 2 hostnames from various browsings in the past, so I imagine there aren't that many servers hanging around doing this. I haven't noticed any lag browsing random sites that shouldn't be in the cache so I would guess that the cache servers aren't that overloaded either. > * are theer multiple points from the users to the pipe out of purdue? That is an interesting point. There are multiple ways for people to get out of Purdue and I'm not really sure which they cache, if not all. I'd imagine both T3s to UUNet as well as the Abilene's. The others probably are not as important given their saturation is always relatively low. But why is this in the bad section? It is just an interesting question to me. > * you probably aren't going to get students to use it if they don't have > to (see good side about being big brother) Sure, but I imagine that is also just a kneejerk reflex. Uninformed people make uninformed decisions. If it saves b/w then I'm happy for it, as long as it isn't constantly causing problems. This was a bit of a problem when they first put it up as was mentioned by the initiator of this thread, but for the most part it seems stable to me. I haven't had many problems with it. Basically from the points above it boils down to this. It saves bandwidth by not having to fetch pages constantly, however it does have to be processed by a web cache which adds a bit of delay and another possible point of failure. Given our situation, where our bandwidth is highly utilized (at least for ResNet), I am in the end pleased that they are caching. I also think is a network-friendly decision they have made, it benefits the Internet as a whole if people use caches. This was perhaps first illustrated with popular browser's offering cache files and is simply taken to the next step, where you get a much higher effectiveness by utilizing the same cache for multiple users. I'l admit the first time I saw that there was a web cache I was against it but having thought through the issue and having looked at our network statistics I've changed my mind to what I feel is a more informed decision. -b From brent@purdue.edu Sun, 24 Sep 2000 15:50:25 -0500 Date: Sun, 24 Sep 2000 15:50:25 -0500 From: Brent Meshier brent@purdue.edu Subject: [PLUG] Web caching For purdueonline.com, which is outside Purdue's network, I greatly benefit from the web caching system (lucre.tcom / dinero.tcom). Although much of our content is dynamic, images are not. Even more to the point, 90% of our traffic originates from within PDN, therefore caching greatly improves performance and lowers bandwidth consumption. The majority of requests look like this: lucre.tcom.purdue.edu - - [24/Sep/2000:15:42:16 -0500] "GET /images/gr.gif HTTP/1.1" 304 - lucre.tcom.purdue.edu - - [24/Sep/2000:15:42:16 -0500] "GET /images/wr.gif HTTP/1.1" 304 - lucre.tcom.purdue.edu - - [24/Sep/2000:15:42:24 -0500] "GET /images/purdueonline_logo.gif HTTP/1.1" 304 - lucre.tcom.purdue.edu - - [24/Sep/2000:15:42:24 -0500] "GET /images/topic_top.gif HTTP/1.1" 304 - lucre.tcom.purdue.edu - - [24/Sep/2000:15:42:24 -0500] "GET /images/topic_roundoff.gif HTTP/1.1" 304 - lucre.tcom.purdue.edu - - [24/Sep/2000:15:42:24 -0500] "GET /images/topics_header/header_tp_sports.gif HTTP/1.1" 304 - lucre.tcom.purdue.edu - - [24/Sep/2000:15:42:24 -0500] "GET / HTTP/1.1" 200 76245 notice the "304"s (not modified).. The only bad side I can see is that you can no longer identify requests uniquely by originating IP (unless you uses cookies of course). Just my 2 cents.. --Brent -----Original Message----- From: plug-admin@csociety.purdue.edu [mailto:plug-admin@csociety.purdue.edu]On Behalf Of David Christopher Hansen Sent: Sunday, September 24, 2000 1:57 PM To: plug@csociety.purdue.edu Subject: [PLUG] Web caching I wanted to know everyone feels about the web caching server that has been installed. For those of you who don't know, there is a machine on campus that caches all of your web requests. It, of course, only affects those who use the university internet connection. Its purpose is to alleviate some of the network load. If 50 people request www.yahoo.com/ , it will only be fetched once, the next 49 people will get a copy. What are the consequences of having this thing in the way? I remember last year when they were installing it, we got lots of strange errors because of it. Pages would fail to load, and site a cryptic error message that the caching server generated. Has anyone noticed any problems? BTW, look at these two pages: one is on campus, the other is not. Look at the REMOTE_ADDR field. http://icdweb.cc.purdue.edu/~hansendc/ct.cgi http://hoohoo.ncsa.uiuc.edu/cgi-bin/test2.cgi ------------------------ | David C. Hansen | dave@sr71.net ------------------------ ____________________________________________________ The Purdue Linux Users' Group (PLUG) mailing list. plug mailing list - plug@csociety.purdue.edu http://csociety.ecn.purdue.edu/mailman/listinfo/plug From lewis1@purdue.edu Sun, 24 Sep 2000 16:37:10 -0500 (EST) Date: Sun, 24 Sep 2000 16:37:10 -0500 (EST) From: James C. Lewis lewis1@purdue.edu Subject: [PLUG] Web caching I think you might have had a few misunderstanding or misconceptions about how web caching works... On Sun, 24 Sep 2000, Jason Dubrow wrote: > * won't work on non-cacheable pages (aka dynamic, aka my.yahoo.com, most > large sites), however it'll work on their images, which is a big plus That's why you use smart cache servers that know when to cache and when not to. Plus... web browsers can request the cache server to update its document. Only about 30% of the web traffic is cachable anyway...and smart cache boxes know that. > * ssl probably won't work Nope, SSL works just fine. Its on a different port than 80, and port 80 requests are redirected to the cache servers (at the router level). Most of the time SSL isn't cached anyway... its encrypted so what good would a cached copy of an encrypted document do you? > * for a university, you'll prolly need a bunch of servers to do it (aka > load would be high) This is already true. Different parts of campus already have their own caching server. Resnet has a different from the different parts of Purdue's network. Obviously one server isn't going to be enough. In general you'd probably want 1 cache server one each incoming pipe. For example, the ISP I work for back home, each of our POPs have a cache box sitting there... > * are theer multiple points from the users to the pipe out of purdue? Huh? I don't know what you mean on this one... If each pipe has a box on it anyway...it won't matter. Remember, purdue isn't stupid enough to just use _ONE_ cache server. > * you probably aren't going to get students to use it if they don't have > to (see good side about being big brother) Since they are probably using a transparent proxy, you're going to use it if you want to or not. So I don't see where you're going with this one. Unless of course you just happen to only visit sites using SSL... or sites that aren't on port 80... -James C. Lewis _ /_ _ _ _ / ) / _ ' _ (_/(///)(-_) (__ . (__(-((//_) PUCC Information Center Consultant jclewis@purdue.edu "Only those who dare to fail greatly can ever achieve greatly." -- Robert F. Kennedy Please place all complaints in this box --> [] From doobie@doobie.org Sun, 24 Sep 2000 17:28:29 -0500 (EST) Date: Sun, 24 Sep 2000 17:28:29 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] Web caching time to tear about the torn apart. On Sun, 24 Sep 2000, Brian Poole wrote: > Alrightly, lets tear this apart ;) > > On Sun, 24 Sep 2000, Jason Dubrow wrote: > > Good side: > > less bandwidth from the web cache to the rest of the world. > > I don't think this point can be overemphasized. Purdue's bandwidth to the > real world for ResNet is _constantly_ peaked this semester even though > they have dedicated a T3 line to it. They _must_ conserve b/w in all > forms if they want anything to get through. Go take a look at PDN's MRTG > if you haven't already, you might appreciate the situation a bit more > (http://www.purdue.edu/PUCC/PDN/stats/summary/offcampus.html). Albeit > there are other steps they should perhaps be taking to reduce the > bandwidth load *cough*mp3s*cough*warez*cough*, but that is another subject > we won't get into. The point is they are trying to reduce the b/w used, > which is an excellent idea and one I am very glad for. I can believe the traffic is quite high, but I don't think,as you said, most of the traffic is web traffic (maybe it is....i don't have the #). it in theory will reduce web traffic, however it lies on the question, what percentage of traffic is web traffic...and how much would it be reduced? forgot the name of the guy....but basically... New traffic level = ((percent web traffic) * (percent of original) + percent everything else) / 100 or something like that...i should know this stuff.... > > > you can big a better 'big brother' (good if you are the univ, bad for > > students) > > You can big a better? erg, that one doesn't quite make sense.. sorry, meant be a better big brother.... > > Maybe 'you can build a better' big brother, but I don't put much faith in > that one, all they can do is a bit of generalization considering all the > traffic that is going through them and I'm sure they could already do > that if they so desired. If they want to spy on you they have always been > fully capable given that they control every point of access. > > > > > Bad: > > * won't work on non-cacheable pages (aka dynamic, aka my.yahoo.com, most > > large sites), however it'll work on their images, which is a big plus > > It works fine, it just passes the traffic through. What it won't do is be > able to cache the page, but thats fine, it is not any worse than what > Netscape does. It just adds the bit of indirection that you speak of > below. This isn't worthy of its own bullet in the bad section.. see above theory....amdel's law...thats the name....now what is the spelling of his name? > > * adds an extra layer of indirection, slower > > Yes, there is certainly a bit of processing it has to wait for, the > question is, does this layer outweigh the speed increase you just gained > from caching pages & images? In my experience web caches _generally_ make > browsing faster after the cache builds. There are limited cases where like > dynamic pages that it doesn't make the page faster, but it still does > cache the images, which is probably the largest part anyways. > > > * ssl probably won't work > > SSL works fine in my experience, I would imagine it just doesn't cache > it. It does? That I am very suprised at ssl encrypts all the stuff interal to the encypted packets...the cache would have to have access to a key to decode it.....does 'big brother cache' steal this information? > > > * for a university, you'll prolly need a bunch of servers to do it (aka > > load would be high) > > Shrug, buying new computers has never seemed to be a problem. I'm sure > they can handle spending a bit of money on new hardware. I've seen 2 > hostnames from various browsings in the past, so I imagine there aren't > that many servers hanging around doing this. I haven't noticed any lag > browsing random sites that shouldn't be in the cache so I would guess that > the cache servers aren't that overloaded either. > > > * are theer multiple points from the users to the pipe out of purdue? > > That is an interesting point. There are multiple ways for people to get > out of Purdue and I'm not really sure which they cache, if not all. I'd > imagine both T3s to UUNet as well as the Abilene's. The others probably > are not as important given their saturation is always relatively low. But > why is this in the bad section? It is just an interesting question to me. data goes out one the first time....out the second the next time....if that is possible with the network setup...do you then keep some kind of coherency between the cache boxes? > > > * you probably aren't going to get students to use it if they don't have > > to (see good side about being big brother) > > Sure, but I imagine that is also just a kneejerk reflex. Uninformed people > make uninformed decisions. If it saves b/w then I'm happy for it, as long > as it isn't constantly causing problems. This was a bit of a problem when > they first put it up as was mentioned by the initiator of this thread, but > for the most part it seems stable to me. I haven't had many problems with > it. Doesn't bother me either way though, I'm not at Purdue anymore, and have a bigger brother than Purdue can ever be watching over me. However when I was doing webdevelopment at Purdue, it was pretty much a pain in the ass for me to do debugging, epscially when I wanted to use my IP address to do different things (aka give me a log on the web [it was secure enough for me, and it wasn't anything I would care if someone got]). Do the math, if you think the reduction could make a 10% decrease in bandwidth/fetch time I think it is well worth it. cache the mp3's would be a more practical way to cut down traffic, but that is another totally different discussion. jason From sdh@purdue.edu Sun, 24 Sep 2000 17:32:25 -0500 Date: Sun, 24 Sep 2000 17:32:25 -0500 From: Seth Heckard sdh@purdue.edu Subject: [PLUG] Web caching Jason Dubrow wrote: > > Good side: > less bandwidth from the web cache to the rest of the world. *snip* > * for a university, you'll prolly need a bunch of servers to do it (aka > load would be high) I believe Purdue has only 4 cache servers, each with one 6GB disk, that does all the caching. They were quite tiny (only 1U size case I suppose) and they were Cisco boxen (didn't check the model number). I can think of three of the hostnames off the top of my head: dinero.tcom, lucre.tcom, and moola.tcom. The explanation that I got was that it was primarily done to speed things up for the end user and not to save bandwidth per se. Supposedly it only makes a 5-10% difference in bandwidth consumed whether the cache boxes are on or not. When you think about it, Purdue's T3s are going to be hit pretty hard no matter what they do, although I can't say that I've ever felt a considerable slowdown in ResNet response. I have noticed that the cache boxes aren't always on though. Seth From doobie@doobie.org Sun, 24 Sep 2000 17:34:36 -0500 (EST) Date: Sun, 24 Sep 2000 17:34:36 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] Web caching *snip* > > I've ever felt a considerable slowdown in ResNet response. I have > noticed that the cache boxes aren't always on though. > obviously you were never at Purdue when they only had 3 t1's, one for resnet/labs, one for ECN, and one in case either of the first two died. From rajak@purdue.edu Sun, 24 Sep 2000 17:45:19 -0500 (EST) Date: Sun, 24 Sep 2000 17:45:19 -0500 (EST) From: Brian Poole rajak@purdue.edu Subject: [PLUG] Web caching > > > less bandwidth from the web cache to the rest of the world. > > I don't think this point can be overemphasized. Purdue's bandwidth to the > > real world for ResNet is _constantly_ peaked this semester even though > > they have dedicated a T3 line to it. They _must_ conserve b/w in all > > forms if they want anything to get through. Go take a look at PDN's MRTG > > if you haven't already, you might appreciate the situation a bit more > > (http://www.purdue.edu/PUCC/PDN/stats/summary/offcampus.html). Albeit > > there are other steps they should perhaps be taking to reduce the > > bandwidth load *cough*mp3s*cough*warez*cough*, but that is another subject > > we won't get into. The point is they are trying to reduce the b/w used, > > which is an excellent idea and one I am very glad for. > > I can believe the traffic is quite high, but I don't think,as you said, > most of the traffic is web traffic (maybe it is....i don't have the #). Mmm, I never said that _most_ of the traffic was web traffic. I said we are saturating our b/w and that we must take steps to alleviate it. Caching _does_ reduce b/w usage. I certainly don't believe most of the traffic is web traffic, I imagine a great deal is the illicit traffic that I mentioned, mp3s & warez. I don't know where you got that I thought that most of the traffic was web. > it in theory will reduce web traffic, however it lies on the question, > what percentage of traffic is web traffic...and how much would it be > reduced? forgot the name of the guy....but basically... ... Ehm, I think in all reality it _does_ reduce the web traffic. I don't see how you can argue that. It doesn't matter how much of the total traffic is web traffic, if it is reduced in any way it does reduce the total traffic at least some. The effectiveness of the web cache to reduce total traffic may lie on how much of the traffic is web traffic, but if it will actually reduce the web traffic simply lies on the effectiveness of the webcache itself. > > SSL works fine in my experience, I would imagine it just doesn't cache > > it. > > It does? That I am very suprised at ssl encrypts all the stuff interal to > the encypted packets...the cache would have to have access to a key to > decode it.....does 'big brother cache' steal this information? 'I would imagine it just doesn't cache it' which concurs with what others have also said, HTTPS traffic is not cached. It uses port 443 which does not go through the web cache. SSL works fine and your 'big brother cache' knows nothing about it. Thus the cache does not have to decode it or anything else silly like that. > jason -b From rajak@purdue.edu Sun, 24 Sep 2000 18:07:36 -0500 (EST) Date: Sun, 24 Sep 2000 18:07:36 -0500 (EST) From: Brian Poole rajak@purdue.edu Subject: [PLUG] Web caching And the saga continues.. > can think of three of the hostnames off the top of my head: dinero.tcom, > lucre.tcom, and moola.tcom. #4 is gelt.tcom > > The explanation that I got was that it was primarily done to speed > things up for the end user and not to save bandwidth per se. Supposedly > it only makes a 5-10% difference in bandwidth consumed whether the cache > boxes are on or not. When you think about it, Purdue's T3s are going to Hmm, interesting.. Could you clarify a bit though? A 5-10% decrease in b/w caused by web traffic or a 5-10% decrease in total b/w usage? If the number is in regards to web traffic, how much traffic total is associated with web traffic? Personally, in either case I'm glad they did it. _Any_ decrease in unneeded b/w usage helps. > be hit pretty hard no matter what they do, although I can't say that Is the fact that they are going to be heavily used an excuse to not save b/w where you can? Of course not. You should always try to reduce usage where possible. > I've ever felt a considerable slowdown in ResNet response. I have > noticed that the cache boxes aren't always on though. > Seth Btw, to start a different thread.. I have heard lots of complaints from other students in the reshalls and friends who have DSL through Purdue. Ping times are higher, transfers are slower, etc. The usage on the ResNet T3 is simply incredible. Look at the difference for a ping to an IP (one of the cnet.com webservers, picked at random) from a machine on campus.. 64 bytes from 216.200.247.135: icmp_seq=0 ttl=244 time=79.283 ms compared to a ping from a machine in a reshall.. 64 bytes from 216.200.247.135: icmp_seq=0 ttl=244 time=164.3 ms or another to kuro5hin.org from campus.. 64 bytes from 209.208.150.45: icmp_seq=0 ttl=242 time=41.566 ms and from a reshall.. 64 bytes from 209.208.150.45: icmp_seq=0 ttl=244 time=696.6 ms I find the difference staggering (This is why I am strongly in support of any effort, even if it's effect is small, to reduce the usage on the offcampus links). But to start the new thread, has anyone heard any future plans to help alleviate the saturation on the resnet DS3? -b From lewis1@purdue.edu Sun, 24 Sep 2000 18:31:22 -0500 (EST) Date: Sun, 24 Sep 2000 18:31:22 -0500 (EST) From: James C. Lewis lewis1@purdue.edu Subject: [PLUG] Web caching On Sun, 24 Sep 2000, Jason Dubrow wrote: > > SSL works fine in my experience, I would imagine it just doesn't cache > > it. > It does? That I am very suprised at ssl encrypts all the stuff interal to > the encypted packets...the cache would have to have access to a key to > decode it.....does 'big brother cache' steal this information? Let's ignore my earlier post were I explained why port 443 isn't cached anyway...and point to this... Web cache's only work in one direction...they don't cache what you send to a site anyway -James C. Lewis _ /_ _ _ _ / ) / _ ' _ (_/(///)(-_) (__ . (__(-((//_) PUCC Information Center Consultant jclewis@purdue.edu "Only those who dare to fail greatly can ever achieve greatly." -- Robert F. Kennedy Please place all complaints in this box --> [] From sdh@purdue.edu Sun, 24 Sep 2000 19:22:22 -0500 Date: Sun, 24 Sep 2000 19:22:22 -0500 From: Seth Heckard sdh@purdue.edu Subject: [PLUG] Web caching Brian Poole wrote: > > The explanation that I got was that it was primarily done to speed > > things up for the end user and not to save bandwidth per se. Supposedly > > it only makes a 5-10% difference in bandwidth consumed whether the cache > > boxes are on or not. When you think about it, Purdue's T3s are going to > > Hmm, interesting.. Could you clarify a bit though? A 5-10% decrease in b/w > caused by web traffic or a 5-10% decrease in total b/w usage? If the > number is in regards to web traffic, how much traffic total is associated > with web traffic? Personally, in either case I'm glad they did it. _Any_ > decrease in unneeded b/w usage helps. I can only assume that they meant a 5-10% decrease in total bandwidth. I doubt they actually keep stats on bandwidth usage attributed to http, etc. (or if they do they aren't on the PDN stats page like everything else.) > > be hit pretty hard no matter what they do, although I can't say that > > Is the fact that they are going to be heavily used an excuse to not save > b/w where you can? Of course not. You should always try to reduce usage > where possible. Point taken. But I can only wonder if the 5% decrease in bandwidth offsets the cost of the cache servers, all the man-hours configuring them, and the day-to-day maintenance that they require. > Btw, to start a different thread.. > > I have heard lots of complaints from other students in the reshalls > and friends who have DSL through Purdue. Ping times are higher, transfers > are slower, etc. The usage on the ResNet T3 is simply incredible. Look at > the difference for a ping to an IP (one of the cnet.com webservers, picked > at random) from a machine on campus.. Purdue does separate resnet/official campus traffic on their network. They even go as far as separating the reshall computer lab and reshall office computers from the rest of the resnet network in the dorms. It's kinda strange to see the difference between the two... the massive stack of hubs going to every dorm room and then the lone switch that is hooked to the PUCC lab and the main office. > I find the difference staggering (This is why I am strongly in support of > any effort, even if it's effect is small, to reduce the usage on the > offcampus links). But to start the new thread, has anyone heard any future > plans to help alleviate the saturation on the resnet DS3? What I heard at the beginning of the year was that they have no immediate plans to add another T3 or anything, but they would have no choice but to add another if it were needed. I know this has been said before, but the real problem with bandwidth lies not in the web traffic, but rather with napster and other "sharing" tools. I'm not saying that they should necessarily ban those though. What PUCC should do is replace all the hubs in the reshalls with 100MB switches. They already have started migrating labs to that. I know that it's being thought about but I'm sure nothing will happen before I graduate. That way the stupid SMB traffic wouldn't interfere with everyone else. Of course, it would not affect the outside link. They recently did "upgrade" the layout in certain reshalls, but all they did is add one switch that all the hubs connect to rather than have the hubs all connected to each other. > -b Seth From lewis1@purdue.edu Sun, 24 Sep 2000 19:49:21 -0500 (EST) Date: Sun, 24 Sep 2000 19:49:21 -0500 (EST) From: James C. Lewis lewis1@purdue.edu Subject: [PLUG] Web caching On Sun, 24 Sep 2000, Seth Heckard wrote: > Point taken. But I can only wonder if the 5% decrease in bandwidth > offsets the cost of the cache servers, all the man-hours configuring > them, and the day-to-day maintenance that they require. Uh... I'm not sure what you're referring to... back home we use CacheFlow boxes that you just plug into the network and you're done. Once your routers are redirecting port 80 to them you never touch them again. If they break, they send you an email saying they need maintance. We have 5 or 6 of them, and I don't think we've ever even looked at them a second time... Other than the fact that they're bright yellow and catch your eye when you walk past them. If you go for one of the new appliances cache servers are nothing...they are cheap bandwidth (in a way of looking at it) -James C. Lewis _ /_ _ _ _ / ) / _ ' _ (_/(///)(-_) (__ . (__(-((//_) PUCC Information Center Consultant jclewis@purdue.edu "Only those who dare to fail greatly can ever achieve greatly." -- Robert F. Kennedy Please place all complaints in this box --> [] From lewis1@purdue.edu Sun, 24 Sep 2000 19:58:01 -0500 (EST) Date: Sun, 24 Sep 2000 19:58:01 -0500 (EST) From: James C. Lewis lewis1@purdue.edu Subject: [PLUG] Web caching (Info) Hey guys... I'm reading these emails and I can tell the majority of you don't know what Web Cache's do... I don't know what PDN is using for Purdue's network, but I'm pretty sure its a Cache Appliance. The ISP I work for back home uses the CacheFlow 500. We have 1 in our office, and 1 at every POP. I don't know why people think Caching is so evil. To me it means more bandwidth for other stuff. If you live on the web, you're wasting the net's resources anyway. CacheFlow's URL is http://www.cacheflow.com and info on the one we use at work is: http://www.cacheflow.com/products/500/index.cfm I'm just posting this because I'm hoping some of you will go and read about them and get some idea of what they do. -James C. Lewis _ /_ _ _ _ / ) / _ ' _ (_/(///)(-_) (__ . (__(-((//_) PUCC Information Center Consultant jclewis@purdue.edu "Only those who dare to fail greatly can ever achieve greatly." -- Robert F. Kennedy Please place all complaints in this box --> [] From psmith@lib.purdue.edu Sun, 24 Sep 2000 20:02:45 -0500 Date: Sun, 24 Sep 2000 20:02:45 -0500 From: Preston Smith psmith@lib.purdue.edu Subject: [PLUG] Web caching On Sun, Sep 24, 2000 at 01:57:00PM -0500, David Christopher Hansen (david_hansen@ibm.net) wrote: > What are the consequences of having this thing in the way? I remember > last year when they were installing it, we got lots of strange errors > because of it. Pages would fail to load, and site a cryptic error > message that the caching server generated. Has anyone noticed any > problems? The one problem we've had in the Libraries is ones with online database vendors who license use to us by subnets. Some we license to *.purdue.edu, but others are licensed for in-library use only, or a certain library, in which case we tell them that subnets X, Y, and Z are library locations. We didn't hear about the webcache until these things starting giving us 'authorization refused' type messages. You guessed it, requests that were from library subnets apeared to the vendors as moola (gelt, deniro, etc) .tcom.purdue.edu. PDN blamed the vendors for being braindead, and the vendor blamed the 'firewall' we appeared to be coming through. So we wound up having connections to these particular vendors excluded from the webcache.. Minor problem, I suppose, pretty easy fix. That's pretty much all we've encountered problem-wise. -Preston -- Preston M. Smith http://www.lib.purdue.edu/~psmith "The only winner in the War of 1812 was Tchaikovsky." -- David Gerrold From sdh@purdue.edu Sun, 24 Sep 2000 20:03:42 -0500 Date: Sun, 24 Sep 2000 20:03:42 -0500 From: Seth Heckard sdh@purdue.edu Subject: [PLUG] Web caching "James C. Lewis" wrote: > > On Sun, 24 Sep 2000, Seth Heckard wrote: > > Point taken. But I can only wonder if the 5% decrease in bandwidth > > offsets the cost of the cache servers, all the man-hours configuring > > them, and the day-to-day maintenance that they require. > > Uh... I'm not sure what you're referring to... back home we use CacheFlow > boxes that you just plug into the network and you're done. Once your > routers are redirecting port 80 to them you never touch them again. Ok, but speaking from experience setting up squid, it's not just plug-and-play. Boxen designed to be easy to set up will deliver, obviously, but they probably sacrifice features for usability. The maintenance that I speak of is stupid stuff like web pages not cooperating with the cache servers. I was working in the Information Center the other day when some sysadmin with some web site called in saying that people at Purdue were complaining that they couldn't access his web page. He said that they were all coming from dinero.tcom. I never found out what happened, but I'm assuming it was a problem with our cache server. Now, of course, it probably took the guy in charge about 30 seconds to disable caching from that site, but it also means that it doesn't work perfectly out of the box. And also, does anyone know why the cache servers are used at times and not used at others? > -James C. Lewis Seth From doobie@doobie.org Sun, 24 Sep 2000 21:13:54 -0500 (EST) Date: Sun, 24 Sep 2000 21:13:54 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] Web caching > > > > I can believe the traffic is quite high, but I don't think,as you said, > > most of the traffic is web traffic (maybe it is....i don't have the #). > > Mmm, I never said that _most_ of the traffic was web traffic. I said we > are saturating our b/w and that we must take steps to alleviate > it. Caching _does_ reduce b/w usage. I certainly don't believe most of the > traffic is web traffic, I imagine a great deal is the illicit traffic that > I mentioned, mp3s & warez. I don't know where you got that I thought that > most of the traffic was web. I said you didn't think web traffic was the problem... mp3 and warez were always the problem when I was in the res halls...that an porn. I should try and drag up some of my log files.....I did a 'tiny' bit of traffic monitoring in my dorm, it was (off the top of my head) 30% porn, 30% warez, 20% mp3's, 10 misc web stuff, and 10% other if memory serves me correctly. Of course as it got later the porn traffic went up. (seriouslly, no joke). > Ehm, I think in all reality it _does_ reduce the web traffic. I don't see > how you can argue that. It doesn't matter how much of the total traffic is > web traffic, if it is reduced in any way it does reduce the total traffic > at least some. The effectiveness of the web cache to reduce total traffic > may lie on how much of the traffic is web traffic, but if it will actually > reduce the web traffic simply lies on the effectiveness of the webcache > itself. lets say you are saturated (but everything is going), so 100% bandwidth is being used. if 20% of the traffic is from the web, and 70 other. A cache that improves the web traffic by 10%, now makes your saturation level at 98%. That just gives you slightly more bandwidth for the mp3's, warez. man I can believe the porn collection these cache server's have after a few hours. jason From doobie@doobie.org Sun, 24 Sep 2000 21:18:43 -0500 (EST) Date: Sun, 24 Sep 2000 21:18:43 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] Web caching (Info) I've never said they are bad/don't work. I've never seen raw numbers, I have just been stating that if a majority of traffic is NOT web traffic it may become useless, or to an extent not to that much. If web traffic accounts for 20% of traffic (ipulled this number out of thin air), a 10% reduction brings you only to 98% of the original traffic. a 50% brings you to 90%. someone else has to decide if that is worth it. if you reduce web traffic, it'll just give you more bandwidth for non-web traffic. jason On Sun, 24 Sep 2000, James C. Lewis wrote: > > Hey guys... I'm reading these emails and I can tell the majority of you > don't know what Web Cache's do... I don't know what PDN is using for > Purdue's network, but I'm pretty sure its a Cache Appliance. > > The ISP I work for back home uses the CacheFlow 500. We have 1 in our > office, and 1 at every POP. I don't know why people think Caching is so > evil. To me it means more bandwidth for other stuff. If you live on the > web, you're wasting the net's resources anyway. > > CacheFlow's URL is http://www.cacheflow.com and info on the one we use at > work is: > http://www.cacheflow.com/products/500/index.cfm > > I'm just posting this because I'm hoping some of you will go and read > about them and get some idea of what they do. > > > -James C. Lewis > _ > /_ _ _ _ / ) / _ ' _ > (_/(///)(-_) (__ . (__(-((//_) > > PUCC Information Center Consultant > jclewis@purdue.edu > > "Only those who dare to fail greatly can ever achieve greatly." > -- Robert F. Kennedy > > Please place all complaints in this box --> [] > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From rjune@ims1.imagestream-is.com Sun, 24 Sep 2000 21:52:35 -0500 (EST) Date: Sun, 24 Sep 2000 21:52:35 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] NFS woes. I'm having some problems nfs mounting /home. Here is the entry in my /etc/fstabs file on the client(tiamet) firbolg:/home /home nfs exec,dev,suid,rw\ 1 1 Here is my entry in /etc/exports file on the server(firbolg) /home *.bravegnuworld.com(rw,no_root_squash) \ 63.86.29.163(rw,no_root_squash) When I try to mount /home on tiamet, I get this in the log on firbolg firbolg mountd[3056]: authenticated mount request from tiamet:939 for\ /home (/home) firbolg mountd[3056]: getfh failed: Operation not permitted Does anyone have any ideas what this might be? From address-supressed@ecn.purdue.edu Sun, 24 Sep 2000 21:58:44 -0500 (EST) Date: Sun, 24 Sep 2000 21:58:44 -0500 (EST) From: Rick address-supressed@ecn.purdue.edu Subject: [PLUG] NFS woes. >> I'm having some problems nfs mounting /home. >> >> Here is the entry in my /etc/fstabs file on the client(tiamet) >> firbolg:/home /home nfs exec,dev,suid,rw\ >> 1 1 >> >> Here is my entry in /etc/exports file on the server(firbolg) >> /home *.bravegnuworld.com(rw,no_root_squash) \ >> 63.86.29.163(rw,no_root_squash) >> >> When I try to mount /home on tiamet, I get this in the log on firbolg >> firbolg mountd[3056]: authenticated mount request from tiamet:939 for\ >> /home (/home) >> firbolg mountd[3056]: getfh failed: Operation not permitted Make sure that /etc/hosts contains tiamet.bravegnuworld.com before just tiamet. Your server thinks the name of 63.86.29.163 is just 'tiamet'. Then, on the server, type: /usr/sbin/exportfs -v -a -r Rick From rjune@ims1.imagestream-is.com Sun, 24 Sep 2000 22:05:07 -0500 (EST) Date: Sun, 24 Sep 2000 22:05:07 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] NFS woes. You all all *over* that rick. thanks, everything is happy and joyous. From rjune@ims1.imagestream-is.com Sun, 24 Sep 2000 22:15:01 -0500 (EST) Date: Sun, 24 Sep 2000 22:15:01 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] NFS woes. I spoke too soon, It mounts(and works), but after two seconds I get these errors. tiamet.bravegnuworld.com kernel: RPC: garbage, retrying 76 tiamet.bravegnuworld.com kernel: RPC: garbage, retrying 76 tiamet.bravegnuworld.com kernel: RPC: garbage, exit EIO /home: Stale NFS file handle I can umount /home, but then when I try to mount it again, I get the previous error. From address-supressed@ecn.purdue.edu Sun, 24 Sep 2000 22:30:23 -0500 (EST) Date: Sun, 24 Sep 2000 22:30:23 -0500 (EST) From: Rick address-supressed@ecn.purdue.edu Subject: [PLUG] NFS woes. >> I spoke too soon, It mounts(and works), but after two seconds I get these >> errors. >> tiamet.bravegnuworld.com kernel: RPC: garbage, retrying 76 >> tiamet.bravegnuworld.com kernel: RPC: garbage, retrying 76 >> tiamet.bravegnuworld.com kernel: RPC: garbage, exit EIO >> /home: Stale NFS file handle It looks like a bad network connection. Do you have a mismatch of half- and full-duplex network hardware somewhere? You might try ping -f to check if you're dropping packets. >> I can umount /home, but then when I try to mount it again, I get the >> previous error. Alternatively, you might have sufficiently confused the kernel NFS server. You can blow it away and restart (on a RedHat derivative) by: /etc/rc.d/init.d/nfs stop /sbin/rmmod nfsd /sbin/rmmod lockd /sbin/rmmod sunrpc /etc/rc.d/init.d/nfs start The 2.2 kernel NFSd has a number of interesting "issues" with larger drives, failure modes, network interruptions, etc. They've been maintaining 2.2 as it is to retain backward compatibility (with the bugs?). If you're using a stock 2.2.x kernel, you might consider getting RedHat's 2.2.16-3 instead. Other distributions have their own patches in place as well. i.e. vendors' releases of knfsd are incompatible with the bugs. 8^) Patches to finally restore kernel NFS stability went into 2.2.18pre9. If you're into compiling your own kernel, you might try that or 2.2.18pre10. Rick From rjune@ims1.imagestream-is.com Sun, 24 Sep 2000 22:46:39 -0500 (EST) Date: Sun, 24 Sep 2000 22:46:39 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] Another fun topic, ISA/pnp I've got an ISA/pnp sound card(SB Vibra 16) That I would really love to use in a system, however it's settings conflict with the NIC in the machine(ne2k/ISA) the NIC is a software selectable job, but I don't have the software, does anyone know of a way to force the config on the vibra16? From rjune@ims1.imagestream-is.com Sun, 24 Sep 2000 22:48:08 -0500 (EST) Date: Sun, 24 Sep 2000 22:48:08 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] NFS woes. Both boxen are already rh6.2, everything was working, but I had to reinstall the server, at anyrate everything is working fine now, I had to take the *.bravegnuworld.com section out of /etc/exports. From cnd@ecn.purdue.edu Mon, 25 Sep 2000 00:43:31 -0500 Date: Mon, 25 Sep 2000 00:43:31 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] other nfs woes So I'm setting up all my nfs exports and mounts for the first time. I've now switched to using the automounter to mount my home directory. Here's what I've got: fstab: /dev/sda1 / reiserfs defaults 1 1 /dev/hda2 /boot reiserfs notail 1 2 /dev/sda2 /usr reiserfs defaults 1 2 /dev/sdb1 /export/home reiserfs defaults 1 2 /dev/sdc4 /jaz vfat user,exec,uid=8997 0 0 /dev/hda3 /mnt/data vfat defaults 0 0 /dev/sdd1 /mnt/disk0 reiserfs defaults 1 2 /dev/sde1 /mnt/disk1 reiserfs defaults 1 2 /dev/cdrom /mnt/cdrom iso9660 ro,nosuid,noauto,exec,user,nodev 0 0 /dev/fd0 /mnt/floppy vfat sync,nosuid,noauto,user,nodev,unhide 0 0 none /dev/pts devpts mode=0620 0 0 none /proc proc pri=1 0 0 /dev/sdb2 swap swap pri=1 0 0 /dev/sda3 swap swap pri=1 0 0 /dev/hda4 swap swap defaults 0 0 Yes, I went overboard on swap. :-) Oh well. But with the priorities set as '1' on all the swaps, I get "distributed swap". I just did this and have noticed an improvement when refreshing loaded gimp images. [falkor] /etc # cat auto.master # $Id: auto.master,v 1.2 1997/10/06 21:52:03 hpa Exp $ # Sample auto.master file # Format of this file: # mountpoint map options # For details of the format look at autofs(8). /misc /etc/auto.misc /home /etc/auto.home [falkor] /etc # cat auto.home falkor -hard,bg,rw falkor:/export/home [falkor] /etc # [falkor] /etc # cat exports /mnt/disk1/Mandrake auryn(ro) /export/home auryn(rw,no_root_squash) 192.168.1.2(rw,no_root_squash) /mnt/disk0/mp3 auryn(ro) 192.168.1.2(ro) Hopefully the formatting on all that didn't get messed up through sending the mail... So here's the deal. In short, homes are mounted in /export/home on machine falkor. /export/home is exported to falkor (localhost) and auryn. auryn automounts the export AOK. /home/falkor will show up when needed (for example, when I log in.) The prob is this. falkor, which the export resides on doesn't really mount it, but creates a symlink from /home/falkor pointing to /export/home. This to me seems dumb, and it's causing some funkiness. When I log in, it puts me in /export/home/cnd as my beginning directory. If I do a 'cd' I get put into /home/falkor/cnd and if I echo $HOME I get /home/falkor/cnd. This is screwy. Can someone please help? How can I get the automounter to actually mount the export and not make a symlink? And let me know if you need more info on putting this mess together. I did this all today, so... -Chris -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From address-supressed@ecn.purdue.edu Mon, 25 Sep 2000 01:03:50 -0500 (EST) Date: Mon, 25 Sep 2000 01:03:50 -0500 (EST) From: Rick address-supressed@ecn.purdue.edu Subject: [PLUG] other nfs woes >> fstab: >> /dev/sda1 / reiserfs defaults 1 1 >> /dev/hda2 /boot reiserfs notail 1 2 >> /dev/sda2 /usr reiserfs defaults 1 2 >> /dev/sdb1 /export/home reiserfs defaults 1 2 >> /dev/sdc4 /jaz vfat user,exec,uid=8997 0 0 >> /dev/hda3 /mnt/data vfat defaults 0 0 >> /dev/sdd1 /mnt/disk0 reiserfs defaults 1 2 >> /dev/sde1 /mnt/disk1 reiserfs defaults 1 2 >> /dev/cdrom /mnt/cdrom iso9660 ro,nosuid,noauto,exec,user,nodev 0 >> 0 >> /dev/fd0 /mnt/floppy vfat sync,nosuid,noauto,user,nodev,unhide 0 >> 0 >> none /dev/pts devpts mode=0620 0 0 >> none /proc proc pri=1 0 0 >> /dev/sdb2 swap swap pri=1 0 0 >> /dev/sda3 swap swap pri=1 0 0 >> /dev/hda4 swap swap defaults 0 0 Did you mean to also give /proc a priority of 1? I don't think that's going to help gimp. >> The prob is this. falkor, which the export resides on doesn't really >> mount it, but creates a symlink from /home/falkor pointing to >> /export/home. This to me seems dumb, and it's causing some >> funkiness. When I log in, it puts me in /export/home/cnd as my >> beginning directory. If I do a 'cd' I get put into /home/falkor/cnd >> and if I echo $HOME I get /home/falkor/cnd. The getcwd() call does a pretty good job of paying attention to the symlink'd directory. For most applications, this works just fine. But you're using bash, which does all kinds of extra gymnastics to really tell you the real physical directory location you're in. I never found a way to turn it off. The other major shells do the right thing. (NOTE: No flame war advocated. There's generally nothing else wrong with bash.) >> This is screwy. Can someone please help? How can I get the >> automounter to actually mount the export and not make a symlink? So you want to mount an NFS export on the same machine? That's not going to make gimp go faster either. 8^) What you really want is something like Solaris which has a loopback mount that points one directory to another--without going through NFS. The Linux 2.4 kernel has a new VFS mechanism that will allow you to "bind" one directory to another. It's effectively a hard link from one directory to another. ...except that it's symmetric. e.g: mount -t bind /export/home/cnd /home/falkor/cnd When you upgrade to 2.4 and have a suitably recent autofs, it'll do the Right Thing automatically. Rick PS: There are a few other new VFS tricks in 2.4. It would make for an interesting talk for anyone who wants to try giving a good practice talk to improve their presentation skills. From cnd@ecn.purdue.edu Mon, 25 Sep 2000 01:10:18 -0500 Date: Mon, 25 Sep 2000 01:10:18 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] other nfs woes Rick wrote: > > >> fstab: > >> /dev/sda1 / reiserfs defaults 1 1 > >> /dev/hda2 /boot reiserfs notail 1 2 > >> /dev/sda2 /usr reiserfs defaults 1 2 > >> /dev/sdb1 /export/home reiserfs defaults 1 2 > >> /dev/sdc4 /jaz vfat user,exec,uid=8997 0 0 > >> /dev/hda3 /mnt/data vfat defaults 0 0 > >> /dev/sdd1 /mnt/disk0 reiserfs defaults 1 2 > >> /dev/sde1 /mnt/disk1 reiserfs defaults 1 2 > >> /dev/cdrom /mnt/cdrom iso9660 ro,nosuid,noauto,exec,user,nodev 0 > >> 0 > >> /dev/fd0 /mnt/floppy vfat sync,nosuid,noauto,user,nodev,unhide 0 > >> 0 > >> none /dev/pts devpts mode=0620 0 0 > >> none /proc proc pri=1 0 0 > >> /dev/sdb2 swap swap pri=1 0 0 > >> /dev/sda3 swap swap pri=1 0 0 > >> /dev/hda4 swap swap defaults 0 0 > > Did you mean to also give /proc a priority of 1? > I don't think that's going to help gimp. haha. NO! I edited the email by hand and messed with the wrong line. It's correct in fstab. My bad... > > >> The prob is this. falkor, which the export resides on doesn't really > >> mount it, but creates a symlink from /home/falkor pointing to > >> /export/home. This to me seems dumb, and it's causing some > >> funkiness. When I log in, it puts me in /export/home/cnd as my > >> beginning directory. If I do a 'cd' I get put into /home/falkor/cnd > >> and if I echo $HOME I get /home/falkor/cnd. > > The getcwd() call does a pretty good job of paying attention to the > symlink'd directory. For most applications, this works just fine. > > But you're using bash, which does all kinds of extra gymnastics > to really tell you the real physical directory location you're in. > I never found a way to turn it off. The other major shells do the > right thing. (NOTE: No flame war advocated. There's generally nothing > else wrong with bash.) > > >> This is screwy. Can someone please help? How can I get the > >> automounter to actually mount the export and not make a symlink? > > So you want to mount an NFS export on the same machine? > That's not going to make gimp go faster either. 8^) I'm wanting to play with NFS for educational purposes. Trying to do it like what some computing networks on campus do... But you are correct in that the gimp won't go faster. I just want to make it easy to get to home dirs. Especially when we get the UberNetwork set up between differnet apts and dorms and campuses and stuff. :-) > > What you really want is something like Solaris which has a loopback > mount that points one directory to another--without going through NFS. > > The Linux 2.4 kernel has a new VFS mechanism that will allow you to "bind" > one directory to another. It's effectively a hard link from one directory > to another. ...except that it's symmetric. > e.g: > > mount -t bind /export/home/cnd /home/falkor/cnd > > When you upgrade to 2.4 and have a suitably recent autofs, it'll do the > Right Thing automatically. When I read the docs on how to get my sound card working in 2.4, I will switch for sure. It's already compiled and installed (with ReiserFS support, yeah ReiserFS!!!) Just need to get the sound figured out. > > Rick > > PS: There are a few other new VFS tricks in 2.4. It would make for an > interesting talk for anyone who wants to try giving a good practice talk > to improve their presentation skills. > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From address-supressed@ecn.purdue.edu Mon, 25 Sep 2000 01:17:38 -0500 (EST) Date: Mon, 25 Sep 2000 01:17:38 -0500 (EST) From: Rick address-supressed@ecn.purdue.edu Subject: [PLUG] other nfs woes Incidentally, to answer your question in the context of the 2.2 kernel, there is a twisted way of solving your problem. You could customize your automount map so that your partition is locally mounted (by automount) as /home/falkor/cnd and export that. Then remotely mount falkor:/home/falkor/cnd on other systems. What this means is that the automount map must be different on every system. Yes, it's a pain-in-the-automount-map but it works. In a neighborhood network which has no central authority, it might be just as well since there would be no reason to distribute a common automount map. Rick From wallacms@cs.purdue.edu Mon, 25 Sep 2000 01:18:05 -0500 (EST) Date: Mon, 25 Sep 2000 01:18:05 -0500 (EST) From: Matt Wallace wallacms@cs.purdue.edu Subject: [PLUG] Another fun topic, ISA/pnp If you have a PnP BIOS you can sometimes force the irq there. You might try setting PnP OS to No in your BIOS and letting the BIOS configure your devices for you. Matt On Sun, 24 Sep 2000, Richard June wrote: > I've got an ISA/pnp sound card(SB Vibra 16) That I would really love to > use in a system, however it's settings conflict with the NIC in the > machine(ne2k/ISA) the NIC is a software selectable job, but I don't have > the software, does anyone know of a way to force the config on the > vibra16? > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > > From cnd@ecn.purdue.edu Mon, 25 Sep 2000 09:16:33 -0500 Date: Mon, 25 Sep 2000 09:16:33 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] PHP - Richard Lynch, Zend Technologies TONIGHT TONIGHT TONIGHT TONIGHT Monday, September 25th, 2000 7pm - 9pm Physics 112 Richard Lynch from Zend Technologies, the authors of the PHP Zend Optimizer, will be speaking about PHP and writing web applications using PHP on Monday night. He is going to develop a simple web application during the presentation. PHP is the PHP Hypertext Preprocessor. It is used for writing dynamically generated and database driven websites. Here's the description from the website: PHP Version 3.0 is an HTML-embedded scripting language. Much of its syntax is borrowed from C, Java and Perl with a couple of unique PHP-specific features thrown in. The goal of the language is to allow web developers to write dynamically generated pages quickly. Related Links: http://www.zend.com/ - Zend Technologies http://www.php.net/ - PHP Main Website Punch and pie will be available for refreshments. -Chris PLUG President -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From cnd@ecn.purdue.edu Mon, 25 Sep 2000 10:36:45 -0500 Date: Mon, 25 Sep 2000 10:36:45 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] other nfs woes Rick wrote: > > Incidentally, to answer your question in the context of the 2.2 kernel, > there is a twisted way of solving your problem. > > You could customize your automount map so that your partition is locally > mounted (by automount) as /home/falkor/cnd and export that. Then remotely > mount falkor:/home/falkor/cnd on other systems. It works, and I did that at first before I set up the automounter, but it's ugly. I also eventually want to set up rdist or some other thinger to distribute config files easily across all the systems. This is all a learning thing, though I do want to take over the world. > What this means is that the automount map must be different on every system. Ick. > Yes, it's a pain-in-the-automount-map but it works. In a neighborhood > network which has no central authority, it might be just as well since > there would be no reason to distribute a common automount map. I am the central authority, and soon I will rule the world. From ab@eas.purdue.edu Mon, 25 Sep 2000 10:53:26 -0500 Date: Mon, 25 Sep 2000 10:53:26 -0500 From: A Braunsdorf ab@eas.purdue.edu Subject: [PLUG] other nfs woes In message <39CF710D.98BF7897@ecn.purdue.edu>, "Christopher N. Deckard" writes: > > It works, and I did that at first before I set up the automounter, but > it's ugly. I also eventually want to set up rdist or some other thinger > to distribute config files easily across all the systems. This is all a > learning thing, though I do want to take over the world. > > > What this means is that the automount map must be different on every system > . > > Ick. Gee if only there were a way to make it easier to rdist files to groups of machines and customize those files by machine or OS or whatever... There is! It's called distrib, and it was developed at PUCC back in the day. I'm currently working on a new version of it, and it's one of the projects I'll be putting part-timers on as soon as I can hire some. Maybe when we get a little further, we could do a presentation on it for PLUG. Most of you guys don't need to manage dozens of UNIX machines (yet!), but when you do, there's nothing like it. ab From pfitzge1@purdue.edu Mon, 25 Sep 2000 11:10:23 -0500 (EST) Date: Mon, 25 Sep 2000 11:10:23 -0500 (EST) From: patrick.n.fitzgerald.1 pfitzge1@purdue.edu Subject: [PLUG] Web caching On Sun, 24 Sep 2000, Jason Dubrow wrote: > That just gives you slightly more bandwidth for the mp3's, warez. > man I can believe the porn collection these cache server's have after a > few hours. > Now we know the real reason for the cache servers: Purdue University Cum-puting Center Presents... Purdue's BEST OF THE WEB Porn site! That's right, all the hot action you can handle. Our staff of 30,000 students collects all the best porn on the internet and delivers it hot 'n' fresh to your computer! Over 10 Terabytes collected every month! Coming soon to http://www.porn.purdue.edu Only $19.95/month, 7 day trial free! Also, for an additional $5/mo, you get free access to our extensive MP3 collection. If one of our students has downloaded it, it's in there! Now over 5 Terabytes of mp3's online. Coming soon to http://www.mp3.purdue.edu --------------------------- As a sideline, I saw the ECN MasPar sitting on the 2nd floor of MSEE with an "X" on the front. If that means it's going to Salvage, I suggest that we (Plug) commandeer it for our own sinister massively parallel research computing needs. It probably won't run Linux, but Who Cares? It's Massively Parallel! And it's big enough that we could probably gut it out and use it as an office. So sayeth, Patrick Fitzgerald Official Plug Videographer and Keeper of the Sacred Flame -- Gravity is a myth, the Earth sucks. --fortune (5) From address-supressed@ecn.purdue.edu Mon, 25 Sep 2000 11:40:43 -0500 (EST) Date: Mon, 25 Sep 2000 11:40:43 -0500 (EST) From: Rick address-supressed@ecn.purdue.edu Subject: [PLUG] Web caching >> As a sideline, I saw the ECN MasPar sitting on the 2nd floor of MSEE with >> an "X" on the front. If that means it's going to Salvage, I suggest that >> we (Plug) commandeer it for our own sinister massively parallel research >> computing needs. It probably won't run Linux, but Who Cares? It's >> Massively Parallel! ...and it has the aggregate computing power of one 400MHz AMD K6-III. >> And it's big enough that we could probably gut it out >> and use it as an office. Probably a more fitting use. Rick From cnd@ecn.purdue.edu Mon, 25 Sep 2000 13:41:59 -0500 Date: Mon, 25 Sep 2000 13:41:59 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] [Fwd: Offtopic: CERIAS] This is a multi-part message in MIME format. --------------EA0C2246D258F6F802B3672D Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Has anyone been to CERIAS meetings or functions? Maybe you can answer this person. -Chris --------------EA0C2246D258F6F802B3672D Content-Type: message/rfc822 Content-Transfer-Encoding: 7bit Content-Disposition: inline Received: from herald.cc.purdue.edu (herald.cc.purdue.edu [128.210.11.29]) by pier.ecn.purdue.edu (8.9.3/8.9.3moyman) with ESMTP id MAA04004 for ; Mon, 25 Sep 2000 12:43:42 -0500 (EST) Received: from csociety.ecn.purdue.edu by herald.cc.purdue.edu with ESMTP for cnd@purdue.edu; Mon, 25 Sep 2000 12:43:30 -0500 Received: from csociety.ecn.purdue.edu (IDENT:mail@localhost [127.0.0.1]) by csociety.ecn.purdue.edu (8.9.3/8.9.3) with ESMTP id MAA31816 for ; Mon, 25 Sep 2000 12:43:18 -0500 Received: from herald.cc.purdue.edu (herald.cc.purdue.edu [128.210.11.29]) by csociety.ecn.purdue.edu (8.9.3/8.9.3) with ESMTP id MAA31808 for ; Mon, 25 Sep 2000 12:43:18 -0500 Received: from [128.211.205.139] by herald.cc.purdue.edu for plug-admin@csociety.purdue.edu; Mon, 25 Sep 2000 12:43:12 -0500 From: "Sumanth Peddamatham" To: Subject: Offtopic: CERIAS Date: Mon, 25 Sep 2000 12:42:38 -0500 Message-Id: MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) In-Reply-To: <200009250617.BAA25340@pier.ecn.purdue.edu> Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Content-Type: text/plain; charset="iso-8859-1" I know this has nothing to do with PLUG, but I was wondering if any of you guys has attended on of the CERIAS meetings. If so, are they interesting? What do they talk about, etc... Thanks, sumanth p. --------------EA0C2246D258F6F802B3672D-- From cnd@ecn.purdue.edu Mon, 25 Sep 2000 14:56:19 -0500 Date: Mon, 25 Sep 2000 14:56:19 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] [Fwd: Leverage OpenGL's power to build mesmerizing black holes] This is a multi-part message in MIME format. --------------441BE90F884CF78D594C02BC Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Thought some of you may be interested. -Chris --------------441BE90F884CF78D594C02BC Content-Type: message/rfc822 Content-Transfer-Encoding: 7bit Content-Disposition: inline Received: from rock.west.ora.com (rock.west.ora.com [204.148.43.8]) by pier.ecn.purdue.edu (8.9.3/8.9.3moyman) with ESMTP id OAA18620 for ; Mon, 25 Sep 2000 14:51:49 -0500 (EST) Received: (from deniseo@localhost) by rock.west.ora.com (8.11.0/8.11.0) id e8PJpHq04627 for cnd@ecn.purdue.edu; Mon, 25 Sep 2000 12:51:17 -0700 (PDT) Date: Mon, 25 Sep 2000 12:51:17 -0700 (PDT) From: Denise Olliffe Message-Id: <200009251951.e8PJpHq04627@rock.west.ora.com> To: cnd@ecn.purdue.edu Subject: Leverage OpenGL's power to build mesmerizing black holes Content-Type: text Sharing articles I think your group might find interesting: If physics or astronomy have ever fascinated you, then I think you'll be interested in an article just published on the O'Reilly Network. Author Chris Halsall describes how you can simulate building a black hole, complete with a mesmerizing visual display. Chris uses OpenGL, an API developed by Silicon Graphics (SGI) in order to make 3D application developers' lives easier. Chris includes the program and makefile with the article, and there are links to other articles about OpenGL. Chris is the President of Open Source Solutions Inc. and a specialist at automating information gathering and presentation systems. Building a Black Hole with OpenGL http://www.oreillynet.com/pub/a/linux/2000/09/15/blackhole.html?page=1 Accompanying Articles: Creating Real Time 3D Graphics with OpenGL http://www.oreillynet.com/pub/a/network/2000/06/23/magazine/opengl_intro.html OpenGL Rendering and Drawing http://www.oreillynet.com/pub/a/network/2000/06/23/magazine/opengl_render.html Preparing to Build an OpenGL Application http://www.oreillynet.com/pub/a/network/2000/06/23/magazine/opengl_build.html Other Links O'Reilly Network: http://www.oreillynet.com O'Reilly Network Linux affiliate: http://oreilly.linux.com Open Source Solutions Inc.: http://www.os-s.com/ :) Denise --------------441BE90F884CF78D594C02BC-- From zamboni@cerias.purdue.edu Mon, 25 Sep 2000 15:37:48 -0500 Date: Mon, 25 Sep 2000 15:37:48 -0500 From: Diego Zamboni zamboni@cerias.purdue.edu Subject: [PLUG] [Fwd: Offtopic: CERIAS] > I know this has nothing to do with PLUG, but I was wondering if any of > you guys has attended on of the CERIAS meetings. If so, are they > interesting? What do they talk about, etc... They are fantastic! Excellent! Wonderful! I am a CERIAS student, so I have to say good things about them... :-) But seriously: I assume you are referring to the CERIAS weekly security seminar. The talks vary a lot in terms of topics, and therefore on whether you would find them interesting or not. Topics range from very technical to very high-level. Most of them are given by people with a lot of experience in their field, and therefore have interesting things to say. In all fairness, the technical level has declined a little bit from the old days of the COAST security seminar, in my opinion. I would suggest you check the web page at http://www.cerias.purdue.edu/secsem/ for the abstract of each seminar and see if you find it interesting. --Diego From wallacms@cs.purdue.edu Mon, 25 Sep 2000 16:46:46 -0500 (EST) Date: Mon, 25 Sep 2000 16:46:46 -0500 (EST) From: Matt Wallace wallacms@cs.purdue.edu Subject: [PLUG] Discounted Orielly books at UBS I saw a sign somewhere on campus today saying that University Bookstore has all their Orielly books for 20% off. I guess this is probably about what you get them on Amazon for, but I hate waiting for mail order books. Just thought I'd pass it on.. Also it said they're having some sort of drawing for free books. -Matt From doobie@doobie.org Mon, 25 Sep 2000 17:02:38 -0500 (EST) Date: Mon, 25 Sep 2000 17:02:38 -0500 (EST) From: Jason Dubrow doobie@doobie.org Subject: [PLUG] Web caching > Now we know the real reason for the cache servers: > > Purdue University Cum-puting Center Presents... > > Purdue's BEST OF THE WEB Porn site! > > That's right, all the hot action you can handle. Our staff of 30,000 > students collects all the best porn on the internet and delivers it hot > 'n' fresh to your computer! Over 10 Terabytes collected every month! hahah! I bet there is more than 10 TB collected a month > > Coming soon to http://www.porn.purdue.edu > Only $19.95/month, 7 day trial free! > > Also, for an additional $5/mo, you get free access to our extensive MP3 > collection. If one of our students has downloaded it, it's in there! Now > over 5 Terabytes of mp3's online. but napster gives me them all for free! Look at all this metallica I just downloaded. > > Coming soon to http://www.mp3.purdue.edu > > --------------------------- > > As a sideline, I saw the ECN MasPar sitting on the 2nd floor of MSEE with > an "X" on the front. If that means it's going to Salvage, I suggest that > we (Plug) commandeer it for our own sinister massively parallel research > computing needs. It probably won't run Linux, but Who Cares? It's > Massively Parallel! And it's big enough that we could probably gut it out > and use it as an office. Man, I wish I was still at Purdue, that would make a good house, I could set it up somewhere, and live in it! Jason From reh@blacksoul.net Mon, 25 Sep 2000 18:53:04 -0500 (EST) Date: Mon, 25 Sep 2000 18:53:04 -0500 (EST) From: Ryan E. Helfter reh@blacksoul.net Subject: [PLUG] Web caching Speaking of which... I have a NetApp with about 40 gigs of HD space on it... Problem is, its only about 9% full... With all the NFS mounting and crap going on ... I figured... why not share :) If you are interested, its a Netapp 540 with 2 drive shelves (1scratch disk) so 12 drives... If anyone is interested in storing offsite, let me know... Please only full CD albums... Also, just to let you know, this might also be a ICECAST (kinda like SHOUTCAST) stream of audio for all of you winamp junkies... any takers, let me know... ry > > Coming soon to http://www.mp3.purdue.edu > > --------------------------- > -- Ryan E. Helfter reh@blacksoul.net Black Soul Networks, LLC -- From redmanr@minga.com Mon, 25 Sep 2000 20:12:59 -0500 (EST) Date: Mon, 25 Sep 2000 20:12:59 -0500 (EST) From: Rick Redman redmanr@minga.com Subject: [PLUG] [Fwd: Offtopic: CERIAS] > Has anyone been to CERIAS meetings or functions? Maybe you can answer > this person. I am a CERIAS graduate. These "meetings" are actually lectures by important people from the security field. It is actually offered as a 1credit hour class (or it was 2 years ago). I would recomend them to everyone who doesnt mind a meeting at 4pm on fridays =) (ifs thats when they still are). You'll never get to hear people like those who talk at those meetings in the "real world" unless you fork over 1000 bucks. Take full advantage of it while you can. As a side note: being a member of CERIAS will get you VERY far in life if you are interested in the security field. -Minga From rajak@purdue.edu Mon, 25 Sep 2000 20:36:24 -0500 (EST) Date: Mon, 25 Sep 2000 20:36:24 -0500 (EST) From: Brian Poole rajak@purdue.edu Subject: [PLUG] [Fwd: Offtopic: CERIAS] On Mon, 25 Sep 2000, Rick Redman wrote: > > Has anyone been to CERIAS meetings or functions? Maybe you can answer > > this person. > > I am a CERIAS graduate. These "meetings" are actually lectures by > important people from the security field. It is actually offered > as a 1credit hour class (or it was 2 years ago). I'm a CERIAS student myself (so to echo Diego's prior statements, "Of course they are the best." ;). The seminars are still offered as a 1 credit hour lecture or you can just show up and listen to the seminars that interest you. > I would recomend them to everyone who doesnt mind a meeting at > 4pm on fridays =) (ifs thats when they still are). They are now, as of this semester, (thank god) on Weds., so it is not quite as incovenient as in years past. I would recommend that you go to the seminars that are of interest to you. I can't say that I have found all of them interesting, but a good number are. The web page is your friend, it has a brief description of the upcoming seminars. > > You'll never get to hear people like those who talk at those meetings > in the "real world" unless you fork over 1000 bucks. Take full advantage > of it while you can. Agreed, CERIAS hauls in some very good speakers who speak on a wide range of topics. > As a side note: being a member of CERIAS will get you VERY far > in life if you are interested in the security field. > > -Minga Shhh.. What are you trying to do to me here Minga, give me competition for my job!? :) -b From cnd@ecn.purdue.edu Mon, 25 Sep 2000 22:49:12 -0500 Date: Mon, 25 Sep 2000 22:49:12 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] [Fwd: linux difficulties] This is a multi-part message in MIME format. --------------86C9190923D1ACB68B96F8DA Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Anyone willing to help? -Chris -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- --------------86C9190923D1ACB68B96F8DA Content-Type: message/rfc822 Content-Transfer-Encoding: 7bit Content-Disposition: inline Received: from herald.cc.purdue.edu (herald.cc.purdue.edu [128.210.11.29]) by pier.ecn.purdue.edu (8.9.3/8.9.3moyman) with ESMTP id TAA16320 for ; Mon, 25 Sep 2000 19:51:20 -0500 (EST) Received: from roger.ecn.purdue.edu by herald.cc.purdue.edu with ESMTP for cnd@purdue.edu; Mon, 25 Sep 2000 19:51:19 -0500 Received: (from groth@localhost) by roger.ecn.purdue.edu (8.9.3/8.9.3moyman) id TAA20355; Mon, 25 Sep 2000 19:51:19 -0500 (EST) From: "Gregory L. Roth" Message-Id: <200009260051.TAA20355@roger.ecn.purdue.edu> Subject: linux difficulties To: cnd@purdue.edu Date: Mon, 25 Sep 2000 19:51:19 -0500 (EST) Cc: groth@ecn.purdue.edu (Gregory L. Roth) X-Mailer: ELM [version 2.5 PL2] MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Hi Chris, My name is Greg and I attended your PLUG callout earlier this semester. I have linux (Red Hat 6.2) installed on a computer in my office, but I'm having some difficulties with getting the network settings set correctly so that I can make use of features like rlogin from other hosts to this one, ssh, etc. Is there someone I might be able to contact that would be interested in helping me get these things ironed out? Thanks, Greg Gregory L. Roth Purdue University -- School of Aeronautics and Astronautics 1282 Grissom Hall, West Lafayette, IN 47907 email: groth@ecn.purdue.edu Phone: 765.496.6370 --------------86C9190923D1ACB68B96F8DA-- From will@physics.purdue.edu Tue, 26 Sep 2000 05:45:23 -0500 Date: Tue, 26 Sep 2000 05:45:23 -0500 From: Will Andrews will@physics.purdue.edu Subject: [PLUG] Extra CPUs Hi guys, I screwed up and ordered the wrong type of chips. If anyone wants to buy brand-new Intel PIII-600E Coppermine (100Mhz FSB) FCPGA (Socket 370 motherboards only) please let me know. I have two, will sell one or both cheap (US$140 each including fan). Let me know by private email. Thanks, -- Will Andrews GCS/E/S @d- s+:+ a--- C++ UB++++$ P+ L- E--- W+ N-- !o ?K w--- O- M+ V- PS+ PE++ Y+ PGP+>+++ t++ 5 X+ R+ tv+ b++ DI+++ D+ G++ e>++++ h! r- y? From cmaddy@purdue.edu Tue, 26 Sep 2000 13:29:00 -0500 Date: Tue, 26 Sep 2000 13:29:00 -0500 From: Craig Maddy cmaddy@purdue.edu Subject: [PLUG] Installfest Would anyone be interested in putting together an information packet for newbies to hand out at the installfest this year? Please let me know if you are willing to help out. Thanks, Craig Maddy From chrismcnett@hotmail.com Tue, 26 Sep 2000 20:28:04 GMT Date: Tue, 26 Sep 2000 20:28:04 GMT From: Chris McNett chrismcnett@hotmail.com Subject: [PLUG] Installfest I'll try to make something up. What all do you want in it? Just a description of Linux, its distributions, its benefits, etc? And do we have a date and time for Installfest yet? >From: Craig Maddy >To: plug@csociety.purdue.edu >Subject: [PLUG] Installfest >Date: Tue, 26 Sep 2000 13:29:00 -0500 >MIME-Version: 1.0 >Received: from [128.46.156.155] by hotmail.com (3.2) with ESMTP id >MHotMailBB9A35F100C4D821979A802E9C9B0D1E0; Tue Sep 26 11:29:07 2000 >Received: from csociety.ecn.purdue.edu (IDENT:mail@localhost [127.0.0.1])by >csociety.ecn.purdue.edu (8.9.3/8.9.3) with ESMTP id NAA00730;Tue, 26 Sep >2000 13:28:53 -0500 >Received: from herald.cc.purdue.edu (herald.cc.purdue.edu >[128.210.11.29])by csociety.ecn.purdue.edu (8.9.3/8.9.3) with ESMTP id >NAA00706for ; Tue, 26 Sep 2000 13:28:51 -0500 >Received: from [128.210.86.240] by herald.cc.purdue.edu with ESMTP for >plug@csociety.purdue.edu; Tue, 26 Sep 2000 13:29:02 -0500 >From plug-admin@csociety.purdue.edu Tue Sep 26 11:32:28 2000 >Message-Id: <39D0EAEC.A4533E03@purdue.edu> >X-Mailer: Mozilla 4.73 [en]C-CCK-MCD (Win98; I) >X-Accept-Language: en >Sender: plug-admin@csociety.purdue.edu >Errors-To: plug-admin@csociety.purdue.edu >X-Mailman-Version: 1.1 >Precedence: bulk >List-Id: The Purdue Linux Users' Group (PLUG) mailing list. > >X-BeenThere: plug@csociety.purdue.edu > >Would anyone be interested in putting together an information packet for >newbies to hand out at the installfest this year? Please let me know if >you are willing to help out. > >Thanks, >Craig Maddy > >____________________________________________________ >The Purdue Linux Users' Group (PLUG) mailing list. >plug mailing list - plug@csociety.purdue.edu >http://csociety.ecn.purdue.edu/mailman/listinfo/plug _________________________________________________________________________ Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com. Share information about yourself, create your own public profile at http://profiles.msn.com. From rjune@ims1.imagestream-is.com Tue, 26 Sep 2000 16:39:03 -0500 (EST) Date: Tue, 26 Sep 2000 16:39:03 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] Installfest I'll help too, I just got a nice new printer(epson SC760) Thanks to everyone who answered that question. :-) On Tue, 26 Sep 2000, Chris McNett wrote: > > I'll try to make something up. What all do you want in it? Just a > description of Linux, its distributions, its benefits, etc? > > And do we have a date and time for Installfest yet? > > >From: Craig Maddy > >To: plug@csociety.purdue.edu > >Subject: [PLUG] Installfest > >Date: Tue, 26 Sep 2000 13:29:00 -0500 > >MIME-Version: 1.0 > >Received: from [128.46.156.155] by hotmail.com (3.2) with ESMTP id > >MHotMailBB9A35F100C4D821979A802E9C9B0D1E0; Tue Sep 26 11:29:07 2000 > >Received: from csociety.ecn.purdue.edu (IDENT:mail@localhost [127.0.0.1])by > >csociety.ecn.purdue.edu (8.9.3/8.9.3) with ESMTP id NAA00730;Tue, 26 Sep > >2000 13:28:53 -0500 > >Received: from herald.cc.purdue.edu (herald.cc.purdue.edu > >[128.210.11.29])by csociety.ecn.purdue.edu (8.9.3/8.9.3) with ESMTP id > >NAA00706for ; Tue, 26 Sep 2000 13:28:51 -0500 > >Received: from [128.210.86.240] by herald.cc.purdue.edu with ESMTP for > >plug@csociety.purdue.edu; Tue, 26 Sep 2000 13:29:02 -0500 > >From plug-admin@csociety.purdue.edu Tue Sep 26 11:32:28 2000 > >Message-Id: <39D0EAEC.A4533E03@purdue.edu> > >X-Mailer: Mozilla 4.73 [en]C-CCK-MCD (Win98; I) > >X-Accept-Language: en > >Sender: plug-admin@csociety.purdue.edu > >Errors-To: plug-admin@csociety.purdue.edu > >X-Mailman-Version: 1.1 > >Precedence: bulk > >List-Id: The Purdue Linux Users' Group (PLUG) mailing list. > > > >X-BeenThere: plug@csociety.purdue.edu > > > >Would anyone be interested in putting together an information packet for > >newbies to hand out at the installfest this year? Please let me know if > >you are willing to help out. > > > >Thanks, > >Craig Maddy > > > >____________________________________________________ > >The Purdue Linux Users' Group (PLUG) mailing list. > >plug mailing list - plug@csociety.purdue.edu > >http://csociety.ecn.purdue.edu/mailman/listinfo/plug > > _________________________________________________________________________ > Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com. > > Share information about yourself, create your own public profile at > http://profiles.msn.com. > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From kuliniew@purdue.edu Tue, 26 Sep 2000 17:18:55 -0500 Date: Tue, 26 Sep 2000 17:18:55 -0500 From: Paul Kuliniewicz kuliniew@purdue.edu Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 I've been trying to get the SoundBlaster Live that came with my computer to work under Linux for quite some time and I haven't had any luck. I downloaded the latest drivers (multiple times) from http://opensource.creative.com and followed the instructions in the readme file to compile the driver and install it. But when I try to install it, I get errors about there being unresolved symbols. Here's a sample of what I get: -- begin copy-and-paste block -- [root@holly emu10k1]# make install mkdir -p /lib/modules/2.2.14-5.0/misc install -c -m 664 emu10k1.o /lib/modules/2.2.14-5.0/misc mkdir -p /lib/modules/2.2.14-5.0/misc install -c -m 664 emu10k1-joy.o /lib/modules/2.2.14-5.0/misc /sbin/depmod -a 2.2.14-5.0 depmod: *** Unresolved symbols in /lib/modules/2.2.14-5.0/misc/emu10k1.o /sbin/modprobe -r emu10k1.o emu10k1-joy.o [root@holly emu10k1]# /sbin/insmod emu10k1 Using /lib/modules/2.2.14-5.0/misc/emu10k1.o /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol unregister_sound_mixer_R7afc9d8a /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol register_sound_midi_R0570088d /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol register_sound_mixer_Re75b8c12 /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol unregister_sound_dsp_Rcd083b10 /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_memset /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_to_user /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_from_user /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol register_sound_dsp_R0c830ef5 /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol unregister_sound_midi_Rfdab6de3 [root@holly emu10k1]# /sbin/modprobe emu10k1 /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_memset /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_to_user /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_from_user /lib/modules/2.2.14-5.0/misc/emu10k1.o: insmod /lib/modules/2.2.14-5.0/misc/emu10k1.o failed /lib/modules/2.2.14-5.0/misc/emu10k1.o: insmod emu10k1 failed -- end copy-and-paste block -- I don't know if this is a problem with the drivers, my kernel, or something else with my configuration. I'm running a more-or-less off-the-ISO Red Hat 6.2 install (plus security fixes etc.) with the 2.2.14-5.0 kernel that came installed. I thought the Red Hat-supplied kernels had sound support compiled in (at least, the Red Hat 6.1 I installed back home did), so I sort of doubt it's a kernel problem. But then again, I don't know what I need to do to fix it. Also, as per the Sound HOWTO, I tried a "cat /dev/sndstat" and it failed, even though /dev/sndstat does exist. /proc/sound also doesn't exist. Of course, being somewhat of a newbie, I don't know if these should already be good-to-go or if I need to successfully get emu10k1 started before they'll work. Any thoughts? It's a shame being unable to listen to my 100% legal MP3 collection while I'm learning to hack Perl. Thanks in advance. From olenik@purdue.edu Tue, 26 Sep 2000 17:27:37 -0500 Date: Tue, 26 Sep 2000 17:27:37 -0500 From: Hayden Olenik olenik@purdue.edu Subject: [PLUG] 40MB file limit with RH7.0 I noticed a problem while I was using RH6.9.5; I couldn't download files more than 40MB, the file would just stop, and say that the file size limit was exceeded. I didn't think much of this because I was getting the error in some alpha quality software that I didn't trust. However, I tried to download the beta of Unreal Tournament from Loki, and got the same message, in Netscape. When I tried to resume the download in ncftp, I realized Linux was saying this, not the programs. I can get around the 40MB limit by downloading to my vfat partition, then extracting from the tar to my ext2 partition. but this is not a solution. Has anyone else had this problem? Is it new (ie as of RH 7.0beta, and only in rh7.0beta and up)? Is it only with Red Hat? Any help would be appreciated. Hayden From wirges@expert.cc.purdue.edu Tue, 26 Sep 2000 17:27:46 -0500 Date: Tue, 26 Sep 2000 17:27:46 -0500 From: Matt wirges@expert.cc.purdue.edu Subject: [PLUG] Web Site && Install Fest --------------0B17A0B87E40B61E53184B29 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit For any of you who are putting together information for newbies for INSTALLFEST, I need some help in getting a linux newbie guide and to finish, enhance,etc... my linux distro guide(I just don't have the time!). So if you are going to whip something up for installfest, let me know if you would also like to adapt it to the website. (I would need it soon, like in the next 1-3 weeks!) -Matt -- ============================================================================== | Matthew Wirges | | Systems Administrator, Student, Programmer, Geek && PLUG Webmaster | | Office Phone: [765]464-1148 | | Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com | | www.sharpwebinnovations.com | ============================================================================== --------------0B17A0B87E40B61E53184B29 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit For any of you who are putting together information for newbies for INSTALLFEST, I need some help in getting  a linux newbie guide and to finish, enhance,etc... my linux distro guide(I just don't have the time!).  So if you are going to whip something up for installfest, let me know if you would also like to adapt it to the website. (I would need it soon, like in the next 1-3 weeks!)
-Matt 
-- 
==============================================================================
| Matthew Wirges                                                             |
| Systems Administrator, Student, Programmer, Geek && PLUG Webmaster         |
| Office Phone: [765]464-1148                                                |
| Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com     |
| www.sharpwebinnovations.com                                                |
==============================================================================
  --------------0B17A0B87E40B61E53184B29-- From sdh@purdue.edu Tue, 26 Sep 2000 18:55:33 -0500 Date: Tue, 26 Sep 2000 18:55:33 -0500 From: Seth Heckard sdh@purdue.edu Subject: [PLUG] Red Hat 7... I've been running 6.9 since the day it came out, and had both of these problems. I figured it was a beta thing, so when 7 came out I did a clean reinstall (kept my /home tho) hoping that they would end up fixed. They didn't get fixed... 1. Keys don't repeat. In 6.9, they repeated fine in console, but not in X unless I pressed Shift-NumLock, which would conveniently disable my numeric keypad and turn it into a mouse (no kidding...). At least the keys would repeat though. In 7, they don't repeat in either console or X, but the Shift-NumLock trick still works in X at least. 2. My CD burner doesn't burn anymore. I'm not exactly sure if this is a hardware problem or a software problem -- I haven't checked it out in a windows box or anything. But it did work fine previously (and yes, I have scsi-generic support compiled in because it's a pos ide drive) here's the errors (using cdrecord): cdrecord: Input/output error. test unit ready: scsi sendcmd: retryable error CDB: 00 00 00 00 00 00 status: 0x2 (CHECK CONDITION) Sense Bytes: F0 00 02 00 00 00 00 0A 00 0D 8A 5E 3A 00 00 00 Sense Key: 0x2 Not Ready, Segment 0 Sense Code: 0x3A Qual 0x00 (medium not present) Fru 0x0 Sense flags: Blk 0 (valid) cdrecord: No disk / Wrong disk! (and yes, there's a disk in there...) Anyone have any idea? Seth From calx@purdue.edu Tue, 26 Sep 2000 19:17:05 -0500 (EST) Date: Tue, 26 Sep 2000 19:17:05 -0500 (EST) From: David calx@purdue.edu Subject: [PLUG] Red Hat 7... Try different brand of CDRs. > 2. My CD burner doesn't burn anymore. I'm not exactly sure if this is > a hardware problem or a software problem -- I haven't checked it out in > a windows box or anything. But it did work fine previously (and yes, I > have scsi-generic support compiled in because it's a pos ide drive) > here's the errors (using cdrecord): > > cdrecord: Input/output error. test unit ready: scsi sendcmd: retryable > error > CDB: 00 00 00 00 00 00 > status: 0x2 (CHECK CONDITION) > Sense Bytes: F0 00 02 00 00 00 00 0A 00 0D 8A 5E 3A 00 00 00 > Sense Key: 0x2 Not Ready, Segment 0 > Sense Code: 0x3A Qual 0x00 (medium not present) Fru 0x0 > Sense flags: Blk 0 (valid) > cdrecord: No disk / Wrong disk! > > (and yes, there's a disk in there...) > > Anyone have any idea? > > Seth > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From sdh@purdue.edu Tue, 26 Sep 2000 19:19:20 -0500 Date: Tue, 26 Sep 2000 19:19:20 -0500 From: Seth Heckard sdh@purdue.edu Subject: [PLUG] Red Hat 7... David wrote: > > Try different brand of CDRs. I've tried both cd-r and cd-rw disks... Seth From rjune@ims1.imagestream-is.com Tue, 26 Sep 2000 19:32:40 -0500 (EST) Date: Tue, 26 Sep 2000 19:32:40 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 insmod soundcore first. From cnd@ecn.purdue.edu Tue, 26 Sep 2000 23:57:49 -0500 Date: Tue, 26 Sep 2000 23:57:49 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] apache virtual hosts I'm setting up virtual hosts, this all works fine. Let's say that I have domain.com and foo.domain.com. Each of these have seperate DocumentRoots. I have this all working fine and dandy. The problem is that http authentication doesn't work on foo.domain.com, but works fine under domain.com. I'm confused as to why this is happening. Does anyone know of a reason that you cannot authenticate when using VirtualHosts??? Thanks, -Chris From dave@sr71.net Wed, 27 Sep 2000 00:01:17 -0500 Date: Wed, 27 Sep 2000 00:01:17 -0500 From: David C. Hansen dave@sr71.net Subject: [PLUG] rm crashing I have a very strange problem, rm keeps crashing, but only some of the time. "rm -f" always does it. Regular rm only does it some of the time, but does it more when I am doing lots of deletions, such as wildcard or recursive ones. One other strange thing I've noticed, du will give the Use% as - when rm is being bad. This is the only other anomaly I can find. Bad: Filesystem 1k-blocks Used Available Use% Mounted on /dev/sda1 877856 640772 192492 - / Good: Filesystem 1k-blocks Used Available Use% Mounted on /dev/sda1 877856 640772 192492 77% / BTW, it is a normal ext2 partition. -- David C. Hansen dave@sr71.net From sergent@etla.net Wed, 27 Sep 2000 05:13:10 -0700 Date: Wed, 27 Sep 2000 05:13:10 -0700 From: Jonathan Sergent sergent@etla.net Subject: [PLUG] rm crashing >I have a very strange problem, rm keeps crashing, but only some of the >time. "rm -f" always does it. Regular rm only does it some of the >time, but does it more when I am doing lots of deletions, such as >wildcard or recursive ones. > >One other strange thing I've noticed, du will give the Use% as - when rm >is being bad. This is the only other anomaly I can find. Type 'dmesg' and see if there are any kernel messages. I suspect rm is generating kernel oopses and not crashing itself... from your description I would guess that it is a hardware induced problem. It could just be a corrupted filesystem, but I would think that fsck would fix that. Maybe try forcing a full fsck while in single user mode. Hope this helps. --jss. From dave@sr71.net Wed, 27 Sep 2000 00:46:33 -0500 Date: Wed, 27 Sep 2000 00:46:33 -0500 From: David C. Hansen dave@sr71.net Subject: [PLUG] rm crashing That appears to have done it. Thanks for the help. I went to single user mode and ran fsck. Here are the filesystem errors I repaired: Deleted inode 512 has zero dtime Setting filetype for entry 'fs-1' in /tmp/.font-uniz (2385) to 6. Block Bitmap differences: -24630 Free blocks count wrong for group #0 (6510, counted=6511) Inode bitmap differences:-67104 Free inodes count wrong for group #4 (11603, counted=11604) directories count wrong for group #4 (324, counted=323) Was anything serious? -- David C. Hansen dave@sr71.net From sergent@etla.net Wed, 27 Sep 2000 05:49:20 -0700 Date: Wed, 27 Sep 2000 05:49:20 -0700 From: Jonathan Sergent sergent@etla.net Subject: [PLUG] rm crashing >That appears to have done it. Thanks for the help. I went to single >user mode and ran fsck. Here are the filesystem errors I repaired: ... Are you not getting the crashes anymore? The FS errors you reported are minor -- nothing that should have caused kernel weirdness. I suspect you still have hardware problems. I hope you didn't check the FS while it was still mounted read-write. Does dmesg (or syslog) show any messages from the times when you were getting bad things from rm? --jss. From dave@sr71.net Wed, 27 Sep 2000 02:24:04 -0500 Date: Wed, 27 Sep 2000 02:24:04 -0500 From: David C. Hansen dave@sr71.net Subject: [PLUG] rm crashing Jonathan Sergent wrote: > Are you not getting the crashes anymore? Nope > I suspect you still have hardware problems. The disk has been rock solid. The whole machine has been down twice since the beginning of the summer. Once because of a power failure, once for a kernel update. > I hope you didn't check the FS while it was still mounted read-write. Nope, remounted ro. > Does dmesg (or syslog) show any messages from the times when you were getting bad > things from rm? No, both were conspicuously silent on any FS problems. -- David C. Hansen dave@sr71.net From vids@mentor.cc.purdue.edu Wed, 27 Sep 2000 03:46:27 -0500 (EST) Date: Wed, 27 Sep 2000 03:46:27 -0500 (EST) From: Vidyut Samanta (aka vids) vids@mentor.cc.purdue.edu Subject: [PLUG] HELP!! (pcmcia problem) Hi this afternoon when I booted up after a normal shutdown and started getting these errors while starting pcmcia: NMI: lock error (debug interrupt?) CPU: 0 EIP: 0010 EFLAGS:0000002 eax: ff .... es... Process insmod(pid:528, process nr:7, stack page=c47c1000) Stack: ... CallTrace: ... .... And then if freezes there I can't use ^c or anything to kill it. I am pretty clue-less about what caused this problem. I haven't changed or installed anything since the last time it was working. Does anyone know what could have caused this and how I can fix it. Thanks in advance for your help. -vids From kuliniew@purdue.edu Wed, 27 Sep 2000 16:14:03 -0500 Date: Wed, 27 Sep 2000 16:14:03 -0500 From: Paul Kuliniewicz kuliniew@purdue.edu Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 >insmod soundcore first. Well, that takes care of a lot of the broken dependencies. After doing an insmod soundcore, I get: -- begin copy-and-paste block -- [root@holly emu10k1]# /sbin/insmod emu10k1 Using /lib/modules/2.2.14-5.0/misc/emu10k1.o /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_memset /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_to_user /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_from_user -- end copy-and-paste block -- Another person mentioned binary drivers on Creative's web site. I couldn't find them. The Drivers download page didn't have "Linux" an an OS option, and their search engine didn't work. I remember running across something once, but I can't find it anymore. Unfortunately, my SB Live didn't work right out of the Red Hat installation. Although sndconfig recognizes the SB Live, it doesn't configure it properly -- the sound test is silent. Maybe I should upgrade to a later version of the kernel? Or is there something else that will take care of those last three unresolved symbols? Thanks again. One of these days I'll get sound under Linux.... From reh@blacksoul.net Wed, 27 Sep 2000 16:33:29 -0500 (EST) Date: Wed, 27 Sep 2000 16:33:29 -0500 (EST) From: Ryan E. Helfter reh@blacksoul.net Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 Just as a side note, do remember that "modprobe" does take care of the module depending on other modules... like sb.o calls soundcore.o and a few others... try doing a "modprobe sb" im sure this probably is the only thing wrong, but if you have incomplete modules, a depmod -a should tell you... just my .02 ry On Wed, 27 Sep 2000, Paul Kuliniewicz wrote: > Date: Wed, 27 Sep 2000 16:14:03 -0500 > From: Paul Kuliniewicz > To: plug@csociety.purdue.edu > Subject: Re: [PLUG] Need Help w/ SB Live under Red Hat 6.2 > > > >insmod soundcore first. > > Well, that takes care of a lot of the broken dependencies. After doing an > insmod soundcore, I get: > > -- begin copy-and-paste block -- > [root@holly emu10k1]# /sbin/insmod emu10k1 > Using /lib/modules/2.2.14-5.0/misc/emu10k1.o > /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_memset > /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_to_user > /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_from_user > -- end copy-and-paste block -- > > Another person mentioned binary drivers on Creative's web site. I couldn't > find them. The Drivers download page didn't have "Linux" an an OS option, > and their search engine didn't work. I remember running across something > once, but I can't find it anymore. > > Unfortunately, my SB Live didn't work right out of the Red Hat > installation. Although sndconfig recognizes the SB Live, it doesn't > configure it properly -- the sound test is silent. > > Maybe I should upgrade to a later version of the kernel? Or is there > something else that will take care of those last three unresolved symbols? > > Thanks again. One of these days I'll get sound under Linux.... > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > -- Ryan E. Helfter reh@blacksoul.net Black Soul Networks, LLC -- From peddamat@purdue.edu Wed, 27 Sep 2000 16:34:50 -0500 (EST) Date: Wed, 27 Sep 2000 16:34:50 -0500 (EST) From: peddamat@purdue.edu peddamat@purdue.edu Subject: [PLUG] Problems with NIC and HPT366 patch... When I apply the HPT366 aka Ultra-DMA66 patch to my 2.2.17 kernel, it breaks the option to install my NIC (Realtek 8139a) when I run a make xconfig. Before I apply the patch, the option is selectable. Any suggestions? sumanth p. From meuserj@purdue.edu Wed, 27 Sep 2000 16:45:30 -0500 Date: Wed, 27 Sep 2000 16:45:30 -0500 From: John C Meuser meuserj@purdue.edu Subject: [PLUG] SMB file sharing I've been trying to share a directory using SMB, and I want it to be a guest share; no username or password required, read only access. But the closest I've come is I can see the shares on Network Neighborhood (on the client computer) but If I try to mount it, it says the share is not found. I've checked the read-write privilages on the directories and read through a ton of documentation, but I still can't get it. Is there something I'm not seeing? John From dajones@purdue.edu Wed, 27 Sep 2000 17:27:43 -0500 (EST) Date: Wed, 27 Sep 2000 17:27:43 -0500 (EST) From: Dave Jones dajones@purdue.edu Subject: [PLUG] Adaptec vs. kouwell? I have an Adaptec AHA-2940 standard 50-pin SCSI card in my 'puter (pci), but just acquired a Kouwell 910U (also stndrd 50pin pci)..I looked at some info on web sites for comparison..but no real useful information (in english anyway)....any thoughts or preferences from those experienced? oh, and using cdrecord to burn off a .iso file.. never used it before and having minimal difficulties heres my command syntax: cdrecord -dummy speed=4 dev=1,0 /home/dave/images/image.iso gives me: Cdrecord 1.8.1 (i686-pc-linux-gnu) copyright blah blah.. scsidev: '1,0' scsibus: 0 target: 1 lun: 0 cdrecord: No such file or directory, cannot open SCSI driver. cdrecord: For possible targets try 'cdrecord -scanbus'. Make sure you are root. "cdrecord - scanbus" gives me: Cdrecord 1.8.1 blah blah.. Using libscg version 'schily-0.1' scsibus1: 1,0,0 100) 'SONY ' 'CD-RW CRX100E ' '1.0j' Removable CD-ROM 1,1,0 101) * (continued to end) I even chown'd root to /usr/bin/cdrecord as suggested by man page when did not work before and this burner is IDE/ATAPI with full SCSI emulation and works beautifully as /dev/scd0 I probably should have put this into 2 threads,..but oh, well. Dave Jones - dajones@purdue.edu "Kill Everything." - KMFDM From wirges@expert.cc.purdue.edu Wed, 27 Sep 2000 17:32:59 -0500 Date: Wed, 27 Sep 2000 17:32:59 -0500 From: Matt wirges@expert.cc.purdue.edu Subject: [PLUG] SMB file sharing --------------876D7062D50931697381B0CD Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit > Make sure you have a share in smb.conf like this: [public] comment= Public Stuff path=/mnt/pub public=yes writable=yes (or no) printable=no And I know you said you checked the permissions, but double check the permissions!! > I've been trying to share a directory using SMB, and I want it to be a > guest share; no username or password required, read only access. But > the closest I've come is I can see the shares on Network Neighborhood > (on the client computer) but If I try to mount it, it says the share is > not found. I've checked the read-write privilages on the directories > and read through a ton of documentation, but I still can't get it. Is > there something I'm not seeing? > > John > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug -- ============================================================================== | Matthew Wirges | | Systems Administrator, Student, Programmer, Geek. | | Office Phone: [765]464-1148 | | Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com | | www.sharpwebinnovations.com | ============================================================================== --------------876D7062D50931697381B0CD Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit
Make sure you have a share in smb.conf like this:
    [public]
        comment= Public Stuff
        path=/mnt/pub
        public=yes
        writable=yes (or no)
        printable=no
And I know you said you checked the permissions, but double check the permissions!!
 
I've been trying to share a directory using SMB, and I want it to be a
guest share; no username or password required, read only access.  But
the closest I've come is I can see the shares on Network Neighborhood
(on the client computer) but If I try to mount it, it says the share is
not found.  I've checked the read-write privilages on the directories
and read through a ton of documentation, but I still can't get it.  Is
there something I'm not seeing?

            John

____________________________________________________
The Purdue Linux Users' Group (PLUG) mailing list.
plug mailing list  -  plug@csociety.purdue.edu
http://csociety.ecn.purdue.edu/mailman/listinfo/plug

-- 
==============================================================================
| Matthew Wirges                                                             |
| Systems Administrator, Student, Programmer, Geek.                          |
| Office Phone: [765]464-1148                                                |
| Email: wirges@expert.cc.purdue.edu || sysadmin@sharpwebinnovations.com     |
| www.sharpwebinnovations.com                                                |
==============================================================================
  --------------876D7062D50931697381B0CD-- From kuliniew@purdue.edu Wed, 27 Sep 2000 18:20:36 -0500 Date: Wed, 27 Sep 2000 18:20:36 -0500 From: Paul Kuliniewicz kuliniew@purdue.edu Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 >Just as a side note, do remember that "modprobe" does take care of the >module depending on other modules... like sb.o calls soundcore.o and a >few others... > >try doing a "modprobe sb" > >im sure this probably is the only thing wrong, but if you have incomplete >modules, a depmod -a should tell you... >just my .02 Here's some more error messages. -- begin copy-and-paste block -- [root@holly /sbin]# ./depmod -a depmod: *** Unresolved symbols in /lib/modules/2.2.14-5.0/misc/emu10k1.o [root@holly /sbin]# ./insmod emu10k1 Using /lib/modules/2.2.14-5.0/misc/emu10k1.o /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_memset /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_to_user /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_from_user [root@holly /sbin]# ./modprobe sb /lib/modules/2.2.14-5.0/misc/sb.o: init_module: Device or resource busy /lib/modules/2.2.14-5.0/misc/sb.o: insmod /lib/modules/2.2.14-5.0/misc/sb.o failed /lib/modules/2.2.14-5.0/misc/sb.o: insmod sb failed [root@holly /sbin]# -- end copy-and-paste block -- BTW, I finally found the binary drivers on Creative's web site. They didn't want to load because they were designed for an older kernel. Even if I forced it, though, they also didn't work. Any more ideas for how to fix this? Maybe I could get the correct files from one of you guys? I don't know. This is getting annoying.... From lava@mail.orchardsoft.com Wed, 27 Sep 2000 19:01:50 -0500 Date: Wed, 27 Sep 2000 19:01:50 -0500 From: Alan LaViolette lava@mail.orchardsoft.com Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 Well I just looked at my setup and see that I am running a very old version of the kernel 2.2.5, I know I should update but I just got it working again after about a year. When I run depmod -a it tells me I have unresolved dependencies, but if I rmmod emu10k1 and then insmod it everything works fine. The onlything I can suggest is since you have the source for the module, open it up and search for best_memset and see if you can gain any clues as to what module it needs. Plus make a check to make shure that you have the correct version for the 2.2.x kernels and not the one for the 2.3.x/2.4 stuff Alan L Paul Kuliniewicz wrote: > >Just as a side note, do remember that "modprobe" does take care of the > >module depending on other modules... like sb.o calls soundcore.o and a > >few others... > > > >try doing a "modprobe sb" > > > >im sure this probably is the only thing wrong, but if you have incomplete > >modules, a depmod -a should tell you... > >just my .02 > > Here's some more error messages. > > -- begin copy-and-paste block -- > [root@holly /sbin]# ./depmod -a > depmod: *** Unresolved symbols in /lib/modules/2.2.14-5.0/misc/emu10k1.o > [root@holly /sbin]# ./insmod emu10k1 > Using /lib/modules/2.2.14-5.0/misc/emu10k1.o > /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_memset > /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_to_user > /lib/modules/2.2.14-5.0/misc/emu10k1.o: unresolved symbol best_copy_from_user > [root@holly /sbin]# ./modprobe sb > /lib/modules/2.2.14-5.0/misc/sb.o: init_module: Device or resource busy > /lib/modules/2.2.14-5.0/misc/sb.o: insmod /lib/modules/2.2.14-5.0/misc/sb.o > failed > /lib/modules/2.2.14-5.0/misc/sb.o: insmod sb failed > [root@holly /sbin]# > -- end copy-and-paste block -- > > BTW, I finally found the binary drivers on Creative's web site. They > didn't want to load because they were designed for an older kernel. Even > if I forced it, though, they also didn't work. > > Any more ideas for how to fix this? Maybe I could get the correct files > from one of you guys? I don't know. This is getting annoying.... > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From kuliniew@purdue.edu Wed, 27 Sep 2000 19:46:10 -0500 Date: Wed, 27 Sep 2000 19:46:10 -0500 From: Paul Kuliniewicz kuliniew@purdue.edu Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 >Well I just looked at my setup and see that I am running a very old version of >the kernel 2.2.5, I know I should update but I just got it working again after >about a year. When I run depmod -a it tells me I have unresolved >dependencies, >but if I rmmod emu10k1 and then insmod it everything works fine. The >onlything >I can suggest is since you have the source for the module, open it up and >search >for best_memset and see if you can gain any clues as to what module it needs. >Plus make a check to make shure that you have the correct version for the >2.2.x >kernels and not the one for the 2.3.x/2.4 stuff Strangely, I can't find any references to best_memset and the rest in any of the source code files for the SB Live driver. A grep only showed references inside the .o files generated by the makefile. I also couldn't find any reference to them in the source code for the kernel. I'm running out of ideas here. After I installed RH6.2, I went and removed a bunch of the stuff I didn't need that the installer put on (like wuftp and sendmail). Could it be possible I removed something that's needed to get sound to work? Other than that, I'm running out of ideas for how to get this to work, and I don't really want to try a reinstallation to fix it, although it would be an excuse to try RH7.0.... Any other possibilities you can think of? From lava@mail.orchardsoft.com Wed, 27 Sep 2000 21:18:50 -0500 Date: Wed, 27 Sep 2000 21:18:50 -0500 From: Alan LaViolette lava@mail.orchardsoft.com Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 Look at the Makefile and see if it is including some other lib in the object or some strange thing like that. I have never seen a best_* function before, it does not have a man page. Alan L Paul Kuliniewicz wrote: > >Well I just looked at my setup and see that I am running a very old version of > >the kernel 2.2.5, I know I should update but I just got it working again after > >about a year. When I run depmod -a it tells me I have unresolved > >dependencies, > >but if I rmmod emu10k1 and then insmod it everything works fine. The > >onlything > >I can suggest is since you have the source for the module, open it up and > >search > >for best_memset and see if you can gain any clues as to what module it needs. > >Plus make a check to make shure that you have the correct version for the > >2.2.x > >kernels and not the one for the 2.3.x/2.4 stuff > > Strangely, I can't find any references to best_memset and the rest in any > of the source code files for the SB Live driver. A grep only showed > references inside the .o files generated by the makefile. I also couldn't > find any reference to them in the source code for the kernel. > > I'm running out of ideas here. After I installed RH6.2, I went and removed > a bunch of the stuff I didn't need that the installer put on (like wuftp > and sendmail). Could it be possible I removed something that's needed to > get sound to work? Other than that, I'm running out of ideas for how to > get this to work, and I don't really want to try a reinstallation to fix > it, although it would be an excuse to try RH7.0.... > > Any other possibilities you can think of? > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From rjune@ims1.imagestream-is.com Wed, 27 Sep 2000 22:46:50 -0500 (EST) Date: Wed, 27 Sep 2000 22:46:50 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 Something is *very* not right, I've got a Live! and it only required soundcore and the emu10k1 modules, Here is where the copy of the drivers I have reside. I've not looked at upgrading in a while, mostly because I don't have any reason to. http://www.bravegnuworld.com/~rjune/emu10k1.tar.gz uncompress, compile against rh6.2 and it should work like I said. On Wed, 27 Sep 2000, Alan LaViolette wrote: > Look at the Makefile and see if it is including some other lib in the object or > some strange thing like that. I have never seen a best_* function before, it does > not have a man page. > > Alan L > > Paul Kuliniewicz wrote: > > > >Well I just looked at my setup and see that I am running a very old version of > > >the kernel 2.2.5, I know I should update but I just got it working again after > > >about a year. When I run depmod -a it tells me I have unresolved > > >dependencies, > > >but if I rmmod emu10k1 and then insmod it everything works fine. The > > >onlything > > >I can suggest is since you have the source for the module, open it up and > > >search > > >for best_memset and see if you can gain any clues as to what module it needs. > > >Plus make a check to make shure that you have the correct version for the > > >2.2.x > > >kernels and not the one for the 2.3.x/2.4 stuff > > > > Strangely, I can't find any references to best_memset and the rest in any > > of the source code files for the SB Live driver. A grep only showed > > references inside the .o files generated by the makefile. I also couldn't > > find any reference to them in the source code for the kernel. > > > > I'm running out of ideas here. After I installed RH6.2, I went and removed > > a bunch of the stuff I didn't need that the installer put on (like wuftp > > and sendmail). Could it be possible I removed something that's needed to > > get sound to work? Other than that, I'm running out of ideas for how to > > get this to work, and I don't really want to try a reinstallation to fix > > it, although it would be an excuse to try RH7.0.... > > > > Any other possibilities you can think of? > > > > ____________________________________________________ > > The Purdue Linux Users' Group (PLUG) mailing list. > > plug mailing list - plug@csociety.purdue.edu > > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From kuliniew@purdue.edu Thu, 28 Sep 2000 13:04:39 -0500 Date: Thu, 28 Sep 2000 13:04:39 -0500 From: Paul Kuliniewicz kuliniew@purdue.edu Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 >Something is *very* not right, I've got a Live! and it only required >soundcore and the emu10k1 modules, Here is where the copy of the drivers I >have reside. I've not looked at upgrading in a while, mostly because I >don't have any reason to. >http://www.bravegnuworld.com/~rjune/emu10k1.tar.gz >uncompress, compile against rh6.2 and it should work like I said. Well, that definitely helps. I no longer get any compile errors, and both soundcore and emu10k1 load properly. However, even though the modules seem to be functioning perfectly, and I'm not getting any error messages, nothing comes out of the speakers! I've tried both mp123 and xmms -- both seem to think they're playing, but I can't hear anything. (And *yes*, my speakers are plugged in and the volume is turned up.) Any ideas how to get this to work? At least I'm not getting any error messages from anything anymore, but it's still not working. From he@expert.cc.purdue.edu Thu, 28 Sep 2000 13:07:57 -0500 (EST) Date: Thu, 28 Sep 2000 13:07:57 -0500 (EST) From: he@expert.cc.purdue.edu he@expert.cc.purdue.edu Subject: [PLUG] Printers. I bought a Lexmark ??40 which is a color jet printer printer with Postscript capability. Sorry I forgot the exact series character, it is some kind of "40". You can go to lexmark's hp and search jet printer with on-printer cpu and postscript support. The "40" is quite cheap and I bought from on-line for $100-$150. It works smoothly with my linux and can print postscript file very well. kevin On Thu, 21 Sep 2000, Richard June wrote: > I'm looking to buy a printer, can anyone recommend one that works with > linux(color isn't required, but would be nice)? if any of these are known > to work specificly, that would be cool. > > Lexmark Z52/Z42/Z31 > HP 648C/842C/895Cse/932C > Epson SC900/SC1160/SC760 > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From henkler@purdue.edu Thu, 28 Sep 2000 15:39:58 -0500 Date: Thu, 28 Sep 2000 15:39:58 -0500 From: Matthew Henkler henkler@purdue.edu Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 On my machine, XMMS defaults to the EsoundD Output by default. For some reason this doesn't load up by default on my machine either (so no sound will play). I either load esd once (type esd at the prompt), or switch the output plugin to OSS, and then it should work. What I don't understand is what the difference is between esd and OSS, and why I would want to use one or the other. I am not sure how to get esd to load up at boot either. matthenkler Paul Kuliniewicz wrote: > > >Something is *very* not right, I've got a Live! and it only required > >soundcore and the emu10k1 modules, Here is where the copy of the drivers I > >have reside. I've not looked at upgrading in a while, mostly because I > >don't have any reason to. > >http://www.bravegnuworld.com/~rjune/emu10k1.tar.gz > >uncompress, compile against rh6.2 and it should work like I said. > > Well, that definitely helps. I no longer get any compile errors, and both > soundcore and emu10k1 load properly. > > However, even though the modules seem to be functioning perfectly, and I'm > not getting any error messages, nothing comes out of the speakers! I've > tried both mp123 and xmms -- both seem to think they're playing, but I > can't hear anything. (And *yes*, my speakers are plugged in and the volume > is turned up.) > > Any ideas how to get this to work? At least I'm not getting any error > messages from anything anymore, but it's still not working. > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug From kuliniew@purdue.edu Thu, 28 Sep 2000 16:53:34 -0500 Date: Thu, 28 Sep 2000 16:53:34 -0500 From: Paul Kuliniewicz kuliniew@purdue.edu Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 >On my machine, XMMS defaults to the EsoundD Output by default. For some >reason this doesn't load up by default on my machine either (so no sound >will play). I either load esd once (type esd at the prompt), or switch >the output plugin to OSS, and then it should work. > >What I don't understand is what the difference is between esd and OSS, >and why I would want to use one or the other. I am not sure how to get >esd to load up at boot either. Amazingly, this doesn't work either. XMMS, along with every other program, is silent whether I start esd or not. Changing XMMS's output setting doesn't help either. Could it be OSS itself is broken somehow? Any chance an upgrade to Red Hat 7.0 could magically fix whatever this problem is? This is really getting annoying, but at least it's less broken than it was before. Oh well. If anyone else has any ideas, I'd love to hear them. Thanks to everyone who offered a suggestion. I didn't realize this was going to be so hard to fix. From lewis1@purdue.edu Thu, 28 Sep 2000 17:09:42 -0500 (EST) Date: Thu, 28 Sep 2000 17:09:42 -0500 (EST) From: James C. Lewis lewis1@purdue.edu Subject: [PLUG] Netscape freezing at start Hey everyone, My friend is having problems with all versions of netscape. I've seen the problem before, but can not for the life of me remember what I did to fix it. When netscape loads it shows the "splash" screen (version info) and then freezes up for a bit. It doesn't matter what the start page is set to (even on blank)... Anyone got any ideas? -James C Lewis _ /_ _ _ _ / ) / _ ' _ (_/(///)(-_) (__ (__(-((//_) lewis1@purdue.edu jlewis@netnitco.net "Only those who dare to fail greatly can ever achieve greatly." -- Robert F. Kennedy Please place all complaints in this box --> [] From jacoby@csociety.purdue.edu Thu, 28 Sep 2000 19:59:55 -0500 (EST) Date: Thu, 28 Sep 2000 19:59:55 -0500 (EST) From: David Jacoby jacoby@csociety.purdue.edu Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 On Wed, 27 Sep 2000, Paul Kuliniewicz wrote: > I'm running out of ideas here. After I installed RH6.2, I went and removed > a bunch of the stuff I didn't need that the installer put on (like wuftp > and sendmail). Could it be possible I removed something that's needed to > get sound to work? Other than that, I'm running out of ideas for how to > get this to work, and I don't really want to try a reinstallation to fix > it, although it would be an excuse to try RH7.0.... I dunno. I have an SB Live! Value in my sacrificial box, and it has worked w/o problem with both 6.(1|2) and 7b. > Any other possibilities you can think of? > > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > -- Dave Jacoby jacoby@csociety.purdue.edu From sdh@purdue.edu Thu, 28 Sep 2000 20:03:44 -0500 Date: Thu, 28 Sep 2000 20:03:44 -0500 From: Seth Heckard sdh@purdue.edu Subject: [PLUG] Computer Society Open Source CD Sale Don't have linux on your machine? Want it? Want FreeBSD too? The Computer Society can help you out! For those of you who don't know, the Computer Society sells linux/BSD CDs. Here's our current selection: Mandrake 7.1 (2 CDs): $10.00 RedHat 6.2: $5.00 Suse 6.4: $5.00 FreeBSD 4.0: $5.00 Slackware 7.1: $5.00 Stop by the Computer Society office in EE014 to purchase them. Check the list of office hours first to make sure there is someone there to sell you a CD (http://csociety.purdue.org/office/indexN4.html). Even if you don't want to buy a CD, stop in and say hi. We're nice people. Honest. Seth From kuliniew@purdue.edu Thu, 28 Sep 2000 21:17:09 -0500 Date: Thu, 28 Sep 2000 21:17:09 -0500 From: Paul Kuliniewicz kuliniew@purdue.edu Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 >Did you try the sndconfig tool that redhat provides. It plays a sample >sound if your sound card is configured correctly. I tried that way back in the beginning. When I had RH6.1, it said something like "Hey! You have a SB Live! We don't support that!" In RH6.2, it says it's configured properly, and then stays silent when the sample plays. Weird stuff, this. I'll cross my fingers and upgrade to 7.0. Maybe something magical will happen and whatever isn't working will be overwritten and be fixed. From cnd@ecn.purdue.edu Fri, 29 Sep 2000 09:37:43 -0500 Date: Fri, 29 Sep 2000 09:37:43 -0500 From: Christopher N. Deckard cnd@ecn.purdue.edu Subject: [PLUG] Presentation: Adding Perl Power to your Database Purdue.pm and PLUG are proud to announce the first in a series of Perl-related talks. This time, Kyle Wright will be talking about using the DBI modules in Perl. WHEN: Monday, October 2, 2000, 7 PM WHERE: Physics 112 TITLE: Adding Perl Power to your Database ABSTRACT: Everyone knows that Perl is a great language for manipulating data, but why should I use it with a database? Simple. Perl DBI allows a developer to write fewer lines of code, faster, and with fewer errors. Why Perl DBI? It is ideally suited for data processing such as parsing, cleansing, filtering, merging and formatting data into the database or creating reports. Perl DBI allows you to combine all the power that you have with writing a regular Perl script as well as being able to manipulate your database. See you there! -Chris Diego Zamboni PLUG President Purdue.pm -- ---------------------------------------------------------------------- Christopher N. Deckard | Lead Web Technician cnd@ecn.purdue.edu | Engineering Computer Network http://triad.dhs.org | http://www.ecn.purdue.edu/ECN/ ---------------------------------------------------------------------- From rjune@ims1.imagestream-is.com Sat, 30 Sep 2000 09:01:07 -0500 (EST) Date: Sat, 30 Sep 2000 09:01:07 -0500 (EST) From: Richard June rjune@ims1.imagestream-is.com Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 Why would you want to use esd, let me count the ways... 1. Multiple concurrent audio streams, something I don't think OSS stuff can dol. 2. Network transparency(mpg123 -s |esdcat -s kicks some ass) ok, so those are the two big reasons, but aren't they big enough? esd isn't a replacement for the OSS drivers, because it isn't a driver, its a daemon, all it does is catch audio streams, and feeds them to the OSS or Alsa driver. On Thu, 28 Sep 2000, Matthew Henkler wrote: > On my machine, XMMS defaults to the EsoundD Output by default. For some > reason this doesn't load up by default on my machine either (so no sound > will play). I either load esd once (type esd at the prompt), or switch > the output plugin to OSS, and then it should work. > > What I don't understand is what the difference is between esd and OSS, > and why I would want to use one or the other. I am not sure how to get > esd to load up at boot either. > > matthenkler > > Paul Kuliniewicz wrote: > > > > >Something is *very* not right, I've got a Live! and it only required > > >soundcore and the emu10k1 modules, Here is where the copy of the drivers I > > >have reside. I've not looked at upgrading in a while, mostly because I > > >don't have any reason to. > > >http://www.bravegnuworld.com/~rjune/emu10k1.tar.gz > > >uncompress, compile against rh6.2 and it should work like I said. > > > > Well, that definitely helps. I no longer get any compile errors, and both > > soundcore and emu10k1 load properly. > > > > However, even though the modules seem to be functioning perfectly, and I'm > > not getting any error messages, nothing comes out of the speakers! I've > > tried both mp123 and xmms -- both seem to think they're playing, but I > > can't hear anything. (And *yes*, my speakers are plugged in and the volume > > is turned up.) > > > > Any ideas how to get this to work? At least I'm not getting any error > > messages from anything anymore, but it's still not working. > > > > ____________________________________________________ > > The Purdue Linux Users' Group (PLUG) mailing list. > > plug mailing list - plug@csociety.purdue.edu > > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > > ____________________________________________________ > The Purdue Linux Users' Group (PLUG) mailing list. > plug mailing list - plug@csociety.purdue.edu > http://csociety.ecn.purdue.edu/mailman/listinfo/plug > From meuserj@purdue.edu Sat, 30 Sep 2000 09:15:12 -0500 Date: Sat, 30 Sep 2000 09:15:12 -0500 From: John C Meuser meuserj@purdue.edu Subject: [PLUG] Need Help w/ SB Live under Red Hat 6.2 > mpg123 -s |esdcat -s Why would you do that? mpg123 has esd support built in, well I guess if you wanted to use a different server. I agree that ESD kicks ass, it will make a half duplex card act like a full duplex. The command I like to use is esddsp . It makes the program in command think it has sound device all to itself. John From sminster@purdue.edu Sat, 30 Sep 2000 11:56:31 -0500 Date: Sat, 30 Sep 2000 11:56:31 -0500 From: Scott Minster sminster@purdue.edu Subject: [PLUG] esd not working (was: Need Help w/ SB Live under Red Hat 6.2) ESD used to kick ass, but no longer for me. A while ago it just stopped working, but I never put the time in to figure out why. The sound card still works (xmms through OSS plays just fine, for example), but esd can't seem to find it. Theses are the errors I get when I try to run esd (or mpg123): [scott@len ~]$ esd audio_alsa: no cards found!Audio device open for 44.1Khz, stereo, 16bit failed Trying 44.1Khz, 8bit stereo. audio_alsa: no cards found!Audio device open for 44.1Khz, stereo, 8bit failed Trying 48Khz, 16bit stereo. audio_alsa: no cards found!Audio device open for 44.1Khz, stereo, 8bit failed Trying 22.05Khz, 8bit stereo. audio_alsa: no cards found!Audio device open for 22.05Khz, stereo, 8bit failed Trying 44.1Khz, 16bit mono. audio_alsa: no cards found!Audio device open for 44.1Khz, mono, 8bit failed Trying 22.05Khz, 8bit mono. audio_alsa: no cards found!Audio device open for 22.05Khz, mono, 8bit failed Trying 11.025Khz, 8bit stereo. audio_alsa: no cards found!Audio device open for 11.025Khz, stereo, 8bit failed Trying 11.025Khz, 8bit mono. audio_alsa: no cards found!Audio device open for 11.025Khz, mono, 8bit failed Trying 8.192Khz, 8bit mono. audio_alsa: no cards found!Audio device open for 8.192Khz, mono, 8bit failed Trying 8Khz, 8bit mono. audio_alsa: no cards found!Sound device inadequate for Esound. Fatal. I know esd used to work, and that the sound card still does work, so how do I connect esd to the sound card? ---- Scott Minster sminster@purdue.edu http://mland.dhs.org/ icq://18777468/ (OK, that's not an established protocol, but it should be :) -----Original Message----- From: plug-admin@csociety.purdue.edu [mailto:plug-admin@csociety.purdue.edu]On Behalf Of John C Meuser Sent: Saturday, September 30, 2000 9:15 AM To: plug@csociety.purdue.edu Subject: Re: [PLUG] Need Help w/ SB Live under Red Hat 6.2 > mpg123 -s |esdcat -s Why would you do that? mpg123 has esd support built in, well I guess if you wanted to use a different server. I agree that ESD kicks ass, it will make a half duplex card act like a full duplex. The command I like to use is esddsp . It makes the program in command think it has sound device all to itself. John ____________________________________________________ The Purdue Linux Users' Group (PLUG) mailing list. plug mailing list - plug@csociety.purdue.edu http://csociety.ecn.purdue.edu/mailman/listinfo/plug From finnegpt@purdue.edu Fri, 01 Sep 2000 14:15:09 -0500 Date: Fri, 01 Sep 2000 14:15:09 -0500 From: Pat Finnegan finnegpt@purdue.edu Subject: [PLUG] meeting topic, various questions [Reply-Message-Removed] The "pport beowulf thing" is called PAPERS - Parallel Adaptor for Parallel Execution and Rapid Synchronization (or something like that). The URL I have for it is http://garage.ecn.purdue.edu/~papers/, but I couldn't get a connection to the machine. There is some detail about it in the Parallel-Processing-HOWTO (available from your favorite HOWTO vendor). [sadly enough] This was the main reason I decided to look at Purdue. Man I was weird back then... -- Pat "The Aztecs had a concept of zero. They would have made excellent computer programmers if the Spaniards hadn't wiped them out." -- Rick Kennell, EE264 - Advanced C Programming