Re: Net Install Requirements..
On Mon, Jun 17, 2002, Matthew Tedder wrote: I'd like to install Debian on an old 8MB i386 with ~500MB of disk space. Is this possible? I did it on a 386sx/20 with 8Mb RAM and 300MB HD with Debian Potato. A fair amount of swapping during the install was involved, but it did work fine after I got it on. Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Where is a Debian Tech. Overview?
On Tue, Jun 18, 2002, Matthew Tedder wrote: Where can I find a Debian technical overview? I am interested so I can figure out how to build my desktop server OS. That is, something that describes: a. The installation process Easiest thing, if possible, is to just try to find a spare machine and install it. Otherwise, you can read the installation manual on the website. b. What minimally must be installed Look into different prioirities of packages, base, important, standard, etc... Also, see Debian Policy Manual on development section of debian.org website. HTH, Daniel c. Where everything is And, does Debian support LSB 100% yet or nearly? Has is the state of LSB compliance? Matthew -- Anything that can be logically explained, can be programmed. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Difficulty mounting ext3 partitions with labels...
Hi, Found the answer to my own above question by examining e2fsprogs source (didn't think to make devfs connection), thus replying to myself for benefit of list archives. Quoting from Richard Gooch's devfs FAQ: Mount by volume LABEL=label doesn't work with devfs Most probably you are not mounting devfs onto /dev. What happens is that if your kernel config has CONFIG_DEVFS_FS=y then the contents of /proc/partitions will have the devfs names (such as scsi/host0/bus0/target0/lun0/part1). The contents of /proc/partitions are used by mount(8) when mounting by volume label. If devfs is not mounted on /dev, then mount(8) will fail to find devices. The solution is to make sure that devfs is mounted on /dev. See above for how to do that. Take care, Daniel On Wed, May 29, 2002, Daniel Freedman wrote: Hi, I'm having trouble mounting my ext3 partitions by referencing their labels. The transcript below details the situation. Also, please note, I've tried the same procedure mounting these partitions as ext2, and also naming them with labels that don't have the leading slash (in case the label code can't handle that, though I believe it should be able to). I can't find anything wrong with the following from either manpages or online documentation, and would very much appreciate any guidance. Thanks so much, Daniel littlebird:~# mount -l /dev/hda5 on / type ext3 (rw,errors=remount-ro) proc on /proc type proc (rw) devpts on /dev/pts type devpts (rw,gid=5,mode=620) /dev/hda6 on /boot type ext3 (rw) /dev/hda7 on /usr type ext3 (rw) /dev/hda9 on /var type ext3 (rw) tmpfs on /tmp type tmpfs (rw) littlebird:~# e2label /dev/hda8 /home littlebird:~# tune2fs -l /dev/hda8 | grep -C2 name tune2fs 1.27 (8-Mar-2002) Filesystem volume name: /home Last mounted on: not available Filesystem UUID: 4d8058ae----9dd6712b923a littlebird:~# mount -L /home /home mount: no such partition found littlebird:~# mount -U 4d8058ae----9dd6712b923a /home mount: no such partition found littlebird:~# mount /dev/hda8 /home kjournald starting. Commit interval 5 seconds EXT3 FS 2.4-0.9.17, 10 Jan 2002 on ide0(3,8), internal journal EXT3-fs: mounted filesystem with ordered data mode. littlebird:~# cat /proc/partitions major minor #blocks name 3 09820440 ide/host0/bus0/target0/lun0/disc 3 1 1 ide/host0/bus0/target0/lun0/part1 3 5 491337 ide/host0/bus0/target0/lun0/part5 3 6 98248 ide/host0/bus0/target0/lun0/part6 3 74883728 ide/host0/bus0/target0/lun0/part7 3 81950448 ide/host0/bus0/target0/lun0/part8 3 9 975208 ide/host0/bus0/target0/lun0/part9 3101421248 ide/host0/bus0/target0/lun0/part10 littlebird:~# mount -V mount: mount-2.11n littlebird:~# uname -a Linux littlebird 2.4.18-686 #1 Sun Apr 14 11:32:47 EST 2002 i686 unknown littlebird:~# -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Default existence of 'ld.so.conf' in Debian base install...
[I sent the following on Monday to debian-devel, but didn't receive any replies, so I thought I'd try debian-user. Thanks for any help.] Hi, I'm not a developer, just a Debian user trying to track down a 'postinst' error in a package on a brand-new Woody install (very sparse, only base + ~10 additional packages). So, my question is: Does Debian policy have anything to say about the guarantee of existence of '/etc/ld.so.conf'. It's clear (from 'dpkg -L libc6') that libc6 does not include it (even though that's were it would appear it should exist, if Debian were to include it by default), and it also appears not to be an explicit 'conffile' [1]. I further searched the Debian Policy manual, but could not find an explicit statement on this issue. Not much on target turned up on the debian-devel list archive. My base Woody install does not have any '/etc/ld.so.conf', and I'm wondering where the appropriate bug should be filed. The package in question above assumes that '/etc/ld.so.conf' exists (tries to 'grep' it), and fails when it can't be found. Should the bug be filed against that package (maybe it should 'touch /etc/ld.so.conf' in the beginning of its 'postinst' script?), or maybe libc6 should include, at the least, a blank '/etc/ld.so.conf'? I'm interested in anyone's thoughts on this. Thanks so very much. Take care, Daniel [1] Reply by Joey Hess: http://lists.debian.org/debian-devel/2001/debian-devel-200106/msg01397.html -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Clear HDD of old OS, etc?
On Fri, May 31, 2002, Dave Price wrote: Hi, I am looking for a quick way to clear an HDD of old data, partitions, etc. I found this on /. thru a google search: dd if=/dev/random of=/dev/hdX When i do this from a console shell after booting from a woody install disk, It does not seem to work ... i.e. I can still see the old partition table in fdisk ... Is there maybe a better dd invocation ? Try adding a 'count=some big number' aloha, dave -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Difficulty mounting ext3 partitions with labels...
Hi, I'm having trouble mounting my ext3 partitions by referencing their labels. The transcript below details the situation. Also, please note, I've tried the same procedure mounting these partitions as ext2, and also naming them with labels that don't have the leading slash (in case the label code can't handle that, though I believe it should be able to). I can't find anything wrong with the following from either manpages or online documentation, and would very much appreciate any guidance. Thanks so much, Daniel littlebird:~# mount -l /dev/hda5 on / type ext3 (rw,errors=remount-ro) proc on /proc type proc (rw) devpts on /dev/pts type devpts (rw,gid=5,mode=620) /dev/hda6 on /boot type ext3 (rw) /dev/hda7 on /usr type ext3 (rw) /dev/hda9 on /var type ext3 (rw) tmpfs on /tmp type tmpfs (rw) littlebird:~# e2label /dev/hda8 /home littlebird:~# tune2fs -l /dev/hda8 | grep -C2 name tune2fs 1.27 (8-Mar-2002) Filesystem volume name: /home Last mounted on: not available Filesystem UUID: 4d8058ae----9dd6712b923a littlebird:~# mount -L /home /home mount: no such partition found littlebird:~# mount -U 4d8058ae----9dd6712b923a /home mount: no such partition found littlebird:~# mount /dev/hda8 /home kjournald starting. Commit interval 5 seconds EXT3 FS 2.4-0.9.17, 10 Jan 2002 on ide0(3,8), internal journal EXT3-fs: mounted filesystem with ordered data mode. littlebird:~# cat /proc/partitions major minor #blocks name 3 09820440 ide/host0/bus0/target0/lun0/disc 3 1 1 ide/host0/bus0/target0/lun0/part1 3 5 491337 ide/host0/bus0/target0/lun0/part5 3 6 98248 ide/host0/bus0/target0/lun0/part6 3 74883728 ide/host0/bus0/target0/lun0/part7 3 81950448 ide/host0/bus0/target0/lun0/part8 3 9 975208 ide/host0/bus0/target0/lun0/part9 3101421248 ide/host0/bus0/target0/lun0/part10 littlebird:~# mount -V mount: mount-2.11n littlebird:~# uname -a Linux littlebird 2.4.18-686 #1 Sun Apr 14 11:32:47 EST 2002 i686 unknown littlebird:~# -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: base files for woody?
On Mon, Jan 14, 2002, Paul E Condon wrote: I requested help at the time on this list. I got a number of private responses that indicated to me that other people had failed and given up. I think there is a problem here. Some people are lucky. Some people are unlucky. This is not the way it should be. Did you remember to also use the four driver disks, in addition to the root and rescue disks? I've installed a number of machines (20) directly as woody without upgrading from stable. HTH, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: RAID repartitioning
On Fri, Jan 11, 2002, George Karaolides wrote: Trying to do this, I got the following error: # mkraid /dev/md11 handling MD device /dev/md11 analyzing super-block couldn't open device /dev/sda11 -- Device not configured mkraid: aborted, see the syslog and /proc/mdstat for potential clues. So I can't make my new partitions into a RAID-5 array. I haven't found any clues in either /proc/mdstat or the log files. Any suggestions? Hi, did you remember to modify /etc/raidtab to let it know about /dev/md11 ? HTH, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Is anyone using woody in a production environment?
On Fri, Jan 11, 2002, Roderick Cummings wrote: Is there a safe and stable way to build/install woody packages onto a potato system other than to dist-upgrade to woody? what's from with dist-upgrade? Only that the new 2.4 kernel-image wont mount my root partition: request_module[block-major-3]: Root fs not mounted VFS: Cannot open root device 303 or 03:03 Please append a correct root= boot option Kernenl Panic: VFS Unable to mount root fs on 03:03 Even though lilo.conf specifies the correct root= option, and fstab is correct as well. Did you remember to add the initrd line to lilo.conf? HTH, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: kickstart HOWTO?
On Tue, Jan 08, 2002, Robert L. Harris wrote: Here at the office I'm trying to get the redhat boxes replaced with debian. Tired of all the boxes being out of date and having to manually update. I'm working on a Debian vs Redhat comparison right now. One thing we need is a kickstart server. It's nice when we get in 5 machine to plug them into the network, throw in a bootable CD and walk away. When you come back it's built, configured 95% and ready to be put in use. I still haven't found a good kickstart howto for debian. Is there such a beast? Can this be done without weeks of work? You probably want FAI. See: http://www.informatik.uni-koeln.de/fai/ HTH, Daniel :wq! --- Robert L. Harris| Micros~1 : Senior System Engineer |For when quality, reliability at RnD Consulting | and security just aren't \_ that important! -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: passwordless ssh on woody failed
On Mon, Jan 07, 2002, Patrick Hsieh wrote: In Debian A, I ssh-keygen the public key, scp and append to Debian B ~/.ssh/authorized_keys It seems that Debian now uses protocol version 2, so maybe you need to add v2 key to ~/.ssh/authorized_keys2. Thanks. I end up with ssh-keygen -t rsa and then appended id_rsa.pub to .ssh/authorized_keys2 on the other machine. It works! Hi, Not sure exactly what's going on here, maybe you just have some config option specifying this in sshd_config, but, according to the release notes on openssh.org, for version 3.0 of openssh, authorized_keys2 is now depracated in favor of authorized_keys. Supposedly, IIRC, support for it might be removed in the future. HTH, Daniel -- Patrick Hsieh [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: starting point for kernel compile
On Mon, Jan 07, 2002, Hans Ekbrand wrote: On Mon, Jan 07, 2002 at 04:15:12AM -0800, David Wright wrote: I'm going to have to compile a custom kernel (need built-in nfs in order to etherboot) for the first time, so I am looking for advice. I would like to start with all the options set as for kernel-image-2.4.17-k6 and then just change the few options I need; is there a way to do this? Note that if you want NFS-swapping, you can not use 2.4.17 since it requires a kernel patch that only exists for some versions (2.4.9 and 2.4.13). I have used 2.4.9 for this reason. Not that I use either, but I thought the consensus was that, while neither option really yet worked great, most people had more success with using nbd, the network block driver for swap rather than swapping over an NFS mount. Any comments from people with experience with both? Also, is there good kernel compilatoin documentation out there? I know the basic steps; what I am looking for is more detailed advice on which options to concentrate on. Linux Terminal Server Project: www.ltps.org has some documentation on compiling kernels for etherboot. (See especially the documentation of version 3.0, it is has a chapter on the subject). -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: starting point for kernel compile
On Mon, Jan 07, 2002, Daniel Freedman wrote: On Mon, Jan 07, 2002, Hans Ekbrand wrote: On Mon, Jan 07, 2002 at 04:15:12AM -0800, David Wright wrote: I'm going to have to compile a custom kernel (need built-in nfs in order to etherboot) for the first time, so I am looking for advice. I would like to start with all the options set as for kernel-image-2.4.17-k6 and then just change the few options I need; is there a way to do this? Note that if you want NFS-swapping, you can not use 2.4.17 since it requires a kernel patch that only exists for some versions (2.4.9 and 2.4.13). I have used 2.4.9 for this reason. Not that I use either, but I thought the consensus was that, while neither option really yet worked great, most people had more success with using nbd, the network block driver for swap rather than swapping s/driver/device/ over an NFS mount. Any comments from people with experience with both? Also, is there good kernel compilatoin documentation out there? I know the basic steps; what I am looking for is more detailed advice on which options to concentrate on. Linux Terminal Server Project: www.ltps.org has some documentation on compiling kernels for etherboot. (See especially the documentation of version 3.0, it is has a chapter on the subject). -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Booting a (network) machine that has no kbd and no monitor attached.
On Sat, Jan 05, 2002, Shaul Karl wrote: I am setting up a machine that will be connected to the local network and will not have a monitor nor a keyboard attached to it. What changes, if any, should be made to the Debian software in order for such machine to be able to boot without complaining about the missing keyboard or monitor? In particular, what modifications are required for the initialization scripts and to the kernel? Are there any boot parameters that should be passed to the kernel about the lack of kbd/monitor? I am aware to the fact that the BIOS setting should probably be changed too. Yet I hope it would be the easier part. Is that so? Some BIOS's are (IMHO, foolishly) written so that the machine will not POST (power on self test) without a video card present. Generally, unlike the keyboard BIOS configuration, usually there's not option on the ones that won't work to make them work, short of getting motherboard manufacturer to change the BIOS or installing a video card and not using it. My preference is to avoid such boards. I'm assuming you've investigated to make sure your board will actually POST headless. HTH, Daniel Note that this machine has enough RAM. It also has its own hard drive. At least initially that machine will not run X at all. Later on it might be used to run the X clients. Short replies with pointers to documentation will be appreciated too. -- -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Supported video cards
On Mon, Jan 07, 2002, Penguin wrote: In general, is it okay to buy an AGP video card for Debian Potato 2.2r4? I really don't want the hassle of trying to upgrade to XFree 4+ (I've tried and failed many times) and I'm not sure if I want woody or sid just yet. AGP in general is fine. Where can I get the list of video cards that Debian Potato supports? I will buy a card from that list. See xfree86.org. Potato includes Xfree86 3.3.6, so if you specifically want support for potato, buy a card from that list. However, Debian 3.0/woody will be released as the new stable within a few months, and it includes Xfree86 v. 4.1, so I'd probably buy a card from the list of supported cards for that 4.1 version of Xfree86, as that will give you access to newer cards (that might even be better bang for the buck). HTH, Daniel Thanks, James -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: OT: Language War (Re: C Manual)
On Thu, Jan 03, 2002, Phil Beder wrote: Thank you!! The diversity of point of view and depth of knowledge of the participants of this group is truly phenomenal. A simple question (in essence where should I start) yielded me not only an interesting variety of response to that question, but a road map, complete with pitfalls and milestones and a vision of where I should end-up. Questions I would have never though to ask were answered, as the ping pong ball of opinion flew around. Why use a low level language like C for GUI applications -vs.- why not when one language will do the trick. The benefits of being able to allocate and access memory locations directly. Old standards -vs.- ANSI 99. Types, Classes, Portablilty!! . . . WOW I understand now why C was creating such a stir back in 1989 when I first started into programming. It's abilities both as a low level and high level language are, I believe, what make it so universally accepted. With C a programmer enjoys the flexibility to write a function many different ways, which means I don't think I would look forward to maintaining C code written by a bunch of programmers with diverse views (but I guess I sure would learn a lot). Thanks for your help, . . . all of you. I'll be sure to avoid the rest of Herb Schildts books (I got a small inexpensive programmers reference of C keywords functions and some common C++ functions that has comes in handy for figuring-out usage and syntax). After what you guys said, Ill bet his ears are ringing. I haven't found the Kernighan and Ritchie book in my local bookstore. I have been using a C for Linux book which seems to get right to the point and I appreciate the direct application to Linux and the gcc compiler. Hi, No comments to add about the language wars, but as far as books go, I find often the easiest method by which to determine what books are worth looking at and what are junk, is by publisher. I don't think it's an anomoly that out of 30+ computer science books I have around, almost all are published by O'Reilly, Addison Wesley, PTR/Prentice Hall, John Wiley, and New Riders. These publishers, IMHO, uniformlly produce much higher quality CS/IT books than most others. I've found generally that Que, Sams, IDG/Hungry Minds, and others (I can't remember right now) are significantly worse, and rarely even consider them. I suggest a similar strategy might prove useful to you as well. I also recommend bookpool.com (standard no affiliation disclaimer) as a great place to shop online, especially for price and customer service. HTH, Daniel I wish I was a good enough programmer to contribute to this great project. Maybe one day when I understand more about Linux I could write a more user friendly help interface with clear syntax, option, and flag usage. Thanks again -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: kernel complaint
On Wed, Jan 02, 2002, dman wrote:
On Wed, Jan 02, 2002 at 04:01:43PM -0800, [EMAIL PROTECTED] wrote:
| I've tried recompiling the kernel,
This error has nothing to do with SMP. You have probably done one (or
more) of the following :
o not included your disk driver in the kernel (you will need
both the driver for your IDE controller and your SCSI
controller)
o not included support for your root filesystem (probably ext2)
o otherwise misconfigured your kernel
The solution is to carefully go through all the options in the
configuration tool and verify that you have included all the necessary
drivers. You can't have your SCSI controller as a module since that
would require reading the disk to load the module that allows reading
the disk to load the module... . Alternatively you can just install
kernel-image-2.4.16-686-smp or kernel-image-2.4.17-686-smp and setup
lilo correctly (RTFM on 'initrd' too).
And, remember that if you use initrd to have initrd-tools installed,
to add the 'initrd=/boot/initrd' (substitute your path to initrd) line
to lilo.conf and rerun lilo, and finally, if you use 'make-kpkg' to
build kernels the debian way with initrd, you have to pass it the
extra flag ('--initrd').
HTH,
Daniel
-D
--
Daniel A. Freedman
Laboratory for Atomic and Solid State Physics
Department of Physics
Cornell University
Re: 2.2 kernel vs. 2.4 kernel, please help!
On Tue, Jan 01, 2002, Penguin wrote: Debian website says 2.4 is not really mature enough yet, does this mean mature enough security wise also? I am trying to figure out what kernel I should use, for the next 6 months at least anyway. Security is my main objective, most other things can go to the wall if they conflict. But I would like my Nvidia card to work with a 2.2 kernel, and I would like to be able to use cdrecord/xcd-roast, and other goodies too. I need to be able to run XFree 4.1. I would like to be able to have sound, using a SoundBlaster 128 Live card. I would prefer iptables, since I know it has a LOG option to record all incoming and outgoing like a packet sniffer for my modem dialup connection - does ipchains with the 2.2 kernel have a comparable option? I want to see if when I log into my POP3 server if my user and pass is sent in clear, etc etc. I want to see everything passing on the wire. Can I use iptables with a 2.2 kernel? Given that I am super paranoid, maybe my old Debian 2.2r2 Potato is the best bet for me. Is there any reason why I may not be able to upgrade X to XFree 4.1 with this version of Potato? If you're superparanoid (as you put it), I would suggest getting an old PC (maybe a Pentium, which can easily be had for $100), and making a dedicated box to serve as a firewall between your workstation and the wild internet (also could nicely double doing impasqing if you have multiple other boxes). On that I'd put either 2.2.20 with ipchains (since it's more mature), 2.4.17 with iptables (easier to configure secure firewall scripts than ipchains, IMHO), or OpenBSD. Even better, maybe, (though I have zero personal experience with the following), might be a linux firewall-specific distribution, such as: Linux Router: www.linuxrouter.org Smoothwall: www.smoothwall.org Gibraltar: gibraltar.vianova.at (which is based upon Debian) Hope this helps and happy new year! -Daniel Thanks :) -- Penguin [EMAIL PROTECTED] Girls are for pleasure; boys are for ecstasy. -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: C Manual
On Fri, Dec 28, 2001, dman wrote: On Fri, Dec 28, 2001 at 09:31:36AM -0800, Paul Mackinney wrote: | Gary Turner muttered: | On Thu, 20 Dec 2001 16:11:33 -0500, Phil Beder wrote: | | Where can I find a good, complete manual for C and C++ programming | languages for the gcc compiler. | ... | Kernighan and Ritchie is 'the' C book, ... Be aware though that KR is the old standard. Their book describes KR C which is very similar, but a bit different than ANSI C. Still, KR is a highly recommended reference by many people. (I don't have it, but my friend's dad has an really old copy of it) The second edition of the KR book, which was published a number of years ago, was updated to comply with ANSI C, and even says so right on the cover :) I have The Complete C++ Reference by Schildt, and it includes a good reference of the standard libraries as well as explaining the languages. I never bought it, but some of my friends said that Pointers on C by Ken Reek is good (he's a professor at RIT). -D I have Pointers on C and agree with your friends that it's quite good. Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Kernel compilation
On Thu, Dec 27, 2001, Jens Müller wrote: I have tried to compile my own kernel 2.4.16. I took the config file k7 from the Debian kernel source, and just activated acpi additionally. Hi Jens! When I installed it, that kernel complained about not being able to mount the root device. Maybe it has something to do with that the Debian kernel has an initrd and mine hasn't? Is an initrd necessary? No, it's not necessary but if your config files assume initrd (in other words, you've compiled it in), you have to make it work, or compile with alternate config files that don't assume initrd. Make sure you have the proper 'initrd=/boot/initrd.img-2.4.16' (or similar path to your initrd image) in your 'lilo.conf' and rerun lilo. Are you using kernel-package to compile your kernel (I'd suggest you do, if you're not)? Have you installed initrd-tools? HTH, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: jumpstart for debian?
On Wed, Dec 26, 2001, Lev Lvovsky wrote: is there any linux/debian equivalent of jumpstart as there is on Solaris? You might want to look into FAI: the Fully Automated Installer (packaged for, and, I believe, developed on, Debian). HTH, Daniel thanks, -lev -- personal site :: www.sonous.com rave site :: raves.sonous.com I'm a DJ! site :: djkgb.sonous.com Progess is the direct result of dissatisfaction. -Mark Rudholm -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Swap Partition/File
On Wed, Dec 26, 2001, Paul A. Thomas wrote: I've two questions tonight which seem simple but not... straight forward. Hi Paul, I downloaded three ISO images and burned them to CD's. 1: I've booted from disk (CD) 1 and 2: both state they are recovery disk/boots... but then allow you to go into the installation applet. The only other ( very limited ) experience I have with Linux is with RH 7.x I don't recall RH Welcome to Debian and Linux! indicating the system was booting to a recovery disk... Am I interpreting the start up correctly? Are both/all of these Cd's intended to allow access to a damaged system? Debian's boot system is generally called 'boot-floppies' (for historical reasons, I believe) even when not actually being used from floppy disks. The first 'disk' for installation purposes is called the 'rescue' disk, and the second is the 'root' disk, then there are usually the driver disks. So, this looks like you're doing fine so far. 2: Related to #1 above: How many CD's should there be for Debian's current release? I have three however in reading the man (manual?) page for cfdisk at debian.org they reference 4 CD's. My concern is working/playing with Debian Linux but not being able to apply what I find in help and MAN files properly as the references are off in the set I obtained. That set of 3 was downloaded from http://www.linuxiso.org/debian.html, for the Intel 386 architecture. I'm not really clear what Debian release you're using. I'm not sure if I should interpret the 'current' release to mean Debian stable (AKA potato), or maybe you mean Debian testing (AKA woody). I'm guessing you probably don't mean Debian unstable (AKA sid), and you probably mean Potato. Don't worry so much about not having everything on CD. It's easiest just to use the first binary i386 CD to get most of the system up, then update your '/etc/apt/sources.list' to include some http/ftp internet download sites, do an 'apt-get update' as root, and then slurp whatever additional packages you need off the net with 'apt-get install packagename'. 3: I assume dbootstrap was started upon bootup with the CD noted above because the commands I am working with are not command prompt commands but echo what the MAN pages and installation manual show for CFDISK. I've ( many times ) deleted and created new partitions on the 40 gig HD I have in this system. I believe my goal here is to create a 'kernel' partition of 5 megs, a 10 meg Swap partition, and whatever for general file use. I can create all of the partitions and can make one of them a 'swap' partition, however the GUI? ( Graphical interface of dbootstrap which uses keyboard command to negotiate through it ), after I write the changes to the drive, tells me I need to create a swap drive. If I try to activate the partition I just created and marked as a 'swap' partition the program states there are no new 'swap' partitions on the drive. I assume by 'kernel' partition you mean a '/boot' partition? 10Meg swap is probably woefully inadequate, unless that's a typo and you mean 10Gig swap. You generally want 2-3x your physical ram for swap. Make sure you've properly changed the swap filesystem type from type linux to type linux-swap. You can do this from within cfdisk after creating the partition. Also see Karsten's partition mini-HOWTO at: http://pw1.netcom.com/~kmself/Linux/FAQs/partition.html It's late. I've spent the day giving massages and a pedicure to Microsoft networks.. so I'm pretty sure the answer is in front of me However I'd prefer not to throw this system out the window ( the descriptions I read indicate I'll prefer Debian over other Distributions ). Could somebody provide a late night hint? Thanks! Paul Hope that helps and take care, Daniel PS Please try to wrap your lines better if possible for readability. Thanks. -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Basic Debian size
On Sun, Dec 23, 2001, Dan Robinson wrote: I'm a Linux pre-newbie, as I'm sure you'll notice. Someone gave me a TI TravelMate 5000 laptop computer (for parts for a different model, but this one works better). It has no CD drive. A friend spent about 4 hours installing Debian, (w/o GUI, which I avoid) using a network connection. It still has major problems, which I may have made worse by trying to use DOS in it's partition. I'm wondering about the possibilities of starting over, preferably preserving the partitions and LILO. How many floppies and files it would take to get a Linux system (kernal, shell, more?) that could then tranfer other files from floppies? Then what's the MINIMUM it would it take to get a reasonably functioning system for editing, email and the web? Dan, Part of this depends if you want a typical minimum reasonably functioning system for editing, email, and the web, such as that provided by a distribution, which is going to general make some assumptions about usage, or if you are willing (and technically proficient enough) to essentially roll your own distribution, giving you the flexibility (that's one of the hallmarks of Free Software and Linux, after all) to seriously reduce the size. For this, you might want to look into Linux from Scratch (www.linuxfromscratch.org). Until a few months ago I was on the net in DOS on a 386, using software that would have worked on a 286. I only upgraded because the software was falling behind, not staying compatible with other Considering you can put together really solid Athlon (1GHz) systems with plenty of RAM (512MB) and disk space (40GB) for ~$400 (sans monitor), you might want to think about upgrading just for reliability of components if nothing else (though possibly, at least in high energy cost areas, simply the energy savings of new equipment, if left on 24/7 as mine (and I think other's) is, could justify the upgrade). systems. At times I wonder how much I've really gained since CPM when I ran my own BBS, at first without a hard drive. I noticed recently that my DOS system disk has about 198K in six files. Then there are many utilities, few of which I've ever used, or even know what they're for (and I don't care for Ws, now in two contexts). In Linux circles I hear mainly about needing gigabytes. I'm wondering if there's a basic Linux system, or other open source systems, within a couple orders of magnitude to DOS in size. Otherwise, can someone tell me how Linux is that much better, or are programmers getting sloppier in writing code? I think the comparison between the Linux systems that are many orders of magnitude greater in disk space usage than the DOS systems you are comparing them to is skewed by the fact that the Linux systems are providing vastly more software capabilities (full compiler/development suites for multiple languages, full networking capabilities with many different servers: ftp, http, irc, ssh, lpr, smtp, etc.), XFree86, etc.) and that is reflected in the usage. I mean, if you really want minimum space, a more accurate comparison to DOS might be something like a Linux rescue disk, which is less than 1.44MB, and you can look into if you're interested. I like Tom's Root Boot disk at http://www.toms.net/rb/. For small linux installations, you might also look into: 1. The Linux Terminal Server Project at http://www.ltsp.org/ 2. http://www.embedded-linux.org/ 3. http://www.linuxdevices.com/ 4. http://www.uclinux.org/ (not probably applicable for your needs, but just for your general info) HTH, Daniel Dan Robinson [EMAIL PROTECTED] Eugene OR 97401 http://www.efn.org/~danrob/ Capitalism is the ultimate pyramid scheme, dependent on ever-growing economy, therefore on ever-growing population, therefore on ever-growing resources. Oops! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: /dev/cdrom problem...
On Sun, Dec 23, 2001, JP Glutting wrote: Hi All, I have a problem with my device configuration. I have a CD player and a CD writer (/dev/hdc and /dev/hdd, respectively), and whenever I boot up it links up the writer on /dev/cdrom1, and the reader on /dev/cdrom2 (/dev/cdrom is, correctly a link to /dev/cdrom1). I can change the links, but they go back to the other configuration after a reboot. Of course, this screws up cd players, rippers, writers, etc. Any info on how to configure this correctly? JP, You don't give much information, but maybe you're using a kernel with devfs mounted automatically (or you, or one of your /etc/init.d scripts, are mounting it manually). Since devfs is a virtual file system (I like to think of its analogy to the /proc fs), any changes made to it are not preserved across reboots (or actually across unmounting and remounting it). If my 'devfs' guess is correct, you can find more information about it at: http://www.atnf.csiro.au/~rgooch/linux/docs/devfs.html Hope this helps, Daniel Thanks, JP -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Dual Processor machine
On Wed, Dec 12, 2001, Jérôme-Georges-Michel BENOIT wrote: Thanks you very much for your advice. do you have a real smp kernel installed? Good question ;-) I guess that I have installed Debian in the basic way: 1] how can I check this point ? [EMAIL PROTECTED]:~$ grep SMP /boot/config-2.4.16 CONFIG_SMP=y [EMAIL PROTECTED]:~$ Replace 2.4.16 above with your kernel version. Alternatively (look for multiple CPU listings like below): [EMAIL PROTECTED]:~$ cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 6 model : 6 model name : AMD Athlon(tm) MP Processor stepping: 1 cpu MHz : 1194.681 cache size : 256 KB fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 mmx fxsr sse syscall mmxext 3dnowext 3dnow bogomips: 2385.51 processor : 1 vendor_id : AuthenticAMD cpu family : 6 model : 6 model name : AMD Athlon(tm) Processor stepping: 1 cpu MHz : 1194.681 cache size : 256 KB fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 mmx fxsr sse syscall mmxext 3dnowext 3dnow bogomips: 2385.51 [EMAIL PROTECTED]:~$ 2] may I configure some files ? 3] may I build a new kernel ? Thanks, Jerome BENOIT -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Installation scripts - a request
On Tue, Dec 11, 2001, Johann Spies wrote: I am not sure whether this is the correct list for such a request but I put it anyway. I have had the privilege to help more than 20 people from a windows background to install Debian Linux in the past two weeks. We found the debian installation procedure not very easy but workable. Hi Johann, Glad that you're helping make people aware of Debian. It's really best for people transitioning from Windows to Linux to have someone such as yourself to guide them through the process. That's why InstallFests (at Linux User Groups (LUG's)) are good ideas, IMHO. A request of many of the people was that there should be more opportunity to go back to a previous screen during the installation scripts - especially while configuring packages like xserver-xfree86. If a mistake was made during the installation process e.g. while selecting packages, it is not always possible to go back and rectify the situation. For a newbie this is a serious problem. While it's not a direct answer to your problem, as it does not enable one to go back and forth between screens in the install process, if you're not aware of it, 'dpkg-reconfigure' is a very nice addition to the Debian packaging tools, essentially allowing you to re-answer all of debconf's install question (and furthermore, allows you to specify the priority of questions you want to reanswer.) This program is part of the debconf package. See: 'apt-cache show debconf' or 'man dpkg-reconfigure' HTH, Daniel Regards. Johann -- Johann Spies Telefoon: 021-808 4036 Informasietegnologie, Universiteit van Stellenbosch The spirit of the Lord GOD is upon me; because the LORD hath anointed me to preach good tidings unto the meek; he hath sent me to bind up the brokenhearted, to proclaim liberty to the captives, and the opening of the prison to them that are bound. Isaiah 61:1 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Serial Terminals
On Mon, Dec 10, 2001, Titty Jacob wrote: Hi All, Has anyone configured linux serial terminals. I have a specialix ISA card and some terminals. How do I configure them. Tt Hi Tt, Please see: http://www.linuxdoc.org/HOWTO/Text-Terminal-HOWTO.html HTH, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: using mkinitrd
On Mon, Dec 10, 2001, Richard Weil wrote: I've compiled a custom kernel using kpkg, but I'm unsure how to make an initrd like the official kernels. I configured a 2.4.16 kernel with support for cramfs and initrd. I tried using mkinitrd by invoking: mkinitrd /lib/modules/2.4.16 and it output lots of junk to the xterm I was in. So, before actually making a mistake that might be difficult to undo, is the correct syntax for mkinitrd: mkinitrd -o /boot/initrd-2.4.16 /lib/modules/2.4.16 Do I need to use anything other than mkinitrd? Richard Hi Richard, I think it would be much easier for you to use kernel-package and then compile your kernel with initrd support as such: make-kpkg --initrd --custom=custom.0.1 kernel-image This will create a deb package that upon INSTALL (not compile) will then call mkinitrd and otherwise take care of everything for you. HTH, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Problems with less,emacs,etc. on VT420 on ttyS0...
Hi, I have an actual VT420 terminal connected to ttyS0 of my Linux box, running Debian woody/testing, and am having trouble using certain programs (e.g. less, jed, emacs, etc.) on the terminal, though others (e.g. lynx, vim, more, etc.) run fine. I've tried configuring the VT420 through its setup both using Xon/Xoff (software) flow control and (probably) the preferred hardware (VT420 calls it modem) flow control (which uses DTR/DSR lines). I'm runing agetty (Debian's getty) to set up the terminal login prompt: getty ttyS0 9600 vt420 (I also use the -h flag when I enable the DTR modem control in the VT420's setup.) The primary problem is that less, for example, automatically stops after displaying the first screen, jed stops immediately upon invocation, and emacs doesn't like to run with Xon/Xoff at all (which I believe is a known issue), but also displays lots of junk (probably buffer overruns) even when I use hardware flow control, which I believe should stop this. I thought it might be something with ncurses not getting along with my flow control, but my TERM environmental variable is properly set at 'vt420', and lynx works fine (and uses, I believe, ncurses). I'm wondering if, at least for the hardware modem flow control stuff, it might be that the linux kernel's serial driver doesn't support DTR/DSR flow control (I'm running 2.4.16 on this box), but I've not been able to get a good handle on whether this is true. Grepping through the serial.c source showed references to DTR/DSR, but none that enabled me to nail down its level of support. I believe that my cable is properly wired (followed vt100.net instructions). I have an MMJ null-modem cable, and connect TXD+ to pin 2, RXD+ to pin 3, DTR to pin 8, DSR to pin 7, and TXD- to pin 5 of my DB9 serial connector on my linux box. Anyway, bottom line, has anyone had success running these types of applications on a vt420 (or other vt) in a similar setup? Thanks so very much for any and all suggestions. Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: UPS software
Mark, I have a APC SmartUPS 1000Net and 1250 and both have served me very well. While APC used to not release any drivers (actually more like software than kernel drivers) for linux, they now not only release software, they do so (for much of it at least) under the GPL. Please see: http://www.apcc.com/go/machine/partners/open_source.cfm (They're even kind enough not to make you register... They say: Download requires registration. If you are not comfortable registering to download the...source code, enter user id: opensource and password: opensource on the subsequent login page.) Also here's their main powerchute software (don't run it, try debian package 'nut' instead, so I don't know how redhat specific it is, although this version must be easier than the specific RPM one.): ftp://ftp.apcftp.com/software/unix/linux/pcplus/453/pcplus_453_redhat.tar If that ftp doesn't work, try the main link to this software: http://www.apc.com/tools/download/sw_kit.cfm?sku=sdw64 I don't know anything about Best to recommend for or against, but as far as where to get very good quality refurbished UPS's (with NEW batteries), I've had great success with PEI at www.4ups.net (they're also at 4ups.com, but that website is more annoying). Standard Disclaimer: I'm not affiliated with PEI in any way. Hope this helps and take care, Daniel On Fri, Dec 07, 2001, [EMAIL PROTECTED] wrote: I am looking at purchasing a UPS and trying to decide between APC and Best (Invensys). Best apparently used to release their source code for their unix driver - which is a big plus. But this is no longer the case. Best provides binary versions of drivers for linux which is good. Apparently APC doesn't provide drivers for linux - but there are reverse engineered ones (with source code). So basically I am after some advice regarding the experiences of others. I will probably get a 750VA unit for my home server and would prefer to have access to source code even if the program is not as flash. Any recommendations / experiences welcome. Regards. Mark. -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: About kernel configuration
On Mon, Dec 03, 2001, Bambang Purnomosidi D. P. wrote: Hi, Is there any way to know that, for examples, my kernel-images have quota support enabled? In this case, consider I am using kernel-images from .deb, not compile it myself. Of course many configuration I need to know beside quota support. Hi, Try: grep CONFIG_QUOTA /boot/config-2.x.y where, e.g., x=4,y=9, etc. (specify your installed kernel) HTH, Daniel TIA -- bpdp -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: make-kpkg and 2.4.16
On Tue, Nov 27, 2001, Adam Shand wrote: Likely you didn't include support for your root fs or the hardware controlling the disk. I once left out support for my IDE controller. It didn't work out too well :-). snip Linux IP multicast router 0.06 plus PIM-SM RAMDISK: Couldn't find valid RAM disk image starting at 0. Freeing initrd memory: 3348k freed request_module[block-major-3]: Root fs not mounted VFS: Cannot open root device hda2 or 03:02 Please append a correct root= boot options Kernel panic: VFS: Unable to mount root fs on 03:02 Did you remember to update your lilo.conf file to make sure the initrd parameter in there points to your initrd (typically either /boot/initrd or /boot/initrd-2.4.16)? HTH, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
XF86Config or hardware issue on Thinkpad 560?
Hi, I'm hoping for some suggestions in narrowing down the search space of the problem I'm encountering running Debian Potato with my IBM Thinkpad 560. I recently exchanged the laptop's motherboard (including DC/DC) as the old mb had faulty cpu/cache memory (unit nominally ran potato fine except always got Sig11 on kernel compiles; new board passes kernel compiles with flying colors). It was an interesting process, I'll explain more if anyone's curious. At first, the unit worked perfectly after this swap. However, after a day or so, the LCD screen image on initial boot into XDM started getting very garbled on the upper right portion (consoles still work fine); LCD is active matrix, BTW. Interestingly enough, if I press Function F7, which is the IBM Thinkpad hardware hotkey to switch between internal video (LCD) and external VGA adapter (for monitor), and then switch back to internal LCD, the screen image is restored, and will continue to work properly (which belies the initial belief that it's a hardware issue, as does perfect operation from the getgo in consoles/boot message). I'm guessing there's a possibility that the onboard Trident video chipset of the new mb might need a slightly different XF86Config setup. However, both motherboards belong to exact same IBM part number: 2640-F0E, and both use nominally the same Trident Cyber9385 chipset (so this makes XF86Config issue harder to believe/understand). Does anyone have suggestions on which road to look down first? I can post laptop's XF86Config if people want to see it, but I've held off on that for now to conserve the list's bandwidth. I've already searched google and list archives on topic without success. Thanks so much for any suggestions. Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Is Redhat that bad - a question.
On Sat, May 19, 2001, David Underwood wrote: I really don't know enough to comment on anything posted in that thread, my Linux knolwedge experiance / knoledge is'nt that extensive. I am aware though that the Debian policy is much stronger in relation to the 'packaging issue'. Presumably this is one of the contibuting factors to it's 'legendary' stability. For this reason I have recently began my journey with Debian. The upshoot of all this is, and please forgive my ignorance, but what exactlly is FHS and where can I find information about it. I am concious of my need to learn more! Regards David Underwood. Hi David, The FHS is the Filesystem Hierarchy Standard. More info at: http://www.pathname.com/fhs/ To quote from the abstract of the 12 Mar 2001 release: This standard consists of a set of requirements and guidelines for file and directory placement under UNIX-like operating systems. The guidelines are intended to support interoperability of applications, system administration tools, development tools, and scripts as well as greater uniformity of documentation for these systems. Along similar veins (but more linux specific and larger in scope) is the LSB --- Linux Standard Base --- info at: http://www.linuxbase.org/ Incidentally, to comment tangentially upon the 'Red Hat sucks' thread: I use Debian on all machines I have control over (have to accept RH on certain lab workstations such as this one from which I read mail) and am immensely impressed with its stability, robustness, and overall engineering. However, the danger of portraying Red Hat Linux as 'sucking' is that it so compresses the space with which to judge software engineering that truly bad stuff is artificially elevated. Besides the fact that it often produces lots of email traffic (read: possible flamewar) with much lower signal-to-noise ratio than normal (here's hoping that this hasn't added to that effect :) ). Hope the above helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: What's up with the list?
On Sat, May 19, 2001, John Willey wrote: Now, I appreciate seeing the word die and Outlook in the same sentence as much as the next guy, but why are all the messages on this list suddenly in German? Is something on the listserve screwed up, or is it just me? Is anybody else suddenly receiving the German version of this list? John John, It's not just you. I've probably received 100 German messages to debian-user in the past hour. Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Debian Distro on Floppies?
On Thu, May 17, 2001, Ken Clarke wrote: Hi, I'm trying to find a Linux Distro that can still be obtained for floppy installation. Yes I know about just hooking up a CD-ROM drive, a modem, or a null-modem connection to this ancient machine I'm working with. I've been getting razzed by my LUG for wanting to do this the old fashioned way, but I guess I'm stubborn like that. Anyway, one of the guys from my LUG is a big Debian fan, and said he thought that there might still be a Debian Distro spanned across floppies instead of a CD, but I can't seem to find either it, or mention of it on the site. Could you confirm whether or not Debian still has such a Distro, and if so, point me in the right direction? Thank you for your time, Ken Ken, Hi. Are you looking just for boot-floppies (in other words to install the base system from floppies) or do you actually want to be able to install all ~4000 packages from floppies? Considering now that the binary i386 distribution spans three CD-ROMS, that looks like somewhere in the neighborhood of 1500 floppy disks to be able to install all packages. Since this seems a little excessive, I'm assuming you want the base system, and then install the rest via ethernet. This is definitely possible and I've done so a few times. Look at debian.org for exact installation instructions and means of creating these base install disks. Hope this helps and take care, Daniel PS Kindly wrap lines at 72 chars, please. -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Installation ?mounting partitions?
On Thu, May 17, 2001, Abner Gershon wrote: I am reinstalling debian 2.2r3 from CD after aborting mid floppy disk installation a last week. My hard drive I already partitioned last week as per partitioning mini-How To recommendations. The thing is I don't remember which partition was intended for what purpose and the only way I can guess this is to know the size of each partition. What is the command to switch to a diffent screen from the installation screen then I can run fdisk and figure out the size of the partitions and then go on to finish my install of Potatoe. Thanks. Abner, Alt+Fx should work, where x is 1,2,3,... Then you can probably execute 'fdisk -l' to get what you need. Hope this helps, Daniel PS If you need to get to a console from X-windows, the above alt-Fx becomes Ctl-Alt-Fx -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Karsten's browser reviews (updated)
On Mon, May 14, 2001, ktb wrote: On Mon, May 14, 2001 at 08:34:22PM -0700, Karsten M. Self wrote: I've updated my review of GNU/Linux web browsing alternatives, largely in light of recent advances by the Mozilla and Galeon teams. I'm looking for feedback, particularly on anything I might have missed. http://home.netcom.com/GNU/Linux/FAQs/browsers.html I might add that I'm quite pleased with Galeon. That URL doesn't resolve for me. Couldn't find a combination that worked. kent http://kmself.home.netcom.com/Linux/FAQs/browsers.html worked for me. Thanks for the comparisons, Karsten! -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: scripted jumpstarts?
On Tue, May 15, 2001, Robert L. Harris wrote: I just started at a new company. They're using alot of Redhat because they can kickstart the installs and script them also, making the installs real easy once the initial scripting is done. Does debian have this capabilty? Any thoughts on how or any GOOD short to the point HOWTO's on setting this up? Robert Robert, Please see: FAI (Fully Automatic Installation) for Debian GNU/Linux URL: http://www.informatik.uni-koeln.de/fai/ I don't think (?) it's officially included in Debian Stable, but should be (?) released with Woody. The above site already has versions you can use that are designed for Potato. Also, archives of debian-dev list have other suggestions, but probably not as powerful as this. Hope this helps. Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: TeX/LaTeX: How do you configure it? How do you get it rolling?
On Mon, May 14, 2001, Usuario Universo Online wrote: Hello- I've downloaded all the packages for minimum Tex/LaTeX installation under Debian. However, my tetex-bin package is still broken (it's C** in dselect), but there's no conflict or abything. It seems it's a matter of configuring. I've already tried texconfigure. I'm stuck. The packages don't work, and I don't know why. Does anyone know of a step-by-step tutorial for setting up TeX/LaTeX under Debian? Or known problems? Thank You. Best regards, Henry L. See 'tasksel', an ncurses-based means of selecting commonly-used tasks. I set TeX/LaTeX up with it using probably ~five keystrokes and it worked perfectly. More info, see 'man tasksel'. This was a Potato box. Hope this helps, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: config of xfree86 with i810
On Sat, May 12, 2001, Derya PALANCI wrote: i dont know the version of Xfree... how can i learn?? at command prompt: dpkg -p xfree86-common | grep 'Version:' This uses dpkg to show the entry for xfree86-common (base package of xfree86) then pipes output through grep to only display the version information. You should see something like 3.3.6 or 4.0.?. Further configuration depends upon which version you have. For 3.3.6 look into configuration options with: 'man XF86Config' or 'man XF86Setup' which will give configuration explanations for two common programs to set up X. I don't know about 4.0.2, but these might have changed somewhat. but i tried the command you wrote and it said bash : xfree86 : command not found Maybe you don't have all the necessary xfree86 packages installed or their location is not in you path (see 'echo $PATH') did i make smth wrong? No, we just haven't yet figured out how to fix it. thanks derya Sure, take care and hope this helps, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: [OT] SR ARQ Markov Model
On Mon, May 07, 2001, MaD dUCK wrote: hey guys, sorry to be posting this absolutely off-topic... reason is that I am desperate (this is academic research, but I ain't cheatin' by asking you) and of all my mailing lists, you guys are simply the smartest. i am looking for a Markov Model of the selective-repeat ARQ used in TCP. but i can't find one anywhere. problem is that it's a little more than simple M/M/1 but noone in this world seems to have ever modeled it and published the stuff on the web... and i can't believe that so i am asking you for pointers... anything? thanks! Martin, Attached is the output of running a search on 'markov model' and 'tcp' on INSPEC. It found 23 journal articles, and the attachment lists references and abstracts for each. INSPEC, of course, is available at 'http://firstsearch.oclc.org', but you generally need your academic institution to set up access, which is not readily available to general public, I believe. The only tangential experience I've personally had with markov models was when I was working on cluster algorithms in monte carlo simulations (2D Ising Model, if you care). I've generally found Kurt Binder's books very good in this field, so thought I might mention 'A Guide to Monte Carlo Simulations in Statistical Physics', if you're not already familiar with it. But then again, I'm not sure how much this has on Markov specifically (as Markov models are more mathematics, than physics, IMHO). Anyway, hope this helps and good luck with the research. Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University Database: INSPEC Query: (kw: markov and kw: model) and kw: tcp Database: INSPEC Copyright: Copyright 2001, IEE Accession No: 6900029 INSPEC Abstract No: B2001-05-6150M-108 C2001-05-5640-071 Author(s): Barakat, C.; Altman, E. Editor: Crowcroft, J.; Roberts, J.; Smirnov, M.I. Corp Source:Inst. Nat. de Recherche en Inf. et Autom., Sophia Antipolis, France Title: A Markovian model for TCP analysis in a Differentiated Services network Source: Quality of Future Internet Services. First COST 263 International Workshop, QofIS 2000. Proceedings (Lecture Notes in Computer Science Vol.1922) p. xi+368, 55-67 Publication:Germany ISBN: 3540410767 Language: English Publisher: Springer-Verlag; Berlin, Germany Sponsor:GMD FOKUS; Deutsche Telekom T-Nova; Cisco Syst.; Nokia; ERCIM; IST Program, Commission of the Eur. Communities Conference: Quality of Future Internet Services. First COST 263 International Workshop, QofIS 2000 Conference Loc: Berlin, Germany; 25-26 Sept. 2000 Year: 2000 Treatment: P Practical; T Theoretical or Mathematical Record Type:Conference Paper Abstract: Many schemes have been proposed to support TCP traffic in a Differentiated Services network. We present an analytical model to study the performance of these schemes. The model is based on a Markovian fluid approach. We provide first a general version of the model, then we specify it to the different proposed schemes. For each scheme, we calculate the throughput achieved by a TCP connection. We compare then their service differentiation capacity under different subscription levels, different reservations, and different round-trip times (18 Refs.) Descriptor(s): Internet; Markov processes; quality of service; telecommunication traffic; transport protocols Identifier: Markovian model; TCP analysis; Differentiated Services network; TCP traffic; scheme performance; Markovian fluid approach; throughput; service differentiation; quality of service; Internet Class Code: B6150M (Protocols); B6210L (Computer communications); B0240J; C5640 (Protocols); C5620W (Other computer networks); C1140J Database: INSPEC Copyright: Copyright 2001, IEE Accession No: 6894210 INSPEC Abstract No: B2001-05-6250F-374 Author(s): Yong Bai; Ogielski, A.T.; Gang Wu Corp Source:Wireless Inf. Network Lab., Rutgers Univ., Piscataway, NJ, USA Title: TCP over IS-707 Source: 10th International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC'99). Proceedings p. 3 vol. xxiii+1560, 1253-7 vol.3 Publication
Re: non-linux disk access
On Tue, May 08, 2001, spider wrote: can i have access to non-linux disks (other partitions than linux partition) ??? anyone can tell me, please, how can i do this? [EMAIL PROTECTED]:~$ mount --help Usage: mount [-hV] mount -a [-nfFrsvw] [-t vfstypes] mount [-nfrsvw] [-o options] special | node mount [-nfrsvw] [-t vfstype] [-o options] special node A special device can be indicated by -L label or -U uuid . Also, see 'man mount'. Example: 'mount -t vfat /dev/hda5 /mnt/win5 Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: webmin
On Sun, May 06, 2001, Jim Darrough wrote: Can I obtain and install Webmin using apt-get? Thanks, Jim Darrough Jim Darrough, ARS KI7AY [EMAIL PROTECTED] http://www.ki7ay.com Hi Jim, Searching at 'http://packages.debian.org', webmin is available in unstable, but not potato. See: 'http://packages.debian.org/unstable/admin/webmin.html' You should be able to do 'apt-get install webmin' if your '/etc/apt/sources.list' includes an unstable entry (and you've already run 'apt-get update' with that entry). You might want to check out the some 27 additional modules related to webmin that are packaged and found via above search. Also, note that it's generally considered risky to mix unstable packages in an otherwise stable potato system without being careful. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Question about Apt-???
On Sat, May 05, 2001, Jim Darrough wrote: Is there a concise list of apt-??? commands I can access? And for tasksel? I am intrigued by this functionality. I will stay with Debian and try to learn it. Jim Darrough, ARS KI7AY [EMAIL PROTECTED] http://www.ki7ay.com Jim, Check out Dwarf's Guide to Debian: http://people.debian.org/~psg/ddg/ Also, in practical daily use I almost always stay with either 'apt-get' or 'dpkg', though other people will have different personal preferences. So you might also want to just read 'man apt-get' and 'man dpkg'. Also see contents of '/usr/share/doc/apt/'. Hope this helps, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: motherboard suggestions OT: ECC memory
On Sat, May 05, 2001, Allan Wind wrote: On 2001-05-04 15:21:16, [EMAIL PROTECTED] wrote: Note, though, that I've heard Athlon doesn't play well with ECC, though it remains to be seen if this will be improved for the Dual Athlons (as it should be); again, hopefully, others can add more. Asus K7V, Athlon @ 800 MHz w/ 256 MB ECC running just fine here (had to flash upgrade BIOS to get ECC support though). /Allan Allan, Glad to hear it! I'm going the Athlon route (hopefully dual) myself soon, and want to get PC2100 ECC memory with it, so I'm glad to know you've made it work. My above comment was just a repetition of something that I'd read on one of the hardware review sites; hopefully, I made it sufficiently clear that I could have been wrong, as it turned out I was. Incidentally, what I really meant by my comment was that I had read that certain chipsets that support Athlons due not play well with ECC. I wonder if this is still possibly true, and your Asus is just an example of a better motherboard which uses a more-competently engineered chipset. Let's see... What are the main chipsets that support Athlon? AMD 760, Via kt133, Ali Magick (?or something), Sis (something I think). Anyone know if there is anything behind what I remember reading with one of these chipsets? Or maybe I'm just completely mistaken... Anyway, take care and thanks for sharing your Athlon success with me. -Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Spontaneous reboot! HW diagnostics?
On Fri, May 04, 2001, Tony Crawford wrote: Hi Gang! After over 100 days of error-free service, my Debian 2.2/i486 dial-up-router-cum-print-server did a spontaneous reboot yesterday. (Aftermath: the CMOS clock had jumped ahead an hour and a half, the nmbd log contained non-printing characters at that point, and 1 e-mail in the exim spool was corrupt.) Before this happens again, maybe somebody can recommend ways to test for decaying silicon? Preferably in-service; if that's not possible then at least without booting a different OS? Hi Tony, memtest86 (available from http://reality.sgi.com/cbrady_denver/memtest86/) will give your memory at least a good workout. It needs to be run directly from boot off of floppy in real mode of processor). Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Minor Technical Difficulties
On Thu, May 03, 2001, will trillich wrote: On Thu, May 03, 2001 at 07:06:35PM -0400, Jonathan D. Proulx wrote: Hi, don't know about the scsi errors, but loosing the graphical login is easy. dpkg --purge xdm if you only want to disable it change the /etc/rcN.d/S99xdm symlinks to /etc/rcN.d/K01xdm ...via the debian-friendly method update-rc.d remove xdm preferably. Hi, Don't we also want the '-f' (force) flag in that update-rc.d command (since I thought the suggestion was to run it without first removing the 'xdm'). Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: motherboard suggestions
On Thu, May 03, 2001, Gregory T. Norris wrote: I need to replace a dual PIII-600MHz motherboard, which apparently got fried. I'd appreciate any suggestions for a good, Linux friendly one. Preferably something which can make use of ECC memory, which the dead one (i840 chipset) couldn't. It's not yet clear if either CPU survived, so feel free to suggest boards which might not support the old processors. Thanx! Hi, Depending upon how much you're looking to spend, you might want to look into a Supermicro board with a Serverworks chipset. A little on the pricey side (~$500+), but that might be worth it for you. I'm personally waiting (still...) for the dual Athlons to come out (promises, promises) which should give super bang for the buck (depending upon final unknown cost of dual motherboard) with PC2100 memory ($115/256Mb from Crucial). Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: motherboard suggestions
On Fri, May 04, 2001, [EMAIL PROTECTED] wrote: On Thu, May 03, 2001, Gregory T. Norris wrote: I need to replace a dual PIII-600MHz motherboard, which apparently got fried. I'd appreciate any suggestions for a good, Linux friendly one. Preferably something which can make use of ECC memory, which the dead one (i840 chipset) couldn't. It's not yet clear if either CPU survived, so feel free to suggest boards which might not support the old processors. Thanx! Hi, Depending upon how much you're looking to spend, you might want to look into a Supermicro board with a Serverworks chipset. A little on the pricey side (~$500+), but that might be worth it for you. I'm personally waiting (still...) for the dual Athlons to come out (promises, promises) which should give super bang for the buck (depending upon final unknown cost of dual motherboard) with PC2100 memory ($115/256Mb from Crucial). Answering my own post (or at least adding to it). Just after posting, I read on theregister (http://www.theregister.co.uk/content/3/18750.html) that AMD will supposedly be introducing the chipsets on this May 15 to support dual Athlons. The implication is that they will ship soon after. Who knows, though. You should judge for yourself, I just thought you might want to know. Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: motherboard suggestions OT: ECC memory
On Fri, May 04, 2001, William Leese wrote: I need to replace a dual PIII-600MHz motherboard, which apparently got fried. I'd appreciate any suggestions for a good, Linux friendly one. Preferably something which can make use of ECC memory, which the dead one (i840 chipset) couldn't. one of those things i was going to look into put never got around to. whats the difference between SDRAM and ECC SDRAM? Hi, ECC (Error Correction Code) has an additional 9th bit for every 8 data bits to catch and correct certain errors (though I've heard it is designed more to catch the relatively rare stray bit flip errors, like from passing cosmic rays, than the nasty errors that result from dying memory, but I don't know the full story). So normal SDRAM is for example 32Mx64 (for 256Mb module), while ECC SDRAM is 32Mx72 (for same 256Mb with ECC capabilities). In many cases you can get ECC capabilities for maybe $8-10 more for a 256Mb module, in which case I think it's a good investment. Note, though, that I've heard Athlon doesn't play well with ECC, though it remains to be seen if this will be improved for the Dual Athlons (as it should be); again, hopefully, others can add more. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Newbie needs help
On Mon, Apr 30, 2001, Rahul Agarwal wrote: Hello, I am new to Unix/Linux. I installed debian linux according to the installation instruction at debian.org. But my ethernet card(SMC EZ Card 10/100 PCI (SMC1211 Series)) was not recognized. Also I dont have a gui installed. I found a driver name at linuxdoc.com for 1211 series but dont know how to download it and install it. Can someone please refer to some document which can help me set up my hardware and gui. I dont know anything about linux/unix. Any help will be appreciated. Rahul Rahul, Karstens' suggestions should help you get set up with your debian install. Also, searching on google and looking at archives of this list at lists.debian.org should fill in the missing pieces. Since you say 'I dont know anything about linux/unix' I would also recommend a very well-balanced Linux intro book by Michael Kofler, entitled _Linux_. It's on its second edition and is published by Addison-Wesley. bookpool.com should carry it, along with many other retailers. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Partitioning prior to dual boot installation of Debian and Windows ME
On Mon, Apr 30, 2001, Abner Gershon wrote: Just to clarify I tried again to use DOS fdisk as found on Windows ME StartUp floppy. I may be overlooking something but it does not appear possible to make more than one primary partition on a single physical hard drive with this application. I have downloaded the floppy disk images of rescue as well as rest of boot floppy images. I intend to repartition using cfdisk. I don't think that should be a problem. Make one primary (hda1), then one extended (hda2) and place all your other logical partitions (hda5 up) in that extended partition. Case in point from an fdisk on a machine at work: Disk /dev/hda: 255 heads, 63 sectors, 1653 cylinders Units = cylinders of 16065 * 512 bytes Device BootStart EndBlocks Id System /dev/hda1 * 164514048+ 6 FAT16 /dev/hda265 1378 10554705f Win95 Ext'd (LBA) /dev/hda565 128514048+ 6 FAT16 /dev/hda6 129 256 1028128+ 6 FAT16 /dev/hda7 257 320514048+ 6 FAT16 /dev/hda8 321 381489951 82 Linux swap /dev/hda9 382 442489951 82 Linux swap /dev/hda10 * 443 454 96358+ 83 Linux /dev/hda11 455 466 96358+ 83 Linux /dev/hda12 467 588979933+ 83 Linux /dev/hda13 589 892 2441848+ 83 Linux /dev/hda14 893 1074 1461883+ 83 Linux /dev/hda15 1075 1378 2441848+ 83 Linux Hope that helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Exhausted by P3V4X+Maxtor harddisk.... looking for ata100 stuff... suggestion?
On Mon, Apr 30, 2001, [EMAIL PROTECTED] wrote: newer VIA ide chipsets seem flaky with linux. i have a Epox MVP3G5(Super7) and it runs perfect. but once i upgraded to a p3 ...ide started getting whacked bad. now the hd could be going too, but if i were you i'd go out and get a promise ide ata66 or ata100 controller for that machine and see if that helps. it has done wonders for both of my VIA based systems at home, and im about to install one at work once i break down and reboot(74 days and counting.. sucks to lose that much uptime). the promise controllers are cheap .. US$30-40/each new for OEM parts. as for transfer rates i don't use hdparm to judge it i use bonnie or bonnie++. i think hdparm may just test the drive itself, which can be misleading once you start testing it at the filesystem level ..not that it matters to me though i want a reliable system, speed is ranked 3rd or 4th on my list. i use the ata100 card from promise btw(not the raid one though) Hi, Just in case your considering buying Promises' 'raid' controller, I thought you might want to know that Promise is being relatively disengenuous about this product. They try to sell it as hardware raid, but in reality all the raid code is implemented in their windows drivers (so they're taxing your main CPU not offloaded onto silicon on the controller). Supposedly these are available for linux as well, but why bother? Buy the cheaper promise non-raid controllers and use the much better native linux software raid code, if that's what you want. Though I have no personal experience, if you want IDE-hardware raid, supposedly 3ware is the place to go (very linux friendly), and I've always thought Mylex is the right SCSI raid solution (but now we're really in a different price range and off-track from your initial question). Hope this adds something of use :) Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: closeing open ports
On Sun, Apr 29, 2001, Michael Earls wrote: that was great info, but i do not need to masq any ips, i just need to limit the ports being open, i have edited inetd.conf, but there were some ports not listed in there. here is a port scan on my box, Mike, Hi. I just joined thread so I can't comment on much, but make sure you're not running portsentry (or understand better its implications), since it will bind to many of your ports to track attempted connects (in other words, you might actually be seeing portsentry and not the listed services on those ports). Hope I haven't missed something else joining this thread now, and, if so, please accept my apologies. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: simple gcc question
On Sun, Apr 29, 2001, Mike G wrote: Hi! How do I disablw warnings in gcc, for a piece of code? I compile with -Wall but there are 3 lines I'd like to skip something like #warn off line 1 line 2 line 3 #warn on I've been searching info and google, and found nothing, pointers to documentation will gladly be dereferenced :) thanks Mike, I don't think you can do what you hope to do. The syntax you're suggesting seems to imply preprocessor directives, but these are completed before the code ever gets to the compiler, so I would think that in order to turn off warnings for a section of your code there would have to be defined ansi c syntax for disabling it in that portion. I'd think that there is not that syntax. My guess is you'll have to either live with the warnings, fix what it's warning you about, turn off -Wall so it doesn't warn you anywhere, or turn on a smaller subset of warnings (see all the -W flags in the manpage) that maybe refines what you want. My guess is you might have to live with it. Of course, maybe I'm just wrong or don't understand your question. Hope this helps. Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: least memory xterm look-alike
On Fri, Apr 27, 2001, Renai LeMay wrote: Can someone pass me their recommendation as to the xterm -lookalike program which is the fastest/uses the least memory - similarly with window managers? regards, Renai Hi Renai, Don't have anything to add to xterm issue. But I really love blackbox as a lean,fast,light windowmanager. Also, bbpager is great. Get them as: 'apt-get install blackbox bbpager' and see http://blackbox.alug.org/ Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: sigh... and now what .. !\*#!!
On Fri, Apr 27, 2001, Adriano Peluso wrote: Hello all, Ok, I run the rsync process and it took half a day. When at 99% I got a file corrupted. Moved while tranferring? message. I looked at the server and the dir 2.2_rev2 I was downloading from had been removed ands substituted with 2.2_rev3 but the new one was still empty. So I tryied with another server and this time it took less than an our but still, when at 99% I got the error message that follows. What's wrong now? Adriano, I don't recall exactly, but I thought there was some bug using rsync method to make a pseudo-image under some versions of Windows (98, I think?) I turned out to be an MS-issue, I believe, and there is a documented work-around. Can't remember what, though, but I'm sure google knows. Also, you might want to just order a mastered disk (probably get one for under $5 US), but I'm not sure about shipping to Italy (though I think some do ship from UK which should be quicker than from US). Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Debian on 386SX/3MB
On Fri, Apr 27, 2001, Joe 'Zonker' Brockmeier wrote: On Fri, 27 Apr 2001, Indraneel Majumdar wrote: Hello, Can anyone please tell me whether it is possible to install Debian on a 386SX with 3MB RAM. I tried to install using the potato floppies but it starts to load the compressed ramdisk from the root floppy and then simply hangs. No furthur output. The floppies are OK and work fine on other systems with more RAM. I believe the stated minimum for installation is 8MB. It's probably not going to be possible to perform an installation with only 3MB. Take care, Zonker -- Hi, When I did a recent install of Potato on my oldest PC (Compaq 386s/20, finally freed from MS subordination :) ), the initial install help screen said that minimum specs for RAM with potato were now 12Mb, IIRC, although they used to be lower. I played around with the install, my computer faithfully and mightily struggled, and it finally managed to install a complete Linux environment (minus X but with development stuff) in 8Mb. Indraneel, you might want to see if you can cheaply put old FPM or EDO memory in the system to expand it a little (it'll probably go above 3Mb) and you can find modules for generally $5 per 4Mb. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Implications of using kernel 2.4
On Thu, Apr 26, 2001, Cheng wrote: I am a relatively new Linux user. I've used RedHat and Slackware before and now I'm switching to Debian. My questions is about the possible problems I may face if I install kernel 2.4 as well as the advantages it'd might have. First, I'm used to installing distro off of bootable CDs. So Instead of getting the lastst version 2.2r3, I downloaded the ISO for 2.2r2 and burned those. I plan to install the security pkgs seperately later. I can't seem to find the ISOs for 2.2r3 and I'm not sure how I can install the distro from FTP download, since not everything will fit on one CD. Second, I'd like to install the 2.4 kernel due to some of the advantages I've heard about it, such as Athlon optimization, more efficient threads, and memory detection. But from what I've read on the site, it'd appears that it would not be a good idea to install the 2.4 kernel with the 2.2r2 distro. It kinda scares me that I have to install packages compiled from an unstable release. Hi, I can't comment much on running 2.4 on potato as I don't do it. See this link off the front page of debian.org: http://www.debian.org/News/2001/20010415 Also an article on debianplanet.org: http://www.debianplanet.org/debianplanet/article.php?sid=192mode=threadorder=0 I'm sure others can add more. I'm using my box to mainly store Java source code and act as a file server for my home lan. If I ever feel confident enough, I'd also like to run a mail server from it. To me stability is more important than speed. And I'd appreciate some guidence as to what I should do/study before I go ahead and dump my redhat 7.0 distro for Debian. I'd read Dwarf's guide to Debian as it spends a lot of time on package management tools (one of Debians real strengths): http://people.debian.org/~psg/ddg/ Also, see debianplanet.org and debianhelp.org and newbiedoc.sourceforge.net (debian-specific) as helpful references in addition to anything found on debian.org. Hope this helps and take care, Daniel PS: Please wrap lines at 72 characters. Thanks. -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: OT: palms vs hp49
On Wed, Apr 25, 2001, Sergio E. Schvezov wrote: * Daniel Freedman ([EMAIL PROTECTED]) wrote: On Tue, Apr 24, 2001, Sergio E. Schvezov wrote: first of all hello and sorry bout the off-topic but i have nowhere to ask, sometime earlier in this month i asked about what came to be octave, it was really cool. ... I've had an HP48 for about 7-8 years, and while I enjoy its RPN and like it as a calculator, it is amazingly(!) slow in processing or anything computationally heavy. I tend to believe the HP49 is not much faster. Contrast that with the 30MHz(???, faster maybe?) RISC processors (strongarm??? others, maybe) that are found in many PDA's Hope this helps, Daniel well first of all as in palm i mean Palm Top or handheld as in the device and i want to know if you know of any that run with something like octave or can replace some of or most of hp49's functions orry about the ambiguity in my grammar. Hi Sergio, Yes. I know what you meant. I was speaking generically about PDA's (Personal Digital Assistants, like Palm) just because there might be others with similar functionality that might be better for you than Palm (with the proprietary Palm OS): for example, the Compaq PocketPC can run Linux with even some X-Windows (or so I've heard), so I guess it should be quite feasible to run Octave in that environment (though I'd be hesitant myself to buy it considering MS would get royalties from the deafult WinCE it ships with). Instead, Sharp is supposedly bringing out a similar PDA that natively ships with some form of Linux. Regardless, I chose not to comment on Palm simply because I know very little about it (including whether or not anything like Octave runs on it). Instaed I shared my (somewhat negative) experiences with the alternative you cited, an HP48/49. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: release info
On Wed, Apr 25, 2001, Artem Litvinovich wrote: Anyone have ideas on when a stable release incorporating the 2.4 kernel will be available? (woody i think) ...pardon the newbieness, just decided on potato over RH7.1 //Artem Artem, Welcome aboard! You can run the 2.4 kernel on potato, though I have no personal experience doing so. For instructions, see the featured link on the main debian.org webpage. As far as release of woody, I recall a recent message (on debian-dev I think) that gave from an optomistic November 2001 timeframe to a pessimistic March 2002 timeframe, IIRC. For more info, search archives (lists.debian.org) on debian-dev mailing list. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Info...
At 988120542s since epoch (04/24/01 08:55:42 -0400 UTC), Jenner Alm?nzar wrote: Hello!!! I just bought a new computer and i decided to install linux debian gnu, first i tested the hardware installing windows 2000 and it worked perfectly but when i start to install linux in the create partition section linux only showed me 8 Gb of hard disk when i got a 40 Gb hard drive. What's the problem? Jenner, Hi. You might want to look into lilo's 'linear' option. I vaguely recall that maybe something similar can just be passed into the kernel from your install disk, but I can't remember completely. See 'man lilo.conf' (or search for similar on google if you don't have working linux system) and also use the F-keys during the initial install screen from the Debian boot disk which will probably help you find a similar command to pass to kernel. Hope this gives you someplace to start. Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: converting programms to windows (gcc for win ? )
On Tue, Apr 24, 2001, Daniel Marquez-Klaka wrote: Hiya, maybe off topic i'm writing some progie's under linux, in C using gcc, of cause. Now i need to convert them to run under windows. My question is if there is a gcc port to windows or does anyone know about a free windows compiler that easy compiles ANSI C created under linux. Daniel Daniel, There's a gcc port to windows, but you might not need to use it. You want to investigate what's known as cross-compiling --- compiling for one architecture while using a compiler (and code) written on another architecture. You might run into problems if certain function calls (like some from systime.h for example) are implemented differently on the architecture for which you're cross-compiling (in your case, win32). I assume a google search should give more info. Or else the book _Programming with Gnu Software_ by O'reilly. Hope this helps, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Elm 2.4
On Tue, Apr 24, 2001, Roger Price wrote: I have just taken responsibility for a server running GNU/Linux. Previously I was a systems administrator for Altos Unix and can see distinct similarities. My predecessor has gone without leaving me any documentation. In particular I need information on administering Elm. Please can you help? Roger, Google gives me the following link to Elm documentation: http://www.instinct.org/elm/#documentation That page also gives the following proviso: 'Sadly, most of the information about elm on the internet is badly out of date. The program is still being slowly developed...' You might want to look into mutt. It's probably much more popular at this time, and (I believe) being developed by one (some?, all?) of the initial developers of Elm, so it should provide some familiarity to elm users. See mutt.org. Hope this helps, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: abiword
On Tue, Apr 24, 2001, Marcelo Chiapparini wrote: Hi Christopher thanks for your answer On Tue, Apr 24, 2001 at 05:03:05PM -0500, Christopher Cyll wrote: Hi Marcelo, The Abiword in Unstable reads Word97 just fine (though complicated MS Word features will not show up). What version of Debian are you running? Potato! Stable is a bunch of revisions behind on Abiword, so if that's what you run, YMMV. Does anyone who uses Abiword on Stable know how well importing works? Sorry about my ignorance, but what does YMMV mean? Marcelo, It means Your mileage may vary, implying a cautionary note in making inductions from one person's experience. Chris, On the rare occasions I'm still forced to deal with .DOC, I've used Abiword 0.7.7 (the one in potato) with success in importing MS Word documents, but these have generally been quite simple. From what I know, MS Word Import is so hard mainly because .DOC it is not a very explicitly defined format, nor does there exist any real reference implementation (supposedly documents in MSDN from ~98 timeframe are incomplete and quite buggy). First of all, it is constantly evolving through versions of Word, and much worse, it is supposedly very dependent upon the overall environment of windows (font settings, etc.). This last point can be empirically verified as many of us probably have had past experience with the same document appearing quite differently on different Windows machines, even when using the same version of Word. Hope this helps and take care, Daniel Thanks! Marcelo -- -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: OT: palms vs hp49
On Tue, Apr 24, 2001, Sergio E. Schvezov wrote: first of all hello and sorry bout the off-topic but i have nowhere to ask, sometime earlier in this month i asked about what came to be octave, it was really cool. Well right now i'm going to have the oportunity to buy a palm or sci calc i saw the vr3 wich runs on a modified linux wich i kind of liked. To connclude my options are to buy the hp49 or a palm, so the question is: are there any scientific programs like ocatve avaible for these devices or should i stick to the hp49? again sorry bout the off topic I've had an HP48 for about 7-8 years, and while I enjoy its RPN and like it as a calculator, it is amazingly(!) slow in processing or anything computationally heavy. I tend to believe the HP49 is not much faster. Contrast that with the 30MHz(???, faster maybe?) RISC processors (strongarm??? others, maybe) that are found in many PDA's that don't cost more than a factor of two or three more in price and it's a relatively easy decision. Also, while the HP's look amazingly sturdy, I've known at least three (from a sample pool of six total) to have broken under light use (HP did NOT replace any under warranty, and they charge for phone support, IIRC). Hope this helps, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: networking 2 pcs
On Mon, Apr 23, 2001, Noah L. Meyerhans wrote: On Mon, Apr 23, 2001 at 05:24:09PM -0500, john connolly wrote: I have two pcs connected by a crossover cable. Both have their nics appropriately configured, (to 192.168.1.1 and 192.168.1.2, resp). I cannot get them to ping each other. I think the problem is that the telnetd service is shut off in both of them. No, that most definitely is not related. ping and telnet don't even use the same protocol (ICMP vs. TCP). My guess is that one or both of the systems is trying to do a reverse DNS lookup on the other when you try to ping. Add an entry for each of the hosts in /etc/hosts (on both machines). Then check /etc/nsswitch.conf to be sure that the hosts line looks like this: hosts: files dns installed--I can't find it anyway. When I do apt-get install /cdrom.../telnetd... I get the message that the telnetd package can't be found, even if I give the entire path the file as found with the find command. I have tried it with and without the .deb suffix. Apt-get does not work on individual files. If you have a single .deb that you wish to install use dpkg (e.g. 'dpkg -i telnetd.deb'). Or, assuming you have /etc/apt/sources.list set up correctly, just run 'apt-get install telnetd'. I don't know what a souces.list entry should look like for cdrom access, as all my installation are done via network. You should probably read 'man 5 sources.list' which will document that file. man apt-get might help you too. John, You can try 'apt-cdrom'. See 'man apt-cdrom'. Hope this helps, Daniel noah -- ___ | Web: http://web.morgul.net/~frodo/ | PGP Public Key: http://web.morgul.net/~frodo/mail.html -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: OT: Converting from pine to mutt
On Mon, Apr 23, 2001, Stephen E. Hargrove wrote: I've beening using pine for quite a while now have several folders in ~/Mail. My main inbox is ~/Mail/mbox, but my spam filters and other procmail recipes have created several other mail folders. Because the filters work so well (no credit to me, I assure you), mbox rarely contains anything. I've been toying around with mutt, and made several small changes to the sample.muttrc (now located at ~/.muttrc). I pointed to ~/Mail/mbox as my inbox. However, when I fire up mutt, I don't seem to have access to any of other mail folders. It seems to be looking only to ~/Mail/mbox. I know I'm missing something pretty simple. I've read through several docs and sample .muttrc files, but it's not the answer isn't making itself clear to me. If anyone could point me to a newbie howto doc online, I'd appreciate it. Steve, There's an automatic web-based muttrc generator at: http://mutt.netliberte.org/ This might help. Also from mutt documentation (to be placed in muttrc): 3.11 Defining mailboxes which receive mail Usage: mailboxes [!]filename [ filename ... ] This command specifies folders which can receive mail and which will be checked for new messages. By default, the main menu status bar displays how many of these folders have new messages. When changing folders, pressing space will cycle through folders with new mail. Pressing TAB in the directory browser will bring up a menu showing the files specified by the mailboxes command, and indicate which contain new messages. Mutt will automatically enter this mode when invoked from the command line with the -y option. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Mathematica on Linux
On Sun, Apr 22, 2001, M G wrote: Have any of you used mathematica for linux? i use the windows version and find it very helpful... but i don't want to pay all that money for a linux version if i should just continue using it in windows.. Thanks, Michael Michael, You may want to look into Maxima, a symbolic computer algebra program under continual development for last 15 years (similar lineage from MIT as commercial Macsyma) and just recently relicensed under GPL (free, then, as in both beer and speech). It's already packaged for testing and unstable, and you can download source from main page: http://www.ma.utexas.edu/users/wfs/maxima.html Also, in different vein, but also very useful for GPL unix-based math use, I assume you're familiar with Octave (www.octave.org) It's packaged for all Debian current releases, relatively analogous to Matlab, and in many rudimentary cases, uses exact same syntax. I've found octave especially to be a pleasure to use. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Problem installing Potato
On Sat, Apr 21, 2001, [EMAIL PROTECTED] wrote:
I am trying to install Debian Potato 2.2r2 on a new system.
All goes fine until it gets to installing the base system.
Then I get:
File error!
There was a problem extracting the Base System from
/instmnt/dists/potato/main/disks-i386/current/base2_2.tgz
At the same time, on console3 (Alt-F3):
repeated error of
4hdc: cdrom_decode_status: status=0x51 { DriveReady SeekComplete Error }
4hdc: cdrom_decode_status: error=0x30
followed by
4hdc: ATAPI reset complete
ending with
4end_request: I/O error, dev 16:00 (hdc), sector 1013544
4end_request: buffer-list destroyed
4hdc: cdrom_decode_status: status=0x51 { DriveReady SeekCompl
It's a 1GHz Athlon on an Abit KT7A
The hard disk is a 20GB set as master on the first IDE channel
The CD/DVDROM is an Acer DVP1640A set as master on the second IDE channel.
I can mount the CD-ROM on Alt-F2.
And, I can install my old RedHat6.0 OK, but of course I want all the new
stuff on the Debian dist.
Perhaps I can overwrite RedHat with Debian...?
I'd rather do a Debian install from scratch.
Has anyone any ideas?
Alan,
First idea, I'd check to make sure that the disk was burned (or
mastered ?) correctly against the master md5sum list present in the
top-level directory of the cd-rom. See 'man md5sum'; the flag you want
is '-c'. If that checks, then other causes can be considered.
Hope this helps and take care,
Daniel
--
Daniel A. Freedman
Laboratory for Atomic and Solid State Physics
Department of Physics
Cornell University
Re: Drive images
On Thu, Apr 19, 2001, Bill Fowler wrote: I want to replicate my Debian system onto a number of other PCs, not all of which have the same size hard drive. Is there a good way to make an image of a partition(s) and replicate to other PCs. Thanks, Bill Fowler Bill, Hi. While it won't preserve your custom configuration options, 'dpkg --get-selections FILE' on master computer, then 'dpkg --set-selections FILE' on slave computers should get you much of the way there. Granted you have to do some base install and disk partitioning yourself, but it's still a big time saver. See 'man dpkg' for more info. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: debian 2.2r3 ?
On Tue, Apr 17, 2001, Rick Commo wrote: Thanks, Did the three steps - now running r3 I guess. Interestingly creating the boot floppy seems to have awry. It asked me if I wanted to format the floppy. I answered yes. At some point it threw up an error message saying mformat not found Rick, Hi. I think it'll let you proceed with formatting the floppy even if it doesn't have the final binary 'mformat' to do the final stage. My guess is that an initial 'apt-get install mtools' should let you do this in the future. Hope this helps, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: sigh...big problems if anyone has time to help me out...
On Mon, Apr 16, 2001, Alvin Oga wrote: hi ya GP... what kind of problems did you encounter ??? - use the debian boot floppy to boot debian.. ( trick is to get it online ... get a (default) supported ethernet card... - than get it online and apt-get update, upgrade... - ebay bids of $80 - $90 for cdroms is ridiculous... - brand new name-brand cdrom drives go for $50-$55 range. ( asus 52x, toshiba 50x, sony 52x ... Hi, I agree with you that he is probably much better off to do a floppy boot then net install or just buy a mastered CD-set at linuxmall.com, cheapbytes, etc. for ~$2-4. However, I think the quote below for eBay pricing refers to external parallel-port connected CD-ROMS for laptops, which are much less common than desktop atapi drives, and therefore, I would imagine, sell used for a lot more (they also are much more expensive new, in the $200-300 range). It's a completely different questions as to whether they are worth it... Take care, Daniel c ya alvin On Mon, 16 Apr 2001, Robert Cymbala wrote: GPswyft writes: Well, here goes. I have installed debian a number of times, never successfully. [...] I install linux from the floppies. [...] I can't do this with a cd because i said before I don't own a cd burner. [...] GP, Consider using an external CD-ROM connected to the parallel port. If you don't mind bidding thru eBay, here are recent final bids ($60-$90): http://search-completed.ebay.com/search/search.dll?ht=1query=bantam+backpack Nice thing about eBay is you can read feedback about sellers. Some sellers have hundreds of positive ratings and zero negatives. I've always had very positive dealings with sellers, except once when I bought directly (never buy something that isn't listed on eBay; if you feel you didn't get what was advertised a negative feedback can be left for the seller; it's powerful, and one gets the feeling that some sellers would rather be whipped with a wet noodle than get a neg. feedback). Here's some detail on how to use a Bantam Backpack CD-ROM to install Debian `potato': http://www.lafn.org/~cymbala/Debian/t4700ct.html Yours, -- Rob Cymbala2nd email: [EMAIL PROTECTED] [EMAIL PROTECTED] GnuPG/PGP: www.Lafn.org/~cymbala/pubkey.html http://www.Lafn.org/~cymbala/airguard.html -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: How to partition hard drive?
On Tue, Apr 17, 2001, will trillich wrote: On Fri, Apr 13, 2001 at 05:57:44PM +0200, Allan Andersen wrote: If it's for personal use I would use something like similar to this: /boot - 16 MB bootable swap - 2 x amount of RAM in the PC / - the rest that's a great first-install concept. how big your partitions are will depend ENTIRELY on what you use your computer for. graphics leans this way, web server leans that way, and gamer's paradise is completely different altogether. there's no set defined best way for all instances. you gotta figure it out for yourself. after you munge and install and remove and configure and add and download and tweak -- for a month -- you'll finally have things running the way you like. THEN you do a du /usr/local du /var du /home du /etc -- just kidding du /usr -- subtract /usr/local, of course to find out how much you've used. i'd rank each as a PERCENTAGE of the entire disk space, unless you feel like keeping a large partition at the end in case of i'd sure like to break off this subtree emergency... then do dpkg --get-selections '*' ~/installed.packages and back up /home and /usr/local, reformat, repartition to reflect your usage percentages: /boot = 10mb or less? / = % from 'du' above /home = % from 'du' above swap= 2 * ram /var= % from 'du' above /usr/local = % from 'du' above /usr= % from 'du' above the partitions that are busiest should be in the middle, IMHO. now you can restore /usr/local and /home, then reinstall your set packages with dpkg --set-selections ~/installed.packages Hi, I think Will makes a good suggestion for this empirically-tuned hard-drive partitioning scheme. The only thing I might add is that the above outlined approach will lose any customization you might have made to config files in /etc (of course dotfiles in your home directories have been backed up). Therefore, I would probably add a backup of the /etc directory to archive these customizations. Debian's smart enough not to mess with config files via 'apt-get upgrade', but, as great as it is, it still can't manage to preserve them through a hard-drive wipe :) Hope this adds something and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Harddisk sugguestion?
On Sun, Apr 15, 2001, Chun Kit Edwin Lau wrote: Hi everyone, I am planning to buy a new harddisk about 10 - 20 GB. Is there any recommandations? I am using P3V4X motherboard from ASUS with VIA chipset (I heard there were some problems with this chipset) Anyway, an ATA66 harddisk is what I want but I guess ATA100 with ide card is welcome too. thanx everyone Edwin Lau ([EMAIL PROTECTED]) Edwin, Hi. Based on current bang for the buck, I'd go for a 40 Gb rather than 10-20Gb. It seems to be the sweet spot in pricing right now. The 10-20Gb start at 60-70, while the 40Gb start at 90. It seems worthwhile to get 4x the capacity for the extra $20-30. As for looking for them: Pricewatch.com is your friend... http://www.pricewatch.com/1/26/2119-1.htm Shows numerous 40Gb IBM drives for less than $100. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: how to avoid the X windiws login?
On Sat, Apr 14, 2001, Duser wrote: Well everytime i make X windows work on my pc on a debian installation the next login i'm forced to have a graphic login (on tty7)wich leads me directly to a working windows system (even if at installation time i told the system not to set xdm to give me the grph login), as you can imagine i'm certainly not a linux guru and i cant menage to avoid an x window session start and eat my memory and my processor time, can you help me? Where is the switch i have to press? Hi Michele, You want 'update-rc.d -f xdm remove' as root. Add the '-n' flag to test it. See 'man update-rc.d' and the output below: [EMAIL PROTECTED]:~$ /usr/sbin/update-rc.d usage: update-rc.d [-n] [-f] basename remove update-rc.d [-n] [-f] basename defaults [NN | sNN kNN] update-rc.d [-n] [-f] basename start|stop NN runlvl runlvl . ... -n: not really -f: force You might also want to familiarize yourself with 'update-alternatives' which similarly manages /etc resources (but completely unrelated to your present need to modify init.d scripts). Hope this helps and take care, Daniel Well thanx for all in advance. Michele PS I like the deb packaging system so much that i'd like to try to make it work on a RedHat (please don't beat me) i have on a shared pc (that's why it's a RH, it's not mine), do you know if i can do that?I mean have apt and dkpg and dselect work on that system (rather than convert my deb packages to rpm with alien)? Sounds like a lot of work, good luck... :) Actually there's a Brazilian (I think) distributor of linux who has worked on something along these lines. Can't remember it's name, maybe Concentric, or something like that. Searching this list's archives or google will give you more info -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: I now have X, still no mouse
On Sat, Apr 14, 2001, Rob VanFleet wrote: On Sat, Apr 14, 2001 at 02:30:17PM -0700, Lorens Kulla wrote: I've tried to make the links, my mouse is on com1 (ttyS0). When I issue the command ln -s /dev/ttyS0 /dev/mouse I get file alreaddy Exists. My mouse is a Logitech Moouseman Model:M-CV46. I seem to be really out of my league. Try 'ls --l /dev/mouse', to see where the existant link is pointing to. If it's not already point to ttyS0 just 'rm /dev/mouse' then 'ln -s /dev/ttyS0 /dev/mouse' or, for future reference, 'ln -sf /dev/ttyS0 /dev/mouse' where '-f' flag is to force -Rob -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Major problems booting Debian 2.2 - Nothing works
On Thu, Apr 12, 2001, Len Cheatham wrote: unreadable html Hi Len, I couldn't really follow your question, as I couldn't easily read your html-formatted email (for the future, you're certain to get more responses with text-only email, with line breaks at ~70 chars). But I'm going to give it a shot anyway. I'm wondering if maybe you haven't downloaded the actual full iso image and instead just burned the first floppy disk image to your disk. For you to get the full iso image, see: cdimage.debian.org You'll have to make use of the pseudo-image-kit to download packages and then use rsync to combine them into an iso image which can be burned onto disk. Fortunately, there are instructions for doing so. Alternatively, you can order pre-pressed disks from linuxmall.com, cheapbytes, etc. usually at $1-2 per disk. Might very well be easier for you. If you make your own, make sure to check the final md5sum of the iso image to make sure you've done the psedo-image-kit process correctly. Hope this helps and take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: corel photopaint
On Fri, Apr 13, 2001, Joris Lambrecht wrote: Does anyone know of a vector based drawin program for linux ? I've been using flash at work but want to do similar things at home ... What about xfig? Quite widely used in the academic fields for drawing vector figures. Doesn't use either popular newer library (gtk or qt; in fact I don't remember what it uses), but it's stable, robust, and overall quite nice. www.xfig.org Hope this helps and Take care, Daniel -Original Message- From: Robert Voigt [mailto:[EMAIL PROTECTED] Sent: vrijdag 13 april 2001 11:33 To: debian-user@lists.debian.org Subject: Re: corel photopaint On Friday 13 April 2001 04:10, Michael P. Soulier wrote: Hey people. We're playing with Corel Photopaint 9 for Linux on a Debian 2.2 box, and we're not getting very far. It installed fine, it starts fine. Unfortunately, as soon as you try to open a new document, it bails on an unhandled exception. It could be a stupid permissions problem, but that doesn't explain the exception. Only Corel can help you with this. Do you know Gimp? I can't think of a reason to use PhotoPaint when there's Gimp. If you try Gimp, I recommend you get the latest version (1.2.1). The version in Debian 2.2 is quite old and lacks some functionality. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: How to partition hard drive?
On Fri, Apr 13, 2001, George M. Butler wrote: Hi all, I know this is a real newbie question. I am in the process of installing Debian Potato r2 on a brand new machine I just built. It has a factory clean 30G IDE hard drive and AMD K6-2/500. I put the Debian CD in the CDRom and booted. I have selected the US qwerty keyboard. Now I am ready to partition the hard drive. I would like some suggestions as to how to best partition the drive. I want this to be a Debian only machine and I would also like to have one partition to do the Linux From Scratch project at a later date. I would like to hear suggestions you might have or experieces. Thanks for your help. George, Hi. You'd probably get the following link suggested to you by its author (Karsten) pretty soon, but just in case you're in a rush to read up, a very well-written guide to partitioning Debian is the following: http://kmself.home.netcom.com/Linux/FAQs/partition.html Take care, Daniel George -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: help
On Thu, Apr 12, 2001, Patel Snehal-QA9876 wrote: I am trying to install linux on an old HP755 which does not have any OS installed on it now. what is the best and easy way to install it in the server. I have read all kinds og different way to do this and non are working for me so far. I would like a way to create a bootable disk with the Linux OS on it and just boot from it. all other servers in my area are running HP-UX10.20 Hi, I answered a similar question yesterday. Here's a repeat: I don't think you can get Linux support for HP PA Risc at the same level of maturity as that for most other platforms that Linux runs on. But check out the project to implement it: http://www.parisc-linux.org Also, you can get alpha-quality iso's at: ftp://puffin.external.hp.com/pub/parisc/cd-images/ Supposedly, Debian will host these upon release. Hope this helps and take care, Daniel Take care, Daniel PS. Please try to use more informative subject lines. Thanks. Help ME! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: my network device
On Wed, Apr 11, 2001, Govaere Jan wrote: Hello I've got an compaq 10/100 TX UTP Controller fon a Compaq Proliant 800. I want to install it and I don't know how ? The dmesg | grep eth0 doesnt't print nothing. Thanks for your help Hi, Google is your friend... It brings up for me an old page of Compaq's (now existing only in Google cache): http://www.google.com/search?q=cache:vcmproapp02.compaq.com/ActiveAnswers/Global/en/solutions.1076/Online.7179/additional_online.7216/default.asp+compaq+10/100+TX+UTP+Controller+linuxhl=en According to this page: At this time, the embedded NIC card on the Compaq ProLiant 1600, 1600R, and 1850R is based on the Texas Instruments ThunderLAN chipset. Similar dual-speed and single-speed PCI cards are available in the marketplace, under several different brand names. Both the embedded and PCI cards are controlled by the TI ThunderLAN (TLAN) device driver written by James Banks of Caldera Systems, Inc. (www.calderasystems.com mailto:[EMAIL PROTECTED]). More info can be found on page at above link. Hope this helps and take care, Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Linux fpr HP Apollo 715/50
On Wed, Apr 11, 2001, Andreas Bartsch wrote: where can I get Linux for this hardware on a bootable cd? Thanks in advance Andreas Hi Andreas, I don't think you can get Linux support for HP PA Risc at the same level of maturity as that for most other platforms that Linux runs on. But check out the project to implement it: http://www.parisc-linux.org Also, you can get alpha-quality iso's at: ftp://puffin.external.hp.com/pub/parisc/cd-images/ Supposedly, Debian will host these upon release. Hope this helps and take care, Daniel -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Debian install on RAID hardware
On Wed, Apr 11, 2001, Rafael Hinojosa wrote: Hello everyone, I'm not exactly sure if this question should be posted to this list, or to another more specific list. If so, do let me know. The organization I belong to just received a new Dell PowerEdge 1400 with the PercRAID 2/DC dual channel RAID card. Current plans are to use this machine to replace an old server. snip If anyone has had any previous experience getting this card to work under Debian, some advice would be appreciated. Raf, Hi. I don't have much advice for you since I've never done linux on Raid, but I'm fairly certain (second-hand: from reading this list) that Debian works well with Mylex Raid (supposedly the best supported Raid on linux in general). You'll have to wait further for more knowledgeable folks to help you with the PercRAID. Take care, Daniel Sorry for the long-winded e-mail; Thanks, --Raf -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: problem with the CD images web pages
On Tue, Apr 10, 2001, D-Man wrote: On Tue, Apr 10, 2001 at 08:26:25AM +0200, Peter Fosseus wrote: ... | traceroute cdimage.debian.org Where did 'traceroute' come from? I can't find it on my (potato) system. I do have 'tracepath' 'tracepath6' and 'traceroute6', but I Hi, [EMAIL PROTECTED]:~$ dpkg -p traceroute Package: traceroute Priority: optional Section: net Installed-Size: 100 Maintainer: Herbert Xu [EMAIL PROTECTED] Architecture: i386 Version: 1.4a5-3 Replaces: netstd Depends: libc6 (= 2.1.2) Filename: dists/potato/main/binary-i386/net/traceroute_1.4a5-3.deb Size: 18620 MD5sum: feba02e20848bdfafa6bf7dd9c594eba Description: Traces the route taken by packets over a TCP/IP network. The traceroute utility displays the route used by IP packets on their way to a specified network (or Internet) host. Traceroute displays the IP number and host name (if possible) of the machines along the route taken by the packets. Traceroute is used as a network debugging tool. If you're having network connectivity problems, traceroute will show you where the trouble is coming from along the route. . Install traceroute if you need a tool for diagnosing network connectivity problems. You can also check 'apt-cache search packagename' to find things in cases when the package name is a little more obscure. Hope this helps and take care, Daniel I'm trying to figure out why my home network isn't working yet. I just got a second box (i486sx , free) that I can install Debian on. I think it will be easier to get 2 debian boxes working first, then tackle the 'doze systems. moving off topic now I want to thank everyone on the list and all the Debian Developers for putting together such a great system. I find that Linux is much more fun to work with, and _much_ easier to configure and troubleshoot than M$ Winblows. The win95 machine at home is having trouble seeing anything (including itself) via samba. The Help system asked questions along the lines of : o is networking configured? o is the computer on? o is the computer plugged in? o do you have a computer? (ok, not exactly, but that's what the questions feel like) and gave a conclusion of : sorry, I don't know what could possibly be wrong When something doesn't work in Debian, it isn't particularly hard to find information (ie howtos, etc) that explain how to get information from the system to locate, and then solve the problem. Debian is a great OS! / Thanks, -D -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: New to Debian....compilation problems
On Tue, Apr 10, 2001, omicron wrote: hi.. On 8 Apr 2001, Moritz Schulte wrote: gcc: Internal compiler error: program cc1 got fatal signal 4 mea culpa. it _was_ 11. Hi, If it was signal 11, you probably want to run memtest86 as it is quite possibly a memory error. 'apt-get install hwtools' Hope this helps and take care, Daniel have you overclocked your system? Well, it is about Signal 11, but perhaps it's useful, too: http://www.bitwizard.nl/sig11/. perhaps i should upgrade my gcc/libc6 ?btw, i'll check that url. Thanx. regards omicron ** omicron Mail:[EMAIL PROTECTED] (Sridhar N) www:omicron.symonds.net pubkeys:omicron.symonds.net/pubkeys C O G I T O E R G O S U M ~~~ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Installing network card driver / packages
On Tue, Apr 10, 2001, Nate Amsden wrote: Adam Bender wrote: I just installed potato on my system. I can't find a driver for my network card in modconf (it's a 3Com 3C905C-TX), and therefore I can't download any use the 3c59x module it should work.. Hi, I may be wrong, but I believe the 3c59x driver has problems with the 3c905C. I've set it up satisfactorily with both 3c905A and 3c905B, but ran it a lot of trouble helping a friend get his 3c905c working with it. I believe 3com confirms that you need a different driver, and I couldn't seem to find it in the kernel (they suggest patching kernel with their code). My friend ended up taking another path. So, in other words, you might have to look further than just the 3c59x driver. Sorry I don't have any specific advice to you. The suggstion I made to him was to buy a Linksys card for $12 which had fine tulip driver support. Hope this helps and take care, Daniel PS the situation seems to get more complicated sometimes due to the confusion between driver's name of 3c59x and card's 3c905x, but they are still all 3com vortex chips. nate -- ::: ICQ: 75132336 http://www.aphroland.org/ http://www.linuxpowered.net/ [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Need a way to install driver / packages
On Tue, Apr 10, 2001, Adam Bender wrote: OK, I just installed the base Debian system, and now I need a way to install packages. My big problem is that there is no driver module for my ethernet card in the kernel. Therefore, I can't use a network to get packages. I have a driver for the card, but the only thing I can do is put it on a DOS partition. So, I need a way to get this driver (and packages) onto the Debian partition. Can I: Somehow mount the DOS partition and read files from there? Format a disk so Linux will read it, and copy the files from DOS to there? Thanks much, Adam Adam, Hi. I responded to your message above about the 3c905c. You seem to be taking my word that you can't get the driver for this card. Have you (independently) verified this? I just want to make sure you don't needlessly make your life more difficult in case I turn out to be wrong or things have changed since last I looked. If it turns out to be true that the 3c905c won't work easily, I'm a little confused by your next question. You say you have a driver on the DOS partition so you want to get it on the ext2fs (linux) partition. You can, with 'mount -t vfat /dev/your DOS partition: hda1, etc /mnt/dos' (see 'man mount' for more info). Bu what will this do you??? The way you've phrased your question, it seems, (IMHO), that the driver you refer to on the DOS partition is in fact a DOS driver; this will do you know good under linux. If you've found a linux driver from 3com's web site (as I mentioned in the last email), you'll probably have to patch the kernel (the way I remember them setting it up). Are you certain it wouldn't be easier to just go down to the CMU campus computer store and buy a cheapo Linksys PCI 100Mbs card for $12 or so (this 'tulip' driver works beautifully on Debian potato)... If this is your first intro to Debian, it'll probably make things much smoother and save yourself a lot of headaches. After you're more accustomed, you can look into Debian's 'kernel-package', patch a kernel, and go back and make your 3c905c work (then you'll even have two NIC's so you can run an IPMASQ box). Let me know what you think... Hope the above helps and take care, Daniel PS By the way, you can do your install off of the CD-ROM's (see cdimage.debian.org), so you don't need network connectivity to fully install Debian. See apt-setup, apt-get, apt-cdrom, etc. -- [EMAIL PROTECTED] www.andrew.cmu.edu/~abender Computer scientists have kernel knowledge. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Need a way to install driver / packages
On Tue, Apr 10, 2001, Gavin Hamill wrote: Hi. I responded to your message above about the 3c905c. You seem to be taking my word that you can't get the driver for this card. That's daft :) We use nothing but 3c905c's at work The module you need is '3c59x' and is part of the standard distribution - yeh I know it's not terribly logical.. but hey who said Linux was ever logical? ;) Hi, Huh... This is weird. I'm well aware of the 3c59x - 3c905[abc] mapping, but I'm surprised that people were able to get that driver to work with the 'c' version of the card. I was only able to make it work with 'a' and 'b' versions. pause But, now, after a google search, i find Donald Becker's page listing support for the 3c905c under the 3c59x driver: http://www.scyld.com/network/vortex.html There is, however, something weird in the wind, as 3com maintains a separate driver '3c90x' (GPL'd) for a subset of the cards (only 3c905B and 3c905C) that Donald's driver handles: http://support.3com.com/infodeli/tools/nic/linux.htm So, it looks like my initial word shouldn't have been trusted after all :). I'd try to make the standard 3c59x driver work, Adam. Take care and hope this helps, Daniel Kind regards, Gavin. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: sources.list and locally generated debs
On Mon, Apr 09, 2001, Chris Gray wrote: On Mon, 09 Apr 2001, Colin Watson wrote: Mario Vukelic [EMAIL PROTECTED] wrote: My locally generated debs, e.g. from kernel compiles, show up as obsolete in dselect. Is there an easy way to tell their locations to the package manager or do I have to mess with Packages.gz files and the like? You pretty much have to mess with Packages.gz, I'm afraid (at least, that's the easiest way I can think of). apt-ftparchive in the apt-utils package in testing/unstable may help. If I understand the question correctly, it seems the canonical answer is to raise the version number of the locally created package. This should keep dselect or apt from thinking that the local package is obsolete. You can change the version of the package by adding to the debian/changelog file. There is a nice mode in emacs to help you do this. Hi, At the risk of talking about something I don't really understand fully myself, I believe there's some functionality called 'epochs' built into dpkg, and denoted by 'make-kpkg --revision=2:myrevision' (note the 'n:' syntax), which has some benefits in this area. I looked into it last time I used kernel-package, but I remember conflicting messages/descriptions from different sources, so I don't recall the full system or the true benefits of epochs. Just wanted to mention it to possibly point you in the right direction. Hope this helps and take care, Daniel Cheers, Chris -- Got jag? http://www.tribsoft.com -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Linux you say????
On Mon, Apr 09, 2001, Brendan King wrote: unreadable html Hi Brendan, See cdimage.debian.org for information on making a bootable CD. Otherwise, you can buy one for a buck or two (plus s/h) from linuxmall.com, cheapbytes, and many other places. In the future, you might want to consider not sending html-formatted mail to the list, as you'll probably get many fewer helpful responses as a lot of people have mail readers that don't particularly like html (including mine: i see all your tags, rather than formatted output). Hope this helps and take care, Daniel PS By the way, in case you're not aware, you might want to look into the recent contraversy regarding MS Hotmail's terms of use, whereeby they own pretty much all intellectual property passing through their service. See http://www.troubleshooters.com/passport/offending_material.htm http://www.theregister.co.uk/content/6/18002.html -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Installing
On Sun, Apr 08, 2001, edu wrote: I've got 2 HD, one with Windows and a new one (empty). I want to install Debia in the new one and leave the old one for Windows. Is this a good way of installing Debian? I'm not sure how can it boot if I install Debian in the second hard drive cause I don't know where will I install LILO. I've already tried partitioning the old HD. If this is the best option (leaving the new HD for mp3 and stuff) I need some help with Fips. I have used defrag but it doesnt work (Fips says that the last cilinder is not free, but I can't find image.idx or mirorsav.fil). Hi, I'd just install Debian on new HD, but in case you're wondering about mirorsav.fil, it's probably a hidden file, so that might be why you can't easily see it. Of course, this doesn't really matter if you go with Debian on new HD. As other poster said, use LILO to create entry in MBR on master HD. HTH and take care, Daniel Any suggestions and comments are welcome. Thanks for your help. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Default Window Manager
On Sun, Apr 08, 2001, Aaron wrote: This may sound dumb, but how do I remove xdm? I am new to debian and don't know how to do this. And once xdm is removed will the system then use .xinitrc and start gnome? Aaron, You want 'update-rc.d -f xdm remove' (run as root). Use 'update-rc.d -n -f xdm remove' to try it out without actually doing anything. See 'man update-rc.d' for more info. 'update-rc.d' is one of many handy Debian-specific tools. You might also want to acquaint yourself with the unrelated, but also helpful, 'update-alternatives', which similarly manages certain resources in /etc. Here's some more info: [EMAIL PROTECTED]:~$ /usr/sbin/update-rc.d usage: update-rc.d [-n] [-f] basename remove update-rc.d [-n] [-f] basename defaults [NN | sNN kNN] update-rc.d [-n] [-f] basename start|stop NN runlvl runlvl . ... -n: not really -f: force [EMAIL PROTECTED]:~$ BTW, this was covered earlier this week on this list. You might want to also consult the list archives at debian.org. Take care and hope this helps, Daniel thanks, Aaron -Original Message- From: ktb [mailto:[EMAIL PROTECTED] Sent: Sunday, April 08, 2001 6:26 PM To: Debian-List Subject: Re: Default Window Manager On Sun, Apr 08, 2001 at 06:53:23PM -0700, Aaron wrote: X is currently being started by xdm I believe. The system start twm as the current default WM and I am not getting any error messages just not the WM I want. I didn't have any .xinitrc files on the system so after reading the other message I created one even though it didn't seem to do anything. Currently my .xinitrc file simply says 'echo gnome-session'. Remove xdm or kill the start scripts. If I remember right you wanted to use 'startx' I don't think xdm reads .xinitrc. I may be wrong about that. I haven't used xdm for a long time. kent -- From seeing and seeing the seeing has become so exhausted First line of The Panther - R. M. Rilke -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Default Window Manager
On Sun, Apr 08, 2001, Aaron wrote: Why shouldn't I be starting X from root? I'll go and add .xinitrc to my users home directory and see what happens. Aaron, Anything that runs as root has root level permissions, with the associated ability to do essentially anything to your system. Most people try very hard to absolutely limit the number of programs run as root. X is a _very_ large program, with I'm sure (even consideering its fine pedigree coming out of MIT's Project Athena) a fair amount of bugs, buffer overflows, etc.; not the type of things you would want to have root privileges. You might find it helpful to read up a little more on some gentle intros to linux/unix that might help answer some of these questions (see linuxdoc.org, Michael Kofler's Linux Intro book published by Addison Wesley, etc.). Otherwise, hope the above helps and take care, Daniel Aaron -Original Message- From: ktb [mailto:[EMAIL PROTECTED] Sent: Sunday, April 08, 2001 6:21 PM To: Aaron Subject: Re: Default Window Manager On Sun, Apr 08, 2001 at 06:55:30PM -0700, Aaron wrote: Good question. I didn't have an .xinitrc file until I created one in /root for the root user and then added the line 'echo gnome-session' to it. I'll check and get back to you You shouldn't be starting X as root. .xinitrc belongs in your home directory not /root. Go ahead and post it to the list. kent -- From seeing and seeing the seeing has become so exhausted First line of The Panther - R. M. Rilke -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Help! Accidentally started deleting /usr
On Mon, Apr 09, 2001, Mark Phillips wrote: Well it sounds like this might get to the stage where we should just agree to disagree, but for now I'll throw in a few more thoughts... snip Indeed, this is basically the approach other people have recently suggested for implementing my own undelete. Perhaps this is why unix hasn't got an undelete --- because people have created their own?? Mark, IMHO, this is exactly it. I responded to your message a few days ago and threw together the outline of a quick script (rm alias plus cron job), that could probably be implemented in under five minutes. I think this is precisely why unix doesn't have an 'undelete'; lots of people don't want or need it (even if it had no adverse effects), and those that do are expected (not unfarily, IMHO) to be able to construct the function to their liking in such a negligible amount of time that it is probably not considered beneficial to include a generic one with any given linux/unix anyway: thought process being that enough people would want/have to tweak a generic 'undelete' function to fit precisely what they want that it's essentially no harder to just write it from scratch. Furthermore, it's probably a general unix/linux philosophy thing; provide robust, flexible tools that are quickly used to allow people to construct their specific implementations, rather than trying to guess/estimate what people want to do. In this case, your tools are alias and cron. Hope this helps and take care, Daniel Cheers, Mark. -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Help! Accidentally started deleting /usr
On Sun, Apr 08, 2001, Mark Phillips wrote: Thanks Gavin for your help! I managed to restore my system!! What I did, was use dpkg --get-selections to find out what packages were installed on my system. I redirected this to a file and then edited it. I turned it into a script which ran apt-get --reinstall install list of packages on every package of the system. I broke the list into several chuncks and did the above apt-get on each chunk. Anyway, I believe the above process should have restored my system to fully working order. Certainly it seems to be working. Hopefully this email will be helpful if anyone else experiences the same problem. By the way, is there any way of setting up an undelete for averting this kind of disaster? My Dad mentioned that Novell Netware has an undelete which basically puts off really deleting stuff for about 5 days --- unless it has to because of lack of space. This sounds like a _very_ useful feature. Hi, I'm sure many people have many different ways of dealing with this. Here are a few, off the top of my head: 1. Back up (onto cdrw/tape/network storage, etc) once or twice a week so there's no great worry if something is deleted. 2. Alias 'rm' to 'rm -i' and then use 'rm -f' when you're certain you want to delete large numbers of files. I've found the extra typing of '-f' is likely to prevent mistakes. 3. Set up some form of two-stage delete system. A very rudimentary one could be to alias 'rm' to a script that instead of deleting files moves them to hidden status (prepends '.') and maybe appends some known unique extensions 'whoami-DEL' (or whatever). Then you can setup a cron job to delete this nightly or the like. Surely, others can make this much more elegant (delete them when free space is low, prevent possible namespace clashes, ensure security issues); I just wanted to throw out an idea. Hope this helps, Daniel Cheers, Mark. _/___/~~ /~~_/~~__/~~__Mark_Phillips /~~_/[EMAIL PROTECTED] /~~\HE___/~~__/~~\APTAIN_ /~~__/~~ __ They told me I was gullible ... and I believed them! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Installing Debian
Aaron, Hi. I think this might have to do with the LBA setting (maybe in your bios) of your hard drive. See the following article for common symptoms such as LI (then lockup, rather than LILO then boot). I know you get 'LIL' rather than 'LI', but it's probably worth looking into: http://www.linuxgazette.com/issue52/tag/15.html (non-debian specific, but still probably on target) Hope this helps and take care, Daniel On Fri, Apr 06, 2001, Aaron wrote: I just recently bought a set of debian cd's and am now trying to install debian on my pc. I have some experience installing Linux using redhat and I have done some freeBSD installs so I am not totally new to this, but I can't seem to make this work. Every time I do an install and then do the reboot I get LIL at the boot prompt. So for some reason Lilo is not being installed correctly on the MBR. When I set up my partitions I have the / partition on sdba1 so I figure that this should get Lilo installed on the first 1024 cylinders. Maybe not though. The other problem I am now having is that I get a message saying the X server can't find my mouse. Can someone tell me what tool to use to configure my mouse? Thanks, Aaron -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: Give me my econ-gnome-ical text mode back!
Hi, On Thu, Apr 05, 2001, Kevin Stokes wrote: I've actually gotten far enough with Debian that I have Gnome up and running, and have gotten to play the squash-bill-gates game, thanks to ya'll. However, when I booted Linux this morning, my nice text login was gone, and instead a graphical login was presented.This is nice and all, but most of the time I would prefer to do my work in good ole' text mode and only go into Gnome when I want to. What setting do I change? su (or sudo, etc.) to root: 'update-rc.d -n -f xdm remove' to test what it will do... 'update-rc.d -f xdm remove' to do it... man update-rc.d (will control scripts run in runlevels) very handy debian tool... Also, the chair in front of the Linux machine is uncomfortable, and sometimes I would prefer to work at my usual machine instead. I think I need to load a 'telenetd' package to use telnet to connect remotely, correct? Is it correct that I use the dselect program to download and configure it?I did start it up, but I'm very worried that it will try to update all sorts of stuff which I just spend a whole day configuring, and perhaps break my X stuff again etc. (You see I'm used Microsoft stuff, which tends to muck itself up just terribly everytime you let it check itself for new hardware.) So my question is, if I just go in there and mark 'telenetd', will it only download stuff that I ask for, and possibly some dependent packages? For many, dselect has fallen out of favor, at least compared to using apt-get, dpkg, etc. (maybe also console-apt, etc., but I haven't yet used such). You might want to look into familiarizing yourself with apt-get and dpkg... you probably should (as root, sudo, etc.): 'apt-get install -s telnetd' (-s flag will _s_imulate the apt-get installation and allow you to observe what would happen) then, if satisfied: 'apt-get install telnetd' would put it on your system I'm not exactly sure what you want this program for, but you seem to want to use telnet as a client in which case you'd want 'telnet' package, rather than 'telnetd'; the 'd' at the end of telnet lets you know it's a daemon, in other words, it's installing a telnet server for you to allow others to telnet to your machine. The real question is: why are you looking to use telnet, rather than 'ssh' (secure shell), which will ensure that your password can't be sniffed, which can be very common with telnet. See openssh.com, part of the openbsd project, and: 'apt-get install ssh' Hope this has been helpful and take care, Daniel Kevin Stokes Pie in the Sky Software www.pieskysoft.com -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
Re: getting /bin files for debian unstable x86
On Thu, Apr 05, 2001, Cyanide Morgoth Calcuterm wrote: = Original Message From Karsten M. Self kmself@ix.netcom.com = on Thu, Apr 05, 2001 at 04:28:45PM -0400, Cyanide Morgoth Calcuterm ([EMAIL PROTECTED]) wrote: snip I wish I could say that. I'm a student I don't have extra cash. What is your data worth to you? snip Actually I have a really easy option assuming that it's packaged right. Basically I assumed (rightly) that all I need are those files. Now I have a system with debian 2.1 which had the various files in /bin but they weren't compatable and so therefore had to think of other options. Then I thought if unstable had a simple tgz file with the base OS then maybe I could easily extract it in winzip and then just copy over the files without fuss or muss. Unfortunately some wonderfully gifted person decided that unstable didn't need a disk section or even a base install and so therefore I don't have access to this easy medium. If only I have access to a clean system with the files in /bin that is on x86 and that is running the version of libc that unstable works under (2.2.x if I remember correctly) then it would work. Unfortunately sometimes I believe in bad luck and I believe that things never work properly the first time. Hi, I'm a little hesitant to get involved in this mildly heated conversation, but maybe I can add something of (marginally) technical interest. If I understand it correctly, you're looking for base install packages like those found on the disks of stable (potato). Now, I've never messed up my system so that I've had to re-install (though I do keep tar-gzipped, encrypted backups on network storage), so I'm not sure if one could replace most of the standard base components of bin from the base install packages. Since you seem to be implying that one can do this on potato but not on woody, I'll assume this to be the case, and proceed to surmise the reason why: Those base install disks, I believe, are generally created in one of the last processes in the evolution of a given distribution. So, in other words, after woody freezes, the boot-disk people will work long and hard to create the boot-disk environment, which would provide you with what you want. So, I believe that you've levied what I think is an unfair criticism against 'unstable', suggesting that it is not planning to create these packages, when I believe that it is just being done at another point in the process. Anyway, it's probably called 'unstable' for a reason. Although you might disagree, I think Karsten was very thorough in outlining a number of options for you to recover/repair your system and possibly prevent a similar event from occuring. The advice I'd listen to, personally, is: Scale your goals and methods to your resources. I would suggest that, if you admittedly don't have the resources (as a student, like me) to perform reliable backup, that you ensure you only use the stable version of Debian as then you'll minimize your chances of needing exactly the resources that you don't possess. Hope this proves helpful. Best of luck in fixing your current predicament, and sorry I don't have more suggestions for your immediate problem at hand. Take care, Daniel -- Daniel A. Freedman Laboratory for Atomic and Solid State Physics Department of Physics Cornell University
