Re: did windows kill my linux?
On Fri, 28 May 1999, Christensen wrote: When I tried fdisk -1 (after a DOS boot) it wouldn't recognize the parameter. (Maybe I have a different version of fdisk?) You need the linux fdisk. It should be available on the rescue disk. Just boot from the rescue disk like normal, and on the color prompt, hit altf2 to get a command line. Then, fdisk -l from there (that's an L, not a one). My only thought is that you may have overlapped a partition. Brandon Brandon Mitchell [EMAIL PROTECTED] -- http://bmitch.dhis.org/ ICQ: 30631197
Re: did windows kill my linux?
Send in the output of fdisk -l Brandon Brandon Mitchell [EMAIL PROTECTED] -- http://bmitch.dhis.org/ ICQ: 30631197
Re: duplicate posts?
On Tue, 25 May 1999, Pollywog wrote: Is everyone else getting duplicate posts? If you are refering to the NDN messages, then yes. Seems someone didn't setup their server correctly (bounces should go to sender, not to list). I just sent a message to the debian list master and the post master at nm.nacion.co.cr. nm.nacion.co.cr is timing out though, seems they are having lots of problems, although it's probably slowing down the loop they are having with our list server. I'm probably the 1000th person to report the problem, so others probably need not bother. Otherwise, no, I'm not seeing dups. Although I did get a cool double download problem with fetchmail recently. I started fetchmail in daemon mode and immediatly after ran fetchmail to wake up the daemon. Whatever lock file that needed to be written hadn't started yet, so there was a cool double downloading going on. Actually, it wasn't too cool, it was very annoying. There can probably be some better file locking and pid maintaining with that program (every once in a while, fetchmail is convinced that fetchmail -q is pointless). Have a good one, Brandon
BUG: /dev/null 0600 was: Re: postgresql installation trouble ...
On Mon, 24 May 1999, Jon Marler wrote: I figured it out ... The permissions on /dev/null by default are set to 0600 root root. I set them to 0666 root root. Problem solved. Whatever did this is a bug. /dev/null should be 0666 by default. If you have an idea on what the offending program is, it would help. If you goofed while logged in as root, shame on you :-) Brandon (who missed the beginning of the thread)
Re: cfs???
On Sun, 11 Apr 1999, caffeine wrote: If you want a working CFS, I recommend compiling your own copy or hassling the maintainer. I don't think bugging the maintainer would help since it appears that cfs is not maintained upstream. It's also hard to find maintainers for crypto packages. The appropriate thread: http://www.debian.org/Lists-Archives/debian-devel-9903/msg02774.html http://www.debian.org/Lists-Archives/debian-devel-9904/msg00011.html And a link to tcfs: http://tcfs.dia.unisa.it Enjoy, Brandon +--- ---+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.dhis.org/ ICQ 30631197 | | Throughout history, UNIX systems have regularly been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | |-- UNIX System Administration Handbook |
Re: parallel computing, rsh, and DHCP environment?
On Fri, 9 Apr 1999 [EMAIL PROTECTED] wrote: OK, but if the addresses change occasionally, then it would still be a mess even with setting up a DNS, yes? (given that the hosts file needs IP addresses?) The suggestion of dynamic dns is probably the best one. Check out dhis.org. Out of curiosity, do DHCP-administered addresses change only when a lease expires without renewal? (These machines are not allowed to stay up constantly.) If your ip is not in use, the lease has _not_ expired, and there are no free ip's to give out, a dhcp server may use your ip. We have a week or two for our ip here (I guess the client sends a renew request every few hours so it knows the client is alive). However, when my ip would change over the weekend, I could tell they ran out of ip's. If we aren't short on ip's, my address will stay for a few months (summer and winter break). Brandon +--- ---+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.dhis.org/ ICQ 30631197 | | Throughout history, UNIX systems have regularly been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | |-- UNIX System Administration Handbook |
Re: Raid issues
The maintainer appears to be Martin Schultze [EMAIL PROTECTED]. There are no bugs reported for this package which is impressive. I expected a please update to upstream release or something. The only thought I have is that raidtools has been replaced by another program or perhaps in the kernel. I claim no more knowledge on the subject, so you may want to send a polite request to Martin. Brandon +--- ---+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.dhis.org/ ICQ 30631197 | | Throughout history, UNIX systems have regularly been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | |-- UNIX System Administration Handbook |
Re: tn3270
On Thu, 25 Mar 1999, Philip Lehman wrote: Looking for a tn3270 app for the console. Could somebody point me in the right direction? tia -phil x3270 works for me. Requires X though, I don't think tn3270 has been pacakged and I was never able to compile it years ago when I was interested in it. HTH, Brandon +--- ---+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.dhis.org/ ICQ 30631197 | | Throughout history, UNIX systems have regularly been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | |-- UNIX System Administration Handbook |
Re: Hitting web site consistently dropped ppp connection?
On 21 Mar 1999, John Hasler wrote: Heikki Vatiainen writes: Escape sequence is probably the correct quess. I just downloaded the page and one of comments had this signature: -- + ++ATH0 ^ Space added by me to protect modem owners. My guess is your modem saw the three plus signs and hung up. Sounds likely, but any modem that does so is defective. The modem should ignore +++ unless it is preceded and followed by a pause. There is a fee they don't have to pay if they don't completely implement the hayes standard. So it depends on who you ask as to if it's defective, the modem is working as it was designed by the manufacturer, but the manufacturer may have been better to pick a different design. Brandon +--- ---+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.dhis.org/ ICQ 30631197 | | Throughout history, UNIX systems have regularly been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | |-- UNIX System Administration Handbook |
Re: Backups
I'd also add the result of dpkg --get-selections to a backup (redir the output to a file). Brandon +--- ---+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.dhis.org/ ICQ 30631197 | | Throughout history, UNIX systems have regularly been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | |-- UNIX System Administration Handbook |
Re: Debian Logo *Idea* Contest! (Delay for painting contest required)
On Thu, 18 Feb 1999, Andreas Tille wrote: So let us define first, *what* we want to show on our logo. This is always the first step in software related things. Wy did we step over it?? Last time I asked what a good open source logo would look like (or some related question) irc got pretty quiet. This seems to be the thing no one knows. Wasn't there an open source logo contest? I think of things like an open box or open book. I also like this poster of a light house I have which seems appropriate. If we get some good ideas, maybe some gimp'ers would be willing to draw them up before this contest is over. Our request to the gimp contest team was too vague and directionless. Without direction, all they can do is lots of different fonts and modifications of the word debian. Thanks for starting this thread, Brandon +------+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.nws.net/ ICQ 30631197 | | Throughout history, UNIX systems have regulary been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | | -- UNIX System Administration Handbook |
Re: slink install messed up hamm partition
On Sat, 13 Feb 1999, John Bagdanoff wrote: So I don't get the SETSERIAL not found anymore on bootup, but I still have to reset the irq to 3 before I reconnect to my isp before every session. I thought at first this change of behavior was due to installing slink from my hamm partition, but I also upgraded most of my hamm to slink packages at about the same time. What might I have changed to make /dev/ttS2 to revert back to irq 4 all the time? I plan to pull out my modem and change the jumper setting, but before I do that, I'd like to understand why this is happening. This has happened to me. Just add serial to /etc/modules. If you don't, serial is autoloaded whenever you need it, but with the default values. After a while, it is unloaded again, unless you have this line in your /etc/modules. HTH, Brandon +------+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.nws.net/ ICQ 30631197 | | Throughout history, UNIX systems have regulary been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | | -- UNIX System Administration Handbook |
Re: 96% packet loss on my network
On Sun, 14 Feb 1999, Paul Nathan Puri wrote: My ifconfig eth0 output on my laptop is the following: RX packets:2175 errors:0 dropped:192 overruns:0 frame:728 TX packets:2590 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 Interrupt:3 Base address:0x300 Here is the ifconfig eth0 output on my pc-server: RX packets:7 errors:34 dropped:0 overruns:0 frame:34 TX packets:48 errors:0 dropped:0 overruns:0 carrier:0 Collisions:8 Interrupt:11 Base address:0xfc00 I get 96% packet loss. Both cards, the pcmcia and the pci, are fast ethernet cards. Strange how the pc-servers numbers are so much smaller. If you hub has diagnostic lights, also looks at those. I'd verify the base addr (different from what I'm used to) and irq on the pc-server. Make sure nothing else is using that irq and that the card is really using that irq. Then do the same for the laptop. It means pulling out manuals, checking jumpers, using isapnptools, and loading/unloading kernel modules. I can't guarantee a hardware problem, but others on the list seem to agree that it's the first step. Good luck, Brandon +------+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.nws.net/ ICQ 30631197 | | Throughout history, UNIX systems have regulary been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | | -- UNIX System Administration Handbook |
Re: 96% packet loss on my network
On Sun, 14 Feb 1999, Paul Nathan Puri wrote: I plugged in my other laptop using the same pcmcia NIC and wire. Same amount of packet loss. Do I conclude that it's either my hub or the wires? Why kind of wire, I seem to remember them labled class 3 through 5 and needing a class 5 for 100 base T. Brandon +------+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.nws.net/ ICQ 30631197 | | Throughout history, UNIX systems have regulary been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | | -- UNIX System Administration Handbook |
Re: 96% packet loss on my network
On Sat, 13 Feb 1999, Paul Nathan Puri wrote: I just set up my network (laptop to desktop). I'm running linux on both. I pinged my desktop and it seems there is 96% packet loss there. Why would this be? random guess Maybe an incorrect irq or an irq conflict? /random guess You may want to post the output of ifconfig. Good luck, Brandon +------+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.nws.net/ ICQ 30631197 | | Throughout history, UNIX systems have regulary been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | | -- UNIX System Administration Handbook |
Re: wu-ftpd-academ
On Tue, 9 Feb 1999, Daniel J. Brosemer wrote: The Debian Weekly News had an announcement that fixes are already out for wu-ftpd-academ but are in /incoming. I seem to remember something about that being accessable only to developers. Is there a place where we can get at it? There is a list of incoming mirrors in the developers corner (on the web site). My personal favorite is llug.sep.bnl.gov. Brandon +------+ | Brandon Mitchell [EMAIL PROTECTED] http://bmitch.nws.net/ ICQ 30631197 | | Throughout history, UNIX systems have regulary been broken into, beaten, | | brutalized, corrupted, commandeered, compromised, and illegally fscked. | | -- UNIX System Administration Handbook |
Re: Graphical dselect-like tool?
On Thu, 4 Feb 1999, Joe Emenaker wrote: I was wondering if there is a more friendly utility than dselect out there. I realize that, since it has to fit on the install floppies, dselect has to remain pretty lean. I also know that there's been a lot of talk about graphical config tools... but they seem to have been just that: *config* tools. Is there anything that's just warm-n-fuzzy version of dselect? Yes, gnome-apt was released recently. It's alpha software, so you are on you have been warned, and it requires a version of gnome that debian hasn't packaged (at least it wasn't avail when gnome-apt was released). See Debian Weekly News (from debian's web site) for more info. HTH, Brandon +------+ | Brandon Mitchell * [EMAIL PROTECTED] * http://www.resnet.wm.edu/~bhmit1 | | The above is a completely random sequence of bits, any relation to an | | actual message is purely accidental. |
Re: hdparm
On Tue, 2 Feb 1999, Alexander N. Benner wrote: Ship's Log, Lt. Brandon Mitchell, Stardate 310199.2331: /etc/rc.boot/hwtools I thought /etc/rc.boot is obsolete ... In slink? If so, someone should do some checking: [EMAIL PROTECTED](p3):/root# dpkg -S /etc/rc.boot setserial, nethack, hwtools, kbd, sysvinit: /etc/rc.boot Thanks, Brandon +------+ | Brandon Mitchell * [EMAIL PROTECTED] * http://www.resnet.wm.edu/~bhmit1 | | The above is a completely random sequence of bits, any relation to an | | actual message is purely accidental. |
Re: 2.0 install problems
On Tue, 2 Feb 1999, M.C. Vernon wrote: On Tue, 2 Feb 1999, ivan wrote: snip (7) Configure device driver modules (I choose to install the modules: cdrom, lp and serial) IIRC you need to include a support for filesystems here; I haven't installed in a while though - probably under block devices. And you can get back to this menu via modconf. HTH, Brandon P.S. another thought is to uncomment auto in /etc/modules which might find the right modules on demand. I don't know how to activate this change without a reboot. +------+ | Brandon Mitchell * [EMAIL PROTECTED] * http://www.resnet.wm.edu/~bhmit1 | | The above is a completely random sequence of bits, any relation to an | | actual message is purely accidental. |
Re: hdparm
On Sun, 31 Jan 1999, Rob Mahurin wrote:
I am trying to set up hdparm to put my hard drive to sleep after a certain
timout (the -S option, I think it is). I thought it would be prudent to go
through and make certain that my hardware reacted OK before I set it up, not
really expecting any problems. hdparm -y /dev/hd[ab] works fine. hdparm
-Y /dev/hda works fine. However, hdparm -Y /dev/hdb produces an error
message something like
hdparm: hdb: drive read status 0xff { BUSY }
Doc, it hurts when I do this. Then don't do that.
If -y is ok, then -S should be ok. If -Y gives you problems, maybe your
hardware doesn't support it.
My third question (which is actually not an hdparm question, but more
general and I haven't been able to find docs): how would I get this
hypothetical timeout to get set at boot time?
/etc/rc.boot/hwtools
Note: I took out -u and -m on my system because they caused some
corruption.
HTH,
Brandon
+------+
| Brandon Mitchell * [EMAIL PROTECTED] * http://www.resnet.wm.edu/~bhmit1 |
| The above is a completely random sequence of bits, any relation to an |
| actual message is purely accidental. |
Re: hdparm
On Mon, 1 Feb 1999, Rob Mahurin wrote: 07:01 $ ls /etc/rc.boot/hwtools ls: /etc/rc.boot/hwtools: No such file or directory I assume I get to create that? Oops, it's part of the hwtools package. I guess the package split because there was a line commented out for hdparm. My part of the file is: #!/bin/sh # Hdparm optimization # Switches on interrupts during transfers and does multi sector transfers hdparm -u0 -m0 -S120 /dev/hd[ab] #hdparm -u1 -m16 -S120 /dev/hd[ab] HTH, Brandon +------+ | Brandon Mitchell * [EMAIL PROTECTED] * http://www.resnet.wm.edu/~bhmit1 | | The above is a completely random sequence of bits, any relation to an | | actual message is purely accidental. |
Re: Problem installing minicom
On Fri, 29 Jan 1999, Michelle Coelho wrote: I even tried what you said, Brandon: dpkg -i minicom*.deb but it gave an error, since it doesn't know where to find the file. So I tried: dpkg -i /dev/fd0/minicom*.deb, and dpkg replied saying: dpkg: error processing /dev/fd0/minicom*.deb (--install): cannot access archive : not a directory Errors were encountered while processing /dev/fd0/minicom*.deb So then I tried dpkg -i /dev/fd0, but I got dpkg-deb:/dev/fd0' is not a debian format archive dpkg: error processing /dev/fd0 (--install): subprocess dpkg-deb--control returned error exit status 2 Errors were encountered while processing /dev/fd0 Ok, to get the packages off of the floppy and install you have two choices (do these as root): a) mcopy a:\* .; dpkg -i minicom*.deb b) mount -t msdos /dev/fd0 /mnt; cd /mnt; dpkg -i minicom*.deb; cd /; umount /mnt Mtools needs to be installed for the first one. Good luck, Brandon +------+ | Brandon Mitchell * [EMAIL PROTECTED] * http://www.resnet.wm.edu/~bhmit1 | | The above is a completely random sequence of bits, any relation to an | | actual message is purely accidental. |
Re: No Password for Root
Out of curiosity, does this problem machine have an /etc/shadow file? I'm thinking shadow passwords may be turned on (via shadowconfig) but that /etc/passwd doesn't have the corresponding 'x' value in the password field. Just a guess, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: printing problems
On Wed, 20 Jan 1999, John Cuson wrote: last week i started working on printing to a networked printer that uses an hp jetdirect card. i can get the print jobs to the printer, but the output is not appropriately formatted. when i simply lpr a test file, it appears that i am not getting line feeds...the first line of text is printed, but is never wrapped. when i try to print a man page by piping the output to lpr as suggested in the printing usage howto i get blank pages, though if i remove the intervening pipe to the col program i do get one character at the far right margin of the page. here are the relevant lines from /etc/printcap: [snip] i've looked around in the faqs and howto's, the 2nd edition of running linux, and the various docs for this distribution (debian), but i'm not finding anything that seems directly pertinent. does anyone have any suggestions? /usr/doc/HOWTO/Printing-HOWTO.gz Section 11.5 HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: emergency disk/mounting files
On Tue, 19 Jan 1999, Bob Nielsen wrote: On Tue, 19 Jan 1999, KTB wrote: I need to change an X config file I screwed up so I can get into Linux. I'm using my boot disk to get to root, so I can bypass X, so I can edit the file. This is the message and prompt I get when booting from the floppy: VFS: Mounted root (ext2 filesystem) read only. sh: /usr/bin/check -sendfile: No such file or directory /# I then use the command umount like so: /# umount / I then tried mounting like so: /# mount /dev/hdb2 / -rw and got this: mount: you must specify the filesystem type I don't understand how you are doing things after unmounting the root partition. Do you actually boot off the rescue disk and get to the color/mono screen? Anyway, in answer to your current question, try: mount -o remount,rw / HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: emergency disk/mounting files
On Tue, 19 Jan 1999, KTB wrote: I don't know that I can explain it very well but I'm booting to begin with from the rescue disk. When boot: comes up I type linux init=/bin/sh There is no color screen I'm not sure what a mono screen is. I just get the regular non-X screen. I tried mount -o remount, rw / and got the following error: EXT2 warning: maximal mount count reached, running e2fsck is recommended mount: warning: cannot change mounted device with a remount [ all kinds of other helpful info snipped ] Ok, I have a feeling you are booting from your harddrive still, but not into multi-user mode because the / partition is your harddrive, not the floppy. You should get a decent result if you type: linux 1 at the lilo prompt. Your / drive is already mounted and it's mounted rw, that's good. If commands are missing when you boot with linux 1, try running mount /usr to get that drive with all your familiar commands mounted. At that point, you should be able to edit files. If you can't change the file, try running chmod u+w filename and use the editor again. If it still doesn't let you change the file, send the new output of mount. Finally, to check a filesystem, run e2fsck /dev/hd?? replacing the ?? appropriately. Good luck, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: start svgalib progs from X
On Sun, 17 Jan 1999, Rob Mahurin wrote: On Sun, Jan 17, 1999 at 05:54:41PM -0500, Brandon Mitchell wrote: There is a little program called switchvt... I hardly remember, but I think it was just a single C file. That's about all I remember, too. (Although I was trying runvt, loadvt, launchvt.) Anyone have ideas on where it's located/hiding, switchvt didn't show up in any packages. TIA, Brandon I know that there's chvt. But that doesn't work from within an xterm: It works as root. But I need a bit more than chvt, and I know I've seen a program/script that does it about a year or more ago. The more that I need is the ability to do this as a user, run a specified program, and switch back to X when the program is done. I was hoping it would be a debian package, but can make do if it isn't. Thanks, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: oddlu corrupted file during slink upgrade
On Sun, 17 Jan 1999, Jason Wright wrote: It's obviously corrupted and bogus: gusgus:~ # ls -l /usr/man/man2/iopl.2.gz prw--ws--- 1 505 38296 0 Jan 2 1970 /usr/man/man2/iopl.2.gz And I can't delete it This looks familiar. I'd suggest: 1) e2fsck -cf /dev/hd? 2) try to move the file 3) make a ramdisk, move the file to the ramdisk, reboot (I get a sneaky and devilish feeling doing that) 4) probably the one you will hear the most: debugfs I've never used 4 and don't trust it personally, but apparently it should do the trick. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: start svgalib progs from X
On Sun, 17 Jan 1999, Joey Hess wrote: Open will do it. Use open -s -w program, it will switch to a VT, run the program, wait for it to finish, and switch back. That's exactly what I was looking for! Thanks Joey, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: Dynamic IP addressing
On Sun, 17 Jan 1999, Jonathan Burley wrote: Does Debian support DHCP/BOOTP IP addressing? Yes, I'm using DHCP via the dhcpcd program now. I've heard of others using bootp. There is a new dhcpcd program that you will need to use for the 2.2 kernels, but I forgot it's name. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
start svgalib progs from X
I want to start qwcl (quake world) from xqf (an X game finder). I remember something that would change the current vt (via chvt maybe) and then run your non-X program, followed by a change back to X. Does anyone know of such a program/script? Thanks, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: start svgalib progs from X
On Sun, 17 Jan 1999, Tino Schwarze wrote: I want to start qwcl (quake world) from xqf (an X game finder). I remember something that would change the current vt (via chvt maybe) and then run your non-X program, followed by a change back to X. Does anyone know of such a program/script? There is a little program called switchvt... I hardly remember, but I think it was just a single C file. That's about all I remember, too. (Although I was trying runvt, loadvt, launchvt.) Anyone have ideas on where it's located/hiding, switchvt didn't show up in any packages. TIA, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: LILO's now working(thanks!), BUT...
Did you check the jumpersettings at the disks? You have to tell a haddisk, whether it's a slave or the primary disk. Only primary disks can be bootable. This is something I was wondering about (I'm new to PCs and Linux): BIOS doesn't show any information about master/slave disks. I played around with the setup but with no result. I've got an ASUSTEK P2B-S mothercard with Ultra-2 SCSI contoller (SCSI storage controller: Adaptec AIC-7890). (in some postings I saw a formatted description of the hardware configuration, is that obtained by a command or from some file?) Looks like you have a SCSI only system (no IDE), which is nice (wish I had one). I believe SCSI cards have their own bios thing for booting, but you'll need to look at the booting messages to figure out how to get into it. On pc's, they are usually made with cheap parts, which means IDE. IDE can only have 2 drives per ide interface (master and slave) (systems now-a-days how two interfaces, making a max of 4 ide drives). SCSI doesn't have any of these restrictions, but it probably won't show up in your normal bios unless the scsi controller is built into the motherboard. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: manually running cron
On Wed, 13 Jan 1999, Michael E. Touloumtzis wrote: On Tue, Jan 12, 1999 at 10:10:33PM -0500, [EMAIL PROTECTED] wrote: I have cron installed, and I'm sure it should be running things that I'm not currently running - like the locate database update. My problem is, I'm on a laptop, and I shut it down when I'm done with it. So, is there a way to get the cron stuff to run manually? Or possibly at bootuptime? I use anacron on my home machine, because it is powered up only a few hours (or even minutes) a day and would miss regularly scheduled cron [snip] But that may NOT what you want for a laptop. The last thing you want if you're running off a battery is for a bunch of (ana)cron jobs to start and suck up all the battery juice by exercising the disk drives! My solution to the battery drain problem would be to install anacron but remove the anacron symlink from /etc/rc2.d/. Then, if you have ac power and don't mind anacron running, type linux 3 at the LILO prompt which will take you to runlevel 3 and do the anacron stuff. You could change the default runlevel in /etc/inittab, and you can change the timeout and other things for the lilo prompt in /etc/lilo.conf. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: slink does not have lib6
It seems the libc6 package was uploaded with a bad version number. I already reported the bug against locales. libc6 is in incoming, so I think the maintainer has another version waiting to be copied into slink and fix this bug. Just be patient, a fix is on the way, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter | On Wed, 13 Jan 1999, Krzysztof Adamski wrote: I was trying to slink, but it seems that lib6 is missing. Is it missing or I'm doing something wrong? I was using ftp.debian.org. K -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: removing ^M from a text file
sed 's/^M//g' badfile goodfile Replace badfile and goodfile appropriately. Generate the ^M in bash with contv contm HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter | On Tue, 12 Jan 1999, M.C. Vernon wrote: Dear all, I had the misfortune to be stuck with MSWrite over the vac, and although I told it to save as plain text, it didn't. I can delete most of the cruft, but there is ^M at the end of each and every line. I've tried catdoc, word2x, and sed, but none of them will remove them. Can you suggest anything, please? please cc me as I'm not on this list ATM. Thanks, Matthew -- Elen sila lumenn' omentielvo Steward of the Cambridge Tolkien Society Selwyn College Computer Support http://www.geocities.com/Area51/Chamber/8841/ http://www.cam.ac.uk/CambUniv/Societies/tolkien/ http://pick.sel.cam.ac.uk/ -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: want help with sound card...
debian:~/sounds/sndkit/dsp$ cat /usr/X11R6/lib/tkdesk/sound/metal.au /dev/audio bash: /dev/audio: Cannot allocate memory I've configured for a 4K (4096) buffer. When I run dmesg, here's what pops up: Sound error: Couldn't allocate DMA buffer [ ... ] /proc/memstat seems to indicate I have plenty of room to spare: total:used:free: shared: buffers: cached: Mem: 31764480 31096832 667648 28934144 675840 14417920 Swap: 91344896 5373952 85970944 MemTotal: 31020 kB MemFree:652 kB MemShared:28256 kB Buffers:660 kB Cached: 14080 kB SwapTotal:89204 kB SwapFree: 83956 kB Believe it or not, you are out of memory to allocate the DMA buffer. This annoys the heck out of me because the kernel doesn't move data to swap or free up cache to allocate a DMA buffer. It seems that you must have MemFree at a reasonable level. Try starting netscape (or another memory hog) and then exiting to push some stuff out to swap, then try the cat. Good luck, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: Apt and packages on hold
On Mon, 11 Jan 1999, James Dietrich wrote: I use the apt method in dselect. [ ... ] However, apt still shows those same five X packages as being the only ones on hold: Updating package status cache...done Checking system integrity...ok The following packages have been kept back xbase xfonts-100dpi xfonts-75dpi xfonts-scalable xfonts-base 0 packages upgraded, 0 newly installed, 0 to remove and 5 not upgraded. Why doesn't it show libc6 and friends as well? I'd like to make sure I'm not overlooking something obvious. libc6 and friends don't need to be updated / don't have a newer version, so apt isn't holding them back. It will be holding them back when a new version is available. At least that's my guess, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: Apt and packages on hold
Package: locales
Version: 2.0.7-19981211-1
[ sent to the bug system, watch the cc's ]
On Mon, 11 Jan 1999, James Dietrich wrote:
On Mon, Jan 11, 1999 at 10:47:46AM -0500, Brandon Mitchell wrote:
On Mon, 11 Jan 1999, James Dietrich wrote:
The following packages have been kept back
xbase xfonts-100dpi xfonts-75dpi xfonts-scalable xfonts-base
0 packages upgraded, 0 newly installed, 0 to remove and 5 not upgraded.
Why doesn't it show libc6 and friends as well? I'd like
to make sure I'm not overlooking something obvious.
libc6 and friends don't need to be updated / don't have a newer version,
so apt isn't holding them back. It will be holding them back when a new
version is available.
OK, here is the line from dselect for xbase:
*== Std x11 xbase3.3.2.3a-8 3.3.2.3a-8. X Window System ...
^ ^???
There's something missing from that version number.
and here is the line from locales:
*== Std adminlocales 2.0.7v-12.0.7-19981 GNU Libc: ...
The 'v' is missing, so it seems to be an older package:
$ perl -e 'if (2.0.7v-1 lt 2.0.7-19981) {print new\n;}'
$ perl -e 'if (2.0.7v-1 lt 2.0.7v-19981) {print new\n;}'
new
$
This is a bug, thanks for reporting it,
Brandon
+--- ---+
| Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ |
| Sometimes you have to release software with bugs. - MS Recruiter |
Re: libc6_2.0.7v
Joel uploaded a bad set of packages it seems. The version number should have a 'v' or something higher in the name. Until then I'd hold off so dselect doesn't try to remove lots of packages with broken dependencies. I've already reported the bug on the locales package, and I think he'll relize that it applies to his other packages as well. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter | On Mon, 11 Jan 1999, Noah L. Meyerhans wrote: According to the package search engine on debian.org, libc6_2.0.7v is the current version of libc6 in slink. The path it gives is: dists/frozen/main/binary-i386/base/libc6_2.0.7v-1.deb However, this file does not exist. The libc6 version in that directory is libc6_2.0.7-19981211-1.deb. The Apt package currently available in slink depends on version =libc6_2.0.7u. The version in slink does not satisfy this dependency. Any suggestions? I'm assuming that the search engine and apt are correct, and that the libc6 available at the above path is wrong. Is this true? Where can I get the correct libc6? Thanks. noah
kernel 2.2.0-pre3 and ppp
Ok folks, I just finished updating my slink system (it was a bit out of date before). However, I still can't get ppp working with a 2.2.0-p3 kernel I recently compiled. The symptoms are: Jan 9 00:18:20 cnhobbes chat[280]: Password Jan 9 00:18:20 cnhobbes chat[280]: -- got it Jan 9 00:18:20 cnhobbes chat[280]: send (??) Jan 9 00:18:21 cnhobbes pppd[279]: Serial connection established. Jan 9 00:18:21 cnhobbes chat[280]: timeout set to 5 seconds Jan 9 00:18:22 cnhobbes pppd[279]: Using interface ppp0 Jan 9 00:18:22 cnhobbes pppd[279]: Connect: ppp0 -- /dev/modem Jan 9 00:18:52 cnhobbes pppd[279]: LCP: timeout sending Config-Requests Jan 9 00:18:52 cnhobbes pppd[279]: Connection terminated. Jan 9 00:18:55 cnhobbes pppd[279]: Hangup (SIGHUP) Jan 9 00:18:58 cnhobbes pppd[279]: Terminating on signal 15. Jan 9 00:18:58 cnhobbes pppd[279]: Exit. My ppp package version is 2.3.5-2. Are others having this problem? What should I upgrade (potato or the kernel)? TIA, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: Xfree86 3.3.3 for next Debian release or sooner?
On Fri, 8 Jan 1999, Chris R. Martin wrote: Just wondering if XFree86 3.3.3 is being considered for the next release (2.2) or not... I would really like to have XF86 3.3.3 since it contains an X-server for the Riva TNT chipset. Yes, as soon as Branden is done with the X in frozen(slink), he will start working on 3.3.3 for unstable(potato). HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: Slow system: problem isolated, need help with solution
On Thu, 7 Jan 1999, Andrew Ivanov wrote: Basicly, if I try to open any application, I will have a 30 second delay between typing it in and actual execution. During all that time CPU is mostly idle, at least 85%+. Top shows no programs using extra resources. Even xterm takes 30+ seconds to open. Telnet/ftp the same. All this happens only when I am dialed up into my ISP. As soon as I poff, things go back to normal. I recompiled a kernel a couple of times, that didnt do any good. What else can I do? Random guess: how long does an nslookup take? Make sure your own host name (both short and long form) are in the /etc/hosts file. E.g.: 128.239.214.7 hobbes.resnet.wm.eduhobbes The loopback address may be better for you (127.0.0.1). I hope I'm right (for your sake), Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: Slow system: problem isolated, need help with solution
On Thu, 7 Jan 1999, Steve Lamb wrote: On Thu, 7 Jan 1999 20:44:36 -0500 (EST), Brandon Mitchell wrote: 128.239.214.7 hobbes.resnet.wm.eduhobbes The loopback address may be better for you (127.0.0.1). Isn't loopback always supposed to be localhost else it breaks some programs? localhost is on the loopback line as well. I don't think there is a limit to the number of symbolic names you can have on a line. Thanks for pointing it out though, that may not be obvious to others. Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: Keyboard and KDM
On Fri, 8 Jan 1999, Becher, Andrew wrote: I'm having a problem with the keyboard and KDM in slink. The keyboard won't respond after the system boots. I understand someone recently posted a fix for this, but I wasn't subscribed then. Could someone please send it to me? If it's the same as a recent xdm problem, try typing linux 4 at the LILO prompt (assuming linux is the name of your kernel). If that fixes it, edit /etc/inittab and change the default runlevel to 4. It's a work, not a real fix. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | The above is a completely random sequence of bits, any relation to | | an actual message is purely accidental. |
Re: configuring serial port
I would guess either a plug-n-play modem or a winmodem. If it's p-n-p, there is a solution. But there is no solution for winmodems. You'll get the best help from the list for this problem, but you may want to let them know what kind of modem you have and everything you can get out of the windows config. Good luck, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter | On Wed, 6 Jan 1999, CUNO wrote: Brandon Mitchell wrote: On Tue, 5 Jan 1999, CUNO wrote: Brandon Mitchell wrote: What are the symptoms? Does the modem dial and connect to the isp? Run plog -f from another window or virtual terminal, run pon, and post what comes out of plog. the modem doesn't make a single sound, nada Report off plog: pppd [231] : ppd 2.3.5 started by root uid0 pppd [231] : tcgetattr: input/output error (5) pppd [231] : Exit I'm guessing that means your serial port is still incorrectly configured. Can someone on the list verify this? What COM port is your modem on in dos/win? You have to run /etc/rc.boot/0setserial after changing the contents. HTH, Brandon According to Win95 my modem is installed on com3 (ttyS2) and my mouse on com2 (ttyS1). The mouse is correct (ttyS1), it works in Debian. I don't understand it anymore, when I look at my modem it is installed on Com3. But When I look at system (Win95) there are only two serial ports Com1 and Com2. How is it possible that my modem is installed on Com3 while I have 2 serial ports according to Win95. I think there must be a com3, because I configured ttyS2 in Debian, and you cant't configure a port that doesn't exists, or is this thought wrong ? I strongly belief that my PC is the problem and not Debian. O, yeah when I boot Debian only the devices for ttyS0 and ttyS2 are loaded. Wvdialconf can't find a modem on ttyS0 (I thought let's give it a try, you never know). Cuno, ( I dont't know what to do anymore ) +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: configuring serial port
Multi part answer: 1) ttyS2 == COM 3 in dos/win 2) edit /etc/rc.boot/0setserial, disable autoprobing and set it manually 3) add serial to /etc/modules.conf so the modules and your changes aren't unloaded when not in use. If you need any details, just ask. I've been doing installs lately to test slink and this is one of my regular steps. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter | On Tue, 5 Jan 1999, CUNO wrote: Hello, I have installed Debian one week ago. Everything is working fine, except my modem (Trust Communicator e56K). When I boot my system the following serial ports are configured, ttyS0 and ttyS1. /dev/ttyS0 at 0x03f8 (irq = 4) is a 16550A /dev/ttyS1 at 0x02f8 (irq = 3) is a 16550A Debian cant't find ttyS2 and my modem is on this serial port. How can I configure ttyS2, so that I can use my modem. Thanx Cuno Sonnemans, Netherlands -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: Upgrading
Libc5 should still be installed (or can easily be reinstalled) after the upgrade. It should coexist with libc6 without any problems. Good luck, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter | On Tue, 5 Jan 1999, Nidge Jones wrote: I run Debian 1.3.1 and wish to upgrade to release 2.0 Having read up on this a little, am I correct in saying that I will be able to do this without breaking everything? I am concearned about the LIB6c stuff, and what upgrading will mean to my old (LIB5) non-debian packages. Will they still work, or will I end up breaking everything from trying to change to 2.0 ? I understand I need to run the autoup.sh script first, then use dselect as normal afterwards. This is fine for upgrading my Debian Packages. But what about other stuff *I* may have installed in /usr/local etc !
Re: configuring serial port
On Tue, 5 Jan 1999, CUNO wrote: Brandon Mitchell wrote: Multi part answer: 1) ttyS2 == COM 3 in dos/win 2) edit /etc/rc.boot/0setserial, disable autoprobing and set it manually 3) add serial to /etc/modules.conf so the modules and your changes aren't unloaded when not in use. Hi Brandon, It works, thanx. When I boot my system I get: /dev/ttyS2 at 0x3e8 (irq = 4) is a 16450 16450 is correct for an internal modem, isn't it Not sure. What line do I put in /etc/modules serial without quotes. This forces the serial module to stay in memory. The complete line I see in 0setserial or only /dev/ttyS2 at 0x... I remember it being fairly well commented. Just comment (with a #) the line that does the auto probing or configuring or whatever. Then go way down toward the bottom of the file where is says manual configuration. Then uncomment the line for ttyS2. When I run pppconfig and filled in everything and I give pon my modem still doesn't work ??? Can you help me on this one What are the symptoms? Does the modem dial and connect to the isp? Run plog -f from another window or virtual terminal, run pon, and post what comes out of plog. Good luck, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: configuring serial port
On Tue, 5 Jan 1999, CUNO wrote: Brandon Mitchell wrote: What are the symptoms? Does the modem dial and connect to the isp? Run plog -f from another window or virtual terminal, run pon, and post what comes out of plog. the modem doesn't make a single sound, nada Report off plog: pppd [231] : ppd 2.3.5 started by root uid0 pppd [231] : tcgetattr: input/output error (5) pppd [231] : Exit I'm guessing that means your serial port is still incorrectly configured. Can someone on the list verify this? What COM port is your modem on in dos/win? You have to run /etc/rc.boot/0setserial after changing the contents. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: xterm-debian info
The reasoning is at the bottom of http://master.debian.org/~branden/xsf.html HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter | On Mon, 4 Jan 1999, tony mollica wrote: Hi. Can someone point me towards information on the 'xterm' vs. 'xterm-debian' configuration and any other documentation?
Re: The Status Of Slink
On Sat, 2 Jan 1999, Allan Bart wrote: I am planning to put together another debian based system, I was wondering if I should wait until the 2.1 version is available or will it really be that easy to upgrade from hamm? Slink is pretty darn stable right now. I need to get a look at the release critical bug list to get an idea of exactly how far we are. However, the 2.1.4 install disks are pretty good (may still be in incoming, mirrored on llug.sep.bnl.gov). There are only some cosmetic changes that are needed as far as I can tell. However, if you want to use hamm, I've also heard the upgrade is going well, it just means downloading twice. Good luck, let me know if you have any problems, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: installing on cdrom-less box
ftp://llug.sep.bnl.gov/pub/debian/dists/stable/main/disks-i386/current/ Your cd will probably have a similar directory. Instructions are in the directory. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter | On Tue, 29 Dec 1998, Chris Frost wrote: I'd like to install debian on a box which lacks a cdrom; I can export the debian cd's via nfs, ftp, whatever. I had heard that you could install debian with floppies and nfs, but haven't been able to find the floppy image(s) to do so.
Re: pkgsel
With respect to pkgsel, I thought our ultimate idea (not really for slink) was to provide actual empty 'metapackages', which are just normal packages with a whole slew of dependancies and a little README/Changelog/etc in /usr/doc/metapkgname . Is this the plan? How about another thought on how to do it. Do a dpkg --get-selections. Then compare what is currently installed to the groups of packages that have been defined. You then show the user if all, some, or none of the group is currently installed. The user then has the choice to change the group to all or none. Finally save the changes (will dpkg --set-selections remove things?), maybe even offer to run apt-get. Just my 5 cents (inflation) :-) Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: Debian install failure on laptop. Please help ...
On Wed, 30 Dec 1998, Robert J. Alexander wrote: I cannot boot any Debian (tecra or whatever) rescue disk on my latest laptop (previous 3 IBM Thinkpads where OK). Try the latest rescue/drivers images in incoming (mirrored at llug.sep.bnl.gov). They use the 2.0.36 kernel and are now zImages so laptops should be easier. Let me know how it goes as they are hot out of the oven (released yesterday) and I haven't heard any reports back from the other testers yet. (grab the 2.1.4 images) HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: Debian install failure on laptop. Please help ...
On Wed, 30 Dec 1998, Robert Alexander wrote: Thank you very much Brandon. Good to hear there is some hope ... I tried your .gov site but despite having many 2.1.4 stuff the rescue are only 2.1.3 ... probably will have to try in some more time ... BTW the .gov site is a mirror. I looked for an Incoming directory on ftp.debian.org but no luck ... where is it hidden ?? I realized that shortly after sending you the message. Looks like Enrique either didn't get the who upload done (he has a hard time connecting from over seas sometimes) or the mirror didn't finish. The real incoming is for developers only (probably on master). Us lay people are stuck waiting for a mirror or the package getting into the actual archive :-) Cheers from Rome (just sunset now ...). Bob Heh, I'm sitting here at work missing the snow that I saw while driving in today. Rome would be a nice change. Happy holidays, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: Can't get full capacity of my disk!
[ cfdisk reports a 6+ gig drive as 2 gigs ] Two things to check 1) press altf2 to get a terminal and type dmesg to see all the kernel bootup messages. Scroll up using shiftpgup and look for where the kernel detects hdc to verify it works correct. 2) try fdisk from that terminal (fdisk /dev/hdc should work). It's not as pretty, but seems to work where cfdisk falls short. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: xdm (slink)
This is a bug in xdm (reported). Try removing the link in /etc/rc?.d that starts xdm. Better solution is removing xdm and installing it again when it's fixed. I'm not at my box now, so I can't give the exact answer, but I want to do the same because it seems to lock up my keyboard (another bug, but this may be limited to Branden's beta build). HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter | On Mon, 28 Dec 1998, Sean M Hollingsworth wrote: I am having some weirdness with the version of xdm in slink. When the package installed and did its setup, xdm started itself. I thought this was a test, like when you install the Xserver, but xdm did not then shutdown. I killed it and eventually rebooted. Now in /etc/X11/config I have no-start-xdm. When my system came back up, xdm started again. I've tried touching /etc/X11/config, recreating it, etc... Short of renaming /etc/init.d/xdm, I can't find any way of stopping xdm now. Anyone have any ideas (other than that the xdm package has a bug, which I suspect)? Sean Oh and if this message comes through as html, let me know; I seem to be having a problem getting netscape to stop. -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: to mirror or not to mirror
On Sun, 27 Dec 1998, Marco Frattola wrote: hi to all the list, merry xmas and happy new year. Same to you. one of the two server (call it the master) periodically receive new files, scattered in many dirs. the second server (call it the slave) has to catch up. they're connected via an ISDN line and i thought they could synchronize during night hours, where telco bill is cheaper. but i never played with mirror, nor i know of any alternative, and i'd like to get advice from those who have already played with such toys. i could also write some ad-hoc sh/perl thing, but i'd rather not reinvent the wheel .. If you aren't saving a ftp site locally, maybe you should consider rsync. I use it nightly to backup a server at work to a secondary server. HTH, Brandon P.S. I think there is also rdist, but I remember needing more functionallity.
Re: Advansys SCSI Iomega Jaz drives
I believe the previous solution was something like: mkdir jaz; mount /dev/sd? jaz With ? being where linux finds your drive. See the output of dmesg for more help. Brandon On Tue, 22 Dec 1998, Kent West wrote: On Tue, 22 Dec 1998 [EMAIL PROTECTED] wrote: I am trying to install Debian on a 2 GB Jaz disk but when the rescue disk looks for my SCSI devices it does not find my Jaz drive. It has no problem identifying both of my SCSI CD-ROM's (a player and a re-writable) and my IBM SCSI hard drive. My Jaz drive is an external drive connected to the external SCSI port on ID 4. Does anyone have any suggestions on how I can get this to work? Tom Persons I've read other posts that said something to the effect of zips or jaz's showing up as the fourth partition even though it's on the first, or something (As you can tell, I'm really hazy on it.) You might check the list archives within the last three months. -- Kent West [EMAIL PROTECTED] KC5ENO - Amateur Radio: When all else fails. Linux - Finally! A real OS for the Intel PC! Life is an ongoing classroom. - Capt. James T. Kirk, Dreadnought -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null
Re: Communicator Installation
On Mon, 21 Dec 1998, Kirk Hogenson wrote: Can't think of anything else that could be wrong. If you can't get it to work, maybe it would help to post the results of ls -l /root so we can see what you've got. KTB wrote: such file or directory. I have used cd /tmp and verified I have such a directory and used, locate communicator and it lists the file as /root/communicator-'the rest of the file name'. I have a book I am One other thought, locate isn't instant. It works by using a database that is generated periodically. I.e. you can move the file and locate will still show it in the old location. Brandon
Re: 2.1.125 and lilo
I'll take a wild guess at this, although I've never seen the error message
before, so be warned...
On Mon, 21 Dec 1998, Marten Berggren - LUB NetLab wrote:
I have been trying to install kernel 2.1.125 straight from the source into
my Debian 2.0-based system. Everything went fine when I first installed it
(I am typing this from that new kernel), but when I tried to install a
version with somewhat tuned settings, lilo refuses to cooperate:
lilo
Device 0x0800: Invalid partition table, 2nd entry
3D address: 1/0/128 (2056320)
Linear address: 2/81/16 (262144)
If would be very grateful if someone could tell me:
1) What this error means? Seems like a mis-alignment between two systems?
I recognize the second series from cfdisk (see below) but not the first
one.
My guess is that you kernel is above the cyl, sector, or whatever limit
that lilo has.
2) How do I correct it? Can I use cfdisk to rewrite the partitiontable while
running from the disk? Besides, cfdisk does not seem to detect any errors?
(See below.)
Ack, um, one suggestion, don't try it. Writing partition tables is a bad
thing on a running system, especially if you are using that drive, or
expecting to keep the data on it. I believe lilo points it's loader to
the kernel on your harddrive. In fact, I've seen someone rm -rf / and
then reboot to find that lilo still worked and the kernel started fine
('course, nothing else worked). You may want to start copying and
removing old copies of files to see if you can get your kernel below the
limit. My personal solution is to make a small partition (10 megs is
plenty) at the front of the drive, and mount it as /boot. (A small /
partition is nice, but not needed for getting your kernel to the right
spot on the drive.) You may do this yourself by shrinking your swap a
little adding another partition, fix up mount points in /etc/fstab, and
rerunning mkswap from a rescue floppy.
3) Why did it happen? Are there any other utility that touches the partition
table besides cfdisk? (lilo, for instance?) If the partition table has not
change but rather lilo:s view of it (through the 2.1.125 kernel), is it
possible to create a partition table that works both for 2.0 and 2.1?
Woops, already answered that, at least if my guess is correct.
Good luck,
Brandon
Re: Custom kernel build question. Help !!
On Tue, 22 Dec 1998, Mahesh Padmanabhan wrote: My question is: 1) How can I cleanly build a new kernel image (zImage) without affecting the Debian installation on my PC ? Use the regular make commands (I think: make mrproper; make menuconfig; make depend; make zImage; But look at the docs to be sure). 2) How do I then transfer this image to a floppy to make a rescue disk out of this image so that I can boot the laptop with it ? Write the rescue image to a floppy and mount the floppy as a msdos disk to /mnt. Copy your kernel over the existing one and run whatever script there is (it will do some rdev stuff). I believe there is also a doc on the disk about what needs to be compiled into the kernel (loopback, filesystems, ramdisk, etc). HTH, Brandon
Re: Linux Newbie Questions.
On Mon, 21 Dec 1998, Holdsworth, Ed wrote: 1. What is the difference between X11 and Xfree86? Will either of these be installed as part of the Debian Linux installation? Do I need to go to the Xfree86 site and install it? I am assuming that the windowing software typically reqires use of X11 or Xfree86. Xfree86 is an implementation of X11. It is provided in debian. 2. I have located packages on the Debian FTP site, but no instructions on how to install a package. Do all the files listed under a package directory ned to be downloaded and do I need to compile with GCC? Instructions on how to install a package would be greatly appreciated. dpkg -i pkgname.deb will install any individual package, no compiling, and no need to install entire directories of packages (though it is possible). Dselect and apt do a much better job of this. They will download and install the packages that you ask for. Also, on your initial install, you can select entire groups of packages. HTH, Brandon
Re: rotating log files
On Sun, 20 Dec 1998, Thomas Adams wrote: Speaking of anacron: I used one of the preset configuration schemes (Scientific Workstation) on one hamm machine and noticed that it has both cron and anacron running. Isn't this kind of pointless? Not at all. Anacron runs when you boot up to see if there is anything you need to run, and will take the appropriate action at that point. Then, when you cron time hits, it updates the time (for anacron) and runs your scripts again. This way, if my system is left off for a week, everything gets back on schedule and it also works perfectly fine if I leave it on for a few years. Cron has other uses too, I like hourly chimes myself. HTH, Brandon
Re: Why does 16 bpp look the same as 24 bpp?
On Sun, 20 Dec 1998, Rick Macdonald wrote: I can't tell the difference between 16bpp and 24bpp 24 colour JPEG files, viewed with xv. 24 color JPEG's or 24 _bit_ color JPEG's? You can get 2^24 colors with 24 bit color. Can anybody explain this? I would have thought that 65K colours would _not_ look as good as 16M! 16M? That's ram, not sure what the conversion is to the number of colors, but that depends on your resolution. Sounds like you may be getting your units messed up. The big advantage of lots of colors isn't one picture, but being able to run lots of apps with lots of colors each. I.e. you can run 2^8 apps using 16 bit color each with 24 bit color. Dang, I need more than one meg on my card. HTH, Brandon
Re: Memory Problem
On Sun, 20 Dec 1998 [EMAIL PROTECTED] wrote: I seem to be having a memory leak on my system, and I'm not sure how to find out which program it is. Does anyone have any ideas/suggestions on how to find it? First, try free to see how much you have, and especially, how much is being used for buffers. I ran a system for 3 months and had to reboot from what I think is buffers taking up so much space (although that makes no sense to me, and I'm sure to get a followup saying that I'm wrong). The worst problem was the buffers taking up enough space to prevent me from allocating dma buffers for my sound card. Next, check ps aux and see what is taking up lots of memory. Kill the offending program and see how that improves memory. I'm anxious to try a 2.1 kernel to see if memory handling has improved much, especially those dma buffers. HTH, Brandon
Re: (off topic) UPS recommendation
On Thu, 10 Dec 1998, Mitch Blevins wrote: I like the Best Patriot UPS's. They are dirt cheap and work well. Best will mail you a cable free of charge (or at least they did for me several times) if you know which one your want. Dang, wish I knew that about the cable. I went out and got my own. Debian has a bpowerd package which will monitor the UPS and shutdown when the battery get's low. The part number of the cable to order is in the bpowerd package. The schematics for the cable are also there in case you are a glutton for punishment and want to build your own. How does bpowerd compare to the ups monitoring software that Best Power provides? Are there any advantages? Oh, to the original poster, go over the slashdot.org and search for ups. There was an ask slashdot question a while back. This lists archives should also have a link (from me :-) Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: Why sudo does it ?
On Fri, 11 Dec 1998, shaul wrote: [01:53:38 shaul]$ sudo -l You may run the following commands on this host: (root) /sbin/halt (root) /sbin/shutdown -r now (root) /sbin/shutdown -h now [01:55:14 shaul]$ sudo reboot Sorry, user shaul is not allowed to execute /sbin/reboot as root on rakefet. I'm guessing that reboot != shutdown -r now. Try running: sudo shutdown -r now HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: Jaz 1Gb on hamm dist
On Sun, 6 Dec 1998, David Stern wrote: Ack, looks like the standard dos/mac formatting trick. Try sdc4: I hate to interrupt a perfectly good thread, but can you give more details about whatever this trick is, and what the deal is with the 4th partition? There is some crazy way they format/partition the disks so they work on both macs and pc's, but it results in you needing to use the 4th partition on pc's. That's about all I know, except that Iomega isn't the only one to do this, Syquest does this too I think. I have an atapi zip drive, and if I hadn't read the messages very early in the bootup sequence with a zip disk in the drive, I never would've known to mount my zip disks on the 4th partition (/dev/hdd4). I read the zip HOWTO, but didn't find any mention of this. Mounting on the fourth partition works, but it seems wrong because fdisk doesn't show anything like that. I've got a zip disk too, which is why I'm familiar with this problem. I thought my partition table looked weird like that too. But I went ahead and repartitioned/reformated the disks for e2fs. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: install hamm from Inet
On Mon, 7 Dec 1998, Ben Bucksch wrote: Downloaded it 2 times (I think from ftp.debian.org via Go!Zilla like all other packages), both times the same error. Now CuteFTPed fom ftp.jp.debian.org and WinZip opens it. Is it possible, that the server-file is corrupted? I guess it's possible. But you said WinZip opened it? Then the file would be ok, but something is going wrong with your extraction on the debian installer. If that's the case, and you've been following the instructions, and using the stuff from hamm, then I don't know what is going wrong. Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: installing linux debian on Toshiba Satellite
On Mon, 7 Dec 1998, Vazquez Jr, E A wrote: Don't expect much help here, I've been trying for DAYS to get a response to my questions but apparently no-one likes laptop installs. It may be that no one is familiar with your hardware. If you are the poster that has the cd rom via parallel port on a laptop, then that's a pretty tricky combo and I don't know if it is supported. Also realize that you get what you pay for, and around here, sometimes it's a bit more than that. I'm trying to install debian on a Toshiba Satellite. I get as far as booting linux.. but then the Toshiba logo comes up and the cycle restarts. Try the tecra boot images. They are the standard response to boot problems like this on laptops. I'm sure I read a post recently about this but I can't remember what it was. You can always check the mailing list archives available on www.debian.org. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: suid script
On Sun, 6 Dec 1998, Jiri Baum wrote: 1) kernel opens the file, finds it suid 2) kernel executes the shell with that uid 3) shell opens the same filename ... I think it's probably the kernel that does the open on step 3, No, it's the shell - it gets passed the filename. If it was the kernel opening the file, there wouldn't be any problem, just like there's no problem with ordinary executables. That explains the weird behavior I was having when I tried getting a counter example working. I've been taking a course in OS and when I implemented this in my kernel I had to do two opens because of the way I was setting up my address space. If I had implemented it right, I could have seeked back to the original open file. But considering I never got to multiuser stuff, this wasn't a problem. I guess this just adds to the list of things that would be nice to have changed in the kernel, the unix way seems better to me. Thanks for the bonk with the clue stick :-) Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: install slink from Inet
On Sun, 6 Dec 1998, Ben Bucksch wrote: the file dists/frozen/main/disks-i386/2.1.0_1998-10-31/linux is missing. What should I do to install slink from Inet? Can I also use the hamm base system with the packages of slink? And later update to the slink base? Ben, I'm not quite sure what file you are looking for. The install images are the resc* and drv* files. You also need the base* file accessable, either as the full tarball on a harddrive or zip, or as the split up files that can be written to a disk. With that said, the resc disk is not ready in the above directory (unless my mirror is out of date). You can get the latest from ftp://llug.sep.bnl.gov/debian/Incoming/ (warning, you will have to look through a lot of files). I have heard one report that the drivers disks wasn't working properly (and a verification was also made). To answer you second question, you can install hamm and upgrade to slink. Doing this will end up replacing everything from the install (except the configs and other necessary stuff). Therefore, there is no reason to try to update to the slink base, it's already been done. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: Jaz 1Gb on hamm dist
On Sun, 6 Dec 1998, BOHICA wrote: In all version of DOS that I own (going back to PC-DOS v2.04) when I boot with the Iomega drivers I can read the cartridges without any problems, however, from Linux I get a mount error that reads: MSDOS filesystem not found or too many devices mounted. Can you send me (and the list too) the output of dmesg and fdisk -l with the cartridge in the drive? This will help verify you are using the correct device. Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: Mouse Question
On Sun, 6 Dec 1998 [EMAIL PROTECTED] wrote: Hey I was wondering if there was a way to get my system to auto detect my mouse so I know who to use while installing X Does anyone know if there is a program out there that does that Install gpm and run gpmconfig. I prefer the trial and error approach to going through the autodetect procedure. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: install slink from Inet
On Sun, 6 Dec 1998, Ben Bucksch wrote: I'm not quite sure what file you are looking for. The install images are the resc* and drv* files. You also need the base* file accessable, either as the full tarball on a harddrive or zip, or as the split up files that can be written to a disk. With that said, the resc disk is not ready in the above directory (unless my mirror is out of date). You can get the latest from ftp://llug.sep.bnl.gov/debian/Incoming/ (warning, you will have to look through a lot of files). I have heard one report that the drivers disks wasn't working properly (and a verification was also made). Running the install.bat from the above dir will fail, lacking a kernel 2 load. Using the 2.0-linux-file ends up root fs not found or so after starting Linux from install.bat. Ok, you want to do a 0 floppy install. I don't have a whole lot of experience with those (my bios didn't support the bootable cd's from 2.0 and I don't use win/dos anymore). Where did you get your install.bat from? To answer you second question, you can install hamm and upgrade to slink. Doing this will end up replacing everything from the install (except the configs and other necessary stuff). Therefore, there is no reason to try to update to the slink base, it's already been done. I don't know much about Linux. I thought dselect will only uograde the (add.) progams and the kernel is made by the install-procedure at the disks. There is no need to reinstall debian (unless you are a crazy tester like me :-) There also isn't much of a reason to upgrade your kernel unless there is a feature that you need/want. If you want the stock debian kernel, download the kernel-image package (I think). I tend to grab the raw kernel tarball and compile my own so I have just what I need (and bootup is much quicker). Which ever way you want to go is fine, but I'd avoid the install disks if you alreay have debian installed. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: install hamm from Inet (was: Re: install slink from Inet)
On Sun, 6 Dec 1998, Ben Bucksch wrote: Ok, you want to do a 0 floppy install. I MUST do a 0 floppy install, lacking a floppy. :-) I start to wonder why this is done now-a-days. They are cheap and convienent for many things. Although, having them may create security problems for some. But I'm babling :-) Now tried 2 install hamm, also failed. dinstall says: File Error: There is a problem extracting the Base System from base2_0.tgz. I believe the utility is called star. Try cd'ing to /target and running star base2_0.tgz from tty2. Maybe you can get a better error message than File Error. I.e. this won't fix your problem, but let us know what the problem is. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
RE: Jaz 1Gb on hamm dist
I wrote: Can you send me (and the list too) the output of dmesg and fdisk -l with the cartridge in the drive? This will help verify you are using the correct device. On Sun, 6 Dec 1998, BOHICA wrote: dmesg - Vendor: iomegaModel: jaz 1GB Rev: H.72 Type: Direct-Access ANSI SCSI revision: 02 Detected scsi removable disk sdc at scsi1, channel 0, id 5, lun 0 Ok, sdc is correct. --- fdisk -l Disk /dev/sdc: 64 heads, 32 sectors, 1021 cylinders Units = cylinders of 2048 * 512 bytes Device Boot BeginStart End Blocks Id System /dev/sdc1 ? 937318 937477 1203315272218546+ 20 Unknown Partition 1 has different physical/logical beginnings (non-Linux?): phys=(356, 97, 46) logical=(937476, 3, 15) Partition 1 has different physical/logical endings: phys=(357, 116, 40) logical=(1203314, 30, 19) Partition 1 does not end on cylinder boundary: phys=(357, 116, 40) should be (357, 63, 32) /dev/sdc2 ? 648482 649505 912677269488144 6b Unknown Partition 2 has different physical/logical beginnings (non-Linux?): phys=(288, 110, 57) logical=(649504, 0, 11) Partition 2 has different physical/logical endings: phys=(269, 101, 57) logical=(912676, 1, 10) Partition 2 does not end on cylinder boundary: phys=(269, 101, 57) should be (269, 63, 32) /dev/sdc3 ? 262490 263179 945973699181456 53 Unknown Partition 3 has different physical/logical beginnings (non-Linux?): phys=(345, 32, 19) logical=(263178, 26, 16) Partition 3 has different physical/logical endings: phys=(324, 77, 19) logical=(945972, 51, 15) Partition 3 does not end on cylinder boundary: phys=(324, 77, 19) should be (324, 63, 32) /dev/sdc4 * 680024 680971 68098110668+ 49 Unknown Partition 4 has different physical/logical beginnings (non-Linux?): phys=(87, 1, 0) logical=(680970, 34, 16) Partition 4 has different physical/logical endings: phys=(335, 78, 2) logical=(680980, 61, 8) Partition 4 does not end on cylinder boundary: phys=(335, 78, 2) should be (335, 63, 32) -- end - Ack, looks like the standard dos/mac formatting trick. Try sdc4: mount /dev/sdc4 /jaz Or if it doesn't detect it: mount -t vfat /dev/sdc4 /jaz mount -t msdos /dev/sdc4 /jaz You may also want to try sdc1 and sdc3 if 4 doesn't work. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: SQUAKE SVGALIB problems
As root: chmod u+s /usr/games/squake It needs root to use svgalib, and permissions are then dropped to play quake. On Sun, 6 Dec 1998, Mark Panzer wrote: Well, I tried to use xquake but my system locked up 3 times after about two seconds of use. I then decided to try out squake however when I run squake I recieve the error: svgalib: Cannot get I/O permissions. I checked the permissions on the lib files and changed them so group and user could read and execute them, but there was no change. Suggestions? Panz -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: SQUAKE SVGALIB problems
On Mon, 7 Dec 1998, Michael Beattie wrote: The binary, /usr/games/squake.real has to be setuid root to use svgalib. I't is not packaged as such because it is a security hole. A `chmod +s /usr/games/squake.real` will fix it. Doh, Michael is right, it's squake.real, not squake. Ignore my previous post. Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
RE: Subjects not so comfortable
I have a small config feeling question? Since we spawn hundreds of messages each day, it would be more comfortable to have a constant string in subjects such as [ DEBIAN ] that allows easy You could filter on part of the footer. I hae my xfmail filter set to work on debian-user string in the message. Of course the proper way is with the X-Mailing-List header :-) Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: suid script
On Sat, 5 Dec 1998, Jiri Baum wrote: The way I remember it is: 1) kernel opens the file, finds it suid 2) kernel executes the shell with that uid 3) shell opens the same filename If some fast file-moving is done between (1) and (3), one can substitute something else for the suid script. Don't forget the user can copy / link a suid script into his home directory. Ahh, link is the thing I was looking for. Otherwise, the person who made the suid script would be responsible for the exploit, which wasn't making sense to me. I think it's probably the kernel that does the open on step 3, but it's no big difference in the point you were making. I wonder how other unix variants that allow suid scripts do this? Or better question: are there any? This has been very interesting, thanks for all the info. Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: suid script
On 3 Dec 1998, Gary L. Hennigan wrote:
Pere Camps [EMAIL PROTECTED] writes:
| I want my users to be able to execute this script:
| The problem is that these programs need root's privileges. I've
| suid the script root:root but still the programs say I don't have he right
| permisions to execute them.
Scripts are not allowed to set UID, it's a security feature. I don't
know where this occurs, but it's pretty low level, perhaps in the
kernel itself or in the shell, and there's no getting around it. There
are just too many holes that allowing scripts to be setuid root would
allow and so that capability is disallowed.
It's in bash (which is also sh on most linux systems), a pain in the a**,
I mean, feature. I don't know of any other shells that do this.
The only way around it is to write a C (or maybe Perl) program, have
that program setuid root and have it call the script. Of course for
what you're doing it might be just as easy to have the C program
perform the operations itself using the system() call.
Instant root shell example in C:
int
main() {
setuid(0);
seteuid(0);
execl(/bin/sh, -sh, 0);
}
Either that or install the sudo package and learn how to use it.
Probably the better solution.
Brandon
+--- ---+
| Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ |
| Sometimes you have to release software with bugs. - MS Recruiter |
Re: suid script
On Thu, 3 Dec 1998, Joey Hess wrote: It's in bash (which is also sh on most linux systems), a pain in the a**, I mean, feature. I don't know of any other shells that do this. No, it's in the kernel. Any executable that starts with #! does this, because the kernel is repsonsible for that magic thing working. Dang, looks like you are right Joey, at least I can't get a counter example working. I have been forced to write csh scripts on linux that are run by suid programs because bash will drop it's privleges to the real user id. So, at least is some aspects, bash is worse than others. Any idea why the kernel does this (if it really does, I'm still not sure of it)? Thanks, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: suid script
On Thu, 3 Dec 1998, Pere Camps wrote: I mean, feature. I don't know of any other shells that do this. Doesn't bash have a setting to avoid this? I haven't RTFMd the manuals, but it would be a sensible set. Possibly when compiling, but not after bash has been made. Although, as Joey pointed out, the kernel also does this which was news to me. Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: /usr/lib/libstdc++-libc6.0-1.so.2: undefined symbol: __register_frame_info
On Mon, 30 Nov 1998, Oliver Thuns wrote: /usr/lib/libstdc++-libc6.0-1.so.2: undefined symbol: __register_frame_info Why is this not fixed now?!? :-( There are fixed libs, but they are not on the debian ftp server. Should not to hard to upload the fixed libs to the ftp server. Nope, it's not hard at all. It just takes a bit of time to get out of incoming: ftp://llug.sep.bnl.gov/debian/incoming/ With that said, the upgrade will break packages built with the bad version of these libs (at least that's my understanding of the thread so far). Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: upgrading slink cripples dselect/apt (and X) !!!
On Sat, 28 Nov 1998, Zack Brown wrote: dselect: error in loading shared libraries /usr/lib/libstdc++-libc6.0-1.so.2: undefined symbol: __register_frame_info Give this one a try: ftp://llug.sep.bnl.gov/debian/Incoming/libstdc++2.9_2.91.59-2_i386.deb Good luck, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: Points to Ponder
On Sat, 28 Nov 1998, Nathan E Norman wrote: On Sat, 28 Nov 1998, BOHICA wrote: : If the author of the utility loadlin.exe (Hans Lerman) has stated on his web : When the SLINK distribution was frozen, why weren't the installation : documents upgraded as well? They should address the fact that the 1.73Mb : floppy will not work for some machines and that a separate ROOT floppy is : also needed in addition to the usual disks if a floppy install is done. It won't work at all, so Enrique is making 1.44 images instead. His net connection blows so they've been a bit slow to appear. If you can do better, please do so. The disks work on a few systems (or so I've heard). But, Enrique finally got the 1440 images in incoming (thanks Enrique). Check out llug.sep.bnl.gov if you don't have an account on master. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
RE: Help with automatic logins with telnet
With the below advice noted, it is possible to do this with expect. I've done this so my school wouldn't delete my account when I don't log into one of their systems for a few days. I then crontabbed the program. But, as Shaleh says, you need to understand the security problems associated with this, and look into other solutions first. On Fri, 27 Nov 1998, Shaleh wrote: This is generally a bad idea. What are you attempting to do? Something like rsh or ssh may be a better option. If you just want to telnet w/o typing your name, there is not a GOOD option. On 27-Nov-98 Brian Maneikis wrote: I need help with automatically logging in to a computer with telnet. I want to use a file with my user name and password stored in it, and have telnet read theinput from the file using telnet login where login is the file with my name and password. I've tried this, but it doesn't work. It just gives me a Connection closed by foreign host before it even tries to read from the login file. I'm not sure what's wrong. Could someone please help me with this. Thanks in advance. BTW please send replies to [EMAIL PROTECTED] instead of replying to the list since I'm no longer subscribed to the list. I hope some of this is clear. -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null -- Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] /dev/null Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: dhcpcd wierdness
Dhcpcd sends an IP release message to the server on normal shutdown. This causes my ip to change. Try shutting it down with a kill -9. Also, you may have luck with the -r option getting you connected faster. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: X broken?
On Sat, 7 Nov 1998, Vaibhav Goel wrote: I updated my hamm to slink yesterday. Everything seems to work fine except for X. I get the following error, when I type startx -- -bpp 32 _X11TransConnect: Cant connect: errno=111 I have a feeling your real error is above that. Maybe fonts, maybe locales. Don't forget that xterm split off the main X dist. But, in regards to your question, yes, X has been broken in slink. It's on my list of the big 3 things that need to be fixed right now. I wish there was a solution better than downgrade to the hamm version. Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Sometimes you have to release software with bugs. - MS Recruiter |
Re: UPS systems and Debian/Linux?
Someone else was wondering the same thing: http://slashdot.org/askslashdot/980721/097233.shtml Personally, I have a best patriot pro for my system, and it works good. It has saved me from lots of little power problems (same circuit breaker as my sterio and two garage door openers) and one 30 sec power failure so far. Checkout www.pricewatch.com and www.pricescan.com for cheap mail order prices. On Sun, 1 Nov 1998, Chris R. Martin wrote: I am looking to get a couple UPS units to protect my (2) Debian 2.0/Linux boxes. I don't need a lot of running time; I just want to be able to shutdown in case of a short power outage. I need a unit with uses a serial port to communicate its status, but statistics, etc are not needed. Also, I don't have a large budget for this. Enjoy, Brandon P.S.: $uptime 10:46pm up 70 days, 3:05, 9 users, load average: 1.06, 1.10, 1.04 +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Dijkstra probably hates me (Linus Torvalds, in kernel/sched.c) |
Re: how do you recompile the kernel?
On Sat, 31 Oct 1998, roadrun37 wrote: 4) so instead i did make dep 5) then make clean 6) make zImage 7) make zdisk 8) make modules 9) make modules_install 10) cp /usr/src/linux/arch/i386/boot/zImage /dev/fd0 Like someone else said, you don't need 6 or 10 with make zdisk. But that doesn't mean 6 or 10 are wrong, just repeated. that should be the end? did i forget something? when i boot up, it sees the SoundBlaster, but it then gives me a slew of messages about the modules, how it can't see the vfat or the gpm mouse devices. Ok, the fact that it sees the sound blaster is good. It means you compiled your kernel and it you were able to boot with the new one. I have a feeling that you didn't select some of the correct things for the modules. For vfat, did you modularize the vfat choice in the filesystems menu? Also for gpm, what kind of mouse is it? If it's a ps/2, you need to have the ps/2 option selected in the character devices menu. Does anyone else have this problem? Can someone give me an answer? (preferably someone with a Debian system?) Let us know how it goes. Brandon (who looks forward to the 2.2 days when sound is a module and kernel compiles don't need to be done by the users, only the speed demons, like me :-) +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Dijkstra probably hates me (Linus Torvalds, in kernel/sched.c) |
Re: Catch-22 - Help!
On Sat, 31 Oct 1998, Jerry E. McGoveran wrote: Then I ran make zImage (takes awhile!) and it aborted with the following error: as86 -0 -a -o bootsect.o bootsect.s make[1]: as86: Command not found ***[bootsect.o] Error 127 ***[zImage] Error 2 [EMAIL PROTECTED](pa):bhmit1$ dpkg -S as86 bin86: /usr/doc/bin86/examples/as86_encap bin86: /usr/man/man1/as86.1.gz bin86: /usr/bin/as86 So you need to install the bin86 package. Peeking into the Makefiles, I found as86 defined so: as86=$(CROSS_COMPILE)as86 -0 -a but I was unable to find a def for CROSS_COMPILE. I ran find / *as86 and there were no files found. You're cut and paste verifies that CROSS_COMPILE is undefined, otherwise, there would have been something before as86. HTH, Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Dijkstra probably hates me (Linus Torvalds, in kernel/sched.c) |
Re: I can't use talk.
On Fri, 30 Oct 1998, M.C. Vernon wrote: Err try installing the hamm ytalk - bo packages use the old libc, so may not work properly. ytalk is the client, I believe the problem is with the server. Debian boxes can talk to debian boxes. But, I think the daemon has changed and the problem is that not everyone uses the same version of the talk daemon. Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Dijkstra probably hates me (Linus Torvalds, in kernel/sched.c) |
Re: HP 5L still giving problems!!
Before going for all these detailed solutions, let me say that my HP 6L works fine with magicfilter. According to my printcap, I used hplj4. If the 6L person still has problems with that, we may want to do version checking to see what differs. I think I started using this printer in bo and I'm up to a semi out of date snapshot of slink (I need to check the archives to see if the x problems finally got fixed). On Fri, 23 Oct 1998, G. Kapetanios wrote: I had the same proble. My HP 6L would not work. I tried magicfilter. ASCII text printed fine but postscript dvi e.t.c. were garbage. [ script workaround sniped ] On Fri, 23 Oct 1998, Jose L Gomez Dans wrote: Hi! First of all, I'd like to thank the response that my previous posting to the list had. Most welcome :D Anyway, I tried one of the most common references people gave me (ie, use magicfilter, and use the ljet4l package). The printer prints now text fine, but when a postscript file is sent to it, it prints out all sorts of weird characters, and something that resembles the file. Does anybody know what's wrong with this? I've got gs (v3.33), and I've got my magicfilter for ljet4l (among others :D). Brandon +--- ---+ | Brandon Mitchell * [EMAIL PROTECTED] * http://bhmit1.home.ml.org/ | | Dijkstra probably hates me (Linus Torvalds, in kernel/sched.c) |
Re: have anyone got dhcpcd working ?
On Sat, 3 Oct 1998, Sergey V Kovalyov wrote: I wonder, have anyone manage to get a working dhcpcd ? I've tried to install it once, but the package seemd to be broken. I use it. My site needs the -r switch (or at least did and I never bothered to change it). Installation script complained about absence of /etc/dhcp directory but refused to create one, so I had to do it by hand and reinstall. I've been using it for a while, upgrading, no fresh installs for over a year. Although the last time I helped someone else this summer, there weren't any problems. Then the daemon itself could not create a socked - it requested type 17, which even did not have a name and hence the kernel tried to load non-existing module, something like net-11 Finally, I gave up. Hmm... never seen that one. Didn't show up in a grep of my logs. Closest I come is a non-fatal netscape caused message: Sep 20 14:18:20 cnhobbes modprobe: can't locate module net-pf-4 So I wonder - did I miss something, or it is indeed broken? I can't really verify either. I can say that it works for me under my specific setup. HTH, Brandon --+-- Brandon Mitchell [EMAIL PROTECTED] | Debian Testing Group Status PGP Key: finger -l [EMAIL PROTECTED] | http://bhmit1.home.ml.org/deb/ Dijkstra probably hates me (Linus Torvalds, in kernel/sched.c)
Re: Are there any Disk Managers compatible with Linux?
On Wed, 23 Sep 1998, The Foxes wrote: I am a newcomer to Linux and looking forward to using it, but my Western Digital 2.1 GB Hard Drive requires the Ontrack Disk Manager to run properly (my CMOS only handles hard drives up to 528 MB). Is there a disk manager that will run under Linux? Any assistance on this matter will be greatly appreciated!! Thanks so much. While linux doesn't need Ontrack installed to use your disk (it bypasses the cmos), it will work perfectly fine if you need to keep it installed for another os. No drivers should be needed. (This is from experience a long time ago with another WD harddrive.) Good luck, Brandon --+-- Brandon Mitchell [EMAIL PROTECTED] | Debian Testing Group Status PGP Key: finger -l [EMAIL PROTECTED] | http://bhmit1.home.ml.org/deb/ Dijkstra probably hates me (Linus Torvalds, in kernel/sched.c)
Re: I can,t seem to get out to the internet.
On Mon, 14 Sep 1998, The Burris's wrote: I have my connect script working. I connect to my isp using pppd call provider after it gets logged in I type pppd in a seperate x window and get the trash. I can ping my providers address I can ping my assigned address I have my DNS numbers in the correct file as nameserver 123.45.56.89 nameserver 234.43.23.32 I cannot get my arena browser and lynx browser to goto websights. also, I cannot ping any other websights. Sounds like your gateway isn't getting set for some reason: route add default gw gateway addr Best of luck, Brandon --+-- Brandon Mitchell [EMAIL PROTECTED] | Debian Testing Group Status PGP Key: finger -l [EMAIL PROTECTED] | http://bhmit1.home.ml.org/deb/ Dijkstra probably hates me (Linus Torvalds, in kernel/sched.c)
