- WLUG - wlug.mailman3.com

Re: [Wlug] APT install question

by Mike Perry

Hmmm weird. I just tried to install cpufrequtils on my box and I didn't run into that problem. Sounds like a fluke to me. Try deleting the package from the cache '/var/cache/apt/archives' and reinstalling as Aaron detailed. I ran Kubuntu for a bit and switched back to Ubuntu as I find the integration is much better. But under the hood they should be the same system. When I was running Kubuntu I did find that it already had a cpufreq tool built into its power management. cpufrequtils is in universe and would be an add on. I know Ubuntu's Gnome package manager, Synaptic, has a "fix broken packages" option which may also fix this problem. Perhaps Adapt has something similar. *Ubuntu also has smart if you are familiar with that. sudo apt-get install smart-pm ~Mike > On Sat, 2006-12-30 at 07:11 -0500, Aaron Haviland wrote: > > Andy Stewart wrote, on Dec 29, 2006 at 21:49 EST: > > > -----BEGIN PGP SIGNED MESSAGE----- > > > Hash: SHA1 > > > > > > > > > HI gang, > > > > > > I'm taking my first adventurous steps in the use of a Debian based > > > system (Kubuntu) on my new laptop (LC2464T). (No, I'm not trashing Suse > > > on all of my other systems). I've never previously used an APT based > > > system, so in this sense, I'm quite the newbie. > > > > I wouldn't suggest *Ubuntu as a canonical APT-based distro. There are > > still bugs there but the eye candy is better. > > > > > I'm getting an error when I install a package, and I'm not sure how to > > > resolve it. I've googled to no avail. Note in the output below that it > > > says that the file /etc/init.d/cpufrequtils does not exist, and yet (I > > > think) this is a file that belongs to the package which I'm trying to > > > install. > > > > Indeed it should be part of the package you're trying to install. Could > > you verify that the files does exist? By this point in the installation, > > the package is fully extracted, just not configured. > > > > > > A quick & dirty version of the dpkg installation procedure (APT is a front-end to dpkg/wget/etc) is: > > > > 1) Extract control information from the package. (You can find this > > stuff in /var/lib/dpkg/info/${PACKAGE}.* > > > > 2) Execute pre-install script, if it exists > > > > 3) Extract package. > > > > 4) Run post-install script, if it exists. (I think almost all packages > > have one) > > > > 5) ... > > > > 6) Profit! > > > > > > <snipped ...> > > > > > Get:2 http://us.archive.ubuntu.com edgy/universe cpufrequtils 002-1 [22.4kB] > > > Fetched 32.7kB in 1s (31.8kB/s) > > > > > Selecting previously deselected package cpufrequtils. > > > > 1, 2 > > > > 3 > > > > > Unpacking cpufrequtils (from .../cpufrequtils_002-1_amd64.deb) ... > > > > 4 > > > > > Setting up cpufrequtils (002-1) ... > > > update-rc.d: /etc/init.d/cpufrequtils: file does not exist > > > dpkg: error processing cpufrequtils (--configure): > > > subprocess post-installation script returned error exit status 1 > > > Errors were encountered while processing: > > > cpufrequtils > > > E: Sub-process /usr/bin/dpkg returned an error code (1) > > > ========== > > > > So at this point, the package should be completely extracted, but the > > postinstall is not processing correctly... the command it's trying to > > execute would be: "update-rc.d cpufrequtils defaults" to add the default > > init rc symlinks... > > > > It might be a bug in the package. It might be a fluke. Usually, try > > re-installing it (as you won't be able to do much else until you get it > > resolved, anyway...) > > > > apt-get install --reinstall cpufrequtils > > > > > > If not... good luck figuring out ubuntu's bug system. Debian's is quite > > nice, but if you've got a ubuntu bug, you *have* to use ubuntu's BTS > > instead. (And it appears there are no current bugs filed for ubuntu's > > cpufrequtils. Interestingly enough, Debian's BTS lists several bugs > > against the package, but none seemingly related to your problem.) > > > > > > > > Hmmm... also, how long ago did this happen? Ubuntu doesn't seem to have > > 002-1 in its archives anymore. The current version is 002-2. > > > > I don't know if you're using a front-end to APT, or taking it plain, but > > you might need to run "apt-get update" to refresh the package lists. > > > > > > That was a bit longer than I meant it to be... good luck! > > > > Alternatively, Some friends of mine in #linuxhelp on OFTC are usually > > around to help. I'm in there as 'guinea-pig' > > > > _______________________________________________ > > Wlug mailing list > > Wlug(a)mail.wlug.org > > http://mail.wlug.org/mailman/listinfo/wlug

Dec. 30, 2006

Re: [Wlug] APT install question

by Aaron Haviland

Andy Stewart wrote, on Dec 29, 2006 at 21:49 EST: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > HI gang, > > I'm taking my first adventurous steps in the use of a Debian based > system (Kubuntu) on my new laptop (LC2464T). (No, I'm not trashing Suse > on all of my other systems). I've never previously used an APT based > system, so in this sense, I'm quite the newbie. I wouldn't suggest *Ubuntu as a canonical APT-based distro. There are still bugs there but the eye candy is better. > I'm getting an error when I install a package, and I'm not sure how to > resolve it. I've googled to no avail. Note in the output below that it > says that the file /etc/init.d/cpufrequtils does not exist, and yet (I > think) this is a file that belongs to the package which I'm trying to > install. Indeed it should be part of the package you're trying to install. Could you verify that the files does exist? By this point in the installation, the package is fully extracted, just not configured. A quick & dirty version of the dpkg installation procedure (APT is a front-end to dpkg/wget/etc) is: 1) Extract control information from the package. (You can find this stuff in /var/lib/dpkg/info/${PACKAGE}.* 2) Execute pre-install script, if it exists 3) Extract package. 4) Run post-install script, if it exists. (I think almost all packages have one) 5) ... 6) Profit! <snipped ...> > Get:2 http://us.archive.ubuntu.com edgy/universe cpufrequtils 002-1 [22.4kB] > Fetched 32.7kB in 1s (31.8kB/s) > Selecting previously deselected package cpufrequtils. 1, 2 3 > Unpacking cpufrequtils (from .../cpufrequtils_002-1_amd64.deb) ... 4 > Setting up cpufrequtils (002-1) ... > update-rc.d: /etc/init.d/cpufrequtils: file does not exist > dpkg: error processing cpufrequtils (--configure): > subprocess post-installation script returned error exit status 1 > Errors were encountered while processing: > cpufrequtils > E: Sub-process /usr/bin/dpkg returned an error code (1) > ========== So at this point, the package should be completely extracted, but the postinstall is not processing correctly... the command it's trying to execute would be: "update-rc.d cpufrequtils defaults" to add the default init rc symlinks... It might be a bug in the package. It might be a fluke. Usually, try re-installing it (as you won't be able to do much else until you get it resolved, anyway...) apt-get install --reinstall cpufrequtils If not... good luck figuring out ubuntu's bug system. Debian's is quite nice, but if you've got a ubuntu bug, you *have* to use ubuntu's BTS instead. (And it appears there are no current bugs filed for ubuntu's cpufrequtils. Interestingly enough, Debian's BTS lists several bugs against the package, but none seemingly related to your problem.) Hmmm... also, how long ago did this happen? Ubuntu doesn't seem to have 002-1 in its archives anymore. The current version is 002-2. I don't know if you're using a front-end to APT, or taking it plain, but you might need to run "apt-get update" to refresh the package lists. That was a bit longer than I meant it to be... good luck! Alternatively, Some friends of mine in #linuxhelp on OFTC are usually around to help. I'm in there as 'guinea-pig' -- Aaron Haviland

Dec. 30, 2006

APT install question

by Andy Stewart

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 HI gang, I'm taking my first adventurous steps in the use of a Debian based system (Kubuntu) on my new laptop (LC2464T). (No, I'm not trashing Suse on all of my other systems). I've never previously used an APT based system, so in this sense, I'm quite the newbie. I'm getting an error when I install a package, and I'm not sure how to resolve it. I've googled to no avail. Note in the output below that it says that the file /etc/init.d/cpufrequtils does not exist, and yet (I think) this is a file that belongs to the package which I'm trying to install. Any and all help would be greatly appreciated - thanks! Here's the command and its output from Kubuntu 6.10. ========== # apt-get install cpufrequtils Reading package lists... Done Building dependency tree Reading state information... Done The following extra packages will be installed: libcpufreq0 The following NEW packages will be installed: cpufrequtils libcpufreq0 0 upgraded, 2 newly installed, 0 to remove and 0 not upgraded. Need to get 32.7kB of archives. After unpacking 254kB of additional disk space will be used. Do you want to continue [Y/n]? y Get:1 http://us.archive.ubuntu.com edgy/universe libcpufreq0 002-1 [10.3kB] Get:2 http://us.archive.ubuntu.com edgy/universe cpufrequtils 002-1 [22.4kB] Fetched 32.7kB in 1s (31.8kB/s) Selecting previously deselected package libcpufreq0. (Reading database ... 88398 files and directories currently installed.) Unpacking libcpufreq0 (from .../libcpufreq0_002-1_amd64.deb) ... Selecting previously deselected package cpufrequtils. Unpacking cpufrequtils (from .../cpufrequtils_002-1_amd64.deb) ... Setting up libcpufreq0 (002-1) ... Setting up cpufrequtils (002-1) ... update-rc.d: /etc/init.d/cpufrequtils: file does not exist dpkg: error processing cpufrequtils (--configure): subprocess post-installation script returned error exit status 1 Errors were encountered while processing: cpufrequtils E: Sub-process /usr/bin/dpkg returned an error code (1) ========== - -- Andy Stewart, Founder Worcester Linux Users' Group Worcester, MA, USA http://www.wlug.org -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org iD8DBQFFldO9Hl0iXDssISsRAq1gAJ46+gxWgO9YomxhdJsp0i02GG30XgCfTll1 k584/OlsBqwCFII50BVox90= =VxI2 -----END PGP SIGNATURE-----

Dec. 30, 2006

Test...please ignore

by Andy Stewart

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 - -- Andy Stewart, Founder Worcester Linux Users' Group Worcester, MA, USA http://www.wlug.org -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org iD8DBQFFldK4Hl0iXDssISsRArctAJ9W/DRFiHEiLoJZo9ApGPvo7dTo6ACfTtZ2 qXhhobwHcadmxsOucQiZAoc= =kqJP -----END PGP SIGNATURE-----

Dec. 30, 2006

Re: [Wlug] Can't find files filling up disk

by Michael Zarozinski

Thank you! It was some daemons that didn't shut down properly! Theo Van Dinter wrote: > On Tue, Dec 26, 2006 at 12:27:33PM -0500, Michael Zarozinski wrote: > >> I'm getting warnings that one of my drives is 90% full but I can't find >> the files filling it up. >> >> /dev/hda3 4.9G 4.2G 512M 90% /var <<<<<<< >> >> But running "du -sh *" from / shows /var as only having 322M (full >> output is below). >> >> Any hints on where I can find the files filling up /dev/hda3??? >> > > Aha! I actually ask this exact question during interviews. :) > > There could be several reasons, but a common reason is daemons which are still > writing to log files which were deleted. Do something like "lsof | grep > deleted" and see what's being used. For instance, on my system: > > python 2323 mailman 3u REG 8,2 0 > 75931863 /var/log/mailman/error.5 (deleted) > > So it appears that there was some issue when log rotating where the mailman > proc didn't close the old file, so while it doesn't exist in the directory any > longer (and therefore won't count via du) it still exists in the filesystem > and takes up space (in this case, 0 bytes, but ...) > > > ------------------------------------------------------------------------ > > _______________________________________________ > Wlug mailing list > Wlug(a)mail.wlug.org > http://mail.wlug.org/mailman/listinfo/wlug >

Dec. 26, 2006

Re: [Wlug] Can't find files filling up disk

by Theo Van Dinter

On Tue, Dec 26, 2006 at 12:27:33PM -0500, Michael Zarozinski wrote: > I'm getting warnings that one of my drives is 90% full but I can't find > the files filling it up. > > /dev/hda3 4.9G 4.2G 512M 90% /var <<<<<<< > > But running "du -sh *" from / shows /var as only having 322M (full > output is below). > > Any hints on where I can find the files filling up /dev/hda3??? Aha! I actually ask this exact question during interviews. :) There could be several reasons, but a common reason is daemons which are still writing to log files which were deleted. Do something like "lsof | grep deleted" and see what's being used. For instance, on my system: python 2323 mailman 3u REG 8,2 0 75931863 /var/log/mailman/error.5 (deleted) So it appears that there was some issue when log rotating where the mailman proc didn't close the old file, so while it doesn't exist in the directory any longer (and therefore won't count via du) it still exists in the filesystem and takes up space (in this case, 0 bytes, but ...) -- randomly Selected Tagline: "I didn't have Scotch Tape for many years, but I sure as hell use it now." - Larry Young, 2002/06/16, discussion about not having email

Dec. 26, 2006

Re: [Wlug] Can't find files filling up disk

by Chuck Anderson

On Tue, Dec 26, 2006 at 12:27:33PM -0500, Michael Zarozinski wrote: > /dev/hda3 4.9G 4.2G 512M 90% /var <<<<<<< > root@games [/]# du -sh * > 322M var I would go to single user mode, unmount /var, and run a filesystem check on it. /sbin/telinit 1 umount /var /sbin/fsck -f /dev/hda3 You can also check for what is using the space on /var. I use du -sk so that I can sort the result in kilobytes: df -h /var df -hi /var du -sk /var/* | sort -n The last one won't catch files that begin with a ., so check for large dot files: ls -la /var du -sk /var/.[!.]* | sort -n

Dec. 26, 2006

Can't find files filling up disk

by Michael Zarozinski

Hi all, I'm getting warnings that one of my drives is 90% full but I can't find the files filling it up. Running "df -ha" shows: Filesystem Size Used Avail Use% Mounted on /dev/hda1 2.0G 789M 1.1G 42% / none 0 0 0 - /proc none 0 0 0 - /sys none 0 0 0 - /dev/pts usbfs 0 0 0 - /proc/bus/usb none 506M 0 506M 0% /dev/shm /dev/hda7 58G 1.9G 53G 4% /home /dev/hda6 2.0G 36M 1.9G 2% /tmp /dev/hda2 4.9G 2.6G 2.2G 55% /usr /dev/hda3 4.9G 4.2G 512M 90% /var <<<<<<< Problem disk none 0 0 0 - /proc/sys/fs/binfmt_misc sunrpc 0 0 0 - /var/lib/nfs/rpc_pipefs But running "du -sh *" from / shows /var as only having 322M (full output is below). Any hints on where I can find the files filling up /dev/hda3??? Thanks, Michael root@games [/]# du -sh * 8.0K aquota.user 6.4M bin 11M boot 168K dev 12K du_var 20M etc 8.0K hardware_status.html?mac=00:0C:F1:D7:9B:4B&action=update&status=RH_POST_KICKSTART &hk=cf36ac8c346755f3d318bbd694a168c7799c74de 1.7G home 8.0K initrd 129M lib 16K lost+found 16K media 8.0K misc 8.0K mnt 8.0K opt 0 pico 902M proc 8.0K quota.user 580M root 16M sbin 4.0M scripts 8.0K selinux 8.0K srv 0 sys 72K tftpboot 48K tmp 2.7G usr 322M var

Dec. 26, 2006

Re: [Wlug] Problem w/gcc

by Keith Wright

> From: Mike Frysinger <vapier(a)gentoo.org> > > On Sunday 24 December 2006 19:32, Keith Wright wrote: > > > From: Mike Frysinger <vapier(a)gentoo.org> > > > > > > if gcc is randomly ICEing, > > > > What's that then? ICE seems to be an acronym > > it's a gcc term: Internal Compiler Error Thank you, though this is a bit disapointing. I was hoping it meant Inexplicable Computer Error, or something else more general purpose. > > This could be your chance to get your entry > > in the jargon file, and if it becomes the > > next big catch-phrase you might even bargain > > to exchange the IP for some free advice. > > the thing run by esr ? yeah i could care less :p I'm guessing you couldn't care very much less without going negative. I was about to object that the jargon file is older than ESR, but that's not strictly true. In any case, he does run it, by the "You touch it, you own it" Hacker rule of IP. (Which is very different from the Lawyer's rule. They never have to Fix the Damned Thing.) Just for that, no free advice for you. -- Keith

Dec. 25, 2006

Re: [Wlug] Problem w/gcc

by Mike Frysinger

On Sunday 24 December 2006 19:32, Keith Wright wrote: > > From: Mike Frysinger <vapier(a)gentoo.org> > > > > if gcc is randomly ICEing, > > What's that then? ICE seems to be an acronym it's a gcc term: Internal Compiler Error > This could be your chance to get your entry > in the jargon file, and if it becomes the > next big catch-phrase you might even bargain > to exchange the IP for some free advice. the thing run by esr ? yeah i could care less :p -mike

Dec. 25, 2006