[00:04] <nOStahl> how can I remove old kernel headers
[00:05] <sarnold> same thing, apt-get purge linux-headers-blah ..
[00:06] <nOStahl> wont let me
[00:06] <nOStahl> keeps coming back saying the depends thing
[00:06] <sarnold> then delete whichever package depends upon them as well..
[00:07] <sarnold> you might need to put it back when you're finally done, of course, so keep track of which packages require the headers..
[00:07] <nOStahl> howdo I delete a package if it keeps not letting me delete them? :P
[00:09] <sarnold> nOStahl: add more package names to the command line
[00:09] <sarnold> any luck finding a pastebin that you can get to?
[00:10] <nOStahl> the terminal access I have to the vm wont let me copy
[00:10] <nOStahl> but scroll up I hand typed it all out painstakingly :)
[00:11] <sarnold> nOStahl: apt-get purge linux-headers-server linux-image-server   shoul dhave taken care of the ones you typed earlier..
[00:12] <nOStahl> it complains about dependancis of the current headers it looks like
[00:12] <nOStahl> and refuses to go further lol
[00:15] <sarnold> so I presume it makes similar claims, <packagefoo> : depends <packagebar> but it is not going to be installed.. -- keep adding <packagefoo>s to the apt-get purge command line until it stops complaining at you :)
[00:20] <nOStahl> k
[00:24] <nOStahl> nope not working
[00:24] <nOStahl> still thinks the disk is full even though its not
[00:24] <sarnold> check df output, perhaps it is...
[00:25] <stgraber> full even though it's not => did you check "df -i" when you run out of inodes your disk is "full" even though there's still bytes left to write to
[00:26] <sarnold> stgraber: oh man, it's been sooooo long since I've seen that, I'd forgotten all about it. :) nice catch.
[00:35] <nOStahl> ya its inods
[00:35] <Patrickdk> yay :)
[00:35] <nOStahl> not yay heh
[00:35] <Patrickdk> evil inodes
[00:35] <nOStahl> how do I take care of inods
[00:36] <Patrickdk> as far as I know, reformat
[00:36] <Patrickdk> is it ext3/4?
[00:37] <sarnold> maybe a fsck will clean out a bunch?
[00:37] <sarnold> look for a huge /lost+found directory on the filesystem?
[00:37] <Patrickdk> ya, a reboot could clear some, if some deleted files are held by open programs
[00:38] <Patrickdk> maildir folders
[00:38] <nOStahl> just rebooted to check that still 100%
[00:39] <nOStahl> how do you empty lost+found?
[00:40] <Patrickdk> rm -rf /lost+found/*
[00:40] <nOStahl> k just didnt know if there was a command associated with managing it
[00:44] <sarnold> look at the files in there, see if you recognize the data, and if you do, mv the file to a new name. if you do'nt, delete the file. hehe.
[00:44] <Patrickdk> well, if he hasn't noticed it missing yet :)
[00:45] <sarnold> Patrickdk: lol
[00:55] <nOStahl> just going to resize the partition heh
[00:55] <nOStahl> wasted too much time already
[00:56] <sarnold> resizing may not add new inodes
[00:57] <sarnold> unless you're re-sizing to make space to make a new /boot or similar..
[00:57] <nOStahl> if it dosnt work then i'll just build a new vm and transfer stuff over I guess
[01:08] <Patrickdk> yes, but exactly how did you run into this issue?
[01:08] <Patrickdk> by default it makes enough inodes for every sector
[01:08] <Patrickdk> you have to specify largefile to cause it to make less inodes
[01:09] <Patrickdk> did you even locate where all the inodes have gone yet?
[01:09] <nOStahl> nope no one knew how
[01:09] <Patrickdk> heh?
[01:09] <Patrickdk> no one knew how what?
[01:10] <nOStahl> how to locate it
[01:10] <Patrickdk> that is rather easy
[01:10] <nOStahl> I ran a litle bash line that located a wordpress install as holding the most inodes
[01:11] <sarnold> heh, php sessions or something?
[01:12] <mastr[x]bennett> ah. server specific. excellent
[01:12] <Patrickdk> no gui talk!
[01:18] <nOStahl> what I dont understand is that that wordpress install is just running 2 images on a slider for a digital display so how is it using so many inodes lol
[01:18] <Patrickdk> my guess?
[01:18] <Patrickdk> it got hacked, and you have all kinds of other websites your hosting now :)
[01:19] <sarnold> Patrickdk++
[01:19] <nOStahl> not easy feat to do when its local networked :P
[01:39] <delinquentme> SO
[01:40] <delinquentme> what kind of files should I be worried about having 777
[01:40] <sarnold> if you've got a FAT filesystem on an SD card or something, those files might be 777. you probably shouldn't have any others on your systems.
[01:41] <delinquentme> sarnold, yeahhh so thats one of the things I've noticed :D
[01:41] <delinquentme> but like .xls files?
[01:41] <delinquentme> or .jpg?
[01:41] <delinquentme> what about .py ?
[01:43] <delinquentme> sudo find . -perm 777 -type f
[01:43] <delinquentme> this will searching within the current dir right?
[01:48] <Valduare> so anything new with ubuntu server 13.04?
[06:26] <akaWolf> Hello! I'm using svnserve v. 1.7.5 and libsasl2-2_2.1.25. I start the svnserve, then I try to connect to localhost repository, and then I've got "svn: E170001: Authentification error"
[06:26] <akaWolf> Can everyone help me?
[06:50] <Senor> why nslookup command  returns "localhost"  for all 192.168.1.x  at  192.168.1.116 ?
[09:31] <Senor> what is eth0:0
[09:31] <Senor> ?
[09:33] <bekks> A virtual IP on eth0
[09:34] <Senor> if I tcpdump all packets throguh its ip ,what -i option should be given out?
[09:34] <Senor> tcpdump -i eth0 ? or -i eth0:0 ?
[09:36] <bekks> Neither nor. -i specifies an interface, regardless of what IP is configured on that interface.
[10:01] <Senor> bekks:chould you tell me what command to type?
[10:02] <bekks> Whats the problem you are trying to solve behind all that?
[10:26] <Senor> I want to detect whether  packets are transfered  through eth0:0 ro realserver
[10:26] <bekks> Then take a look at your netmasks (of all interfaces) and your routing table.
[10:29] <Senor> then ?
[10:31] <bekks> Then you will see wether packets have to pass eth0:0
[10:33] <Senor> ok,I try to vivsit eth0:0 ,expecting it transfers my request to other server ,then get the reply from that server ,but I have no received reoly from it
[10:33] <Senor> what is  about the reaon?
[10:33] <Senor> reason
[12:56] <qiyong> why my apache ps shows as: /usr/sbin/apach
[12:56] <qiyong> not apache2
[12:56] <qiyong> W: Failed to fetch http://mirror/ubuntu/dists/precise-security/multiverse/binary-i386/Packages  404  Not Found
[12:56] <qiyong> why amd64 box fetching i386 files? i don't have any deb-i386 in sources.list at all.
[13:10] <jtv> qiyong: Your apache may up as /usr/sbin/apach just because there isn't room... try piping the output through "| cat"
[13:10] <jtv> I don't know why an amd64 box would fetch i386 files, unless of course the installed OS is i386.
[13:11] <qiyong> jtv: pgrep shows the name with path
[13:11] <qiyong> jtv: the path shouldn't showup
[13:11] <jtv> Why not?
[13:13] <qiyong> did you try?
[13:13] <jtv> It may be a matter of which ps options you use...  I like "ps -ef" (POSIX flavour).  For me, the paths normally show up.
[13:13] <jtv> You didn't say which options you use.
[13:13] <jtv> But some programs override the string that shows up in ps to be something shorter.
[13:14] <qiyong> just bare ps
[13:14] <StathisA> i can only install 12.04 on a VM for testing and learning...am i losing much from 13.04?
[13:14] <qiyong> and other procs don't have paths
[13:16] <jtv> That doesn't mean it's wrong though.  If apache sets its ps text to its full path, then that's the right thing for ps to show.
[13:22] <qiyong> it breaks my scripts
[13:22] <qiyong> another 12.04 shows normally
[13:22] <jtv> That sounds as if there may be something weird about the apache you have installed.
[13:24] <mardraum> StathisA: for testing and learning, probably not
[13:24] <mardraum> StathisA: detail more about what you are trying to achieve?
[13:29] <qiyong> why amd64 box fetching i386 files? i don't have any deb-i386 in sources.list at all.
[13:33] <StathisA> well tbh not much, just messing around with cli...probably some headless file server/torrent downloader in the end
[13:38] <mardraum> StathisA: you'll be fine
[13:38] <StathisA> mardraum: ok cheers!
[13:42] <qiyong> i have pkg names as foo:i386
[13:42] <qiyong> is that right?
[13:48] <mardraum> qiyong: it is for a 32 bit pkg, yes?
[13:48] <qiyong> mardraum: anyway my box is amd64
[13:48] <qiyong> i guess the pkg db is screwed up
[13:49] <mardraum> no, it's probably that you installed something needing 32 bit pkgs
[13:53] <qiyong> damn, aptitude shows pkgs in arhcitecture: i386
[13:53] <qiyong> that status is 'p' though
[14:17] <qiyong> where can I talk about apt internals?
[14:21] <rbasak> qiyong: there's #debian-apt on OFTC. There are relatively few people involved in apt internals though.
[14:21]  * rbasak used to lurk there
[14:52] <kearneykid> hey i have a HP ProLiant ML330 G3 Server and i want to install ubuntu server on it. i have tried version 13.04 and 10.04
[14:52] <kearneykid> when installing both give me a blank screen
[14:54] <kearneykid> any ideas?
[14:58] <kearneykid> marlinc any idea?
[14:59] <marlinc> What?
[14:59] <kearneykid> hey i have a HP ProLiant ML330 G3 Server and i want to install ubuntu server on it. i have tried version 13.04 and 10.04
[14:59] <kearneykid> when installing both give me a blank screen
[15:02] <marlinc> I've got no idea to be honest
[15:02] <kearneykid> no problem thanks
[15:31] <Koheleth> Hi, how do I install and use a new language pack on my server, its currently French but I need it in English
[15:31] <Koheleth> 12.04 LTS
[15:38] <rbasak> Koheleth: install language-pack-en. I'm not sure if you'll need it, but "locale-gen en_US.UTF-8" (or whichever variant) won't hurt.
[15:44] <qiyong> what is var/log/user.log for?
[16:12] <qiyong> service or start/stop, which toolset is the go?
[16:13] <RoyK> doesn't matter
[16:41] <genii> qiyong: service will do both upstart and sysvinit, start/stop as far as I'm aware only does upstart
[16:56] <RoyK> genii: oh, didn't know that
[17:50] <adidas> I have two ubuntu servers. One is meant to be a fileserver with a RAID 1 configuration (12.10), and the second is a web/app server (12.04.2) running uwsgi-python which is serving up a python/pyramid application. I was able to mount a directory that contains images in the RAID server on the web server using sshfs, but for some reason the web app cannot write images to the mounted RAID directory. It can read the images though. I mounte
[17:50] <adidas> d the directory as www-data and when I am acting as www-data via su I can write to the RAID directory. However, the web app gets a permission error when it attempts to write an image (it can create directories so that's a mystery to me). I am not sure how to troubleshoot this since  ps reports that the app is running as www-data.
[17:51] <hachre> adidas: it must be a max filesize issue, or some additional restrictions that are set up on this uwsgi-python thing
[17:52] <adidas> thank you!
[17:53] <adidas> I was able to get it to work once with an image from a phone. Then I attempted to upload larger images. I kept testing with the larger images with no luck.
[17:54] <adidas> I just tried it again with that same cell phone image (just under a MB) and it worked!!
[17:54] <adidas> thanks so much
[17:55] <hachre> :)
[17:55] <hachre> np
[17:56] <adidas> I guess thats what touch worked too lol
[18:56] <StathisA> i'm running a headless server, and i need something to help me backup stuff from a NAS to another NAS on scheduled intervals...anybody knows if a package can do this?
[18:57] <StathisA> smth like Cobian for instance, but headless and for ubuntu
[18:57] <StathisA> also has to NOT use Rsync
[19:01] <melmoth> whats wrong with rsync ? that s exactly what is it for.
[19:02] <StathisA> RSync tends to keep the source file permissions, ending up with unusable backups - unless i chown them (as an admin)
[19:02] <andol> StathisA: Well, rsync has plenty of flags in regards to what metadata to keep, or not to keep.
[19:02] <melmoth> well, that s not really a problem, you scn script that easily. and i bet there may be an option about that as well
[19:03] <melmoth> like --no-perms
[19:03] <StathisA> hmm..ok i might try this up again then...any interfaces (web maybe) that'll help?...i like CLI but sometimes i need some GUI
[19:04] <melmoth> dont know of any.
[19:04] <melmoth> i launch my backup in some bash script in a screen session.
[19:04] <andol> StathisA: While I don't have any good gui for you, I'd like to point you in the direction of rsnapshot. It gives you a nice config file wrapping, provides snapshots based on hard links, etc.
[19:06] <StathisA> i could Cron some bash i guess, but i'd like some reporting on whether it completed - and email notification when it doesnt. i probably could work on some scripting but it be nice to have it newb-style too