UbuntuIRC / 2008 /10 /17 /#ubuntu-server.txt
niansa
Initial commit
4aa5fce
[01:31] <acemo> how can i make umask work for a user that doesn't has a home folder (like www-data)
=== beau__ is now known as citizen42alpha
=== PrivateVoid_ is now known as PrivateVoid
=== RoAk is now known as RoAkSoAx
[03:13] <mathiaz> zul: how do you use the ec2 pluging for vmbuilder?
[03:13] <mathiaz> zul: I don't see the help messages
=== mcasadevall is now known as NCommander
[03:41] <uvirtbot`> New bug: #284744 in ec2-ami-tools (multiverse) "should depend on libopenssl-ruby" [Undecided,New] https://launchpad.net/bugs/284744
[04:31] <trimeta> I'm having trouble with my LVM-RAID1 setup.
[04:32] <trimeta> I had an existing volume group with one physical volume (which was actually a RAID1 device) and two logical volumes.
[04:32] <trimeta> I just added a second physical volume (also actually a RAID1 device), extended one of my logical volumes to fill the volume group, and grew the filesystem.
[04:33] <trimeta> When booting from a LiveCD, I can see my volume group with no difficulty, mount the logical volumes, etc.
[04:33] <trimeta> However, when booting from the system, I get the following errors:
[04:34] <trimeta> Anyone here, or should I not bother posting the full error?
[04:36] <ScottK> trimeta: This channel is much more active during the European/US business day.
[04:37] <trimeta> ScottK: Fair enough.
[04:37] <trimeta> Do you know much about RAID or LVM, though?
[04:46] <trimeta> In particular, it appears (though I could be entirely wrong here) that for some reason Ubuntu isn't properly initializing one of my RAID1 devices at boot.
[04:49] <ScottK> trimeta: Nope.
[05:22] <trimeta> I'll check in the main channel, there're always people there, at least.
=== freaky_t is now known as fReAky_t
=== fReAky_t is now known as freaky_t
[07:29] <CrummyGummy> Hiya, I'm looking to bond two interfaces together. I've done it in centos like this -> http://tinyurl.com/5du6yu but all the info I've found on the subject for Ubuntu points to installing ifenslave. The Centos way looks cleaner to say the least. Am I correct in this?
=== beau__ is now known as charliechaplin
[07:45] <_ruben> CrummyGummy: this is how i've done it (one of many ways):
[07:45] <_ruben> http://paste.ubuntu.com/58691/
[07:56] * CrummyGummy looks
[08:05] <CrummyGummy> _ruben: I'll do it that way.
[08:05] <CrummyGummy> Thanks\
[08:58] <uvirtbot`> New bug: #284807 in dhcp3 (main) "Network is unreachable" [Undecided,Incomplete] https://launchpad.net/bugs/284807
[09:00] <tonyyarusso> Say, where could I find a summary of the new features in Intrepid specific to server usage?
[09:49] * delcoyote hi
[10:37] <kraut> moin
[12:36] <uvirtbot`> New bug: #284738 in samba (main) "[Intrepid] wrong icons for samba shares" [Undecided,New] https://launchpad.net/bugs/284738
[12:48] <c00l2sv> hi, i've got a problem
[12:49] <c00l2sv> I can't boot my machine after I did an upgrade of some packages
[12:49] <c00l2sv> that modified initrd
[12:49] <c00l2sv> image
[12:49] <c00l2sv> any advices please help
[14:11] <ScottK> Who is going to be the server team rep at the release meeting today?
=== enmand_ is now known as enmand
[14:40] <zul> ScottK: dendrobates probably would be my guest
[14:40] <zul> s/guest/guess/g
[14:41] <dendrobates> zul: yup
[14:42] <ScottK> dendrobates: (Assuming your the Server Team rep a the release meeting) - Clamav has a new RC out. They don't go final until 3 Nov, so we're kind of stuck. Generally their RC -> Final is just security fixes. So I'd like to discuss uploading the RC and then putting the final in -updates (assuming it all tests well, etc.)
[14:42] <ScottK> If we upload the RC soonish, that gives us a better chance of having any breakage detected and fixed in their final release.
[14:43] <ScottK> I'll be at the meeting for MOTU, so I can discuss.
[14:45] <dendrobates> ScottK: it is late, but I am open to it.
[14:46] <ScottK> dendrobates: It's 0.94 -> 0.94.1 so history says it should be pretty compatible.
[14:46] <ScottK> It's almost all bug fix.
[14:47] <dendrobates> ScottK: We might get some push back.
[14:48] <ScottK> Agreed, so we discuss it and see.
[15:01] <maestrolinux> http://s2.ar.bitefight.org/c.php?uid=19732
[15:12] <slestak> i have a 8.04.1 server running in a vm as a lamp server. it is a new install. mysql5 is running, but i cannot seem to log in (in order to set the root passwd). mysqladmin -u root dumps usage info instead of putting me into the mysql shell
[15:14] <slestak> i have googled and looked at the ubuntu server guide, but rtfm doesnt seem to be effective for me righ tnow :)
[15:14] <zul> hi mathiaz you were asking about vmbuilder/ec2 last night?
[15:14] <sommer> slestak: you should have been prompted to set the password during the mysql-server install
[15:14] <slestak> maybe the root user doesnt exist yet?
[15:15] <sommer> slestak: you should be able reset it using sudo dpkg-reconfigure mysql-server
[15:15] <slestak> sommer: i dont remember that prompt. i mustve passed it. ok, tyvm I will try that.
[15:15] <slestak> i never fully grasped the user and security model of mysql. i need to study
[15:17] <sommer> you can get pretty fined grained per table per database per user permissions with mysql
[15:20] <slestak> ive found imo, i can get more done faster with postgres, but not all products i want to use support it :)
[15:21] <slestak> is it super important to have a super hard, guarded mysql root password, or is the root user not a su like it is for *nix
[15:22] <sommer> slestak: I think so, there's usually important stuff in the db :)
[15:22] <slestak> sommer: but it is not recommended it be the same pwd as the server right
[15:23] <slestak> is it accepted to be the same as the host server?
[15:24] <sommer> I usually make it different from the server
[15:24] <sommer> doesn'
[15:24] <sommer> t necessarily have to be though
[15:25] <ScottK> dendrobates: I just realized the release meeting starts at the same time I have to pick up a kid from school, so if Server Team gets up before I get there, please just insert a placeholder for clamav so we can come back to it when I get there.
[15:36] <cpu_temp> I`m running ubuntu server. How can I permanently change "ulimit -n". "ulimit -n" gives "1024" and "cat /proc/sys/fs/file-max" gives "65535". So, what to belive?
[15:38] <henkjan> cpu_temp: /etc/security/limits.conf
[15:38] <cpu_temp> henkjan: wrote "* hard nofile 65535"
[15:39] <cpu_temp> henkjan: I also tryied the sysctl method -> ulimit -n still gives 1024
[15:50] <dendrobates> ScottK: ok
[15:55] <slestak> sommer: i verified mysql is running with ps -ef but the dpkg-reconfigure command we discussed returns nothing and propmts for nothing.
[16:02] <ivoks> slestak: what's the problem?
[16:02] <mathiaz> nijaba: I've filed a bunch of bugs against vmbuilder
[16:03] <mathiaz> nijaba: as a result of yesterday's testing
[16:03] <mathiaz> nijaba: someone should go through them and set an importance and figure out which one should be targeted for intrepid
[16:06] <nijaba> mathiaz: I did see them this morning. I've still been trying to fix the grub error 2. I am at the point where, comparing images created with uvb and pvb show NO significant difference in /boot, but one boots and the other does not, so I think I will wait for soren's return on that one
[16:07] <nijaba> mathiaz: I'll be triagging the others, but it seems that you mentioned a couple others that are not in the bug list (such as a problem with --dest), didn't you?
[16:10] <sommer> slestak: sorry, try: sudo dpkg-reconfigure mysql-server-5.0
[16:27] <zul> mathiaz: #284762 -- why?
[16:27] <mathiaz> nijaba: dest -> I've filed a bug
[16:27] <mathiaz> bug 284762
[16:27] <uvirtbot`> Launchpad bug 284762 in vmbuilder "ec2 plugin should be shipped as a separate package" [Undecided,New] https://launchpad.net/bugs/284762
[16:27] <mathiaz> zul: otherwise python-vm-builder has to be depend on python-boto and ec2-ami-tools
[16:28] <mathiaz> zul: which is not really required
[16:28] <mathiaz> zul: in order to be able to run vmbuilder
[16:28] <zul> mathiaz: probably not going to happen for intrepid
[16:28] <mathiaz> zul: well - for intrepid there is nothing for now
[16:28] <mathiaz> zul: so we can pretty much do what we want there
[16:29] <mathiaz> zul: there is also a missing dependency on ec2-ami-tools
[16:29] <mathiaz> zul: the openssl ruby library
[16:29] <mathiaz> zul: I've filed a bug for it
[16:29] <zul> mathiaz: k
[16:30] <zul> sorry about that
[16:33] <zul> yeah the ari/aki options can be renamed
[16:34] <zul> mathiaz: i dont think its a good idea to hard code the aki and ari because that information can change and become out of date
[16:36] <maestrolinux> http://s2.ar.bitefight.org/c.php?uid=19732
[16:53] <mathiaz> nijaba: why does python-vm-builder depend on kvm?
[16:56] <slugg1> hi. i've installed ubuntu server on my old computer. i included openssh during installation, and after installation i've installed openjdk, which didn't work for glassfish so i installed java from sun instead and then glassfish. That's all that's installed. I'm having problems with running glassfish, the start script will run normally - but the glassfish process will die, either instantaneously or after a while. After being up a few day
[16:58] <slugg1> i've altered the $PATH to include glassfish, and $JAVA_HOME to point to the correct java
[16:58] <slugg1> that should include every modification i've done to the system
[17:00] <slugg1> (all actions have been done through ssh)
[17:02] <slugg1> oh, and the system is updated
[17:10] <nijaba> mathiaz: hmmm good question... Can't see of any reasons at first sight
[17:32] <CMD_L1N3> hello everyone
[17:32] <CMD_L1N3> I am reinstalling ubuntu-server on my home machine....
[17:33] <CMD_L1N3> while i am back at school my father wants a GUI for it so he doesn't have to do alot in CLI
[17:33] <CMD_L1N3> what would be the best GUI to install?...something like Xfce?
[17:34] <[Solars]> thats a personall preference
[17:35] <CMD_L1N3> well. i want something that is not demanding and will be okay to leave up for a long time
[17:35] <CMD_L1N3> i don't really know a lot about window managers. i was seeing if anyone else suggests one
[17:37] <[Solars]> try them all, and uninstall the ones you don't want
[17:37] <slugg1> https://help.ubuntu.com/community/ServerFaq question#3 might be interesting for you..
[17:37] <uvirtbot`> New bug: #244568 in postfix (main) "saslauthd run control script not using OPTIONS from /etc/default/saslauthd" [Undecided,New] https://launchpad.net/bugs/244568
[17:51] <uvirtbot`> New bug: #285069 in libnss-ldap (universe) "openldap liblber and active directory" [Undecided,New] https://launchpad.net/bugs/285069
[18:00] <lamont> so why was 244568 assigned to postfix, I wonder?
=== MenZa is now known as MenZaxPhoenix
[18:03] <ScottK> lamont: Me too.
[18:59] <uvirtbot`> New bug: #267932 in samba (main) "Panel Launcher Crashes when Network File is Opened" [Undecided,New] https://launchpad.net/bugs/267932
=== MenZaxPhoenix is now known as MenZa_Aeris
[19:27] <sergevn> I have a very strange problem, when my server has an uptime for like 3 days ( random) I cannot connect to the machine again, I can open tcp sessions but they dont give the proper replay back.
[19:27] <sergevn> ssh, httpd, smtp, they al dont work anymore
[19:28] <sergevn> someone has some advice or had simmilar problem?
[19:28] <henkjan> check hardware
[19:29] <sergevn> problem is, i dont have physical access atm
[19:29] <sergevn> what hardware could it be? NIC?
[19:30] <henkjan> memory maybe
[19:37] <sergevn> hmm
=== daishadar is now known as nxmehta
[19:41] <daturan> sergevn, when you bring the machine back on line does the syslog show anything interesting? and does using the -v switch on the client when you try to ssh?
[19:51] <sergevn> daturan: by syslog you mean dmesg?
[19:52] <sergevn> daturan: it's not only ssh, it's all the network daemon's :)
[19:52] <daturan> well I would check the /var/log/syslog
[19:53] <daturan> yea I just had a similar issue,,
[19:53] <sergevn> well there doesnt seem to be anything spectacular in syslog
[19:54] <sergevn> ill start with disabling nfs, maybe that's a problem
[19:54] <daturan> yea I would try using verbose when ssh won't connect, at least you'll see what you're not getting,, can you telnet to 22 when ssh and the other services aren't working?
[19:55] <sergevn> yeah, i can telnet, that is the strange thing, alle the tcp! network daemon's dont give a reply back
[19:55] <sergevn> only an empty row :)
[19:56] <daturan> yea, that's weird, I just lost a bunch of .so files so all kinds of services failed
[19:57] <sergevn> the only thing that is left to try
[19:57] <sergevn> is changing the the router where it's behind now
[19:57] <daturan> I did an strace on ssh ,,, that's how I found what was missing,, I don't know how you'd do it for server services
[19:57] <sergevn> but dont think that is problem, because when i reboot the server, everything is fine
[19:58] <daturan> yea,, like you said it does look like it may be a hardware issue,, and dmesg should have shown something,, I dont' really know at this point
[19:58] <sergevn> agree. very frustrating, it's my backup server
[19:58] <sergevn> reboot is ghetto solution but..
[20:00] <daturan> hey it works,, at least till you can replace the hardware
[20:00] <daturan> can you leave a ssh session open? or does the problem cut off existing connections?
[20:01] <sergevn> havent tried that one yet.
[20:01] <sergevn> good one
[20:01] <daturan> yea at least you could poke around, it's tough when you don't have phy access
[20:02] <sergevn> sunday, wtf-problems day
[20:02] <sergevn> thanks for the help, im off.
[20:02] <daturan> c ya
[20:11] <uvirtbot`> New bug: #285132 in apache2 (main) "Apache2 has trouble handling (ip6-)localhost" [Undecided,New] https://launchpad.net/bugs/285132
[20:48] <kimi__> ubuntu 8.04.1 kvm virtio : if guest system have more then 3.5gb network stop working
[20:53] <maestrolinux> http://s2.ar.bitefight.org/c.php?uid=19732 the best of games
=== jarias_ is now known as jarias
[20:56] <kimi__> ubuntu 8.04.1 kvm virtio : if guest system have more then 3.5gb network stop working
[21:01] <ivoks> kimi__: this might not be related to RAM
[21:04] <kimi__> ivoks: when i change memory less then 3.5gb virtio works - greater then 3.5gb vritio stops working
[21:05] <ivoks> 32bit host?
[21:05] <kimi__> 64bit
[21:05] <kimi__> when i change to network device to rtl8136 - it works
[21:05] <ivoks> heh...
[21:05] <ivoks> could you report the bug?
[21:06] <ivoks> so we could track it...
[21:06] <ivoks> most of the 'virtual' guys aren't here
[21:06] <kimi__> second problem is - virsh console cant connect to guest system
[21:08] <kimi__> kvm 77 fix something with the memory >4gb
[21:11] <ivoks> 'night all
[21:11] <kimi__> ivoks
[21:12] <kimi__> you have a time for me when i can reach more guys from virt. team ?
[21:12] <ivoks> well, report the bug
[21:12] <ivoks> that's the best way to work on the issue
[21:12] <kimi__> ok thx
[21:13] <ivoks> np
[21:13] <ivoks> take care
=== jarias_ is now known as jarias
[21:18] <ivoks> lol
[21:18] <ivoks> i've actually found a job offering
[21:18] <ivoks> required: good knowledge of ubuntu server
=== daishadar is now known as nxmehta
[21:36] <mathiaz> zul: the bug with ec2-bundle is due to self.vm.filesystems[0]
=== ghaleb is now known as Vecam
[21:43] <Vecam> Hello, is there a way to encrypt folders as to be read by one user only, even the administrator can't read users data ?
[21:43] <hads> encfs?
[21:44] <Vecam> hads, I don't know if this can be intergrated with samba
[22:29] <zul> mathiaz: which bug are you talking about? the one with kvm and vmbuilder?
[22:29] <mathiaz> zul: yes
[22:29] <zul> k
[22:29] <mathiaz> zul: true - it may be easier to keep the raw images around
[22:30] <mathiaz> zul: my use case is that I had a failed upload to S3/
[22:31] <mathiaz> zul: so I just wanted to redo the upload where it was left
[22:31] <zul> mathias: can you do that?
[22:32] <mathiaz> zul: well - if I had the parts around, probably.
[22:32] <mathiaz> zul: as I would only have to run ec2-bundle-upload
[22:32] <mathiaz> zul: in my use case I'd have to run ec2-create-bundle (or whatever it's called) again
[22:33] <zul> ah...ok well im really off for dinner now :)
[22:33] <mathiaz> zul: ok
[22:33] <zul> Katie is starting to scream ;)
[22:36] <mathiaz> erichammond1: are you testing the ec2 plugin for vmbuilder?
[23:05] <mathiaz> nijaba: I've briefly looked at the configuration files handling
[23:06] <mathiaz> nijaba: the issue is that only command line options have their configuration value looked up in config file
[23:06] <mathiaz> nijaba: the problem is in set
[23:06] <mathiaz> nijaba: the problem is in set_default in vm.py
[23:07] <nijaba> mathiaz: already fixed
[23:08] <mathiaz> nijaba: hm - nevermind. Now that I think of it I was wrong
[23:08] <mathiaz> nijaba: anyway - you've fixed it - kwel.
[23:08] <mathiaz> nijaba: have you pushed your changes to trunk/ ?
[23:08] <nijaba> mathiaz: few minutes ago
[23:08] * mathiaz pulls
[23:09] <mathiaz> nijaba: rev 241 is the latest?
[23:09] <mathiaz> nijaba: I though you had merge my branch
[23:10] <nijaba> mathiaz: 243 is the latest, but lp is lagging
[23:10] <nijaba> mathiaz: also read my comment to https://bugs.launchpad.net/vmbuilder/+bug/284614
[23:10] <uvirtbot`> Launchpad bug 284614 in vm-builder "exec option not supported in configuration files" [Low,Fix committed]
[23:10] <mathiaz> nijaba: oh right - probably because LP is down ATM
[23:10] <nijaba> mathiaz: but 240 has the fix for it
[23:12] <nijaba> anyway, I am done with slaughtering bugs for today :) Good night everyone
[23:12] <mathiaz> bye nijaba