[00:06] robbiew, Daviey either of you guyses around? [00:06] does the server install come with langpacks by default? [00:06] hey all, micahg is asking becuase of bug #800857 [00:06] Launchpad bug 800857 in firefox "language packs pull in Firefox on upgrade" [High,In progress] https://launchpad.net/bugs/800857 [00:08] basically, if you are running natty, and you have lang pack installed ....and you do a dist-upgrade [00:08] the result is that Firefox will get installed [00:08] it's a packaging bug that micahg has already fixed and is uploading the fix now [00:08] is this something we need to urgently let server users know about? [00:16] SpamapS, hey [00:17] rickspencer3: howdy! [00:18] he SpamapS [00:18] I just wanted to draw your attention to this issue ^ [00:18] * SpamapS looks at the server seeds [00:19] I don't see any direct dependency on it [00:20] ii language-selector-common 0.34.1 Language selector for Ubuntu Linux [00:20] Thats all I see on my natty servers [00:21] SpamapS: there's no language-pack-* installed? [00:21] micahg: no there's no need [00:22] ok [00:23] SpamapS, I can't invite you to #ubuntu-mozillateam [00:24] rickspencer3: I don't think its necessary [00:24] SpamapS, well, they are asking questions about server [00:24] ah [00:26] New bug: #795500 in open-vm-tools (multiverse) "package open-vm-dkms 2011.03.28-387002-0ubuntu1.1 failed to install/upgrade: open-vm-tools kernel module failed to build" [Undecided,Incomplete] https://launchpad.net/bugs/795500 [00:49] getting /bin/sh: if-up: not found failed to bring up br0 [00:53] trying to get a bridge for kvm to run off of my bond 5 interfaces in [00:55] http://pastebin.com/7PkiKeFu [01:04] bah..... took out the pre-up if-up bond0 works great [01:21] hey all! [01:34] hallyn: SpamapS: so arch (and uname -m) reports x86_64 inside the i386 lxc [01:35] if you guys have ideas for fixing that (for example, running /sbin/init under linux32) that would be good [01:37] lifeless: yes, but go ahead and gcc a file that does sizeof(int) [01:37] lifeless: arch and uname -a use uname(2) output. we can't forge that. [01:37] but you're running 32-bit, and compiling 32-bit apps [01:37] hallyn: man linux32 [01:38] hallyn: we very much can forge uname output :) [01:39] jcool, guess that's how schroot does it [01:39] i see no reason why we can't have lxc-start run linux32 [01:39] will have to see how it does it [01:40] lifeless: i'll hack that in tomorrow [01:40] would you like a bug report ? [01:40] lifeless: how about just a comment on the existing one? [01:40] sure [01:40] lifeless: thanks for the tip :) [01:40] somebody wasn't thinking [01:41] * hallyn points at himself [01:42] commented [01:42] thx [01:45] does it really just set sys_personality? [01:46] guess i'll just do that [01:53] lifeless is everywhere [01:53] hallyn: sudo lxc-start -n lucid-test-lp -d usr/bin/linux32 sbin/init [01:54] hallyn: gets me thr right arch when I then ssh into it ;) [01:54] Ursinha: hola :> [01:54] lifeless: exactly, except i thin ki'd prefer to use setarch manually [01:54] lifeless: i don't understand, though, why uname -a works then [01:54] lifeless: hiya [01:54] hallyn: be my guest, I'm just hacking around till you catch up :> [01:54] in the kernel code, it doesn't look like it should work [01:55] hallyn: uname -a [01:55] Linux lucid-test-lp 2.6.38-9-generic #43-Ubuntu SMP Thu Apr 28 15:23:06 UTC 2011 i686 GNU/Linux [01:56] lifeless: right, and strace it and it just calls uname(). Which should just copy the uts_ns->utsname contents [01:56] now maybe glibc is helping out [01:57] * lifeless is happy w/magic [01:57] heh [02:04] hallyn: oh yeah, where should I report the lxc-stop no working thing ? [02:04] I presume its the cgroup thing ? [02:05] lifeless: against lxc. it's a cgroup thing, but a bug in lxc [02:05] new bug, or in the 'broken by cgroup' bug ? [02:05] what is the broken by cgroup bug? [02:05] https://bugs.launchpad.net/ubuntu/+source/lxc/+bug/784093 [02:05] Launchpad bug 784093 in lxc "lxc broken by cgroup-bin" [High,Confirmed] [02:06] yeah i think it's the same thing. Just a different aspect of it. What he describes inthe description would be worked around by SpamapS' config. [02:07] but it is that bug, just comment in there that lxc-stop doesn't work. [02:07] I have spamaps config [02:07] right [02:07] will do [02:07] thx [02:07] killing my client now, ttyl [02:07] ciao === JasonnAWAY is now known as Jasonn [02:15] It's not "broken"; the user just needs to mount ANOTHER cgroup with "lxc" instead of "none" as its source device [02:16] At least, that's my ten-second diagnosis [02:17] IIRC lxc looks through /proc/mounts for cgroup mountpoints, it picks the one called "lxc" or otherwise the first one it finds. === Jasonn is now known as JasonnAWAY === lifeless_ is now known as lifeless [04:11] lifeless: do'h, that was easier than i thought, lxc already does the right thing, you just need 'lxc.arch = i686' in the config file. [04:11] will push that to ppa tomorrow. gnight. [04:13] You do? [04:13] Pretty sure I'm not doing that [04:15] hallyn: ahha! [04:15] hallyn: thanks for digging === medberry is now known as med_out [05:34] I installed samba4 package and it's missing the samba-tool to create new users in the samba4 active directory. === koolhead11| is now known as koolhead11 [06:35] What's with the recent update to grub that requires you to say "Yes" to continue without installing grub in order to continue with the upgrade, when in fact, grub is installed and being upgraded? [06:35] (lucid) === larsemil1 is now known as larsemil [11:43] how can I change the default shell in ubuntu? [11:43] not chsh, but the default used by useradd [11:44] erm - found it - /etc/default/useradd [11:47] New bug: #801098 in kvm (main) "Unable to create VM when .iso file is stored on external USB drive" [Undecided,New] https://launchpad.net/bugs/801098 === NG_ is now known as ng_ [13:06] New bug: #801115 in tomcat6 (main) "package tomcat6 6.0.28-10ubuntu2 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/801115 [13:10] jamespage: I didn't do it this time since I didn't want to step on what you'd already done, but for bugs like 789810 where it's clearly a misconfigured system I tend to turn them into questions and follow-up that way. It gets them off the bug list, but the user still gets help. [13:14] ScottK: that sounds like a good idea - I tend to subscribe to the bug just in case the user needs any help but that does sounds like a better method. [13:15] jamespage: It might be something to discuss at a server team meeting as a team 'best practice'. [13:16] ScottK: good thought.. [13:16] Thanks. [13:17] I believe that a large fraction of server bugs are really support requests (even if the user filing them doesn't know this) [13:17] ScottK: The only issue could be, is the first response normally is "it looks like this, can you confirm?" <-- at that state, it could still be a bug. [13:18] Daviey: That's true for some things, but some like the postfix error 75 invalide hostname ones are clearly misconfigured systems. [13:18] ScottK: Yeah, that is true. Or issues that are never going to get fixed at the distro level. [13:18] Or never going to be able to provide enough information to have a useful bug report. [13:19] In the case of 789810 it's also clearly a misconfigured system and the error log even tells you exactly how to fix it. [13:20] ScottK: BTW, Ursula is going to be doing some analysis on our bugs to help see where the issues are. The idea of a 'best pratice' is a great one. [13:20] Getting SpamapS to do his work item on the postfix exit 75 bug pattern would be a help... [13:20] ScottK: what are your thoughts on things like bug 800651? [13:20] Launchpad bug 800651 in virt-manager "Failed to create new machine" [Undecided,New] https://launchpad.net/bugs/800651 [13:20] err, wrong one [13:21] bug 795500, rather [13:21] Launchpad bug 795500 in open-vm-tools "package open-vm-dkms 2011.03.28-387002-0ubuntu1.1 failed to install/upgrade: open-vm-tools kernel module failed to build" [Undecided,Incomplete] https://launchpad.net/bugs/795500 [13:23] Daviey: You're kinder than I am. I'd have marked that one invalid since it's not an Ubuntu kernel and so he's doing something unsupported. [13:23] Daviey: Who's Ursula? [13:24] ScottK: From the Launchpad QA team, joining ~ubuntu-server. [13:24] Oh. I didn't know. [13:25] (you all picked the one day a week I'm pretty reliably offline and in $work meetings for the server team meeting) [13:25] ScottK: it hasn't been announced yet. *oops*. [13:26] ;-) [13:26] ScottK: The meeting time and date is more of a problem than ever before, as those that have to be there are on a wider TZ split than previous cycles. [13:27] Yep. [13:27] And near the begining of the week IMO, is better - as there is still time to work on stuff before people forget over the weekend :) [13:27] Since I'm kind of in the middle of the split, time is less of an issue for me than the day. [13:28] Although, if it enables more people to attend - the idea of rotating days wouldn't hurt so much. [13:29] Perhaps Tue/Wed rotation. [13:29] I'm not sure what's best. [13:30] Rotation can (equally) confuse people and cause them to miss the meeting. [13:30] true. [13:30] I'm not pushing to have it changed for my sake. I know whatever day/time is picked will be bad for someone. [13:31] ScottK: Well trying to be as accomodating to everyone. [13:31] Sure. I appreciate that. [13:42] New bug: #801123 in openssh (main) "sshd init script does not use the option "--pidfile"" [Undecided,New] https://launchpad.net/bugs/801123 [13:46] morning [13:49] jdstrand: i am going to put up the new libvirt for you so you can have a look at this libvirt problem that im having (newer libvirt) if that is ok [13:50] how could i get the cpu utilisation of modules please ? (kernel space, not user land/procss) [13:52] hey zul [14:02] th0mz, build a profiling kernel? [14:03] patdk-wk: like with systemtap ? [14:03] somebody point me this link [14:03] dunno :) [14:04] it's been like a decade since I last cared to do it [14:04] erf [14:04] oki ;) [14:08] zul: joy [14:08] jdstrand: exactly [14:10] zul: where is it [14:12] jdstrand: ~chucks/libvirt on chinstrap [15:38] hello everyone [15:38] anyone can help me ? [15:38] Ask a question, potentially get an anwer. [15:38] answer even. [15:39] i already have install ids [15:39] how can i alert that ids alert file automatic.. [15:39] means.. [15:40] if i use ehco command.. [15:40] where need i put that file to autorun everytime alerts detect [15:48] Daviey: hey [15:48] Daviey: can you push http://people.canonical.com/~serge/lxc_0.7.4-0ubuntu11/lxc_0.7.4-0ubuntu11.dsc for me? [15:48] In return, I promise, within the next month, to apply for upload rights to the server set :) [15:58] Daviey, hallyn: uploaded [15:58] stgraber: thanks! [15:58] np, and yes, apply for PPU :) [15:58] stgraber: haha, now i dn't have to apply for upload rights? :-) [15:58] zounds! [15:59] hm, no zul [15:59] is he out looting or something? [15:59] :) [16:00] you just had to mention his nick for him to appear [16:00] lol [16:00] irc is not race-free it seems [16:01] zul: when I get the tempalte stuff straightened out, i'm gonna have to look into poweroff with lxcguest not always working. I think we're overmounting under /var again. (you needed that for libvirt or something) [16:01] zul: so i'm just sayin', if you have a few minutse where you're bored and want to look, feel free. Else I'll get to it. Should probably open a bug [16:01] hallyn: i didnt think i have a problem with but i havent tested recently [16:02] zul: it was sporadically happening for meyesterday, but it's possible the laptop was just objecting to stress [16:02] hallyn: ah ok [16:02] SpamapS: ok, I think I"m going to add '-t [debootstrap|uec]' to the lxc-ubuntu template and pull in your lxc-natty-uec stuff [16:03] so long as you're nto doing any special post-processing, this should be fine [16:03] hallyn: may I ask what you are referring to? (the stress thing) [16:03] hallyn: I was wondering, can't we just ship with a generic lxc-ubuntu template script and have symlinks for the various versions? then we can just check $0 to know what $dist to use [16:03] jdstrand: once in awhile yesterday, when I did 'poweroff' in a container, it wouldn't shut off, i'd have to lxc-stop it. [16:04] I see [16:04] stgraber: that's waht i have now :) check lxc-users mailing list, or ppa:serge-hallyn/lxc-natty, [16:04] hallyn: nice! [16:04] jdstrand: i didn't have time to watch the pattern, was creating containers like mad, [16:04] hallyn: I hit that infuriating guest instability bug yesterday [16:04] hallyn: (kvm) [16:04] jdstrand: oh, in kvm? [16:04] man [16:04] hallyn: and thought you started hitting it [16:05] wish we could get to the bottom of that [16:05] yes [16:05] no [16:05] i've never had it [16:05] anytime I go on wireless, boom. though it has happened other times, that is a very easy trigger here === med_out is now known as med [16:06] stgraber: code is at https://code.launchpad.net/~serge-hallyn/ubuntu/oneiric/lxc/update-templates fwiw === med is now known as medberry [16:06] New bug: #801187 in openssh (main) "should use GTK3" [Low,Confirmed] https://launchpad.net/bugs/801187 [16:07] jdstrand: remind me, you've reproduced this with qemu.git ? [16:07] If so, I"d say we should pull in aliguori and Avi [16:09] out for a bit, bbl [16:19] hallyn: not qemu.git. I am currently still on natty [16:39] guys, is there any sort of centralized mgmt tool for ubuntu-server akin to Red Hat's spacewalk? (spacewalk.redhat.com) [16:41] jbwiv: have a look at http://landscape.canonical.com [16:41] zul: fyi, I can reproduce the libvirt issue. what is happening is the dynamic profile isn't being created [16:41] zul: I don't know why yet [16:42] jdstrand: cool [16:42] jdstrand: good im not going crazy then [16:42] well, that isn't the word I would use [16:42] jdstrand: heh [16:45] zul: was hoping for something open source ;-) any idea what pricing landscape has? [16:45] jbwiv: no check the website [16:46] zul, yes, did. no pricing listed. I'll contact them. thx [16:48] quick question about TFTPD if no one is busy? [16:48] altice, don't ask to ask ;) [16:49] haha, sry, just being polite [16:49] anyways [16:49] so i loaded up xinet and TFTPD and the client, and can turn up xinet services (I have made the tftp file in the xinet dir to turn on tftpd server), however.......I cannot get it to actually turn up or work? [16:49] followed a tutorial online [16:52] also, if I do a "netstat -anp" I can see that socket 69 has been opened.....so, not quite sure where that leaves me to T-shoot the prob [16:53] anyone have any ideas? or feel like pointing me to some reading material that an obvious google search has not given me? [16:54] @jbwiv ; looks like everyone is busy hmm? [16:55] altice, busy, or not sure of the answer ;) [16:56] nahhh, I'm sure someone knows. I'm sure it's something silly I'm doing like in the tftp server config file or something [16:57] oh well, I'll try another room :) someone msg me if you have any thoughts on it [16:57] thanks all [17:05] Just upgraded from 10.10 to 11.04 and now the system halts during boot (during or after Stopping System V runlevel compatibility) [17:05] I never get an prompt on that line [17:06] Starting grub-loader, booting to rescue-mode and choosing resume normal boot works though [17:09] hi! im looking for some help installing a usb nic to server 11.04 [17:09] open computer, shove into slot, turn computer on, use nic [17:09] oh usb :) [17:10] lol :3 [17:10] shove in slot, hope linux has drivers for it :) [17:10] what model? [17:10] lemme check... [17:11] its a cisco linksys. not totally sure on the exact model? [17:11] exact model helps [17:12] i'll search it [17:12] it doesn't say on it? [17:14] I have a very old one at home [17:15] usb300m. not much more than that :\ [17:16] that the new/current one [17:16] pretty sure [17:17] i think it saw it when it asked to which if to config during intsall, but i dont know how to access it from there [17:17] it should just work [17:18] since its usb, will it not show up in ifconfig? [17:18] only active ones show up I think by default [17:18] what does lsusb show? === ng_ is now known as NG_ [17:20] the usual linux ports, then... [17:21] port005: uniclass tech [17:21] grep eth /var/log/messages [17:22] no such directory [17:22] heh? [17:24] there is no log directory in var [17:24] logs? [17:25] grep eth /var/logs/messages [17:25] na it's log [17:25] what is wrong with your system? [17:25] i dont know D: [17:25] its a fresh install of 11.04 so i havent done anything to it yet [17:26] there were more entries on lsusb, should i post them too? [17:26] lsusb | pastebinit [17:26] will do it [17:26] assuming you have pastebinit installed [17:27] im not on the serv atm, i have to roll back and forth from it :x [17:28] thet other entires were from asix electronics and genesys logic. 2.0 usb 4 port hub [17:31] oh, that asix is the one i think [17:37] zul: I think I found the commit that broke stuff [17:37] jdstrand: oh? url [17:39] zul: http://libvirt.org/git/?p=libvirt.git;a=commitdiff;h=12317957ecd6c37a2fb16275dcdeeacfe25c517e#patch9 [17:39] zul: problem is, it is a redesign issue [17:39] zul: so I need to dig in more [17:39] (ie, it isn't something we could just revert) [17:39] jdstrand: cool thanks [18:54] zul: hey! I just noticed bug 801259 [18:54] Launchpad bug 801259 in python-lockfile "[MIR] python-lockfile" [Undecided,New] https://launchpad.net/bugs/801259 [18:55] zul: a bit weird as the package is already in main :) [18:55] stgraber: yep [18:55] zul: I had it promoted for LTSP: bug 686045 [18:55] Launchpad bug 686045 in python-lockfile "[MIR] python-lockfile" [Undecided,Fix released] https://launchpad.net/bugs/686045 [18:55] doh! [18:55] so apparently you did all the paperwork for nothing ;) [19:10] Hey guys. Does anyone know if there is a way to specify interface order in /etc/network/interfaces for which devices will be brought up first and brought down first? I'm asking because I am putting a bridge on a vlan on a bond and therefor the bridge needs to come up first and the vlans needs to come up second and then the two bridges need to be brought up [19:11] jetole: have you tried with the standard config? [19:12] if that doesn't work, I guess not marking them as "auto" might be the thing, and then upping them manually in a separate script later [19:13] jetole: from "man interfaces": "ifup brings the named interfaces up in the order listed" [19:13] http://manpages.ubuntu.com/manpages/hardy/man5/interfaces.5.html [19:14] jetole: have you looked at ifenslave ? [19:14] jetole: its specific purpose is to handle complicated bonding configurations. === Guest57215 is now known as CrazyGir [19:18] RoyK, warzauwynn and SpamapS: thanks. [19:18] warzauwynn: I didn't realize it brought up the interfaces in the order listed so that is great to know [19:19] SpamapS: yes I have to use ifenslave for bonding but I'm not sure how that would related to which order the interfaces are brought up or it would have anything to do with bridging and vlans [19:19] jetole: me neither, but that's the beauty of google. http://www.google.com/search?q=ubuntu+interfaces+order [19:22] jetole: hrm, you're right, its not useful for bridges.. I just lumped all those wwords into 'bonding' in my head.. [19:23] jetole: from 'man interfaces' "ifup brings the named inter‐ faces up in the order listed. [19:23] hahahaha [19:23] * SpamapS reads backwards [19:23] warzauwynn: well done :-P [19:23] :D [19:26] lynxman: Regarding -stomp, were you looking at removing the hard coded paths? [19:26] (and using the unversioned symlink?) [19:26] Daviey: yeah, apart from trying to make it build properly [19:26] Daviey: the unversioned symlink is by far not a problem though :) [19:27] lynxman: Are you happy to carry on, or wanna push a branch with what you have? [19:27] Daviey: I think we got it almost hammered now [19:27] SpamapS: yeah thats what warzauwynn mentioned but I didn't realize. I have read that man page a handful of times but not any time recently. I didn't remember ever seeing that and didn't think to read it again which was a mistake on my part so sorry guys [19:28] lynxman: rocking [19:28] Daviey: now the packages are there, but erlang client fails :) [19:28] lynxman: erland is overated. :) [19:30] Daviey: deffo [19:31] erlanG even more so [19:31] Daviey: yeah I don't know who that Erland guy is :D [19:31] :) [19:34] erlang is over rated [19:35] Hi [19:36] So I am trying to get natty to pxe, which I have done for all other versions without issue, but for some reason, when I go through the install from the pxe images, the system fails to boot. [19:37] I thought it might be my preseed or something has changed with preseed between maverick and natty, but I also did a manual install using the pxe images and the system is still failing to boot after manual install [19:37] Anyone know if I am overlooking something simple here? [19:37] Seems odd, since I have been able to do this without issue on older versions of ubuntu [19:37] I got the images from mirrors.kernel.org === Aison is now known as Aison_StinkSauer [19:43] Hmm, [19:43] dead channel? [19:52] is there a rackspace irc channel? [19:53] hm, there is [19:56] zul: ok, after discussing with upstream, I have a preliminary patch [19:57] zul: I'm going to have to test it and have them review it [19:57] jdstrand: ack [19:57] zul: so it is going to be a while [19:57] jdstrand: no worries [19:57] zul: they reorganized things in an incompatible way with the apparmor driver [19:57] hah === NG_ is now known as ng_ [20:17] New bug: #660428 in nova (universe) "nova-objectstore start errors after install" [Medium,Fix released] https://launchpad.net/bugs/660428 === JasonnAWAY is now known as Jasonn === ng_ is now known as NG_ === Aison is now known as Aison^ANGRY === RudyValencia- is now known as RudyValencia === NG_ is now known as ng_ [21:55] Anyone have any good resources online (Google hasn't turned up anything useful for me) for troubleshooting boot issues for ubuntu? I'm doing a standard server install with software raid1 (set up following the official ubuntu docs) and I am unable to boot after getting past the grub boot screen. It proceeds to a black screen and then after another 20 seconds or so the monitor enters standby mode (amber light) [21:55] Forgot to mention - 11.04 [21:56] and x64 (on an asus TS100-E6 [21:56] New bug: #801338 in amavisd-new (main) "package amavisd-new-postfix 1:2.6.5-0ubuntu2 failed to install/upgrade: subprocess installed post-installation script returned error exit status 1" [Undecided,New] https://launchpad.net/bugs/801338 === ng_ is now known as NG_ [22:01] I actually just attempted SSH connection to it and I am able to connect. So it does look to be just an issue with the vga output. [22:02] I'm just really surprised to discover that because I figured there wouldn't be any vga issues because I'm just at runlevel3 (no gui launched) [22:02] I guess I'll go google a bit on that. [22:02] Wow, IRC is useful. [22:03] Is anyone alive in here? [22:09] Anyone else had experience installing ubuntu on an asus ts100-e6 server system? [22:09] It's installed sucessfully, but then after getting past the grub boot screen I get no vga output. Black screen. I am able to SSH into the box. === NG_ is now known as ng_ [22:29] stgraber: this is getting comical. i'm about to have to send out a patch 5/2 [22:34] yay, finally got it all working again. [22:34] now on to cgroups-bin [22:40] is there an easy way to install a package and its dependencies in some directory? (so that it can be used in a chroot. It shouldn't be a complete install) [22:50] hallyn: hehe :) === ng_ is now known as NG_ [23:41] http://ftp-master.debian.org/new/mysql-5.5_5.5.13-1.html [23:41] So... close... :-D [23:43] -8 ? [23:47] RoAkSoAx: ping