[00:08] adam_g: did something happen to the openstack-ci? [00:09] zul, not that i know of. its probably just pegged from parallel builds [00:09] adam_g: gah [00:17] adam_g: just doing a local test for cinder === freeflying_away is now known as freeflying === freeflying is now known as freeflying_away === freeflying_away is now known as freeflying [02:56] Hola. I was curious if anyone might have an opinion on how to best solve the following problem. On a new server being provisioned via bash script there are some sysctl.conf changes made by a third party that `sysctl -p` dislikes because there are a couple unknown keys and this causes the bash script to exit with an error due to the -ex specification on `#!/bin/bash -ex` [02:56] So, I was thinking either wrapping the additional entries being made to sysctl.conf that are known keys and the sysctl -p in a separate bash script without -ex OR using sed or something to search and replace those unknown keys with a commented out version OR something else a kind soul might be able to recommend? [03:00] aha.. even better yet.. providing -e to sysctl will ignore errors about unknown keys.. roger! [04:46] styol: you could put "set +e" before the lines that are allowed to fail, to disable errexit, and then "set -e" afterwards to put it back [08:55] ewdwed [08:56] Hello guys [08:56] I got a little question for you [08:56] I noticed that the motd is not applyed on the first login on the machine : always on the second login time, is that normal ? [09:18] nobody has the answer ? === klaas_ is now known as klaas [09:56] Blargh, I'm on a UEFI motherboard. How do I set up a boot loader for Xubuntu 12? [09:57] I'm staring at a Grub command line off my pen drive that I barely got to load. [09:57] !uefi | w0rp [09:58] w0rp: UEFI is a specification that defines a software interface between an operating system and platform firmware, it is meant as a replacement for the BIOS. For information on how to set up and install Ubuntu and its derivatives on UEFI machines please read https://help.ubuntu.com/community/UEFI [10:05] I hate UEFI now because it cost me time. [10:19] <3 boot-repair-disk [10:27] zul, up yet? [10:35] zul, doing neutron rc1 [11:15] hi, how to install and share a printer without http interface? [11:15] jamespage: just woke up ill talk to you about babel after i drop liam off (need to have breakfast as well) === freeflying is now known as freeflying_away === Jare_ is now known as Jare [11:22] !cups | hXm [11:22] hXm: Printing in Ubuntu is done with cups. See https://help.ubuntu.com/community/Printers - https://wiki.ubuntu.com/HardwareSupportComponentsPrinters - http://linuxprinting.org - Printer sharing: https://help.ubuntu.com/community/NetworkPrintingWithUbuntu [11:25] cups requires http interface [11:30] zul, https://code.launchpad.net/~james-page/swift/fixup-cache-perms/+merge/189044 [11:31] when you start [12:31] jamespage: https://code.launchpad.net/~zulcss/cinder/babel-ftbfs/+merge/188963 [12:42] zul, did you upload the horizon fix from yesterday? if so please can you merge the branch [12:42] zul, +1 on that cinder fix [12:43] jamespage: not yet i was going to wait until horizon is out [12:43] zul, OK _ its is - I was doing that - I'll merge your branch if thats OK [12:44] jamespage: yep [12:47] i have a question, if my ubuntu server got compromised and i install a new version, is there a way to make sure the hacker didnt alter files i can bring over from the compromised machine to the new one? [12:48] g0tcha: probably check from a good known backup [12:48] very old backup === freeflying_away is now known as freeflying [12:50] zul: https://code.launchpad.net/~james-page/horizon/havana-rc1/+merge/189067 [12:51] jamespage: +1 [12:53] zul, ta - uploaded [12:53] zul, so thats keystone and horizon done right? [12:53] I'm looking at neutron [12:55] ceilometer, keystone, glance and horizon [12:57] jamespage: ill get nova [13:02] jamespage: do you want to do cinder when its out, ill get heat [13:05] zul, I think alot of the neutron test failures are due to patching of neutron.conf [13:06] jamespage: oh? [13:06] zul, yeah - we set a core plugin and I think that conflicts with some of the test cases [13:07] jamespage: *sigh* lovely [13:12] jamespage: or zul [13:12] jcastro, hell [13:13] o [13:13] blimey that was freudian [13:13] http://askubuntu.com/questions/353127/migration-from-generic-openstack-to-ubuntu-openstack [13:13] heh [13:13] jamespage: i was about to say [13:13] recommendations here? [13:13] jcastro, run away screaming? [13:13] seriously thats hard [13:13] it is [13:13] afaik we don't have any kind of recommendation for this [13:14] all I've ever heard is "you learned your lesson, now you get to start over." [13:15] jcastro, yeah - there are just 1000 things that could go wrong during a migration [13:16] * jamespage thinks about it a bit [13:16] what is 'generic openstack'? [13:19] zul, that swift bug I fixed was causing 503's on serverstack [13:19] not seen that before [13:19] jamespage: cool! i like testing things before moving them out to our users ;) [13:21] * jamespage sighs [13:21] another new neutron plugin - embrane [13:21] is it me or does everyone have a new SDN solution these days [13:25] jamespage: btw this will make you happy: python-ceilometer | 2013.2~b3-0ubuntu1 | saucy | all [13:25] w00t [13:29] zul: hola [13:29] koolhead17: hi [13:29] zul: long time. how have you been? [13:29] koolhead17: good, busy, you? [13:30] zul: same2same [13:33] jamespage: https://code.launchpad.net/~zulcss/heat/2013.2.rc1/+merge/189076 [13:33] zul, dude - you can edit the existing changelog entry - thats already uploaded [13:34] jamespage: erg? [13:34] look at the diff [13:34] damn it...gimme a sec [13:46] jamespage: https://code.launchpad.net/~zulcss/heat/rc1/+merge/189081 andhttps://code.launchpad.net/~zulcss/nova/2013.2.rc1/+merge/189082 [13:47] zul, ack on heat [13:47] zul, nack on nova [13:48] zul, just as a detail I've been describing these as 'New upstream release candidate." [13:49] jamespage: actually nova b3 ubuntu2 never got uploaded [13:50] zul, urhg [13:56] jamespage: nova fixed [13:57] zul, pushed? can't see it [13:57] jamespage: just did [13:57] jamespage: i didnt create a new changelog entry for rc1 because there is no ubuntu2 for nova [13:58] zul, fine but you still have a typo [13:58] jamespage: wait there is [13:58] med_: hey, sorry - did you straighten the dhcp issue out? === psivaa is now known as psivaa-afk [14:04] jamespage: ok *now* its fixed [14:05] zul, I still don't see it here - https://code.launchpad.net/~zulcss/nova/2013.2.rc1/+merge/189082 [14:07] hi guys/gals.. was there an update to the repos that removed/renamed older packages? [14:07] E: Unable to locate package libxml2dev [14:07] jamespage: pushed to an rc1 branch i had its there now [14:08] zul, the un-uploaded changelog entry is back [14:08] jamespage: yeah it got stuck in proposed [14:08] OK [14:09] zul, +1 [14:09] jamespage: thanks [14:09] lol [14:09] im retarded [14:10] had -'s in my puppet manifest then i replaced, which took away the -'s from ssl-cert & other packages [14:10] thanks anyways === psivaa-afk is now known as psivaa [14:12] hallyn, yep, I needed to flush my precise [14:12] new containers were dhcping fine [14:14] jamespage: alright we are just waiting for cinder and neutron correct? [14:14] zul, glance done? [14:15] yes - I see it is [14:18] med_: cool [14:18] yep [14:18] I'd never done that before. [14:19] Greetings. [14:20] Can anyone offer any pointers to info as to why iptables -A INPUT -m conntrack --ctstate ESTABLISHED,RELATED -j ACCEPT is better/worse than iptables -A INPUT -m state --state ESTABLISHED,RELATED -j ACCEPT [14:20] zul, https://code.launchpad.net/~james-page/neutron/havana-rc1/+merge/189092 [14:20] I'll look at the tests again but for now.... [14:21] jamespage: +1 [14:30] How does havana end up in saucy if they release on the same date? Very tight coupling and upstream cohesion? [14:30] med_, well the rc1's are going in now [14:30] nodz. [14:30] albeit fixes that should be release [14:30] it will get zero-day sru'ed [14:31] which reminds me I need to line up an SRU team person todo that [14:31] jamespage, I'm mostly asking in case I need to answer anyone else. [14:31] med_, the answer is probably on release day +1 [14:31] ack. [14:35] zul, OK - started on tidying the CA staging area [14:35] python-wsme uploaded from saucy - should be good now [14:36] jamespage: ack [14:48] zul, http://people.canonical.com/~jamespage/ca/havana/ [14:48] for lesscpy [14:49] +1 [14:51] zul, actually ply is not required [14:51] its enough in precise already [14:51] jamespage: sweet === freeflying is now known as freeflying_away [14:58] jdstrand: sadly, commit fbe2e26c15af35e4d157874dc80f6a19eebaa83b in qemu breaks a bunch of the qa-regression-tests test-qemu.py checks :( (i'll push a fix... just a pain) [14:58] (it changes a bunch of 'file=whatever' to remove the 'file=') [15:54] hallyn: yuck. that is the type of stuff I don't understand why people change. it was fine before... [15:55] hallyn: I guess you'll be making that shange to be saucy+ specific? [15:55] jdstrand: "HMP is intended for humans, so we'll make sure no non-humans can parse it" is the commit message's position [15:55] yes [15:55] jdstrand: well i'm comparing qemu version to 1.6 [15:55] sounds good [15:56] hallyn: is there an interface for non-humans to parse? [15:56] jdstrand: there is. i wasn't sure if we should use that, or if we actually wanted the HMP to get verified [15:56] my dog would love a simplified interface, for example [15:56] one woof for 32-bit, two for 64-bit [15:56] hallyn: I'll let you make that call. I haven't updated/used that script in quite a while [15:57] jdstrand: for now i'm just gonna make it work as is, but the duct tape is coming loose so pretty soon i'll make a patch proposal to change it i think [15:57] (i wont' just push that without discussion -somewhere :) [15:57] hallyn: in general, if there is a mchine-readable interface, the tests could all be made to use it. then have a couple of tests for HMP [15:57] sounds good [15:58] but yeah, doesn't have to be done now [15:58] jdstrand: thanks, ttyl [15:58] np [16:06] OK, I'm going nuts trying to figure this out on my own. Thus I reach out to the community. I have a Bugzilla 4.4 installation on my Ubuntu 12.04 server [16:08] OK. [16:08] And I have Bugzilla configured to send emails whenever a bug is submitted or changed. But it won't do it and I believe the problem is with postfix. My mail.log shows connection time out. [16:08] I have port 25 open [16:09] So what gives? [16:11] Can you send mail without having bugzilla initiate it? [16:13] I've been trying to figure out how to do that with the mail command, but haven't figured it out? do you have a quick few lines I can try? [16:14] mail someone@somewhere.com [16:14] press ctrl-d to finish [16:15] that was easy :) [16:15] ok, so I got "send-mail: fatal: chdir /Library/Server/Mail/Data/spool: No such file or directory" [16:16] I notice that it says send-mail there. isn't that different than postfix? [16:16] seriously setting up email correctly has been the most challenging thing for me. [16:17] jarkinox: How did you install postfix? [16:18] What's send-mail? [16:18] What's /Library/Server/Mail/...? [16:18] Neither of these are standard components in Ubuntu. [16:18] during the server installation and also with apt-get [16:18] shit [16:18] jarkinox: what's the output of the command "which sendmail"? [16:19] hold on, sorry guys. I ran that command on my mac terminal [16:19] switching to ubuntu [16:19] too many windows open [16:20] ok, so I sent email through ubuntu and it seems to have worked ok [16:20] but haven't received the email yet [16:20] checking mail.log [16:21] Oct 3 09:20:16 Ubuntu postfix/smtp[32192]: connect to XXXXXXXXXXXXXXXXX:25: Connection timed out [16:22] rbasak: /usr/sbin/sendmail [16:22] So it's not a postfix problem then. Find out why you can't connect to that server on port 25. [16:22] it's a gmail server [16:22] It may be that your ISP blocks connections. That's not uncommon. In that case, your ISP probably provides a smarthost, and you'll need to configure postfix with that. [16:22] how do I figure that out? [16:24] Or you could configure the server to send everything through your gmail account. That might be easier for you. [16:24] http://askubuntu.com/q/228938/7808 can help you with that. [16:24] awesome [16:25] thank you rbasak [16:25] i think that would be easiest [16:26] man this is overwhelming [16:26] ugh === matsubara is now known as matsubara-lunch [17:55] hi.. I was wondering if anyone can point to some resource for silently installing ldap client on Ubuntu 12.04 [17:55] I'm looking for a script [17:59] mallu, you only need to do apt-get install ldap-utils [17:59] well... that is prompting for ldap server information [18:00] that's just installing stuff [18:01] there is no question asked during this install. If you mean 'silently configure the ldap client so that it's pointing to an ldap server", that would be specific [18:02] and it depends on what you want to do [18:03] mallu: hopefully helpful: http://manpages.ubuntu.com/manpages/lucid/man7/debconf.7.html [18:03] there is no debconf question asked [18:04] since -motu and -bugs are all failing to respond to my question, anyone here know the SRU process and can tell me what happens if a package in precise-proposed gets verification-failed on the SRU bug? [18:06] TheLordOfTime: I'm pretty sure that package dies. No progress happens until someone either re-tests and confirms that it -does- fix the problem, or upload a new version .. [18:07] sarnold: yeah, that's what I thought (and at the same time as you answered, hggdh answered in -bugs). [18:07] sarnold: my main question is what happens to the package in -proposed, does it just get removed or what? [18:08] TheLordOfTime: hrm, I doubt it'd get removed right away, that'd be too easy for some jerk to waste hours of buildd time... [18:08] usually, it will be demoted after a while by "itself". If it is a critical failure, the archive managers should be pinged [18:09] i see. === matsubara-lunch is now known as matsubara [19:03] hallyn: ping [19:05] zul: uh, nobody here now but us chickens. bok bok. please call back later. bok bok [19:05] hallyn: riiight...do you know why policykit got added as a dep for libvirt? [19:06] zul: hm. no. there *was* a policykit-related bruhaha recently, [19:06] hallyn: ok cool [19:06] zul: either it was part of the licensing problem, or it there was a patch in libvirt-security about it. [19:06] lemme check git log [19:07] hallyn: yeah saw it mdeslaur bumped it [19:07] * zul shakes his fist at mdeslaur [19:07] heh. that won't work, i'm looking at qemu log [19:07] 'bumped' meaning refused it? [19:07] yeah, i don't know, sorry [19:08] zul: commit ecbb3d51b5f195a73377840d35072b7014da2aa0 claims to fix building without policykit [19:08] (sep 27 2013) [19:08] i'd frankly prefer to build without it, but... [19:09] hallyn: ditto makes backporting to the CA a bit more interesting [19:10] all right i need a change of venue - biab [19:25] hallyn: i cant find that commid id [19:28] zul: you can't find ecbb3d51b5f195a73377840d35072b7014da2aa0 in the libvirt src tree? [19:28] hallyn: no found it [19:28] http://libvirt.org/git/?p=libvirt.git;a=commit;h=ecbb3d51b5f195a73377840d35072b7014da2aa0 [19:31] is that in the package? [19:32] zul: and when you said 'it became a dependency' what did you mean? it won't compile without it, or debian added it as a build-dep? [19:32] hallyn: libsystemd-login-dev is a dependency of policykit-1 and its not in precise (and i dont want to go down that rabbithole ;)) [19:37] zul, policykit backports from raring and folsom built fine in precise folsom/grizzily -staging [19:38] libsystemd-login-dev? [19:43] adam_g: libsystemd-login-dev isnt a dependency for policykit-1 for raring [19:43] zul, oh, you're having trouble with the saucy backport? [19:43] yeah [19:43] zul, why did we backport libvirt for H? [19:43] adam_g: i really dont want to backport systemd [19:44] adam_g: because people wanted a newer libvirt [19:44] zul, for what features, though? [19:44] adam_g: and there is an open cve for 1.1.1 right now [19:44] in the CA [19:44] Hey, do any of you guys know if there is an Ubuntu Server live image, or can point me to something on building a Server live image? [19:45] zul, what were the feature requirements that drove the bump in libvirt, though? IIRC, we did it last cycle for ceph rbd support [19:45] i believe thats right [19:46] zul: so adding that patch to saucy's libvirt, and dropping the polkit dep; does that work? [19:46] hallyn: working on it [19:46] (we can still do the same thing for saucy) [19:46] cool [19:46] why do we have a policykit-1 dep anywya? [19:51] i'm quite certain i didn't ad it [19:51] zul: [19:51] [ Guido Günther ] [19:51] * [05e9a39] build-depend on policykit so polkit auth works with virsh [19:51] as well [19:51] from 0.6 changelog [19:52] well screw that ;) [19:52] zul: it's in precise too [19:54] hallyn: building a testdeb [19:55] need to go do dad duty [19:55] o [19:55] o/ [19:55] that's me with two heads [20:31] hallyn: buids fine without polkit === matanya_ is now known as matanya [21:21] zul, nova-novncproxy is borked [21:21] its blocking the autopkg tests [21:21] File "/usr/lib/python2.7/dist-packages/websockify/websocketproxy.py", line 86, in __init__ [21:21] websocket.WebSocketServer.__init__(self, *args, **kwargs) [21:21] TypeError: __init__() got an unexpected keyword argument 'no_parent' [21:24] zul, you might want to cherry pick reverse of https://github.com/openstack/nova/commit/3eb67b811ae2442bd86781d9f1c4078a982cfe84 [21:29] zul: well i guess we shoudl at least run wqa-regression-testing against it before pushing to saucy [21:30] jamespage: ack ill look at it tonight [21:30] hallyn: ill put it in a ppa tonight [21:30] zul, marverllous [21:30] * jamespage goes back to fixing neutron [21:30] jamespage: have to take liam to scouts/beavers tonight [21:31] zul, funny - I did that this evening as well! [21:31] jamespage: heh [21:46] all right the only qemu test failure that still worries me is test_nic failing to connect over ssh. The rest I'm gonna fix in the testsutie tonight. THis one, gotta figure out what's going on. === matsubara is now known as matsubara-afk [22:09] hm ,looks like i82551 is the problem