[00:09] <hallyn> smb: by chance do you have the fix for bug 1541810 in your xenial libvirt merge?
[00:10]  * hallyn d/ls the trees to look at debdiff in any case
[00:33] <hallyn> smb: cgmanager-dev should not be needed for libvirt in xenial
[09:22] <smb> hallyn, the service should now read the right file I think, cgmanager could try to drop. But it would help to get this via email
[09:23] <lordievader> Good morning.
[09:30] <phpcoder> hello
[09:30] <phpcoder> https://gist.github.com/anonymous/3275dc5e67b64c8be18b
[09:30] <phpcoder> i cannot remove mysql-server-5.6
[09:30] <phpcoder> how can i remove it brutally?
[09:32] <lordievader> Can you reinstall the package to see if it creates the script? (apt-get install --reinstall)
[09:39] <phpcoder> lordievader, nothing
[09:39] <phpcoder> same thing...it can not creates the scrupt
[09:40] <phpcoder> lordievader, should i remove it brutally_
[09:40] <lordievader> No, that is a bad idea.
[09:41] <halvors> Hi. The IPv6 connectivity of my server is now working propatly.
[09:41] <phpcoder> lordievader, eh but i cannot install it
[09:41] <phpcoder> or reinstall
[09:42] <phpcoder> lordievader, how can i do that
[09:42] <lordievader> What version of Ubuntu are you running?
[09:42] <halvors> The roouting table looks just fine.
[09:42] <phpcoder> lordievader, 15.10
[09:42] <lordievader> phpcoder: Could you comment on bug 1455818 and mention you found it on 15.10 too? (Besides the bug report contains a workaround)
[09:45] <phpcoder> lordievader, i am not registered there
[09:46] <lordievader> Do you mind making an account?
[09:48] <phpcoder> lordievader,
[09:49] <phpcoder> i see another user worte: "I just ran into this issue on my new install of Ubuntu Server 15.10."
[09:49] <lordievader> I know.
[09:49] <lordievader> But still, please do comment that it affects you too.
[10:09] <purefan> Hello! I read http://www.cyberciti.biz/faq/linux-cve-2016-0728-0-day-local-privilege-escalation-vulnerability-fix/ and decided to upgrade the kernel of my server from 3.13-0-68-generic to something newer, but I dont want to do a dist-upgrade, just update the kernel. I found http://kernel.ubuntu.com/~kernel-ppa/mainline/ but dont know which one to choose...
[10:09] <phpcoder> lordievader, i am doing it
[10:15] <purefan> Ohh... and the ppa for trusty isn't here http://ppa.launchpad.net/kernel-ppa/ppa/ubuntu/dists/
[10:39] <dholbach> hiya
[10:39] <dholbach> can you guys please help out a bit with sponsoring?
[10:39] <dholbach> http://reqorts.qa.ubuntu.com/reports/sponsoring/
[10:39] <dholbach> we have a number of server/cloud related requests
[10:43] <dholbach> roaksoax, ^ do you think you can please bring this up with somebody who can do a broader call for help?
[10:44] <dholbach> (or anyone else really...)
[10:45] <dholbach> Beret maybe too: ^
[14:06] <Beret> dholbach, yeah, I'll help
[14:06] <dholbach> thanks a lot
[14:06] <dholbach> :-)
[14:12] <Spyware> Hello, I want to use zabbix and I hesitate between ubuntu and debian server, which distro you advise me? Thank you
[14:21] <coreycb> jamespage, ddellav:  ceilometer has a new dep on gnocchiclient but I'm going to leave it in universe and move it to Suggests.  I looked at MIRing gnocchi and gnocchiclient, but gnocchi in particular at this time is a rabbit hole of MIRs.
[15:46] <gcmacmalloc>  I have the following issue with the cloudimg 14.03 on openstack: ALERT!  /dev/disk/by-label/cloudimg-rootfs does not exist.  Dropping to a shell!
[15:46] <gcmacmalloc> anyone can confirm or infirm this ?
[16:44] <rizzuh> I installed 14.04 in OEM mode. After installing and configuring, there is no oem-config program to run. What do I do next to reset the installation?
[17:36] <SpamapS> rbasak, jamespage, utlemming, smoser: My old friends, I have a question for you about this MIR: https://bugs.launchpad.net/ubuntu/+source/libiscsi/+bug/1271653
[17:36] <SpamapS> would the server team be willing to triage for libiscsi? I'd like to enable libiscsi support in qemu so we can run nova in an unprivileged container.
[17:56] <jamespage> SpamapS, so that does userspace based iscsi target access?
[17:57] <SpamapS> jamespage: correct!
[17:58] <SpamapS> jamespage: I also wonder if it might be useful for something like LXD or Docker to support that so it could inject a device node into the container without that container needing privileged access.
[17:59] <SpamapS> jamespage: qemu already has support for libiscsi, but it has to be turned off in Ubuntu.
[18:01] <jamespage> SpamapS, might have context there as well - are you working on getting bits of openstack running in containers?
[18:01]  * jamespage read that into "I'd like to enable libiscsi support in qemu so we can run nova in an unprivileged container"
[18:02] <jamespage> SpamapS, I'd be supportive of having libiscsi in main for qemu enablement
[18:02] <SpamapS> jamespage: yes. :)
[18:03] <jamespage> SpamapS, we should compare notes
[18:04] <jamespage> :)
[18:04] <SpamapS> jamespage: I'm actually proxying Jessie Keating from BlueBox, who you may see, commented on that MIR.
[18:05] <jamespage> SpamapS, I saw
[18:05] <sarnold> is that my old buddy who worked at rhat for a while?
[18:05] <SpamapS> sarnold: indeed
[18:06] <SpamapS> we're both over at IBM now. :)
[18:06] <sarnold> SpamapS: cool :)
[18:09] <jamespage> hah - this has been in since kilo
[18:11] <jamespage> hallyn, zul, rbasak: ^^ I think this would make a nice additional generally for openstack - make sense to you guys as well?
[18:12] <hallyn> ?
[18:12] <jamespage> hallyn, libiscsi enablement in qemu - needs MIR for libiscsi
[18:13] <zul> jamespage: +1
[18:13] <jamespage> hallyn, mterry has +1'ed but it still needs security team review
[18:13] <hallyn> +1 fromme,
[18:13] <hallyn> of course
[18:13] <hallyn> hopefully it's gone up not downhill :)
[18:14] <jamespage> apparently so
[18:16] <jamespage> I feel its going to be pretty light to added -server and -openstack teams for triage...
[18:16] <jamespage> SpamapS, just needs  a sec team +1 now
[18:17] <SpamapS> jamespage: thanks. I had poked them earlier and they suggested they're really busy right now, but I'd ask you guys to weigh in first so they can prioritize appropriately.
[18:17] <sarnold> it's on our short list :)
[18:17] <jamespage> \o/
[18:17] <jamespage> sarnold, hopefully with neutron-vpnaas ?
[18:17]  * jamespage crosses fingers :)
[18:17] <SpamapS> jamespage: you are the best, thanks. :)
[18:18] <SpamapS> is neutron-vpnaas still a forked l3 agent?
[18:18] <jamespage> kinda
[18:18] <jamespage> forked is harsh - it depends on the core l3-agent code and adds stuff I think
[18:20] <sarnold> jamespage: interesting, I hadn't looked into it before.. yeah I can see why that'd be useful. it's not currently on our shortlist but we'll try to get as many done before relesae as we can
[18:35] <SpamapS> jamespage: btw, I opened this bug for tracking the qemu change https://bugs.launchpad.net/ubuntu/+source/qemu/+bug/1544216
[18:51] <kulyzu> is there a way to alias a command, like u[0-9]+ will be anything the user type like u1 u2 u3 ...u333
[18:52] <hallyn> SpamapS: ok you sold me at unpriv containers
[18:56] <sarnold> I have trouble seeing how libiscsi would help lxd beyond being an image store
[18:56] <sarnold> actually 'mounting' any luns for container use would require the kernel's involvement, right?
[19:09] <SpamapS> hallyn: right? ;-)
[19:10] <hallyn> SpamapS: sarnold: that's where sforshee comes in handy i guess
[19:10] <SpamapS> sarnold: Is there no way to have a user space daemon serve as the backing for a block device?
[19:10] <hallyn> or just qemu-nbd
[19:14] <sarnold> SpamapS: like a fuse but for block devices? dunno.. -can- fuse do that?
[19:15] <sarnold> SpamapS: the only other usrespace iscsi support im' aware of is stgt, and that looks like it's primarily an iscsi target, and perhaps long abandoned..
[19:17] <SpamapS> sarnold: yeah, no, targets aren't what this is about anyway, this is about clients.
[19:17] <sarnold> yeah
[19:18] <sarnold> i've only ever seen the initiators built into nics or the kernel -- or libiscsi with qemu :)
[19:20] <patdk-wk> or, ipxe :)
[19:20] <sarnold> oh :) nice
[20:37] <tmartins> hey guys... When using cloud-init with OpenStack Heat, is it possible to have 2 "user_data" sections at the same time?
[20:38] <tmartins> Like this: http://pastebin.com/7HiDpjWr ?
[20:55] <ikonia> tmartins: I don't think so
[20:55] <ikonia> (I don't KNOW though)
[20:57] <tmartins> :-/
[20:57] <tmartins> cloud-init is very limited...   :(
[20:58] <nacc> tmartins: i'd ask in #cloud-init?
[20:59] <tmartins> Mmm... Good one...   =P
[20:59] <ikonia> it is very limited
[20:59] <tmartins> yeah...   =/
[20:59] <ikonia> I think thats it's design though - simple tasks to configure an image
[21:00] <ikonia> it would end up being a workload to configure a server if it wasn't light and simple
[21:01] <tmartins> basically, initially, I'm creating a user, then, I'll run a script, that will call Ansible
[21:01] <tmartins> simpler -> complex
[22:12] <coreycb> beisner, can you promote 1.4.1-0ubuntu1.2~cloud1 to juno-updates?
[22:12] <coreycb> oslo.messaging
[22:12] <beisner> o/ howdy coreycb - sure, looking ...
[22:14] <coreycb> thanks beisner
[22:21] <coreycb> beisner, can you also promote neutron 1:2014.1.5-0ubuntu2~cloud0 and 1.3.0-0ubuntu1.4~cloud0 to icehouse-updates?
[22:21] <beisner> coreycb, hah, i was just reading serge's bug comment.   where's my neutron?
[22:21] <coreycb> beisner, ah yeah that
[22:22] <beisner> coreycb, ok oslo.messaging 1.4.1-0ubuntu1.2~cloud1 pushed from proposed to juno-updates in UCA
[22:22] <coreycb> beisner, thanks
[22:29] <beisner> coreycb, ok neutron 2014.1.5-0ubuntu2~cloud0  pushed from proposed to icehouse updates in UCA re: bug 1318721
[22:36] <beisner> coreycb, fyi, promoted oslo.messaging 1.3.0-0ubuntu1.4~cloud0  icehouse-updates in uca for same bug ^
[22:36] <coreycb> beisner, awesome, thanks!
[22:36] <beisner> coreycb, yw