16:01:15 <odyssey4me> #startmeeting openstack_ansible_meeting
16:01:15 <cloudnull> o/
16:01:15 <openstack> Meeting started Tue Oct  2 16:01:15 2018 UTC and is due to finish in 60 minutes.  The chair is odyssey4me. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:16 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:01:19 <openstack> The meeting name has been set to 'openstack_ansible_meeting'
16:01:19 <spatel> sure!! i am hungry :)
16:01:19 <spotz> You're late by 1 whole minute odyssey4me!
16:01:28 <d34dh0r53> o/
16:01:30 <odyssey4me> spotz: :p
16:01:34 <spotz> \o/
16:01:36 <spotz> :)
16:01:45 <odyssey4me> #topic rollcall
16:02:07 <hwoarang> o/
16:02:22 <odyssey4me> I'll give it 5 mins to give time for people to switch gears. :)
16:02:28 <prometheanfire> o/
16:02:42 * cloudnull gears switched
16:04:18 <nicolasbock> o/
16:05:30 <prometheanfire> nicolasbock: signed your keys/ids
16:06:06 <odyssey4me> alright...
16:06:14 <odyssey4me> #topic what happened since last meeting
16:06:29 <odyssey4me> mchandras: Proposed distro job for ceph deployments and needs votes to get it in. Also restored lxc and ceph deployments for SUSE. Also needs votes :)
16:06:41 <openstackgerrit> Merged openstack/openstack-ansible-os_keystone master: Update deprecated option  https://review.openstack.org/607084
16:06:49 <odyssey4me> mnaser: please fill this etherpad https://etherpad.openstack.org/p/osa-stein-ptg
16:07:00 <odyssey4me> evrardjp: Making Ocata "extended maintenance" (This would add ocata-em tag, not closing the branch). See also: https://governance.openstack.org/tc/resolutions/20180301-stable-branch-eol.html . No real value for us, what do we point to for upstream repos (IMO, the ocata-em upstream tags when available, else the stable branch it is attached to).
16:07:26 <odyssey4me> does anyone have any particular highlights to raise?
16:08:04 <cloudnull> none from me
16:08:10 * cloudnull has no ocata clouds
16:08:41 <odyssey4me> I think those items may be a bit dated, because we discussed them in last week's meetings.
16:08:44 <odyssey4me> But meh.
16:09:35 <odyssey4me> alright, let's move on then
16:09:38 <odyssey4me> #topic bug triage
16:09:51 <odyssey4me> it's a reasonably short list
16:09:53 <odyssey4me> #link https://bugs.launchpad.net/openstack-ansible/+bugs?search=Search&field.status=New
16:10:08 <odyssey4me> #link https://bugs.launchpad.net/openstack-ansible/+bug/1782388
16:10:08 <openstack> Launchpad bug 1782388 in openstack-ansible "Installing Multipath But Not Enabling In Nova Causes Volume Attachment Failures" [Undecided,New]
16:10:19 <odyssey4me> bgmccollum: you around?
16:10:57 <odyssey4me> interesting note at the bottom of that bug
16:11:08 <odyssey4me> looks like that might be something actionable if anyone wants to pick it up?
16:11:20 <odyssey4me> confirmed/medium ?
16:11:40 <hwoarang> sounds good to me
16:11:56 <cloudnull> ++
16:12:04 <cloudnull> I'd say high
16:12:13 <cloudnull> basically cinder + nova is broken atm
16:12:29 * cloudnull if im reading that bug correctly
16:12:34 <odyssey4me> fair point - we should perhaps instrument setting things up the right way using some sort of driver model
16:12:51 <odyssey4me> but perhaps also detect whether multipath is implemented, and set the right things
16:13:13 <odyssey4me> anyone got some time to work that out?
16:13:46 <cloudnull> In my lab I use NFS so I haven't seen this issue but i suspect this is an LVM specific issue
16:13:59 <cloudnull> probably not something we'd see in ceph
16:14:04 <odyssey4me> yeah, very likely
16:14:18 <odyssey4me> ok, moving on
16:14:23 <cloudnull> so maybe those options need to be used when the volume backend is LVM
16:14:26 <odyssey4me> https://bugs.launchpad.net/openstack-ansible/+bug/1794337
16:14:27 <openstack> Launchpad bug 1794337 in openstack-ansible "Apache and Horizon configs not automatically turning off SSL when overridden" [Undecided,New]
16:14:29 <cloudnull> bgmccollum it'd be great to confirm
16:15:36 <spotz> Oh that one was mine;)
16:16:04 <spotz> So the hack works, just not sure it's the actual fix we want to implement
16:16:12 <odyssey4me> cloudnull: yeah, I commented the suggested approach in the bug.
16:16:28 <cloudnull> ++
16:16:42 <odyssey4me> spotz: thanks for noting it - I've seen several confused users trying not to use SSL and it doesn't work, so this is pretty important
16:16:55 <odyssey4me> I'd say confirmed, high.
16:18:06 <odyssey4me> Any objections?
16:18:17 <odyssey4me> And anyone available to pick that up?
16:18:44 <hwoarang> no objections from me
16:18:50 <spotz> I'll pick it up if someone wants to point me in the right direction:)
16:19:06 <spotz> Cause hacking by spotz may not be the best idea:)
16:19:23 <hwoarang> spotz: sounds good. we can go through reviews if you post something on gerrit
16:19:48 <odyssey4me> spotz: I expect there may just need to be a conditional check around some of the content.
16:20:22 <spotz> It'll give me something to look at at the horse show this weekend
16:20:30 <odyssey4me> thanks spotz
16:20:46 <odyssey4me> next up https://bugs.launchpad.net/openstack-ansible/+bug/1794525
16:20:46 <openstack> Launchpad bug 1794525 in openstack-ansible "Upgrade fails when having cinder_qos_specs defined" [Undecided,New]
16:21:12 <openstackgerrit> Nicolas Bock proposed openstack/openstack-ansible-os_octavia master: Added suse vars  https://review.openstack.org/607304
16:21:32 <hwoarang> oh that reminds of something
16:21:46 <odyssey4me> :/ it'd be nice to replace this with a module instead
16:21:47 <hwoarang> https://review.openstack.org/#/c/561218/
16:22:23 <odyssey4me> does the openstack cli support the qos functions?
16:22:50 <odyssey4me> the bug seems confirmed then
16:23:07 <hwoarang> possibly yeah
16:23:17 <odyssey4me> I'd say high priority then, given this is broken functionality.
16:23:48 <hwoarang> ok
16:24:25 <odyssey4me> I can pick this up after the meeting.
16:24:47 <odyssey4me> next up https://bugs.launchpad.net/openstack-ansible/+bug/1794613
16:24:47 <openstack> Launchpad bug 1794613 in openstack-ansible "Rocky setup hosts fails to recreate destroyed container" [Undecided,New]
16:25:06 <skiedude> Question about the internal/external LB IP that you define. If I'm not planning on having this publically accessible and its only internal, can the internal/external just be the same IP?
16:25:31 <noonedeadpunk> it seems he's missing lxc_hosts group on re-creation
16:26:03 <odyssey4me> skiedude: we're in a bug triage meeting - we'll be done in 30 mins if you're happy to hang tight
16:26:09 <odyssey4me> noonedeadpunk: that is true
16:26:10 <skiedude> ah ok, cheers
16:26:12 <cloudnull> could that be related to the new lxc2 vs lxc3 config migrations  ?
16:26:34 <odyssey4me> noonedeadpunk: oh actually, there's no limit in the recreation
16:27:01 <spotz> I'll answer skiedude in DM, I know this one:)
16:27:58 <odyssey4me> cloudnull: yeah, it may be related - not sure
16:28:08 <odyssey4me> anyone got a chance to try recreating the issue
16:28:56 <noonedeadpunk> hm, interesting thing is his environment...
16:29:46 <odyssey4me> ok, I'll leave this as a new bug given it's unconfirmed
16:29:54 <noonedeadpunk> probably we should ask for more details to have a chance for reproducing?
16:30:23 <odyssey4me> well, it seems to me that the steps are there - if someone else can't reproduce, then ask
16:30:52 <noonedeadpunk> do we suggest that he uses master branch?
16:31:26 <noonedeadpunk> as with queens/rocky I never faced with such problem. But generally I agree with your point
16:31:36 <odyssey4me> I think this was rocky IIRC.
16:31:49 <odyssey4me> ioni has been testing Q->R upgrades
16:32:26 <odyssey4me> ok, moving on: https://bugs.launchpad.net/openstack-ansible/+bug/1794688
16:32:26 <openstack> Launchpad bug 1794688 in openstack-ansible "Rocky upgrade nova_oslomsg_notify_host_group' is undefined" [Undecided,New]
16:32:58 <odyssey4me> this is definitely true, and some work needs to be done to get ceilometer working again
16:33:15 <openstackgerrit> Merged openstack/openstack-ansible stable/pike: Bump SHA to fix flipping screens in spicehtml5  https://review.openstack.org/606949
16:33:58 <noonedeadpunk> I may try to work on it
16:34:06 <odyssey4me> noonedeadpunk: that'd be great, thank you
16:34:14 <odyssey4me> confirmed/medium?
16:34:56 <noonedeadpunk> I suppose yes
16:35:35 <odyssey4me> ok, that's it for today
16:35:39 <odyssey4me> #topic open discussion
16:35:48 <odyssey4me> Anyone got anything to raise for discussion?
16:36:15 <spotz> We need to start a Berlin etherpad if we haven't yet, so we know who's coming, sessions and so on
16:36:44 <odyssey4me> spotz: cool, could you fire one up and add it to the agend for the next meeting?
16:36:57 <spotz> odyssey4me: Yep
16:37:23 <prometheanfire> spotz: gimme link when done
16:38:12 <spotz> https://etherpad.openstack.org/p/OSA-berlin-planning prometheanfire
16:38:37 <odyssey4me> thanks spotz
16:38:42 <noonedeadpunk> May I also participate in discussion with it
16:38:42 <openstackgerrit> German Eichberger proposed openstack/openstack-ansible-os_octavia master: Break out config files for each octavia service  https://review.openstack.org/603228
16:38:48 <noonedeadpunk> ?
16:38:53 <odyssey4me> noonedeadpunk: of course
16:39:31 <odyssey4me> alright, if there's nothing else I'll close the meeting
16:40:04 <odyssey4me> #closemeeting openstack_ansible_meeting
16:40:16 <odyssey4me> #endmeeting openstack_ansible_meeting