16:04:06 <evrardjp> #startmeeting openstack_ansible_meeting
16:04:07 <openstack> Meeting started Tue May 30 16:04:06 2017 UTC and is due to finish in 60 minutes.  The chair is evrardjp. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:04:09 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:04:12 <openstack> The meeting name has been set to 'openstack_ansible_meeting'
16:04:46 <andymccr> o/
16:04:48 <evrardjp> #topic rollcall -- knock knock
16:04:55 * mhayden toots
16:05:19 <openstackgerrit> Nolan Brubaker proposed openstack/openstack-ansible-os_glance master: Run service test during glance upgrade  https://review.openstack.org/467666
16:05:23 <spotz> \o/
16:05:27 <palendae> Here and half paying attention
16:05:47 <spotz> Hey I noticed we missed last week:)
16:06:59 <evrardjp> no there was a bug triage last week, which brings us to:
16:07:08 <evrardjp> #topic last week AP
16:07:17 <evrardjp> odyssey4me to confirm/triage https://bugs.launchpad.net/openstack-ansible/+bug/1687369
16:07:19 <openstack> Launchpad bug 1687369 in openstack-ansible "ocata installation fails for aodh,ceilometer and designate when sqlalchemy version is 1.1.9 version " [Undecided,New] - Assigned to Jesse Pretorius (jesse-pretorius)
16:07:38 <evrardjp> it's been 3 weeks, should we release it from you odyssey4me?
16:07:42 <odyssey4me> I haven't had any chance yet - perhaps unassign to let someone else take it
16:07:58 <evrardjp> ok let's keep it for this week triage then
16:08:21 <evrardjp> let's start
16:08:27 <evrardjp> #topic this week's bugs
16:08:33 <evrardjp> #link https://bugs.launchpad.net/openstack-ansible/+bug/1694265
16:08:34 <openstack> Launchpad bug 1694265 in openstack-ansible "repo build fails multiple architecture environment" [Undecided,New]
16:09:19 <evrardjp> Isn't power and amd64 already working?
16:09:23 <evrardjp> I'm surprised of this
16:09:41 <ThiagoCMC> Guys, I removed "debug: true" out from my user_variables.yml, how to apply it?
16:09:42 <andymccr> hmm
16:10:08 <andymccr> im surprised by that too evrardjp
16:10:51 <evrardjp> I think this deserves more input
16:10:54 <andymccr> yeah
16:10:57 <andymccr> what version is that etc
16:11:06 <evrardjp> It looks like u'repo_servers_16.04s390x' doesn't exist, so I guess the architecture is s390x
16:11:22 <andymccr> hmm
16:12:22 <evrardjp> I've answered
16:12:28 <evrardjp> marking incomplete
16:12:29 <evrardjp> ok?
16:12:47 <evrardjp> Any objections?
16:12:57 <evrardjp> 5
16:12:58 <evrardjp> 4
16:13:00 <evrardjp> 3
16:13:01 <evrardjp> 2
16:13:02 <evrardjp> 1
16:13:06 <evrardjp> Done, next!
16:13:16 <evrardjp> #link https://bugs.launchpad.net/openstack-ansible/+bug/1694222
16:13:17 <openstack> Launchpad bug 1694222 in openstack-ansible "When deploying nova with spice, sometime Spice port binding is failing" [Undecided,New]
16:14:26 <evrardjp> do we have spice in our gates?
16:15:24 <evrardjp> maybe we should have this as a scenario
16:17:02 <evrardjp> Opinion anyone?
16:17:13 <evrardjp> could someone confirm?
16:17:38 <andymccr> lemme check
16:17:44 <andymccr> i think default changed to novnc
16:18:18 <andymccr> hmm nope
16:18:21 <andymccr> spice is on by default
16:18:35 <evrardjp> that's what I thought too. Novnc is only in RPC IIRC :p
16:18:57 <evrardjp> I think that's why mhayden had issues recently with keymaps
16:18:58 <andymccr> i thought upstream nova default changed but probably not then
16:19:08 <evrardjp> oh
16:19:17 <evrardjp> my bad, that's not what I meant
16:19:28 <evrardjp> that would be good to track too
16:19:46 <andymccr> that error is still weird
16:19:47 <evrardjp> in any case, i'd be inclined to say it's a configuration issue
16:19:54 <evrardjp> did you see the IP?
16:20:03 <andymccr> yeah
16:20:20 <evrardjp> we need more info about what that IP is, at least I'd say
16:20:33 <evrardjp> I ask on the bug and mark it incomplete
16:20:37 <evrardjp> ok?
16:21:21 <andymccr> yeah
16:21:24 <andymccr> that sounds sensible
16:22:06 <evrardjp> ArchiFleKs: are you there by any chance?
16:22:34 <evrardjp> 5
16:22:40 <evrardjp> 3
16:22:44 <evrardjp> 1
16:22:54 <evrardjp> #link https://bugs.launchpad.net/openstack-ansible/+bug/1693234
16:22:54 <openstack> Launchpad bug 1693234 in openstack-ansible "Using galera_my_cnf_overrides brakes cluster config" [Undecided,New]
16:23:13 <evrardjp> that sounds like a config template bug
16:23:30 <ArchiFleKs> evrardjp: hi
16:25:03 <andymccr> evrardjp: yeah agreed. thats weird though - but should be easy to test
16:25:04 <evrardjp> we moved to another bug now, but don't hesitate to scroll, and if you want we can talk about your bug later
16:25:06 <andymccr> would say tahts high though
16:25:13 <evrardjp> agreed.
16:25:32 <evrardjp> let's mark it as confirmed and high because we don't have a test for that
16:25:47 <evrardjp> even if it works
16:27:11 <evrardjp> cloudnull: do you have time to take a config_template issue?
16:28:51 <evrardjp> ok let's roll
16:28:52 <evrardjp> #link https://bugs.launchpad.net/openstack-ansible/+bug/1692044
16:28:53 <openstack> Launchpad bug 1692044 in openstack-ansible "Tempest failing on os_nova role on CentOS 7" [Undecided,New]
16:29:13 <evrardjp> mhayden: did you got more chance to dig into this deeper?
16:31:56 <evrardjp> I can assume this is confirmed?
16:32:10 <evrardjp> I trust this.
16:32:59 <evrardjp> fun stuff triaging alone.
16:33:38 <evrardjp> on top of it I can't change statuses on launchpad
16:33:50 <evrardjp> I'll mark it as confirmed high when LP is working again
16:34:11 <evrardjp> #link https://bugs.launchpad.net/openstack-ansible/+bug/1691749
16:34:12 <openstack> Launchpad bug 1691749 in openstack-ansible "Generated no_proxy env string can be too long for pam_env" [Undecided,New]
16:34:30 <andymccr> this
16:34:31 <andymccr> hmm
16:35:48 <andymccr> so it looks like its just docs left - it sounds like we just need some work done on the docs
16:36:20 <evrardjp> " Introducing an alternative method" ?
16:36:26 <jmccrory> think containers get FQDNs now, maybe no_proxy could be wildcarded
16:36:50 <andymccr> hmm
16:37:02 <evrardjp> now we are talking.
16:37:32 <evrardjp> jmccrory: should we prepare something for this?
16:37:39 <evrardjp> Spec/change/anything?
16:38:01 <evrardjp> It would be good to not lose this valuable comment in the depth of irc logs.
16:38:03 <jmccrory> maybe doc change, don't know if it's mentioned or used within any roles
16:38:30 <evrardjp> Could I assign this to you, with low priority, and point to our conversation in the bug?
16:38:32 <jmccrory> sure
16:38:38 <evrardjp> great
16:38:42 <jmccrory> i'll try to test out somewhere as well
16:39:16 <evrardjp> cool
16:39:19 <evrardjp> thanks!
16:39:23 <evrardjp> #link https://bugs.launchpad.net/openstack-ansible/+bug/1691130
16:39:24 <openstack> Launchpad bug 1691130 in openstack-ansible "fails at repo_build : Build package requirements file" [Undecided,New] - Assigned to Jesse Pretorius (jesse-pretorius)
16:40:05 <evrardjp> I'll mark this as incomplete, this way we'll see if we get the author's response
16:40:11 <andymccr> sounds good
16:40:24 <evrardjp> at least when LP works again
16:40:25 <evrardjp> :p
16:40:28 <evrardjp> next
16:40:30 <evrardjp> #link https://bugs.launchpad.net/openstack-ansible/+bug/1691120
16:40:31 <openstack> Launchpad bug 1691120 in openstack-ansible "use find instead of grep" [Undecided,New]
16:40:37 <evrardjp> confirmed wishlist
16:40:42 <evrardjp> low-hanging-fruit
16:40:47 <evrardjp> ok?
16:40:49 <andymccr> agreed
16:40:50 <jmccrory> yep
16:41:19 <evrardjp> oh it seems that LP takes my status changes again... Woot!
16:41:24 <evrardjp> anyway
16:41:36 <evrardjp> #link https://bugs.launchpad.net/openstack-ansible/+bug/1687369
16:41:37 <openstack> Launchpad bug 1687369 in openstack-ansible "ocata installation fails for aodh,ceilometer and designate when sqlalchemy version is 1.1.9 version " [Undecided,New]
16:42:14 <andymccr> seems like it was a deploy error
16:42:30 <evrardjp> sounds like it
16:42:35 <andymccr> maybe checked staging docs?
16:42:36 <evrardjp> Invalid?
16:43:02 <andymccr> yeah
16:43:14 <evrardjp> #link https://bugs.launchpad.net/openstack-ansible/+bug/1692750
16:43:15 <openstack> Launchpad bug 1692750 in openstack-ansible "deprecate glance_api_version option" [Low,New] - Assigned to zhongshengping (chdzsp)
16:43:52 <evrardjp> Could we move this as confirmed?
16:43:58 <andymccr> yeah sorry that was my bad last week
16:44:01 <andymccr> i should've marked it as confirmed
16:44:13 <evrardjp> there was maybe a reason :p
16:44:16 <evrardjp> ok next
16:44:27 <evrardjp> #link https://bugs.launchpad.net/openstack-ansible/+bug/1692017
16:44:29 <openstack> Launchpad bug 1692017 in openstack-ansible "[WARNING]: Failure using method (v2_runner_on_ok) in callback plugin" [Low,New]
16:44:43 <evrardjp> confirmed. I've that in many places
16:44:45 <andymccr> its just an annoying warning
16:44:47 <andymccr> yeah
16:45:02 <evrardjp> low though, it's a callback it's just for displaying things
16:45:29 <evrardjp> I'll take it
16:45:38 <evrardjp> it bothers me too
16:46:09 <evrardjp> that's it for the current list. ArchiFleKs want to discuss your bug again?
16:46:50 <evrardjp> Or any other topic, everyone?
16:46:58 <evrardjp> if not I'm closing the deal for today.
16:47:14 <evrardjp> 60 seconds!
16:47:20 <andymccr> the countdonw is on! :P
16:47:31 <ArchiFleKs> evrardjp: actually it is a compute node
16:47:37 <evrardjp> andymccr: let's do it in the street.
16:48:16 <evrardjp> so the thing is we didn't see that in gates, could you tell us more that can help the triage?
16:48:27 <evrardjp> For a fix, we could discuss that after the triage meeting
16:48:36 <evrardjp> but at least know if we can confirm it or leave it as is.
16:49:14 <ArchiFleKs> i'm using ocata, I had this bug at first, then I switched back to novnc and at some point I upgraded osa and switch back to spice it was working again and maybe a week ago i updated to ocata branch and redeployed and ran into the same issue
16:49:35 <evrardjp> are you sure it's not already running on the node?
16:49:35 <evrardjp> :p
16:49:53 <evrardjp> that would be a good reason to fail: the port is already used.
16:50:01 <ArchiFleKs> evrardjp: i'm not sure it is related to osa
16:50:14 <evrardjp> It probably is linked to your host right now
16:50:14 <ArchiFleKs> actually yes but libvirt is supposed to do the port automaticly
16:50:29 <evrardjp> indeed
16:50:40 <evrardjp> it doesn't sound OSA related at the end
16:50:53 <evrardjp> I'll mark this as invalid then.
16:51:05 <evrardjp> We can discuss that in our channel after triage
16:51:17 <odyssey4me> if you had vnc set when building an instance, that will remain configured for the VM for its lifetime
16:51:26 <odyssey4me> so the port will be a novnc port
16:51:27 <ArchiFleKs> i believe you have a port range that starts at 5900  and increase
16:52:01 <odyssey4me> you would have to change the db entry if you want to convert from novnc to spice or vice-versa
16:52:08 <evrardjp> Could you at least output a list of the ports already assigned on this machine?
16:52:22 <ArchiFleKs> odyssey4me: i'll check that, i did not had time to check but i recall running into this when we first deployed and they were no instances
16:52:28 <odyssey4me> once the db entry is changed, you'd have to shut down and possibly even migrate the instance to another host to make the instance actually get the new config
16:52:50 <ArchiFleKs> that's why we switched back but yes maybe right now there is a mixup between spice and novnc
16:52:52 <odyssey4me> the console type is typically set for the lifetime of the instance
16:53:12 <evrardjp> ok let's conclude it's not osa related
16:53:33 <evrardjp> Thanks everyone for your today's triage attendance!
16:53:41 <evrardjp> #endmeeting