17:00:07 <gibi> #startmeeting nova notification
17:00:08 <openstack> Meeting started Tue Dec 12 17:00:07 2017 UTC and is due to finish in 60 minutes.  The chair is gibi. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:09 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:12 <openstack> The meeting name has been set to 'nova_notification'
17:00:15 <gibi> hi!
17:00:22 <mriedem> o/
17:00:57 <gibi> This is the last meeting for 2017
17:01:17 <mriedem> let's make it a good one
17:01:30 <gibi> as next week this time I will be on a the company Christmas party
17:01:57 <gibi> so there are couple of notification related bugs
17:02:20 <gibi> bug 1736976
17:02:22 <openstack> bug 1736976 in OpenStack Compute (nova) "test_live_migration_actions functional test randomly fails with "AssertionError: The migration table left empty."" [Undecided,In progress] https://launchpad.net/bugs/1736976 - Assigned to Balazs Gibizer (balazs-gibizer)
17:02:29 <gibi> you reported
17:02:43 <mriedem> i saw your update on that
17:02:44 <gibi> I've just pushed a fix
17:03:22 <gibi> The problem seems obvious to me
17:03:42 <gibi> as the test is too fast to try to get the migration object
17:04:01 <mriedem> +2
17:04:07 <gibi> cool
17:04:21 <gibi> while I played with this test I noticed another failure mode
17:04:22 <gibi> 
17:04:25 <gibi> bug 1737781
17:04:27 <openstack> bug 1737781 in OpenStack Compute (nova) "test_live_migration_actions randomly fails with testtools.matchers._impl.MismatchError: 6 != 4" [Undecided,In progress] https://launchpad.net/bugs/1737781 - Assigned to Balazs Gibizer (balazs-gibizer)
17:04:39 <gibi> so I pushed a fix for that as well
17:04:52 <gibi> #link https://review.openstack.org/#/c/527444/
17:05:17 <gibi> the next thing on my list is the third bug
17:05:27 <gibi> bug 1737201
17:05:28 <openstack> bug 1737201 in OpenStack Compute (nova) "TypeError when sending notification during attach_interface" [High,Confirmed] https://launchpad.net/bugs/1737201 - Assigned to Balazs Gibizer (balazs-gibizer)
17:05:55 <mriedem> as you've seen my comments in there, i'm not sure how that can happen
17:06:02 <gibi> if I ignore the question "how can this happen" then the fix seems easy
17:06:05 <mriedem> and i'm only seeing it in the check queue on a patch that's having other timing issues
17:06:14 <mriedem> yeah the fix is easy, it's just weird
17:06:25 <gibi> anyhow I agree with you that a simple fix adding an extra if cannot hurt
17:06:26 <mriedem> granted, there are places in the nova/network/model.py code where it's checking to see if self['network'] is set
17:06:48 <mriedem> but from what i can tell in the neutron api code, we will always set network if we have a vif
17:07:11 <mriedem> i held off on pushing that fix because i wasn't sure if it would paper over something worse,
17:07:18 <mriedem> like maybe a timing issue where we overwrite the info_cache values
17:07:43 <gibi> could be. I think I will not try to reproduce it with a functional test just a unit test
17:08:13 <gibi> anyhow that was all about bugs I know about
17:08:18 <mriedem> i'm sure it's trivial to reproduce the TypeError with a unit test
17:08:26 <gibi> yeah :)
17:08:56 <mriedem> like, i wondered if https://review.openstack.org/#/c/403568/ did something
17:09:07 <mriedem> like wipe out the info_cache when it shouldn't have
17:09:43 <gibi> it was merged 5 months ago and we only see one occurence of the problem so far
17:09:53 <gibi> it is hard to tell
17:10:09 <gibi> anyhow fix is simple so I'm not affraid of committing it without a real reproduction
17:10:19 <gibi> as it is just a safeguard
17:10:38 <gibi> moving forward
17:11:10 <gibi> the bp service-create-destroy-notification is now Done Done as the followup patch also merged
17:11:54 <gibi> during last week I noticed a new patch proposing another notification
17:11:57 <gibi> #link https://review.openstack.org/#/c/526251/
17:12:08 <mriedem> yeah i agree that's a specless bp for rocky
17:12:11 <mriedem> good call
17:12:15 <gibi> cool
17:13:10 <gibi> OK, that was all I wanted to mention
17:13:36 <mriedem> from your latest ML update,
17:13:36 <gibi> I will be active until 20th of Dec then on vacation
17:13:47 <mriedem> it links a few patches which look like they need a rebase, e.g. https://review.openstack.org/#/c/403660 - has failing testes
17:13:49 <mriedem> *tests
17:14:06 <mriedem> ok i'm done after the 22nd
17:14:13 <mriedem> back on jan 2
17:14:24 <gibi> hm, I missed -1 from Zuul
17:14:29 <gibi> I will check it tomorrow
17:14:52 <mriedem> sounds good
17:14:53 <gibi> I will be back on 8th of Jan
17:15:03 <mriedem> going anywhere?
17:15:21 <gibi> just visiting my and my wife's family
17:15:32 <mriedem> cool, same here
17:15:36 <mriedem> enjoy
17:15:43 <gibi> I hope will enjoy too
17:15:57 <gibi> if you have nothing else then let's close this
17:16:02 <mriedem> nope - ttyl
17:16:09 <gibi> cool, bye
17:16:13 <gibi> #endmeeting