08:03:46 <d0ugal> #startmeeting mistral
08:03:47 <openstack> Meeting started Fri Oct  5 08:03:46 2018 UTC and is due to finish in 60 minutes.  The chair is d0ugal. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:03:48 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
08:03:50 <openstack> The meeting name has been set to 'mistral'
08:03:55 <d0ugal> Morning everyone! It's the Friday office hour.
08:04:20 <d0ugal> ^ rakhmerov, apetrich, bobh, mcdoker181818, akovi, hardikjasani
08:04:47 <d0ugal> rakhmerov: apetrich is on vacation this week - he is finally moving into a flat in Berlin. He has some trouble with his move which has taken up lots of his time lately
08:05:40 <rakhmerov> hey
08:05:49 <rakhmerov> ok
08:06:21 <rakhmerov> therve: not sure I understand.. Why does it think I try "mistral.db.sqlalchemy" ?
08:06:34 <d0ugal> rakhmerov: because python 2 does relative imports by default
08:06:41 <rakhmerov> having a hard time to understand really..
08:06:45 <d0ugal> rakhmerov: I guess you must be importing from mistral.db.something?
08:06:58 <rakhmerov> aah...
08:07:00 <rakhmerov> gosh
08:07:07 <rakhmerov> so what is the right way?
08:07:57 <d0ugal> Looking at the patch...
08:08:45 <d0ugal> rakhmerov: from __future__ import absolute_imports
08:08:52 <d0ugal> rakhmerov: from __future__ import absolute_import
08:09:10 <d0ugal> https://www.python.org/dev/peps/pep-0328/
08:09:11 <rakhmerov> what's this? :)
08:09:13 <d0ugal> If you want more details
08:09:26 <rakhmerov> ok, I'll read it
08:09:28 <rakhmerov> thanks a lot
08:09:38 <d0ugal> rakhmerov: That will make the importing behaviour like it is in Python 3 - absolute by default
08:09:45 <d0ugal> You don't really need to read it :)
08:10:01 <rakhmerov> interesting..
08:10:03 <d0ugal> but basically, that means "sqlalchemy" will only ever be the top level package and not one relative to your location
08:10:10 <rakhmerov> I feel very uneducated now )
08:10:27 <d0ugal> lol
08:10:34 <d0ugal> I guess you have been lucky to not hit this before :)
08:11:14 <rakhmerov> yes, that's why I'm so confused
08:11:28 <rakhmerov> so where do you think I need to insert this thing?
08:11:41 <d0ugal> rakhmerov: at the top of the utils.py
08:11:43 <rakhmerov> somewhere in the initialization of the application?
08:11:53 <rakhmerov> ooh, the module where I do import?
08:11:54 <rakhmerov> ok
08:12:03 <d0ugal> I added a comment :)
08:12:11 <rakhmerov> ok
08:12:13 <rakhmerov> thanks
08:12:16 <d0ugal> It's a file specific change, it doesn't impact anywhere else
08:12:38 <rakhmerov> ooh, funny :)
08:12:41 <rakhmerov> "HERE" :))
08:12:45 <d0ugal> lol
08:12:57 <d0ugal> Like other __future__ imports it needs to happen before any other import
08:14:18 <rakhmerov> d0ugal: I'm ready to discuss anything else
08:14:28 <d0ugal> rakhmerov: Great, I'm not sure I have anything specific
08:14:38 <d0ugal> I have just been catching up this week
08:14:43 <therve> rakhmerov: Or __import__("sqlalchemy")
08:14:45 <rakhmerov> ok
08:14:50 <d0ugal> therve: ugly :)
08:14:51 <rakhmerov> therve: ok
08:14:57 <therve> Meh :)
08:15:03 <rakhmerov> ok, anyway, thanks to both :)
08:16:36 <rakhmerov> d0ugal: just FYI, working on https://review.openstack.org/607807. Very good change. But I'll have to postpone it for now and get back to it slightly later.
08:16:58 <rakhmerov> d0ugal: my goal is now to fix "join" logic because it's very very slow on some use cases
08:22:25 <d0ugal> rakhmerov: sounds good, I'll take a look at that change
08:27:48 <d0ugal> We have no new bugs this week :)
08:28:30 <rakhmerov> d0ugal: I created one today but assigned it to myself immediately
08:28:48 <d0ugal> oh, good
08:28:54 <d0ugal> I did mean "untriaged" bugs
08:29:16 <rakhmerov> https://bugs.launchpad.net/mistral/+bug/1796242
08:29:16 <openstack> Launchpad bug 1796242 in Mistral "Retry DB transactions on sqlalchemy.exc.OperationalError" [High,In progress] - Assigned to Renat Akhmerov (rakhmerov)
08:29:18 <rakhmerov> yep
08:29:18 <openstackgerrit> Renat Akhmerov proposed openstack/mistral master: Add sqlalchemy.exc.OperationalError to the retry decorator  https://review.openstack.org/608171
08:29:39 <rakhmerov> d0ugal: also one thing, we need to do something to make CI more stable
08:29:51 <rakhmerov> it's painful now to push patches through
08:30:34 <rakhmerov> d0ugal: please also review https://review.openstack.org/#/c/603593/, very small
08:30:52 <d0ugal> will do!
08:31:18 <rakhmerov> ok
08:31:46 <d0ugal> Is it the mysql and rally CI jobs that fail most?
08:31:54 <d0ugal> or something else?
08:32:00 <d0ugal> oh, I see a tripleo failure on one...
08:32:07 <rakhmerov> yeah :)
08:32:15 <rakhmerov> rally too, yes
08:32:43 <rakhmerov> another thing: https://review.openstack.org/#/c/607974/
08:32:54 <rakhmerov> POST_FAILURE on devstack-kombu
08:33:12 <d0ugal> I never understand the POST_FAILURE result
08:33:13 <rakhmerov> it seems kind of sporadic
08:33:20 <rakhmerov> here and there pretty often
08:33:31 <rakhmerov> maybe there's a general issue with the CI infra
08:33:34 <rakhmerov> dunno
08:35:02 <d0ugal> The tripleo failure I see on thrash|g0ne's patch seems legit
08:35:32 <d0ugal> actually, I'm not sure now
08:35:41 <d0ugal> I wish these logs were earlier to read.
08:36:46 <d0ugal> http://zuul.openstack.org/builds.html?project=openstack%2Fmistral
08:37:15 <d0ugal> The failures do indeed seem sporadic
08:37:20 <d0ugal> Lots of failures on different jobs.
08:39:15 <rakhmerov> yes
10:04:00 <d0ugal> #endmeeting