08:00:01 <d0ugal> #startmeeting mistral
08:00:02 <openstack> Meeting started Fri Jun  8 08:00:01 2018 UTC and is due to finish in 60 minutes.  The chair is d0ugal. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:00:03 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
08:00:06 <openstack> The meeting name has been set to 'mistral'
08:00:16 <d0ugal> Who is around for the Mistral office hour today?  https://etherpad.openstack.org/p/mistral-office-hours
08:00:42 <d0ugal> rakhmerov, apetrich, bobh, mcdoker181818 ^
08:01:12 <d0ugal> As usual, add your IRC nick to line 16 in the etherpad if you want a reminder ping in the future!
08:03:42 <d0ugal> I am going to be proposing the Rocky Milestone 2 releases today.
08:10:19 <d0ugal> and by today, I mean "now" :)
08:10:30 <d0ugal> https://review.openstack.org/573610
08:27:26 <therve> d0ugal: Is it a good time to chat about https://review.openstack.org/572807 ?
08:27:44 <d0ugal> therve: Sure
08:28:02 <d0ugal> I'm just doing milestone 2 administrivia
08:28:17 <therve> d0ugal: https://review.openstack.org/573306 I think is the fix, hopefully that'll be confirmed
08:28:31 <therve> But it requires a heatclient release, so I was wondering if we can find a workaround
08:29:00 <d0ugal> Right
08:29:06 <d0ugal> Do we know where it is called from?
08:29:45 <therve> Yes, serialization code
08:30:21 <d0ugal> That is what I suspected. Got a link to relevant line handy?
08:30:51 <therve> Hum I don't have the full traceback anymore
08:31:29 <therve> https://github.com/openstack/mistral/blob/master/mistral/serialization.py#L106 was the one in mistral I remember
08:31:34 <therve> Then it got in jsonutils
08:32:20 <d0ugal> Right
08:32:35 <d0ugal> therve: I guess a workaround would be for the mistral actions to set loaded=True?
08:33:23 <therve> d0ugal: Hum maybe. Just the get_stack one though
08:33:39 <d0ugal> Right
08:33:51 <d0ugal> It would be a fairly nasty hack, but I don't have any other immediate ideas
08:34:16 <therve> d0ugal: So there is no way in the workflow to operate on the result of the action before it's serialized?
08:34:59 <d0ugal> therve: No, I don't think so
08:35:24 <therve> Right
08:35:38 <therve> In which case the client release is probably the best way then
08:36:40 <d0ugal> Yeah
08:36:54 <d0ugal> I guess my work around in the workflow would be good enough for now?
08:37:14 <d0ugal> Until the release I mean
08:37:19 <therve> I guess
08:37:25 <therve> It doesn't prevent the double request
08:37:41 <d0ugal> Well, it means we have one slow inefficient request :)
08:38:40 <d0ugal> but it does mean we store that output in the action execution table in mistral.
08:47:46 <openstackgerrit> Merged openstack/mistral-extra master: fix tox python3 overrides  https://review.openstack.org/572969
08:52:37 <openstackgerrit> Merged openstack/mistral-dashboard master: fix tox python3 overrides  https://review.openstack.org/572968
09:01:10 <d0ugal> #endmeeting