16:00:01 #startmeeting blazar 16:00:02 Meeting started Thu Aug 1 16:00:01 2019 UTC and is due to finish in 60 minutes. The chair is priteau. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:03 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:05 The meeting name has been set to 'blazar' 16:00:23 #topic Roll call 16:00:45 Hi jakecoll 16:01:12 Good morning priteau 16:01:22 Hi diurnalist 16:01:26 Heyyo 16:01:38 jakecoll: Good morning to you. It's end of the afternoon for me ;-) 16:01:57 oh right 16:02:10 #topic Upstream contributions 16:02:43 Let's start with a quick recap of the various threads and action items 16:03:11 First, we have the network segment reservation feature 16:03:35 I saw that jakecoll finally got tests to pass, great news! https://review.opendev.org/#/c/668749/ 16:03:41 That is pushed and passing all the tests. I'm almost done with the spec file, which should be pushed up today. 16:03:50 That's good to hear 16:04:24 Do you think there are any valuable alternative approaches worth mentioning in the spec file? I just listed those as none. 16:05:42 I am checking the floating IP spec for what we said there 16:06:29 I suppose one alternative would be to have Blazar create ports on a Neutron network it would own, but that would be clunky to use. 16:07:38 I think no alternatives will be just fine 16:07:49 :+1: 16:08:21 Other action items were: 16:08:24 * priteau jakecoll submit floating IP update patch 16:08:27 That's done 16:08:49 https://review.opendev.org/#/c/672763/ 16:09:19 I would like to have Masahito's review of it 16:09:27 * priteau priteau find required soft-delete changes 16:09:43 I've added requested changes as a comment in https://review.opendev.org/#/c/585807/ 16:10:00 There were two 16:10:05 1) Change the current patch series to "resurrect" soft-delete compute hosts when recreated by the operator 16:10:30 This is so we can implement soft-delete on the compute host table as well, while keeping host ID unchanged. 16:10:43 2) Implement Glance's approach for DB purge: https://docs.openstack.org/glance/latest/admin/db.html#purging-the-database 16:11:01 This is so operators can clean up soft-deleted data regularly 16:12:52 jakecoll: If you've got time to look at these two required changes, feel free to do so as I don't have cycles to work on this until mid-August 16:13:39 yep, should be able to get to it this afternoon 16:14:33 I see you've been productive and also posted the DB query optimisation patch: https://review.opendev.org/#/c/672994/ 16:15:26 jakecoll: Would it be possible to split this patch in two, one for the host allocation query change, and the other for the API change? 16:16:18 I can help you with some git magic to do that 16:16:19 Yea, that can be done 16:16:36 Thanks. 16:18:01 I think we've covered the various upstream efforts? 16:19:09 Besides the slow queries and allocation API limitations, are there other blockers for the reservation calendar? 16:19:49 I pushed allocation support to the blazarclient as well. 16:20:27 right, https://review.opendev.org/#/c/673106/ 16:20:58 We need allocations support for floating ip and networks. 16:22:01 Jason's was talking about updating redmine. I should be able to start on that next week unless there are other priorities on Chameleon. 16:22:21 Updating redmine? 16:23:23 The TACC collaboration server? 16:24:55 I am just not sure what was the link with what you were discussing. Anyway, if we can get your API change for host allocation merged, next step would be to add allocations for other types of reservations, it shouldn't be much work. 16:26:06 Oh, working on floating ip and network allocations API support. 16:26:39 Nice. Please keep patches coming if you can :) 16:26:45 will do 16:27:45 #topic AOB 16:27:49 Anything else to discuss today? 16:28:35 Nothing I'm aware of 16:29:49 We can end early then, thanks a lot for joining! 16:30:13 #endmeeting