21:00:49 #startmeeting nova_cells 21:00:52 Meeting started Wed Jun 24 21:00:49 2015 UTC and is due to finish in 60 minutes. The chair is alaski. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:00:53 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:00:56 The meeting name has been set to 'nova_cells' 21:01:05 anyone around for cells today? 21:01:10 o/ 21:01:10 \o 21:01:26 excellent 21:01:29 o/ 21:01:34 #topic Tempest 21:01:42 * alaski hands it over to melwitt 21:02:02 the cells job is still looking in good shape goo.gl/b7R8wq 21:02:29 we have had a ML thread about whether to make it voting right now http://lists.openstack.org/pipermail/openstack-dev/2015-June/067723.html 21:02:53 and I think we'll do a vote about whether to make it voting at tomorrow's nova meeting 21:03:27 awesome 21:03:53 cool 21:03:53 at which time I can un -W the voting patch https://review.openstack.org/#/c/190894/ and make it depend on the project-config change for the regex https://review.openstack.org/#/c/194410/ 21:04:08 if it gets voted to turn on, I mean 21:04:25 heh, or just be a rebel 21:04:34 :) 21:04:37 yeah!! 21:04:49 other than that, I opened a new bug for a rare-ish failure https://bugs.launchpad.net/nova/+bug/1466696 21:04:49 Launchpad bug 1466696 in OpenStack Compute (nova) "Cells: Race between instance 'unlock' and 'stop' can cause 'stop' to fail" [Low,Confirmed] 21:05:05 with the corresponding logstash query http://goo.gl/2reb35 21:05:24 oooh 21:05:32 #link https://bugs.launchpad.net/nova/+bug/1466696 21:05:59 I think that's all the status I have on the tempest job 21:06:33 cool, thanks 21:06:43 that looks like a tricky bug too 21:07:03 yeah. I couldn't immediately think of a way to address that one 21:07:56 out of order messages are tough with the system in place now 21:08:15 well, more eyes will be good 21:08:21 right 21:08:40 #topic Specs 21:09:01 https://review.openstack.org/#/c/190147/ is the only open spec I'm aware of now 21:09:24 I have uploaded a spec last night about the database connection switching https://review.openstack.org/#/c/194935/ 21:09:39 nice 21:10:09 added it to the priorities etherpad 21:11:01 I was wondering about the spec deadline that's like... tomorrow. we don't yet have all the specs we need to accomplish a simple 'nova boot' with cells v2 do we? can we not have anymore specs after tomorrow? 21:11:24 I'm not really sure what's going to happen with that 21:11:38 * dansmith staggers in late 21:11:57 We add the migration plan to #190147 21:12:07 I think we have enough for a boot, but with some db tables in the cell vs in the api where we'd want them 21:12:18 melwitt: I sincerely wonder if by any chance some freeze exception can occur 21:13:00 alaski: hm, okay 21:13:51 melwitt: it may not be pretty, but I think we have all the pieces at least proposed as specs 21:14:30 bauzas: we may need to attempt that, but we're also getting to a point where we have enough work to last the cycle I think 21:14:45 alaski: agreed 21:14:50 speaking from my personal velocity recently :( 21:15:40 belmoreira: great. It looks like edleafe still wanted a bit of clarification, but it's looking good 21:16:21 anything else on specs? 21:16:24 we are waiting for more reviews before a new iteration 21:16:35 belmoreira: the overall glance at the spec was good IMHO 21:16:51 belmoreira: I'll vote on that soon 21:17:00 belmoreira: okay. I'll look at that again 21:17:04 bauzas: great 21:17:57 #topic Open Discussion 21:18:26 anything people would like to discuss? 21:18:31 alaski: can you be more verbose in your comment on #192098 :) 21:18:49 have no idea what you are asking for :) 21:20:03 belmoreira: ahh, rather than having an integer field there with just the id, have an ObjectField there will the cell_mapping object 21:20:35 because the only use of cell_id will be to get the object later, so it might as well be loaded at the start 21:21:21 so get_by_host will need to join the cell_mapping table, then load it in _from_db_object 21:21:28 ahhh... ok 21:21:51 right 21:22:00 is that change tracked in the etherpad? 21:22:08 bauzas: not sure 21:23:13 it is now 21:23:18 anything else? 21:23:18 coolness 21:23:24 speaking of etherpads, I noticed the etherpad we keep referencing in our specs is somehow now in chinese https://etherpad.openstack.org/p/kilo-nova-cells 21:23:45 (according to google translate anyway) 21:23:58 wow 21:24:01 :) 21:24:06 haha, yeah 21:25:30 ??? 21:25:37 I wonder if we can revert that back somehow 21:26:25 I'll see if I can do something. I can see the old content in the timeslider so at worst a copy-paste can be done 21:27:25 looks like it went back and forth a few times 21:27:27 melwitt: thanks 21:28:15 that seems like it for the day, unless there's a last minute topic? 21:29:00 See you all next week. 21:29:07 #endmeeting