13:00:28 #startmeeting watcher 13:00:30 Meeting started Wed Nov 22 13:00:28 2017 UTC and is due to finish in 60 minutes. The chair is alexchadin. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:31 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:33 The meeting name has been set to 'watcher' 13:00:39 hi! 13:00:41 hi 13:00:45 Agenda: https://wiki.openstack.org/wiki/Watcher_Meeting_Agenda#11.2F22.2F2017 13:02:20 Yumeng__: let's discuss some things 13:02:28 #topic Open Discussions 13:02:33 Shouldn't we deprecate Ceilometer support in future release? 13:02:55 Yumeng__: what do you think about it? 13:03:14 first of all, I talk about datasource 13:05:01 alexchadin: in Sydney we discussed that 13:05:20 NEC use ceilometer for POC 13:05:43 and goncci for production 13:06:17 Yumeng__: ceilometer-api is deprecated since Ocata 13:07:00 And ZTE use ceilometer for production, but we are considering replace ceilometer with gnocchi cause ceilometer is slow 13:07:38 I think we can keep that for a while since ceilometer is still used in production 13:08:11 Stop maintain ceilometer 13:08:29 Yumeng__: I've seen this patch today: https://review.openstack.org/#/c/522183/ 13:08:47 Support and maintain gnocchi 13:10:19 Yumeng__: many things we are planning to do with Watcher works with Gnocchi mostly (and Monasca sometimes) 13:10:45 Yumeng__: I'd propose to set gnocchi as default datasource at least 13:11:12 agree 13:12:33 ok. 13:12:43 and to mark ceilometer as deprecated in future releases (by using appropriate logging) 13:12:43 It’s ok for me 13:12:57 ok 13:13:06 #action alexchadin mark ceilometer as deprecated 13:13:18 It would be good to test HA in Watcher (Pike, master) using haproxy + keepalived (other proposals?) 13:13:33 suzhengwei: I've seen your thread with hidekazu about HA 13:14:09 suzhengwei: you've done it with Ocata, right? 13:14:17 yes 13:14:37 suzhengwei: did you use haproxy + keepalived? 13:14:55 yes, but watcher doesn;t support haproxy. 13:15:21 suzhengwei: yeap, I've seen your bug 13:15:47 suzhengwei: I'll try to investigate it and to run Pike Watcher in HA mode 13:15:59 ok 13:16:24 let's move on to Review Action Items 13:16:25 do you have some clue? 13:16:38 suzhengwei: not yet 13:16:43 #topic Review Action Items 13:16:49 Watcher specs 13:16:57 exclude locked instances needs to be updated #link https://review.openstack.org/#/c/506078/ 13:18:40 cluster-maintenance-strategy needs reviews #link https://review.openstack.org/#/c/464602/ 13:19:01 I can review again 13:19:46 me too, although it's mostly LGTM 13:19:59 #action Yumeng__ alexchadin review https://review.openstack.org/#/c/464602/ 13:20:13 Add storage capacity balance Strategy needs to be updated #link https://review.openstack.org/#/c/501659/ 13:20:33 Ops... I will update 13:20:36 Yumeng__: +1 13:20:52 #action Yumeng__ submit new PS https://review.openstack.org/#/c/501659/ 13:21:10 From today I can spare more time in review and commit 13:21:19 :) 13:21:24 Yumeng__: it's nice! :) 13:22:34 Add audit scoper for storage data model spec needs reviews #link https://review.openstack.org/#/c/517169/ 13:22:55 Yumeng__: suzhengwei: aditi works on it 13:23:15 it would be great if you take a look at it 13:23:37 ok 13:24:52 #action suzhengwei review https://review.openstack.org/#/c/517169/ 13:26:17 do we need to continue listing Watcher patches? 13:26:31 https://wiki.openstack.org/wiki/Watcher_Meeting_Agenda#11.2F22.2F2017 there are some of them 13:26:53 most of them need reviews 13:27:08 suzhengwei: option to rollback action_plan when it fails https://review.openstack.org/#/c/517179/ would be good to have unit tests 13:27:48 suzhengwei: marker patches should be based on https://review.openstack.org/#/c/519945/ 13:28:02 I will update it 13:28:15 suzhengwei: you may rebase them to pass py-jobs 13:29:09 do you have any questions? 13:29:17 they are all ok if common marker filter is merged. 13:29:21 or topics to discuss :) 13:29:32 suzhengwei: yeap, just try to rebase them 13:29:34 a serial of commits. 13:30:35 then I'd propose to end up this meeting 13:30:44 this commit should be merged firstly.https://review.openstack.org/#/c/521847/ 13:30:52 hidekazu and aditi should join us next time 13:31:20 suzhengwei: licanwei did the same commit 13:32:03 suzhengwei: take a look: https://review.openstack.org/#/c/519945/ 13:33:28 yes, I see. But the common filer for marker should be merged. So licanwei's commit or this commit is merged, other commits would be ok autoly. 13:34:17 commom filter for marker need to be merged firstly. 13:34:41 suzhengwei: I've rebased one of your commit on licanwei's 13:34:49 suzhengwei: let's see what we get 13:35:47 suzhengwei: is it ok? 13:36:09 i think it will not work 13:36:55 suzhengwei: why? 13:37:23 now your audit template commit based on licanwei's commit 13:37:31 not add common filter of marker 13:37:33 you both have the same utils.py file 13:39:50 I separate the common marker filter as one commit, https://review.openstack.org/#/c/521847/ 13:40:19 we can discuss in waycher chanel. 13:40:20 suzhengwei: yeap, and other commits which should use this commit 13:40:24 yeap 13:40:29 thank you for coming and have a good day 13:40:37 #endmeeting