04:01:36 #startmeeting masakari 04:01:37 Meeting started Tue Jun 2 04:01:36 2020 UTC and is due to finish in 60 minutes. The chair is samP. Information about MeetBot at http://wiki.debian.org/MeetBot. 04:01:38 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 04:01:40 The meeting name has been set to 'masakari' 04:02:19 #topic Critical Bugs and Patches 04:02:39 Any critical bugs or patches need to discuss here? 04:03:02 https://review.opendev.org/#/c/702328/ 04:03:03 patch 702328 - masakari - not wait when instance evacuate error - 3 patch sets 04:03:18 #link : https://review.opendev.org/#/c/726761 04:03:19 patch 726761 - masakari - Check if host belongs to a segment - 3 patch sets 04:04:51 tpatil: #726761 looks good. I will approve it. Thanks. 04:05:07 samP: Issue with the unit tests 04:06:34 suzhengwei: Are you facing any issue in addressing unit test review comment? 04:07:26 tpatil: Are you referring to #702328? 04:07:27 Do you mean patch 702328? 04:07:27 https://review.opendev.org/#/c/702328/ - masakari - not wait when instance evacuate error - 3 patch sets 04:07:30 samP: Let me know what you think about exception class name? is InstanceEvacuateError ok with you? 04:08:08 suzhengwei: Yes 04:08:12 To me both is ok. 04:08:18 samP: #726761, ok, thank you 04:09:13 But the unit test, I have no idea how to test the inner function in a function. 04:10:34 suzhengwei: Any other option to test your changes? I will see if I can find a solution to this problem 04:12:15 Both are OK to me, However, InstanceEvacuationFailed make more sense 04:12:30 The current unit test is my only option. If you have better test, please free to update it. 04:12:50 samP: Or InstanceEvacuationError is also fine to me 04:12:53 ok, I will use InstanceEvacuationFailed. 04:13:36 suzhengwei: Ok, will update the unit tests if I can find a better option 04:13:53 tpatil: thank you. 04:14:35 suzhengwei: tpatil: thanks 04:16:57 Other than that, code LGTM. 04:17:22 ^ #726761 04:18:46 tpatil: is this OK to merge? 04:18:48 https://review.opendev.org/#/c/728104/ 04:18:48 patch 728104 - masakari (stable/ussuri) - Monkey patch original current_thread _active - 1 patch set 04:19:24 Yes, I have confirmed the changes by applying the patch as well 04:19:40 tpatil: OK then, Thanks 04:19:43 728104 is common issue, we can merge it. 04:20:21 suzhengwei: agree, I have seen same changes are also applied in other OpenStack services. 04:24:06 tpatil: would you like to discuss about the new API changes? 04:24:17 I mean host API 04:24:32 about Victoria work item: Evacuate non-recovery (’HA_enabled = False’) instances in shutoff status at host failure except specified tenants (tpatil) 04:24:55 samP: you mean: Command parameter support to segment list command to filter out segments based on host input parameter 04:25:05 #link : https://etherpad.opendev.org/p/masakari-victoria-workitems 04:25:26 Please check lines #24-30 04:25:28 tpatil: yes 04:26:39 If we add a new API to solve this problem, the API URI syntax for host APIs would be inconsistent 04:26:49 so I would prefer to implement #1 option. 04:27:41 Need feedback on option #1 and #2 04:27:50 yes, you are correct, #2 seems awkward 04:29:09 suzhengwei_: How about you? which option sounds good to you? 04:29:53 patch : https://review.opendev.org/#/c/707415 04:29:54 patch 707415 - masakari - Add host query parameter support - 4 patch sets 04:30:02 option #1 is implement in above patch 04:30:14 s/implement/implemented 04:30:32 I will check the patch. #1 seems like a good solution. 04:30:39 Thanks! 04:32:55 Any other issues or patches or items need to discuss here? 04:33:42 Spec: https://review.opendev.org/#/c/702427/ 04:33:43 patch 702427 - masakari-specs - Evacuate non-recovery (`HA_enabled = False`) insta... - 9 patch sets 04:34:05 I have replied to suzhengwei comments 04:34:20 samP: Could you please add your comments too? 04:35:14 tpatil: sure, I will review and add my comments. 04:35:28 samP: Thanks 04:35:35 I have draft a proposal for long-term evolution. It metions the HA_enabled. We can disscus next together. 04:36:20 mentions 04:38:01 "long-term evolution", what is this about? 04:38:34 Promotion for host failure workflow. 04:39:15 I will add it to watcher-spec today. Please refer to it if you have time. 04:39:51 OK, thanks. Sure. Would you like to discuss it as next item or in our next meeting? 04:40:16 yes 04:42:59 Another work item: Modify masakari-hostmonitor in order to run it inside container 04:43:19 As discussed in last meeting, I have added few pointer describing the current issue to complete this work item 04:45:04 in etherpad: https://etherpad.opendev.org/p/masakari-victoria-workitems 04:45:20 Lines #16-23 04:46:20 it seems crm itself uses systemctl command and the purpose of modify host-monitor is to remove usage of systemctl 04:48:39 Another issue observed is crm node list results ERROR if pacemaker_remote is installed since it removes 'pacemaker' package 04:48:53 Thanks, we have to find a workaround to work with pacemaker_remote 04:51:14 https://review.opendev.org/#/c/732477/ 04:51:14 patch 732477 - masakari-specs - promotion for large scale hosts failure - 1 patch set 04:51:55 I recommend to take it as a proposal of long-term evolution. 04:52:39 suzhengwei_: thanks. I will check that. we can discuss this on our next meeting 04:52:51 ok 04:54:12 tpatil: I need to look into pacemaker_remote and crm packages. I will try to find some workaround for this. 04:54:43 samP: Thanks 04:56:57 We only have 4 mins left. Please use #openstack-masakari or open-dev ML for further discussion. 04:57:45 Any other items? 04:57:56 samP: no 04:59:05 OK Then, running out of time. Thank you all for attending to today's meeting. 04:59:05 no 04:59:18 suzhengwei_ tpatil: thanks 04:59:23 #endmeeting