03:01:14 #startmeeting Masakari 03:01:15 Meeting started Tue Jun 26 03:01:14 2018 UTC and is due to finish in 60 minutes. The chair is tpatil. Information about MeetBot at http://wiki.debian.org/MeetBot. 03:01:16 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 03:01:18 The meeting name has been set to 'masakari' 03:01:36 Hi All 03:01:47 sorry Im late 03:01:55 Hi 03:02:02 Hi samP 03:02:15 Please take the pilot seat 03:02:15 not started yet, right? 03:02:28 tpatil: sure.. 03:02:32 already started 03:02:43 ah.. yes thanks 03:02:49 OK then, 03:03:25 #topic bugs and patches 03:04:11 any bugs or patches need to discuss? 03:04:25 Louie Kwan has uploaded a new PS 03:04:38 I'm busy reviewing other patches. Need help in reviewing his patch 03:04:46 #link https://review.openstack.org/#/c/534958/ 03:04:47 patch 534958 - masakari-monitors - Introspective Instance Monitoring through QEMU Gue... 03:04:59 tpatil: great.,, I will take a look on that 03:05:05 tpatil: thanks for the review 03:05:16 Thank you 03:05:35 I have proposed a fix for LP bug https://bugs.launchpad.net/masakari/+bug/1773132 03:05:35 Launchpad bug 1773132 in masakari "masakari-engine runs recovery twice for one notification when disconnection with rabbitmq" [Undecided,In progress] - Assigned to Tushar Patil (tpatil) 03:05:48 #link https://review.openstack.org/#/c/576042/ 03:05:49 patch 576042 - masakari - Avoid recovery from failure twice 03:05:56 Please take a look at this patch as well 03:06:31 tpatil: sure 03:06:49 tpatil: thanks for the fix 03:08:02 Other LP bug https://bugs.launchpad.net/masakari/+bug/1773765 03:08:03 Launchpad bug 1773765 in masakari "There is a possibility that 'running' notification will remain" [Undecided,New] 03:08:20 It is little tricky to fix this issue becoz we don't have much information about which instances are already evacuated 03:09:06 Masakari doesn't save any info about which instances are evacuated and how many are still pending during host failure notification 03:09:24 hence it's difficult to re-execute running status notification 03:11:15 humm...I think we can store some of these info in masakari, and with current status form nova, we are able to calculate it. 03:11:40 But, not sure is this the best way to deal with. 03:12:01 yes, it's a big change 03:12:29 Can we focus on this issue in next release? 03:13:07 If we decide to maintain basic info in masakari, we can also add code to show the current progress of host failure notification 03:13:46 like source host failure, out of 10 instances, 5th instance evacuation is in progress 03:13:47 tpatil: Indeed. 03:14:41 toabctl: IIRC, nova does have api for get the status for migration, 03:17:20 samP: Yes, but Masakari will still need to maintain basic info to query status from nova 03:18:09 tpatil: agree. In any cause, it seems like this fix is too heavy for this release. 03:18:23 samP: Correct 03:19:10 Let's postpone it to next release. Otherwise, some one raise urgent flag..! 03:20:42 any other bugs or patches? 03:21:07 One minor issue 03:21:11 #link https://review.openstack.org/#/c/574614/ 03:21:12 patch 574614 - masakari - Segment description allows multiline characters 03:21:28 Description doesn't allow multiline characters 03:21:52 I will review this patch today 03:23:19 LGTM 03:24:02 need to rebase 03:24:19 thanks for the fix 03:24:39 Ok 03:25:11 I have voted + 2 on this patch https://review.openstack.org/#/c/567825/ 03:25:12 patch 567825 - masakari - Enable mutable config in Masakari 03:25:40 Need someone to review and approve this patch 03:26:17 sure, I will take a look 03:26:21 tpatil: thanks 03:27:50 Let's move to status update for features and sub-projects 03:28:13 (1) Horizon Plugin 03:28:46 I have reviewed 3-4 patches and posted comments 03:29:39 I want to review, but it will take some time to get there...;) 03:29:47 ok 03:29:50 but I will be their soon 03:30:24 tpatil: thanks for reviews 03:30:44 (2) recovery method customization 03:31:01 specs to be reviewed 03:31:27 I read the spec. I will put my comments soon, No critical comments, basically LGTM 03:31:40 Implementation is almost finished. Working on adding unit tests. Expect to push patch in next week 03:31:51 tpatil: sure, thanks a lot 03:32:46 (3) Ansible support for Masakari 03:33:13 Any issues? 03:33:19 This task is currently on hold due to unavailability of dev. environment 03:33:39 we had discussed about this issue in last-to-last meeting 03:33:40 tpatil: ah, yes. my fault 03:34:48 I will try to provide server init method asap 03:34:49 Still many items are pending. Looks like we will need to postpone it to the next release 03:34:59 samP: Thanks 03:36:10 tpatil: are you referring to ansible? 03:36:30 yes 03:37:34 tpatil: monitor part is bit difficult and I think it is OK to postpone it to next release 03:38:28 samP: Agree 03:39:13 if some one came up with a quick solution, then we can reconsider to put it into to this release. 03:39:27 ok 03:40:09 Otherwise, let's consider it on next release 03:40:19 #topic AOB 03:40:28 I want to bring up this issue report by Torin on openstack ML 03:40:30 #link http://lists.openstack.org/pipermail/openstack/2018-June/046620.html 03:40:40 s/report/reported 03:41:56 execute method is failing with 124 error 03:42:08 humm.... seems like Corosync faild to sync 03:43:08 timeout returns 124 error, but if you pass --preserve-status option to timeout it return 0 03:44:06 In that case, I don't think it will raise ProcessExecutionError 03:45:06 Do we expect host monitor process to be run by superuser? 03:46:12 becoz run_as_root=True is passed to utils.execute method 03:46:13 https://github.com/openstack/masakari-monitors/blob/master/masakarimonitors/hostmonitor/host_handler/handle_host.py#L113 03:47:36 tpatil: case here is, tcpdump can not be run by the normal user. 03:49:03 If you have any info about this issue, can you please reply to the above ML thread 03:49:21 tpatil: sure, I will join to discussion. 03:49:53 I need to read the thread carefully. I will reply to thread 03:50:20 tpatil: thanks for answering to thread 03:51:07 ah, and this is openstack ML, not the dev 03:51:30 yes 03:51:43 OK got it. 03:52:02 After upgrading to Queens, Torin is not able to run host monitor process 03:52:21 That is critical. 03:52:50 tpatil: I will take a look on this issue. 03:53:27 Thanks 03:54:15 Any other items to discuss? 03:54:34 Otherwise, let finish today's meeting. 03:54:35 That's all from my end for now 03:55:18 Please use openstack-dev ML or IRC #openstack-masakari on freenode for further discussion.. 03:55:31 thank you for attending today's meeting 03:55:50 tpatil: thanks 03:55:56 #endmeeting 03:56:01 Thank you, Bye 03:56:09 ah, you have to end it.. 03:56:19 yes will do 03:56:22 #endmeeting