04:00:14 <tpatil> #startmeeting Masakari
04:00:15 <openstack> Meeting started Tue Mar 10 04:00:14 2020 UTC and is due to finish in 60 minutes.  The chair is tpatil. Information about MeetBot at http://wiki.debian.org/MeetBot.
04:00:16 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
04:00:18 <openstack> The meeting name has been set to 'masakari'
04:00:24 <tpatil> Hello Masakari members
04:00:44 <tpatil> Today Sampath san is joining late
04:00:53 <tpatil> Roll call?
04:02:32 <suzhengwei_> tpatil: hi
04:02:47 <tpatil> suzhengwei_: Hi
04:03:00 <tpatil> #topic Critical Bugs
04:03:54 <tpatil> #link : https://bugs.launchpad.net/masakari/+bug/1858762
04:03:56 <openstack> Launchpad bug 1858762 in masakari "“UNKNOWN” host_status notification may cause unsafe evacuation" [Critical,Confirmed] - Assigned to Shilpa Devharakar (shilpasd)
04:03:58 <tpatil> Any update?
04:06:07 <suzhengwei_> Is Shilpa here?
04:06:17 <tpatil> No
04:06:40 <tpatil> This bug seems like a simple fix, need to check the host status and execute the workflow
04:06:55 <tpatil> Moving ahead
04:07:19 <tpatil> Any other bugs you want to bring up for discussion now?
04:07:48 <suzhengwei_> https://review.opendev.org/#/c/708325/
04:09:12 <tpatil> suzhengwei_: I have checked your comment and it does  makes sense to me
04:10:02 <tpatil> I have discussed about it with Sampath san and looking forward for his comment on the same.
04:10:42 <suzhengwei_> thx. another one, https://review.opendev.org/#/c/702328/
04:11:00 <suzhengwei_> it need some review
04:12:20 <tpatil> Regarding the first patch, could you please add unit tests for your changes? I  know the changes are simple, but we need unit tests for bug fixes
04:13:07 <suzhengwei_> ok, I will try to add some unit tests later.
04:13:19 <tpatil> suzhengwei_: Thanks
04:13:43 <suzhengwei_> https://review.opendev.org/#/c/701489/
04:13:59 <tpatil> https://review.opendev.org/#/c/702328/, I will review this patch in this week. Sorry lately I'm busy working on another project so I couldn't spend time on this project.
04:14:34 <suzhengwei_> This is a simple bug, too
04:15:19 <tpatil> suzhengwei_: Agree, will review this patch as well in this week.
04:15:32 <suzhengwei_> thx
04:16:38 <tpatil> #topic: Ussuri work items
04:16:43 <tpatil> #link : https://etherpad.openstack.org/p/masakari-u-workitems
04:17:27 <tpatil> Enable/Disable evacuation segment wise
04:18:02 <tpatil> Spec: https://review.opendev.org/#/c/705893/
04:19:03 <tpatil> suzhengwei_: I have seen you have uploaded a new PS. Will review it soon.
04:19:26 <suzhengwei_> thx
04:20:30 <tpatil> suzhengwei_: You also have implemented the spec, so we will plan to plan this feature in U release.
04:20:42 <tpatil> s/plan/land
04:21:15 <suzhengwei_> yes, I will finish it before U release.
04:22:26 <tpatil> suzhengwei_: Thanks
04:22:44 <tpatil> # link: https://review.opendev.org/#/c/704811/
04:23:42 <tpatil> suzhengwei_: I have seen your comment on above patch. Also, discussed with Sampath san offline. Unfortunately, he is not available today yet
04:24:19 <tpatil> IMHO, it's ok to add a new API GET /hosts and GET /hosts/<hostuuid>
04:24:54 <suzhengwei_> You can leave some comment or message under the patch.
04:25:32 <tpatil> Question arise, whether  it's ok to continue existing API GET /segments/{segment_id}/hosts
04:25:43 <tpatil> or deprecated it
04:26:31 <suzhengwei_> Maybe we can talk about it sometime later, when more people are here.
04:26:57 <tpatil> because supporting both these API would be redundant
04:27:19 <tpatil> suzhengwei_: Agree
04:28:00 <tpatil> suzhengwei_ : Do you have anything else to discuss about U work items?
04:29:00 <suzhengwei_> https://review.opendev.org/#/c/710191/
04:30:14 <suzhengwei_> this is a simple issue, it is not a bp. I just want to draw some notices.
04:31:15 <tpatil> From masakari point of view, I think what's important is nova-compute process and for the rest of the processes, masakari doesn't take any actions
04:32:13 <suzhengwei_> yes, but it will try to restart the failed processes in monitor.
04:32:56 <tpatil> The current code is not suitable if you are monitoring too many processes.
04:33:06 <suzhengwei_> it is not good in line flow when two or more process is failed.
04:33:35 <tpatil> I don't think you need a spec for this change
04:34:10 <suzhengwei_> No, it doesn't.
04:35:09 <tpatil> You can allocate a pool of thread to do the job. definitely we cannot create one thread per process. just  in case operator adds too many processes for monitoring.
04:35:20 <tpatil> s/thread/threads
04:36:40 <tpatil> suzhengwei_: Feel free to work on these changes. No bp/spec need for this change.
04:36:52 <tpatil> s/need/needed
04:37:25 <suzhengwei_> The monitor process list contain nova-compute, libvirt and ovs in my env.
04:38:11 <suzhengwei_> OK, I will give a patch directly.
04:38:36 <tpatil> suzhengwei_: Thanks
04:41:49 <tpatil> That's all I wanted to bring up in this meeting
04:42:04 <tpatil> If no other topics left for discussion let's close this meeting early
04:43:51 <tpatil> suzhengwei_: Thank you for joining the meeting
04:44:31 <tpatil> See you all then in the next meeting. Thank you
04:44:42 <tpatil> #endmeeting