12:00:23 <mtembo> #startmeeting watcher
12:00:23 <opendevmeet> Meeting started Thu Jun 26 12:00:23 2025 UTC and is due to finish in 60 minutes.  The chair is mtembo. Information about MeetBot at http://wiki.debian.org/MeetBot.
12:00:23 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
12:00:23 <opendevmeet> The meeting name has been set to 'watcher'
12:00:49 <mtembo> Hello everyone, who's is around today?
12:00:54 <dviroel> o/
12:01:31 <chandankumar> o/
12:02:34 <rlandy> o/
12:03:36 <jgilaber_> o/
12:06:54 <mtembo> Welcome everyone. We will start with todays agenda.
12:07:02 <mtembo> #link https://etherpad.opendev.org/p/watcher-eventlet-removal
12:09:35 <dviroel> mtembo: this is the evenlet removal etherpad, not sure if you want to share the agenda link instead
12:10:01 <mtembo> sorry about that.
12:10:33 <mtembo> Todays agenda.
12:10:42 <mtembo> #link: https://etherpad.opendev.org/p/openstack-watcher-irc-meeting
12:10:46 <dviroel> :)
12:11:49 <mtembo> #Topic:  Call for reviews
12:11:54 <mtembo> #link https://review.opendev.org/c/openstack/watcher/+/951699
12:12:07 <mtembo> #link: https://review.opendev.org/c/openstack/watcher/+/952017
12:12:49 <dviroel> i added both, this is related to the move of some integrations to experimental features
12:13:04 <dviroel> we discussed about that in the past and in the ptg
12:13:13 <dviroel> both are open for reviews
12:13:52 <dviroel> a new doc page is being added to cover the missing details about services integrations (the ones tha are not datasources for instance)
12:14:05 <dviroel> so pls, take some time to review/approve/comment
12:16:05 <dviroel> if there are no comments about that, we can move to the next reviews..
12:16:16 <dviroel> which I also added
12:16:19 <dviroel> the spec ones
12:16:32 <dviroel> #link https://review.opendev.org/c/openstack/watcher-specs/+/947282 (Adds spec for extend compute model attributes)
12:16:32 <mtembo> Spec for open reviews
12:16:37 <mtembo> #link:https://review.opendev.org/c/openstack/watcher-specs/+/947282
12:16:49 <dviroel> this first one is about extending compute model attributes
12:17:10 <dviroel> sean and alfredo already reviewed, I did a small update yesterday
12:17:24 <dviroel> it requires more eyes, but should be close to be merged
12:18:26 <dviroel> ack, next one
12:18:42 <dviroel> #link https://review.opendev.org/c/openstack/watcher-specs/+/949804 (Support multitenancy with Prometheus datasource)
12:19:06 <dviroel> i am reviewing, sean is also reviewing this one
12:19:45 <dviroel> it is good for all to take a look, on how prometheus datasource communication will happen when we have Aethos service as reverse-proxy
12:19:58 <dviroel> please add  your comments and concerns there
12:22:08 <dviroel> ok
12:22:14 <mtembo> Is there anything anyone would like to add on tho this before we move on
12:22:16 <mtembo> ?
12:23:09 <dviroel> no, lets continue
12:23:25 <mtembo> Moving on to the next topic
12:23:33 <mtembo> #topic: Eventlet removal updates
12:23:34 <mtembo> #link: https://etherpad.opendev.org/p/watcher-eventlet-removal
12:23:44 <dviroel> some updates on this topic
12:24:02 <dviroel> if you go to the evenlet removal etherpad, line 70
12:24:12 <dviroel> i added some comments about latest  issue found
12:24:25 <dviroel> there is a problem, with watcher, when moving to the new threading backend of oslo.service
12:25:11 <dviroel> since decision-engine starts 2 services within the same service manager, when moving to the threading backend this will create 2 processes
12:25:26 <dviroel> the problem is that both services access the same in-memory data model, which can't be shared across 2 process
12:25:42 <dviroel> one possible solution is to merge both services, like in:
12:25:51 <dviroel> #link https://review.opendev.org/c/openstack/watcher/+/952499
12:26:04 <dviroel> nd get everything in a single process
12:26:11 <dviroel> i am also testing it in a WIP chnage, where we disable monkey_patching and enable oslo.service threading backend
12:26:19 <dviroel> this second patch is still WIP, but already give us some results:
12:26:31 <dviroel> #link https://review.opendev.org/c/openstack/watcher/+/952257
12:26:59 <dviroel> this second patch adds a new job that disables monkey-patching in the decision-engine
12:27:17 <dviroel> going to work more in the second patch
12:27:36 <dviroel> but the first one is already open for review/comments/concerns
12:28:53 <dviroel> any comments or concerns?
12:28:53 <mtembo> Thank you dviroel.
12:30:16 <dviroel> thanks mtembo, I think that we can move on then
12:30:19 <mtembo> #topic: Priority major tech debt to be worked on in upcoming weeks
12:30:42 <rlandy> thanks mtembo
12:30:49 <mtembo> welcome rlandy
12:31:10 <rlandy> I know there are still on going discussions but I thought we could mention some main areas of tech debt we plan to address
12:31:25 <rlandy> high level
12:31:50 <rlandy> looks like there is agreement to:
12:32:01 <rlandy> run code coverage and asses dead code
12:32:45 <rlandy> fix some high level bugs
12:32:58 <rlandy> and
12:33:19 <rlandy> address some other areas of concern:
12:33:50 <rlandy> like API validation and improve testing for the non-happy path
12:34:03 <rlandy> any major other area anyone would like to highlight?
12:34:26 <chandankumar> rlandy: adding watcher intree functional tests
12:34:58 <chandankumar> like https://github.com/openstack/nova/tree/master/nova/tests/functional
12:35:30 <chandankumar> It will be much needed for reproducing bugs and regression.
12:36:03 <chandankumar> Another area we missed is moving from service client to openstacksdk.
12:36:10 <dviroel> +1 ^
12:36:20 <chandankumar> that's it which I can remember.
12:36:37 <dviroel> another one that was filed
12:36:48 <dviroel> #link  https://bugs.launchpad.net/watcher/+bug/2111100 Migrate JSON Schemas to a newer version
12:36:59 <dviroel> which can be considered too
12:38:05 <chandankumar> yes
12:38:15 <dviroel> there is other one, that may require more discussions or refactoring:
12:38:15 <dviroel> #link https://bugs.launchpad.net/watcher/+bug/2111021 Replace DB polling method for checking cancelled actions/action plan
12:38:26 <dviroel> but not so high priority at this point
12:39:40 <chandankumar> rlandy: one more thing I just remembered we need a doc around explaining audit template scope relation with audit.
12:39:59 <chandankumar> I will raise a bug for the same.
12:40:39 <rlandy> thank you for those additions
12:41:54 <mtembo> Do we have any more contributions here ?
12:42:21 <rlandy> thanks mtembo - we can move on
12:42:35 <mtembo> Moving on the next topic
12:42:45 <mtembo> #topic: Bug Triage
12:43:00 <mtembo> #link: https://bugs.launchpad.net/watcher/+bug/2111100
12:43:26 <dviroel> this is the outdated json schema bug
12:43:39 <dviroel> we can triage this one
12:43:57 <dviroel> there are links to where the code is using the old schema validation
12:44:36 <dviroel> i think that we could set it to Medium importance, not sure what others think
12:44:57 <chandankumar> +1 to medium
12:45:45 <mtembo> thank. status moved to triaged
12:46:27 <mtembo> importance set to medium
12:46:30 <dviroel> tks
12:46:58 <mtembo> moving on.
12:46:59 <mtembo> link: https://bugs.launchpad.net/watcher/+bug/2111021
12:47:20 <mtembo> #link: https://bugs.launchpad.net/watcher/+bug/2111021
12:47:30 <dviroel> this is the improvement mentioned before
12:48:51 <dviroel> we can triage too, not sure regards the importance yet
12:49:17 <dviroel> we can set as Wishlist I think
12:49:26 <dviroel> since it is not a real bug .. yet
12:50:09 * dviroel at least we didn't find any issue with current implementation so far :)
12:50:17 <jgilaber_> dviroel: this will still work as it is with the evenlet changes right?
12:51:02 <dviroel> jgilaber_: not sure yet, i didn't start to work on the applier yet
12:51:11 <dviroel> there is code there that needs to be replace, for sure
12:51:19 <dviroel> like the way that it spawn threads
12:51:36 <dviroel> but the looping and checking the db state, not sure if we really need to touch there
12:52:12 <dviroel> i probably can answer you with more details
12:52:22 <dviroel> once i actually start to change there :/
12:53:00 <jgilaber_> sure, I just wanted to point out, that even if it's something important to improve, it's not critical right now
12:53:39 <dviroel> ack, tks jgilaber_
12:54:05 <mtembo> Thank you. The bug has been updated to to triaged and to wishlist. Moving on to the last bug.
12:54:22 <mtembo> #link: https://bugs.launchpad.net/watcher/+bug/2115058
12:55:49 <dviroel> "if the 'internal' availability zone does not have any storage pools, then the storage model will list all existing volumes, instead of no volumes"
12:55:56 <dviroel> ouch
12:56:08 <jgilaber_> yep, it's a corner case
12:56:23 <dviroel> for sure, not the expected behavior
12:56:34 <jgilaber_> I was playing with the audit scope while testing something else and I noticed this
12:57:04 <jgilaber_> it didn't look to me like it would be a really complex fix, but I think it's an important issue to fix
12:57:19 <dviroel> +1
12:57:39 <jgilaber_> so I'd say it's a high level
12:57:46 <dviroel> me too
12:58:29 <dviroel> it would produce a wrong action plan, and there is no other way to solve the issue without the fix
12:59:37 <mtembo> Bug set as traiged and priority high.
12:59:38 <mtembo> #topic Chair for next meeting 3rd July 2025
13:00:11 <dviroel> I can be the next one
13:00:33 * dviroel checks if is not a holiday here
13:00:44 <dviroel> yeah, Jul 3rd is fine
13:00:58 <mtembo> Thank you dviroel for taking it up.
13:01:01 <mtembo> thank you everyone for attending.
13:01:06 <dviroel> tks mtembo o/
13:01:25 <mtembo> #endmeeting
13:01:30 <jgilaber_> thanks mtembo!
13:02:21 <mtembo> #endmeeting