Thursday, 2026-03-05

chandankumarsean-k-mooney: dviroel Hello, Please have a look at https://review.opendev.org/c/openstack/watcher/+/963857, 3 watcher-tempest-plugin patches related to zone migrations depends on it. thank you!07:46
dviroelhi all, watcher meeting will start in 15m, pls add your topics to our meeting agenda:11:45
dviroelhttps://etherpad.opendev.org/p/openstack-watcher-irc-meeting#L2911:46
dviroel#startmeeting watcher12:00
opendevmeetMeeting started Thu Mar  5 12:00:10 2026 UTC and is due to finish in 60 minutes.  The chair is dviroel. Information about MeetBot at http://wiki.debian.org/MeetBot.12:00
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.12:00
opendevmeetThe meeting name has been set to 'watcher'12:00
dviroelhi o/ - who's around for today's meeting?12:00
jgilabero/12:00
rlandyo/12:01
amoralejo/12:01
dviroelcourtesy ping: sean-k-mooney chandankumar12:02
dviroelso let's start with today's meeting agenda12:02
dviroel#link https://etherpad.opendev.org/p/openstack-watcher-irc-meeting#L2912:02
dviroelas usual, feel free to add your own topics to the agenda12:02
dviroelthere is a Review topic to place your changes that requires attention12:02
dviroelthere is also a topic for Bugs, if you want to discuss any12:03
dviroellets start from the top12:03
dviroel#Announcements12:03
dviroel#undo12:03
opendevmeetRemoving item from minutes: #link https://etherpad.opendev.org/p/openstack-watcher-irc-meeting#L2912:03
dviroelah12:03
dviroel#topic Announcements12:03
dviroelok, so here is again the meeting agenda:12:04
dviroel#link https://etherpad.opendev.org/p/openstack-watcher-irc-meeting#L2912:04
dviroelfirst announcement12:04
dviroellets look at Gazpacho schedule12:04
dviroel#link https://releases.openstack.org/gazpacho/schedule.html12:04
dviroelwe should be in preparation for RC1, which is target to next week12:05
chandankumaro/12:06
dviroelso for now we should focus on reviewing and merging important/critical fixes that we want include in this release12:06
dviroelit seems that we already have a list of bugs in the Review topic, so we can look them in a few12:06
dviroels/bugs/patches12:07
dviroelIt is also the week to get our release highlights merged in releases repo12:07
dviroel#link https://review.opendev.org/c/openstack/releases/+/97887812:07
dviroelthat not the right one12:08
dviroel#undo12:08
opendevmeetRemoving item from minutes: #link https://review.opendev.org/c/openstack/releases/+/97887812:08
jgilaber#link https://review.opendev.org/c/openstack/releases/+/97601612:08
jgilaber^^ that is the watcher patch12:08
dviroelah thanks jgilaber 12:08
dviroelplease take a look and review, we need this merge this week12:09
dviroeli would ask our release liasions to vote once they agree12:09
dviroelit lgtm 12:09
dviroeland for the recors, we also merge our release prelude12:10
dviroel#link https://docs.openstack.org/releasenotes/watcher/unreleased.html#prelude12:10
dviroelthanks rlandy for proposing both patches12:11
amoralejwould https://blueprints.launchpad.net/watcher/+spec/skip-actions-in-pre-condition deserves a mention?12:11
dviroelamoralej: i think that is valid yes, it is a important feature 12:11
amoralejthat's user visible change12:11
amoralejalso, i'd mention, now we can safely run multiple instances of the decision-engine and applier (because of the service monitors)12:12
amoralejI'll comment12:13
dviroelamoralej: yeah, please add a comment in the patch12:13
dviroelwe can see if rlandy can update that for us12:13
dviroelother we can prpose a new PS12:13
dviroelthank amoralej++12:13
dviroelone more announcement in the list 12:14
rlandywill do12:14
dviroelPTG will be handled virtually on Apr 20-2412:15
dviroelmake your registration if you plan to join us:12:15
dviroel#link https://openinfra.org/ptg/12:15
dviroelI also created a new etherpad for the PTG, so we can start adding topics that we plan to discuss12:16
dviroel#link https://etherpad.opendev.org/p/watcher-2026.2-ptg12:16
dviroelin the following weeks we can start proposing the dates that we want for watcher, and we may want to avoid conflicts with other projects (if possible) like cyborg and nova12:17
dviroelany other announcement before we continue?12:17
sean-k-mooneycyborg will have ptg session too but im expecting 1 cross project sesssion with nova12:17
sean-k-mooneyi dotn know if we need one with watcher12:18
sean-k-mooneybtu we can seee if there are any topics that cross over12:18
sean-k-mooneyat some point watcher will need awarenes of instnace with cybrog devices12:18
sean-k-mooneyas you canot currently live migrate tehm12:18
dviroeli was more worried about conflict since most of the people would like to join both12:18
sean-k-mooneyso it will have to fall back to cold migration12:18
rlandyare there any other cross over teams?12:19
rlandyobservability?12:19
rlandywhere we want a combined session12:19
sean-k-mooneythe telemetry team tend to have a very short ptg12:19
sean-k-mooneyusually on one day for 2-4 hours12:20
dviroelyeah, maybe with observability to discuss about the future of datasources and possibly new metrics for watcher12:20
sean-k-mooneybut yes they reached out12:20
dviroelbut I don't see yet a cross session with cyborg, i don't want to overlap only12:20
dviroelboth project sholdn't need the full week :)12:21
dviroelfow now, we can start adding the topic to the etherpad and see how much we plan to discuss too12:21
dviroelany other comment or annoucement?12:22
dviroel#topic Reviews12:23
dviroelI see that we have some reviews open that are candidates for this releae12:23
dviroelwant to highlight them jgilaber ?12:23
dviroel#link https://review.opendev.org/c/openstack/watcher/+/95876612:23
jgilaberyes, thanks dviroel 12:24
dviroel( Remove watcher_notifications from default value) 12:24
jgilaberI think the first one is a good candidate since it's a trivial fix,although not terribly important12:24
dviroelthe 'openstack.notifications' was added as default in previous release right12:25
jgilaberyes12:25
amoralejyes, it should be safe12:26
jgilaberthe next three I would be ok with posponing them until we get the stable branch created, but I would like to get them in soon to enable volume migration testing12:26
dviroeli think that we are fine with the first one too12:26
amoralejIt'd be good to the zone_migration ones in this cycle imo, otherwise, we should backport them as those are bug fixes12:26
jgilaberand the last one12:27
jgilaber#link https://review.opendev.org/c/openstack/watcher/+/97807712:28
jgilaberI discovered the bug while working on the openstacksdk changes for nova_helper12:28
jgilaberit's a pre-existing bug but I think it would be good to get the fix12:28
dviroelthe zone_migrations  one I need to revisit, but it is a good fix to get into the release yes, and it open for a while now, but I want to look with more details12:29
dviroelthe last one is an important fix too12:29
dviroelI will take a look too12:30
jgilaberthanks dviroel! that all from me12:30
dviroeland there is one more12:31
dviroel#link https://review.opendev.org/c/openstack/watcher/+/978077 (Handle missing fields building storage model XML or list)12:31
* dviroel is slow today, maybe needs more coffee12:31
amoralejyes, i added that, it fixes low bug, we can defer to next release although it's pretty trivial change12:32
* dviroel fixing link again12:32
dviroel#undo12:32
opendevmeetRemoving item from minutes: #link https://review.opendev.org/c/openstack/watcher/+/97807712:32
dviroel#link https://review.opendev.org/c/openstack/watcher/+/960265 Handle missing fields building storage model XML or list)12:32
dviroelyeah, we can focus in the other from the list for now12:33
dviroelthanks for proposing this one amoralej 12:33
amoralejsure, np12:33
dviroelI will also take a look on https://review.opendev.org/q/project:openstack/watcher+status:open+branch:+master12:34
dviroelto see if we are missing something, but I think that's it12:35
dviroelany other reviews that you folks want to cover?12:35
dviroel#topic Bugs12:36
dviroelquick looking at LP bugs, there is one missing triage12:36
dviroel#link https://bugs.launchpad.net/watcher/+bug/214221912:37
jgilaberyes I opened this one recently12:37
dviroelas mentioned in the bug, same issue from zone_migration right12:37
jgilaberexactly12:37
dviroelwhen dealing with many instances/volumes this can turn to an additional problem12:37
jgilaberthe storage capacity balance strategy has been tested quite less than zone migration12:38
jgilaberso we missed the problem12:38
dviroelthe zone_migration related bug is12:38
amoralejdid you check if we have what we need in the storagemodel to leverage on it instead of cinder api?12:38
dviroel#link https://bugs.launchpad.net/watcher/+bug/211995712:38
jgilaberamoralej, I did not check that12:38
jgilaberI submitted the report quickly so we have a tracker12:39
dviroelit is important to check that before removing yes12:39
amoralejjust in case we have to extend the model for the migration12:39
sean-k-mooneywe are currently not reporting the voluem info in the api12:39
sean-k-mooneyso addign info for volumes wont affect it12:40
jgilaberit gets pools, snapshots and volume types12:40
sean-k-mooneyas i noted before i dont think we shoudl continue to expose teh model in teh api12:40
jgilaberI don't think we have information about snapshots and volume types in the model12:40
sean-k-mooneyso even if we add new values for the ocmptue side im not sure we shoudl include them in teh api respocne12:40
dviroelin last ptg we decided to freeze the datamodel api12:40
sean-k-mooneyyep12:40
amoraleji didn't mean to get the info in the api but in the model itself12:41
jgilabernote that in this strategy it only uses the storage model12:41
dviroelbut I didn't get a chance to move forward with this, but we should before adding any new attribute in compute model12:41
amoralejso that the strategy has all the info required12:41
dviroeljgilaber: yeah, it a problem only in compute model12:41
sean-k-mooneyi still also belvie we shoudl be moving the model out of memory and into a graph db or sql db sonner rather then later12:41
sean-k-mooneywe likely can add these fiels btu as it stands each desiion engione will have its own model in memory12:42
sean-k-mooneyso they can have diffent views of the cloud12:42
amoralejyep, that's a good discussion, but a different one :)12:42
dviroelcorrect, maybe a good discussion for the next PTg12:42
sean-k-mooneywell i bring it up mainly becasue fo size. i think we can likely add the reqruied fiels in thsi case12:43
dviroel+112:43
sean-k-mooneybut we will evenutlly have issues with the size of hte model in the curernt pattern12:43
amoraleji finally could get real size data for the model, so that we can discuss in the ptg12:43
sean-k-mooneyack12:43
dviroelamoralej: nice thanks12:44
dviroelso for the bug, i would set to medium, as we did for the zone_migration one12:44
amoralej+112:44
jgilaber+112:44
amoralejand tags: decision-engine12:45
jgilaberand strategies12:45
amoralejright12:46
dviroelgood point :) 12:46
dviroelanything else to cove in this topic?12:47
jgilabernot from me12:47
dviroelok, so we can move12:48
dviroel#topic Open Discussions12:48
dviroelanything that yo folks want to bring?12:48
dviroelI just want to highlight that 2024.2 jobs are failings12:48
dviroeli rechecked a 2024.2 ptach yesterday12:49
dviroelhttps://review.opendev.org/c/openstack/watcher/+/950264/1#message-af40543ff5aef0f527d1ec235f85368ab8dd236a12:49
dviroelit seems that most of them are failing with "No module named 'pkg_resources'"12:49
jgilaberthe problem seems to be coming from keystonemiddleware12:50
dviroelso we would have to manually pin setuptools like we did for 2025.2?12:50
sean-k-mooneyno12:51
sean-k-mooneywe can do that for the docs job but not the rest in general12:51
sean-k-mooneythis need to be fixed in the midelware package with an actully release on stable 2024.212:52
sean-k-mooneyand htat need to happen vai the requiremetn repo with a uc bump12:52
amoralej2024.2 is moving to EoL next month btw12:52
sean-k-mooneyyep but there will be final release before then12:53
dviroelsean-k-mooney: is there someone working on that, do you know?12:53
amoralejthere are ongoing fixes in requirements and middleware repo?12:53
amoralejyeah, that was my question actually12:53
sean-k-mooneydviroel: im not sure we were backporting the tempest pin but im unclear about the tox jobs12:53
sean-k-mooneyamoralej: yes there are12:53
sean-k-mooneywe have been trying to fix this in the sable btrnaches for 4-6 weeks12:54
sean-k-mooneywe shoudl follow up in the qa or requirement channel12:54
dviroelack, we have some backports in 2024.2 that we could merge before the final release..12:55
dviroelI can follow up there and see where we are12:55
amoralejyep, that'd be good12:55
sean-k-mooneyhttps://review.opendev.org/c/openstack/requirements/+/976602 might fix it12:55
dviroelack, still not passing: https://review.opendev.org/c/openstack/requirements/+/976602?tab=change-view-tab-header-zuul-results-summary12:57
dviroelbut lets see how it goes12:57
dviroelthanks for the updates sean-k-mooney 12:57
dviroelanything else that you folks want to discuss in the last 3 min?12:58
sean-k-mooneyhttps://review.opendev.org/c/openstack/requirements/+/976903 need to be cherry picked first i think to 2024.212:58
dviroeloh, interesting12:58
sean-k-mooneyya we need to fix the requriement repo to not depend on pkg_resouce first12:58
sean-k-mooneylets follow up with the requirements folks in there channel12:59
sean-k-mooneyand see what the plan is12:59
sean-k-mooneywe can do that seperatly12:59
dviroelsean-k-mooney++12:59
dviroeloh last minute12:59
dviroel#topic Volunteers to chair next meeting12:59
dviroelsomeone want to chair next week?13:00
dviroelotherwise I can take it13:00
jgilaberI can do it dviroel 13:00
dviroelnice, thanks jgilaber 13:00
dviroelok let's wrap up for today13:01
dviroelwe will meet again next week13:01
dviroelthank you all for participating13:01
dviroel#endmeeting13:01
opendevmeetMeeting ended Thu Mar  5 13:01:20 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)13:01
opendevmeetMinutes:        https://meetings.opendev.org/meetings/watcher/2026/watcher.2026-03-05-12.00.html13:01
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/watcher/2026/watcher.2026-03-05-12.00.txt13:01
opendevmeetLog:            https://meetings.opendev.org/meetings/watcher/2026/watcher.2026-03-05-12.00.log.html13:01
chandankumarthanks dviroel !13:01
amoralejthanks dviroel++ 13:02
opendevreviewJoan Gilabert proposed openstack/watcher master: Fix check for volume type in cinder helper migrate  https://review.opendev.org/c/openstack/watcher/+/96385716:15
opendevreviewJoan Gilabert proposed openstack/watcher master: Fix small issues in cinder_helper unit test  https://review.opendev.org/c/openstack/watcher/+/97902916:15
opendevreviewMerged openstack/watcher master: Add tests for get_dst_pool_and_type method  https://review.opendev.org/c/openstack/watcher/+/96471716:18
opendevreviewJoan Gilabert proposed openstack/watcher master: Fix zone migration dst pool and type selection  https://review.opendev.org/c/openstack/watcher/+/96471816:28
jgilaberdviroel, sean-k-mooney thanks for the reviews! I've respined the zone_migration fix to address your comments16:30
sean-k-mooneyhttps://review.opendev.org/c/openstack/watcher/+/964718/9..10/watcher/tests/unit/decision_engine/strategy/strategies/test_zone_migration.py#157116:31
sean-k-mooneyshoudl that not have been self.assertEquial as you had in the regession test?16:32
jgilaberno that is correct. In the patch I changed a bit some of the tests and that one covers a case with two inputs for the same pool but different destinations16:38
jgilaberand the matching input does not specify the dst_type so it's None16:38
sean-k-mooneyack16:38
rlandyhttps://review.opendev.org/c/openstack/releases/+/976016 - updated16:50
opendevreviewMerged openstack/watcher master: Remove watcher_notifications from default value  https://review.opendev.org/c/openstack/watcher/+/95876618:52
opendevreviewMerged openstack/watcher master: Fix check for volume type in cinder helper migrate  https://review.opendev.org/c/openstack/watcher/+/96385719:30
opendevreviewMerged openstack/watcher master: Fix zone migration dst pool and type selection  https://review.opendev.org/c/openstack/watcher/+/96471819:57
opendevreviewMerged openstack/watcher master: Fix small issues in cinder_helper unit test  https://review.opendev.org/c/openstack/watcher/+/97902919:57

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!