Thursday, 2025-07-10

« Wednesday, 2025-07-09 Index
opendevreviewHo Minh Quang Ngo proposed openstack/watcher-tempest-plugin master: Add tests for disable migration in host maintenance  https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/95421400:29
opendevreviewJoan Gilabert proposed openstack/watcher master: Enable storage model collector by default  https://review.opendev.org/c/openstack/watcher/+/95132310:35
amoralejwatcher meeting is in 20 minutes, remember to add your topics to the agenda https://etherpad.opendev.org/p/openstack-watcher-irc-meeting11:40
amoralejit's meeting time!12:00
dviroel\o/12:01
chandankumaro/12:01
amoralej#startmeeting watcher meeting - 10-July-202512:01
opendevmeetMeeting started Thu Jul 10 12:01:33 2025 UTC and is due to finish in 60 minutes.  The chair is amoralej. Information about MeetBot at http://wiki.debian.org/MeetBot.12:01
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.12:01
opendevmeetThe meeting name has been set to 'watcher_meeting___10_july_2025'12:01
rlandyo/12:01
amoralejcourtesy ping: sean-k-mooney jgilaber 12:02
amoralejlet's start with the agenda topics12:03
amoralej#topic Eventlet Removal Updates12:03
amoralej#link https://etherpad.opendev.org/p/watcher-eventlet-removal12:04
amoralej#link https://review.opendev.org/c/openstack/watcher/+/952257 12:04
amoralejall yours dviroel 12:04
dviroelo/12:04
jgilabero/12:04
dviroelthe watcher patch mentioned is is marked as wip12:04
dviroelbut it has some progress12:04
dviroellastest issue found was with the continous audit handler12:05
dviroel#link https://review.opendev.org/c/openstack/watcher/+/952257/6/watcher/decision_engine/messaging/audit_endpoint.py12:05
dviroelsince it is being started in audit endpoint construcotr12:06
dviroelit was causing a problem with its jobs running on a different process12:06
dviroellast patch set I am testing it now in the decision engine service12:06
dviroelthe one that we are creating here:12:07
dviroel#link https://review.opendev.org/c/openstack/watcher/+/952499 (Merge decision engine services into a single one) 12:07
dviroelI created new scenarios tests, just to validate continous audit, which are missing from our plugin today12:08
dviroel(we only have api tests)12:08
dviroel#link https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/95426412:08
dviroelI think that we can discuss more if we should duplicate tests to cover continous audits, in a different test file12:08
dviroelor if we can change existing tests to cover continous audits12:09
dviroelthe thing is:12:09
chandankumardviroel: ++ thank you for adding the scenario tempest for adding continous audit.12:09
amoralejyep, that was a gap12:09
dviroelwe need to create compute resources to populate the model, since the continous audit thread will need these model to execute a strategy12:10
dviroelbut int the end, we don't need to execute the action plan, to validate the continous audit code12:10
dviroelthat's why this tempest change is not really executing the action plan12:10
amoralejyes, no need12:11
amoralejbut actually, we may use the nop or sleep actions12:11
amoraleji guess those do not need any model?12:11
dviroelcorrect12:11
dviroelwe need strategies that consume model info12:11
dviroelto reproduce the isse that I found12:11
amoralejahhhhh, sorry i was missing the point totally12:12
amoralejnow i got it12:12
dviroelthe issue with continous audit running in a different process, in threading mode :) 12:12
dviroelso in the end, it is important for the continous audit consume a real model info12:12
amoralejyeah, i wasn't understanding, I do now12:12
amoralejyeap12:13
amoralejeven if we don't run the action plan12:13
dviroelyes12:13
amoralejmakes sense12:13
dviroelI was able to reproduce the issue in CI here12:13
dviroel#link https://review.opendev.org/c/openstack/watcher/+/95436412:13
dviroelresults:12:13
dviroel#link https://77b011d758712ead8b20-de6b79a0bbc85dd849a2bc7008d89fe0.ssl.cf2.rackcdn.com/openstack/a493d2d23f764b198bd434dbfe5451fd/testr_results.html12:14
dviroelsee that we also need 2 tests to reproduce the isse12:14
dviroelsince the first run will trigger a model update12:14
dviroelbut not the second test, it will not get an update model12:14
dviroelin the, just to point that it is important to test the continuous audit in CI12:15
dviroelI think that was jgilaber that raised that in previous meetings/chat12:15
dviroeljgilaber++12:16
dviroelok, so I will continue my work in the decision-engine patch, now working with fixing/adding more units tests12:16
dviroelto be ready for a review too12:16
amoralejactually, one test we may do is create an workload balance or something like that, in first execution, no vms are created, so actionplan is created empty, then create two vms in the same host, and following execution should cratee an non-empty one12:16
dviroelright, we can do everything in a single test, to reproduce the issue, it is a good idea12:17
amoralejit may be tricky about timing given that it's hard to predict exactly the time when the model is updated, but i think it'd be doable12:18
dviroeli will give a try, it won't take too much time12:18
amoralejwe may need an interval > than 10 seconds :)12:18
sean-k-mooneyo/12:19
amoralejwe may use any other strategy, anyway host_maintenance, etc... maybe that would be better, but the idea is the same12:20
amoralejso the continuous handler will be thread within the same process ?12:21
dviroelyes, correct12:21
amoralejah, here https://review.opendev.org/c/openstack/watcher/+/952257/6/watcher/decision_engine/service.py12:23
dviroelyep, and it will ne started together with the other handlers/schedulers12:23
dviroeland not in the audit endpoint constructor12:23
amoralejlooks much cleaner12:24
dviroeland note that the continuous audit handler actually uses the backgroung scheduler12:24
dviroelwhich is also created/init there12:25
dviroelany other question/comments on that? we can continuing our discussion in the patch12:26
amoralejthanks for the update12:26
dviroelone more thing12:26
amoralejand thanks for the work you are doing on that12:26
dviroelthere is a related patch ready for reaview:12:26
dviroel#link https://review.opendev.org/c/openstack/watcher/+/95249912:26
dviroelwhich most of you already reviewed/approved12:26
dviroelbut I had to rebase12:27
dviroeland lost your votes12:27
dviroelbecause we merged this12:27
dviroel#link https://review.opendev.org/c/openstack/watcher/+/949641 (Move eventlet command scripts to a different dir)12:27
dviroelwhich is also part of the effort12:28
dviroelty all12:28
dviroelthat's everything I have12:28
amoralejack12:29
amoralejlet's check if there is any recent bug12:29
amoralej#topic bug triage12:29
sean-k-mooneyah that makes sense ill lop back to that today, fyi i wont be aroudn tomorrow to review12:30
amoralej#link https://bugs.launchpad.net/watcher/+bug/211630412:30
amoralejthat's about croniter, is set as triaged already by chandankumar 12:31
chandankumarI checked the backlog no bug was found, so added it.12:31
sean-k-mooneyya so that was a left over form epoxy12:32
sean-k-mooneywe should fix that and backport it when we have time12:32
sean-k-mooneyit became less urgent because a new maintianer took it over12:32
sean-k-mooneybut for the very minimal usage we have to parse the interval 12:32
chandankumaryes, I am working on a fix, will assign it to myself12:32
sean-k-mooneythere is not reason to keep it as an extra dep12:33
amoraleji have one question12:33
amoralejaccording to croniter doc there is a syntax that i was not aware 12:34
amoralejsat#1,sun#2 = # 1st Saturday, and 2nd Sunday of the month12:34
amoralejis that standard cron?12:34
sean-k-mooneyim not sure12:35
sean-k-mooneyi think maybe yes12:35
amoralejand will apscheduler know how to manage that?12:35
sean-k-mooneythe reals quetion is what do we docuemnte as supported12:35
sean-k-mooneyim pretty sure we dont say conitor format12:35
amoralejright we say "Can be set either in seconds or cron syntax"12:36
amoralejactually my question is if there are different flavors of cron formats :)12:36
amoralejas i had never seen that12:36
chandankumarhttps://apscheduler.readthedocs.io/en/3.x/modules/triggers/cron.html will take care of triggering based on cron format12:37
sean-k-mooneyyou can use sun as an aliase12:37
sean-k-mooneyso i think this is just an alteriive encodeing 12:37
sean-k-mooneyhttps://linux.die.net/man/5/crontab12:38
sean-k-mooney0-7 (0 or 7 is Sun, or use names) 12:38
amoralejyes, my doubt was the #1 or #212:38
amoralejas 1st saturday of the month, i.e.12:39
sean-k-mooneyi am not sure12:39
sean-k-mooneynormaly i use the / sysntax12:39
sean-k-mooneybut # might eb a normal thing12:39
amoralejyep12:39
amoralejanyway, it's just a minor detail12:40
sean-k-mooneygoogle say it is12:40
sean-k-mooneyhttps://www.netiq.com/documentation/cloud-manager-2-5/ncm-reference/data/bexyssf.html12:40
sean-k-mooneybut only for day of the week12:40
sean-k-mooneyDay of the Week12:41
sean-k-mooney 12:41
sean-k-mooneyYes12:41
sean-k-mooney 12:41
sean-k-mooney1-7 OR SUN-SAT12:41
sean-k-mooney 12:41
sean-k-mooney, - * ? / L #12:41
sean-k-mooneyamoralej: so i think we are good12:41
amoralejgood12:41
amoralejthanks for checking12:41
amoralejso i think there are no more bugs to discuss about12:42
amoralej#link https://bugs.launchpad.net/watcher/+bug/2115058 12:42
amoralejthere is also that, reported by jgilaber, but also marked as triaged12:42
jgilaberwe have a couple untriaged 12:43
jgilaber#link https://bugs.launchpad.net/watcher/+bug/210885512:43
jgilaber#link https://bugs.launchpad.net/watcher/+bug/210899412:43
sean-k-mooneythe internal az is a specal az in nova12:43
sean-k-mooneyits not a real one and i dont think it has any equivlent in cinder12:43
amoralejright, sorry i was checking only the last ones12:44
sean-k-mooneymy guess is there is some special handling for internal12:44
jgilaberack sean-k-mooney then I'll try at some point to check creating a new az and see if the problem still persits12:44
amoralejinternal az is kind of default one if there is no explicit ones?12:44
sean-k-mooneysince it not an az you are everm ent to use in an api request12:45
sean-k-mooneyamoralej: no that default az is called nova12:45
sean-k-mooneyinterenl is used for thigns that are not comptues like the metadta api12:45
sean-k-mooneyits a weird legacy thing12:45
amoralejah, got it12:45
amoralejmaybe we could even exclude from the model, then12:46
sean-k-mooneyso for nova that likely ok i am not sure if they use internal for the same thing in cinder12:47
sean-k-mooneysince this is about the stroage model we shoudl check with them first12:47
amoralejgood point12:47
amoralejbut, for the regular case, az names for cinder and nova should not match?12:48
sean-k-mooneyhttps://docs.openstack.org/api-ref/compute/#id29112:48
sean-k-mooneyincase your interested the schduler and condocotrs in nova are part of the Internal zone12:48
amoralejwrt https://bugs.launchpad.net/watcher/+bug/2108855 now that the spec is merged, can we set it as triaged?12:50
sean-k-mooneywe can close it as invlid with a link to the spec since it was a feature request not a bug12:51
sean-k-mooneyill do that now12:51
amoralejack, thanks12:51
amoralejabout bugs.launchpad.net/watcher/+bug/210899412:52
sean-k-mooneyim not entily sure about htat but its for 2023.2 which is now end of life12:54
sean-k-mooneyso i think we just close it unless we see the problem in later relesaes12:54
jgilaberI did try to reproduce it at some point12:54
jgilaberbut I couldn't12:54
sean-k-mooneyany objection if i set it to wont fix.12:55
amoralejor invalid12:55
jgilaber+1 from me12:55
amoralejthe reported didn't replay in almost 2 months...12:55
sean-k-mooneyit went eol 3 months ago12:55
amoraleji think it's good to move to some other state12:56
amoralejactualy, maybe incomplete12:56
sean-k-mooneyso the branh its reproted for is eol so we cant fix it on 2023.2 even if we wanted too12:58
sean-k-mooneythat why i was saying wont fix rather then incomplete12:58
sean-k-mooneybut i can update it if you liek12:58
amoraleji just did :)12:58
amoralejdamn, we did it in parallel :)12:59
sean-k-mooneyyours came second12:59
sean-k-mooneyso you won12:59
sean-k-mooneyits fine we can leave it as it is12:59
amoralejok12:59
amoralejso i think that was it about bugs12:59
amoralej#topic volunteers to chair next meeting13:00
amoralejany?13:00
rlandyI'll do it - it's been a while13:00
amoralej#action rlandy will chair next week13:00
amoralejjust in time13:00
amoralejunless someone has some last minute item, i'm closing the meeting13:01
amoralejthank you all for joining!13:01
amoralej#endmeeting13:01
opendevmeetMeeting ended Thu Jul 10 13:01:43 2025 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)13:01
opendevmeetMinutes:        https://meetings.opendev.org/meetings/watcher_meeting___10_july_2025/2025/watcher_meeting___10_july_2025.2025-07-10-12.01.html13:01
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/watcher_meeting___10_july_2025/2025/watcher_meeting___10_july_2025.2025-07-10-12.01.txt13:01
opendevmeetLog:            https://meetings.opendev.org/meetings/watcher_meeting___10_july_2025/2025/watcher_meeting___10_july_2025.2025-07-10-12.01.log.html13:01
opendevreviewJoan Gilabert proposed openstack/watcher-tempest-plugin master: Add test for volume retype with zone migration  https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/95462515:19
opendevreviewDouglas Viroel proposed openstack/watcher-tempest-plugin master: WIP - Add scenario tests with continuous audit  https://review.opendev.org/c/openstack/watcher-tempest-plugin/+/95426420:17
« Wednesday, 2025-07-09 Index

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!