Tuesday, 2026-05-12

*** jgilaber_ is now known as jgilaber11:06
jgilaberHi all, kind reminder the IRC meeting will start in ~30 minutes, add your topics to the agenda https://etherpad.opendev.org/p/openstack-cyborg-irc-meeting#L7613:31
jgilaber#startmeeting cyborg14:01
opendevmeetMeeting started Tue May 12 14:01:07 2026 UTC and is due to finish in 60 minutes.  The chair is jgilaber. Information about MeetBot at http://wiki.debian.org/MeetBot.14:01
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.14:01
opendevmeetThe meeting name has been set to 'cyborg'14:01
jgilaberHi! Who is around today?14:01
sean-k-mooneyo/14:01
jgilabercourtesy ping: amoralej  bogdando rlandy  chandankumar14:01
rlandyo/14:01
amoralejo/14:02
jgilaberwhile we give folks a minute, feel free to add last minute topics to the agenda https://etherpad.opendev.org/p/openstack-cyborg-irc-meeting#L5114:02
jgilaberlet's get started 14:03
jgilaber#topic cyborg tempest job definition14:03
jgilaberI noticed last week that the tempest job definition is stored in the plugin repo instead of the service repo14:04
jgilaberwhich I think is and odd pattern14:04
sean-k-mooneyit a pattern that exists in some plugins14:04
sean-k-mooneybut its not my perfered approch14:04
jgilaberI proposed a patch to move the definition to the main cyborg repo14:04
jgilaber#link https://review.opendev.org/c/openstack/cyborg/+/98762014:05
sean-k-mooneyso we may need to rename them 14:05
jgilaberI wanted to ask for opinions before going forward and propose the same change in the stable branches14:05
sean-k-mooneybecause we may not be able ot remvoe the defintion in the tempest plugin due to stable branches14:05
jgilaberthere are several options, what I did was to copy the job definition but name it cyborg-tempest-base14:06
jgilaberand then move the cyborg-tempest job as a child of this new base job14:06
jgilaber#link https://review.opendev.org/c/openstack/cyborg-tempest-plugin/+/98762114:07
sean-k-mooneyack14:10
sean-k-mooneysorry was distracted14:10
sean-k-mooneylets continue to review this on gerrit14:10
sean-k-mooneybut i think the overall move makes sense14:10
sean-k-mooneyi woudl like to move all the job definiton severually14:11
jgilabersure, if there are no further comment/objections let's move on14:11
sean-k-mooneyas i said i dont know if we just want to keep the alis in the old place14:11
sean-k-mooneyits more a case fo ensurign we dont brake the stable branches14:11
sean-k-mooneyi dont really have any objections to the move14:12
sean-k-mooneywe just need to figure out the sequcne to make it work14:12
sean-k-mooneythe only one that really shoudl be in the tempest repo are the stable branch specific ones14:12
jgilaberagreed, we can discuss the details like the job names in the patches, I'll push changes to the stable branches after the meeting14:13
jgilaber#topic Bugs14:14
sean-k-mooneywell that kind of my point14:14
sean-k-mooneywe shoudlnot be modifyign the stable branches14:15
sean-k-mooneyanyway yes bugs14:15
jgilaberwe have a new bug to triage14:15
jgilaber#link https://bugs.launchpad.net/openstack-cyborg/+bug/215179214:15
jgilaberthe bug report indicates two errors returned by the client14:17
sean-k-mooneyya so...14:17
sean-k-mooneythat api feature was approvced in bobcat but never marked as implemtned14:17
jgilaberone was fixed in 2025.2 14:17
sean-k-mooneyso im not sure if it actully works14:17
sean-k-mooneyand yes there is a bug in the clinet <> sdk interaction as well14:17
sean-k-mooneylooking at commente 2 par tof this seam to be verion dependent14:18
sean-k-mooneyi.e it may require a newer verison fo the sdk then is allwoed in 2025.114:19
sean-k-mooneyi htink we are going to have to look at when the supprot was added for this in the sdk adn client14:20
sean-k-mooneysee if it works or not on master14:20
sean-k-mooneyand then determim how to proceed14:20
jgilaberthe "attribute list" seems to work for version 4.7.2 so it should work on master too 14:21
jgilaberChandan did reproduce the enable/disable error on master14:21
jgilaberlooking at the code it seems the api controllers are implemented https://github.com/openstack/cyborg/blob/master/cyborg/api/controllers/v2/devices.py#L157-L18314:21
sean-k-mooneyyes they are which shoudl be the last part of the feature14:22
sean-k-mooneybut i dont know if the rest of the feature was compelted14:22
sean-k-mooneyhttps://review.opendev.org/q/topic:%22bp/disable-enable-device%2214:23
sean-k-mooneyin theory the feature as added by https://review.opendev.org/c/openstack/cyborg/+/89130914:23
jgilaberthe code looks complete14:24
sean-k-mooneythere are not tempest tests14:24
sean-k-mooneyso we dont actully know if this worked end to end14:25
jgilaberit looks like it's missing completely from the openstacksdk proxy https://github.com/openstack/openstacksdk/blob/master/openstack/accelerator/v2/device.py14:26
sean-k-mooneyyes https://review.opendev.org/c/openstack/openstacksdk/+/89488814:26
jgilaberno reference of any variable to track if the device is enabled or disabled14:26
sean-k-mooneywas never merged14:26
sean-k-mooneyhttps://review.opendev.org/c/openstack/python-cyborgclient/+/894887 didnt have a depens on14:26
sean-k-mooneybut it requires the sdk change14:27
jgilaberthat explains the bug14:27
sean-k-mooneyso this feature may work at the the api level14:27
sean-k-mooneybut we need to fix thet sdk, add tempest test and test this proeprly end to end in ci14:27
sean-k-mooneywe can mark this as vaild and medium for the client14:28
jgilaberagreed14:28
sean-k-mooneybut not for cyborg itself at least for now14:28
sean-k-mooneyits really a feature not a bug for the sdk14:29
sean-k-mooneyso we can deal with that seperately14:29
sean-k-mooneyill do that now14:29
jgilaberack, thanks sean-k-mooney 14:29
jgilaberthat is everything we had in the agenda14:29
jgilaberany topics for the open discussion?14:30
sean-k-mooneyam honestly tere are a few thing but i kind fo dont have the mental energy to look at them now14:32
sean-k-mooneyso not form me14:32
sean-k-mooneyi guess a quick update14:32
sean-k-mooneythe cve's we found have now been fixed and backported14:32
sean-k-mooneyand the release patch merged yesterday14:32
sean-k-mooneyso the release shoudl be done14:32
sean-k-mooneyi have some followups to work on like removing the test skip14:32
sean-k-mooneyand the docs improvment we talksed about14:33
sean-k-mooneythose are the thing i have not had time to refelct on yet14:33
sean-k-mooneyso  they are on my todo list jsut not someitng i have any really update on14:33
sean-k-mooneythats about it 14:33
jgilaberthanks for all the work there sean-k-mooney 14:33
jgilaberI got a notification in the bug reports  that the fixes were included in new releases so it looks like everything is done 14:34
jgilaberso last thing for today is a call for volunteers to chair next meeting14:35
jgilaberI can do it next week if there are none14:35
sean-k-mooneyif you can that woudl be awsome14:36
jgilaberok so it looks like we can close the meeting for today14:36
jgilaberany last minute comment?14:37
jgilaberclosing then, thanks!14:37
jgilaber#endmeeting14:37
opendevmeetMeeting ended Tue May 12 14:37:59 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)14:37
opendevmeetMinutes:        https://meetings.opendev.org/meetings/cyborg/2026/cyborg.2026-05-12-14.01.html14:37
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/cyborg/2026/cyborg.2026-05-12-14.01.txt14:37
opendevmeetLog:            https://meetings.opendev.org/meetings/cyborg/2026/cyborg.2026-05-12-14.01.log.html14:37

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