Earlier  
Posted Nick Remark
#openstack-cyborg - 2026-03-17
14:26:19 chandankumar yup sounds good
14:26:58 rlandy done
14:27:00 rlandy next
14:27:04 jgilaber yes, and cofirmation that the issue is related to cyborg and other other environmental factors
14:27:31 rlandy #link: https://bugs.launchpad.net/openstack-cyborg/+bug/2011754 accelerators not reconnected to vm after host reboots
14:28:19 rlandy This might be something we may want to keep open until we are at a point we can reverify
14:28:27 rlandy maybe in a few months?
14:28:52 jgilaber yes, we should try to reproduce this
14:28:55 chandankumar here hosts means compute note reboots?
14:29:15 jgilaber I would guess so
14:29:20 chandankumar ok
14:29:27 jgilaber I'm not sure if this could be a problem in cyborg or in nova
14:29:40 jgilaber but it could have been fixed
14:29:40 chandankumar I will try to take a look at this one
14:30:01 rlandy ok chandankumar - pls mark it accordingly after your first look
14:30:11 rlandy otherwise I can mark it triaged?
14:30:11 chandankumar it would be good opporunity to try cyborg in multi node env
14:30:27 chandankumar sure, will do that
14:30:30 rlandy ok - great
14:30:44 rlandy last one for today:
14:30:50 rlandy #link: https://bugs.launchpad.net/openstack-cyborg/+bug/2144308 Image verification code broken due to missing pieces
14:31:09 jgilaber I found this one last week while revieweing some code
14:31:39 jgilaber I noticed some code missing and chandankumar has verified in a devstack env
14:31:42 rlandy chandankumar just updated it
14:32:04 jgilaber in short some code to verify glance images was copied from nova, but the copy was incomplete
14:32:49 jgilaber short term we should copy the missing bits and hopefully build some testing for it
14:33:00 jgilaber the kind of error tells that this has never been tested
14:33:24 chandankumar yes, I will copy the missing bits and verify in my env
14:33:51 jgilaber thanks chandankumar++
14:34:28 rlandy status for this bug?
14:34:34 chandankumar triaged
14:34:42 jgilaber imo this would be a high or critical bug, since it can break production deployments
14:35:00 rlandy triaged, high
14:35:20 rlandy ok?
14:35:27 jgilaber +1
14:35:53 chandankumar +1
14:36:08 rlandy moving on ...
14:36:19 rlandy that's all the bugs we have listed
14:36:26 rlandy any more to raise?
14:37:43 rlandy ok ... moving on
14:37:53 rlandy #topic: Open Discussion
14:38:57 rlandy #topic: PTG preparation: https://etherpad.opendev.org/p/cyborg-2026.2-ptg
14:39:22 rlandy PTG is planned for 20-24th April 2026
14:39:36 rlandy so far we only have one topic added
14:40:08 rlandy suggestion to add topics during the week and confirm during next week's call
14:41:47 jgilaber +1
14:41:52 chandankumar +1
14:41:57 opendevreview OpenStack Release Bot proposed openstack/cyborg master: Update master for stable/2026.1 https://review.opendev.org/c/openstack/cyborg/+/980961
14:42:19 rlandy ok - that's all for the points on the etherpad
14:42:28 rlandy anything else to discuss?
14:42:47 chandankumar nope
14:43:05 opendevreview chandan kumar proposed openstack/cyborg master: Add Grenade upgrade support for Cyborg https://review.opendev.org/c/openstack/cyborg/+/979802
14:43:15 rlandy ok ...
14:43:33 rlandy #topic: Volunteers to chair next meeting
14:44:06 rlandy I'll do it :)
14:44:37 rlandy ok - looks like we are done for today
14:44:41 rlandy thank you all
14:44:47 jgilaber thanks rlandy!
14:44:52 chandankumar thanks rlandy !
14:45:00 rlandy #endmeeting
14:45:00 opendevmeet Meeting ended Tue Mar 17 14:45:00 2026 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)
14:45:00 opendevmeet Minutes: https://meetings.opendev.org/meetings/cyborg_irc_call_march__17__2026/2026/cyborg_irc_call_march__17__2026.2026-03-17-14.02.html
14:45:00 opendevmeet Minutes (text): https://meetings.opendev.org/meetings/cyborg_irc_call_march__17__2026/2026/cyborg_irc_call_march__17__2026.2026-03-17-14.02.txt
14:45:00 opendevmeet Log: https://meetings.opendev.org/meetings/cyborg_irc_call_march__17__2026/2026/cyborg_irc_call_march__17__2026.2026-03-17-14.02.log.html
#openstack-cyborg - 2026-03-18
04:37:02 opendevreview chandan kumar proposed openstack/cyborg master: Add Grenade upgrade support for Cyborg https://review.opendev.org/c/openstack/cyborg/+/979802
05:10:55 opendevreview chandan kumar proposed openstack/cyborg master: Add Grenade upgrade support for Cyborg https://review.opendev.org/c/openstack/cyborg/+/979802
09:12:58 opendevreview chandan kumar proposed openstack/cyborg master: Add NVMeVirt emulated NVMe device support for devstack https://review.opendev.org/c/openstack/cyborg/+/977083
13:20:11 opendevreview chandan kumar proposed openstack/cyborg master: Add NVMeVirt emulated NVMe device support for devstack https://review.opendev.org/c/openstack/cyborg/+/977083
#openstack-cyborg - 2026-03-19
19:35:16 fungi sean-k-mooney: is https://docs.openstack.org/cyborg/latest/configuration/drivers.html the new guide referred to in the 2026.1/gazpacho highlights?
19:37:18 sean-k-mooney yes its still not very compelte but it was better then the 0 content that was more or less there before
19:38:32 sean-k-mooney fungi: did you have any specific questions of feed back?
19:41:30 fungi sean-k-mooney: nope, just going back through on a final pass of highlights now that more of them have merged, flagging things for the marketing/pr folks to consider working into the press package
19:42:19 fungi they're overly eager to say ai-related things about cyborg, because media hype train, and i thought at least that's something useful to show in that regard while you're still getting ramped up on deeper development efforts
19:44:10 sean-k-mooney ya well this guide is mostly ai generated form the souce code and config option help text but framed to be a litte more parseable
19:44:23 sean-k-mooney its ceretnly either then lookign at the config smaples https://docs.openstack.org/cyborg/latest/configuration/sample-config.html
19:44:36 fungi to marketing people, cyborg == gpu == ai
19:45:05 sean-k-mooney fungi: basiclly i created it when i realisted that there was no doc on hwo to deploy with a specific driver or what depencies it needs
19:45:19 fungi so whenever they want to say things about ai, cyborg always comes up
19:45:29 sean-k-mooney ya
19:45:50 sean-k-mooney that proably because its called out in the ai white paper so prominently
19:45:55 fungi but hey, if it brings attention to the project and maybe more help, who am i to argue? ;)
19:46:14 fungi as long as we don't say anything that's untrue, it's cool
19:46:56 sean-k-mooney :)
19:47:12 sean-k-mooney https://www.openstack.org/openstack-for-ai-white-paper implies cyborg is more useful today then it really is
19:47:28 sean-k-mooney but that does nto mean it cant evovle to have a larger role in the future
19:47:35 fungi i'm still enough of an engineer that i don't like to lie about things, but i feel like there are exciting things to say while sticking to the facts
19:48:22 fungi and sometimes we just need to make seemingly-unexciting things exciting
19:48:46 fungi like... driver documentation
19:49:08 fungi (from my perspective, it's great when anyone cares about documentation at all, so that's a clear win)
19:50:14 sean-k-mooney ya specs are grate but oen of the thing is want to improve in general is the intree devleoper docs and user/admin ones too to capture more of the why and how it wors type content
19:50:33 sean-k-mooney the user/admin ones need to be more how you use it focused
19:51:09 sean-k-mooney but i think improving the contributor docs is also important so that future contibutor undersand how and imporantly why thigns work
19:51:13 sean-k-mooney the way they work
19:51:42 fungi that would all be great for project updates, absolutely
19:51:55 sean-k-mooney i partly considered if moving the cyborg-spec into the docs like placement did would be good long term but that was too low a priority to change right away
19:52:24 fungi from a "what's in 2026.1/gazpacho?" angle, there's less to cover for cyborg at the moment, hence gravitating toward the driver config guide
19:53:00 sean-k-mooney ya the main things that change were oslo.db comatablity, eventlet removal and a cople of bug fixes
19:53:14 fungi of the highlights, the only other maybe user-facing one was placement integration
19:53:24 sean-k-mooney partly because the focus was "get somethign releaseable" rather then adding end user value via new features ectra
19:53:44 fungi but that's a bit harder to get marketing people to turn into a good press blurb

Earlier   Later