| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-cyborg - 2026-04-28 | |||
| 14:25:19 | sean-k-mooney | its https://github.com/openstack/os-traits/blob/master/os_traits/owner/__init__.py | |
| 14:25:25 | sean-k-mooney | that was added 4 years ago | |
| 14:25:53 | sean-k-mooney | if it was new in 2026.2 you woudl have to check fi the standard trait exsitns in placment before reporting it | |
| 14:26:14 | sean-k-mooney | as we have to assuem placment could be at least 2 release older then cyborg | |
| 14:26:54 | sean-k-mooney | we also need to acont for the fact the cybrog compute agent can be up to 1 slrup older | |
| 14:27:23 | sean-k-mooney | in thei case we cant assume its newer then 2026.1 and we cant assume it has this backport | |
| 14:27:54 | chandankumar | ah in that case, I keep this to master only. | |
| 14:27:57 | sean-k-mooney | so when we translate teh device spec into a placment request group | |
| 14:28:09 | sean-k-mooney | we cant unconditonly assume its updated | |
| 14:28:12 | chandankumar | I did not consider these assumption | |
| 14:28:30 | sean-k-mooney | so this will need more work to make sure we dont break upgrades | |
| 14:28:39 | sean-k-mooney | the reporting of the triat is fine | |
| 14:28:56 | sean-k-mooney | the requesting of it need to check the compute-agent verion and gate it on the min version | |
| 14:29:06 | sean-k-mooney | or a config option | |
| 14:29:38 | sean-k-mooney | given we are not reprotign it today we must not be requesting it either | |
| 14:29:51 | sean-k-mooney | lets loop bakc to that in more detail on teh review or outside the meeting | |
| 14:29:59 | chandankumar | sure | |
| 14:30:29 | chandankumar | thank you sean-k-mooney! | |
| 14:30:40 | chandankumar | Since there is no more bugs, moving to open discussion | |
| 14:30:50 | chandankumar | #topic Open Discussion | |
| 14:31:22 | chandankumar | Feel free to bring topic here | |
| 14:32:41 | sean-k-mooney | i guess now is as good a time as any to say | |
| 14:32:59 | sean-k-mooney | i will be on pto friday and monday and im thinking of also taking next tuesday | |
| 14:33:12 | sean-k-mooney | so i may not be be here next week for the meeting | |
| 14:33:47 | chandankumar | sure, enjoy your time off! | |
| 14:33:50 | chandankumar | Moving to last one | |
| 14:34:05 | chandankumar | #topic volunteer to chair for next meeting | |
| 14:34:42 | chandankumar | Any volunteer up for chairing the next meeting? | |
| 14:34:50 | jgilaber | I can do it | |
| 14:34:58 | chandankumar | thank you jgilaber ! | |
| 14:35:14 | chandankumar | Now closing the meeting early, thank you all for attending! | |
| 14:35:18 | chandankumar | #endmeeting cyborg | |
| 14:35:18 | opendevmeet | Meeting ended Tue Apr 28 14:35:18 2026 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | |
| 14:35:18 | opendevmeet | Minutes: https://meetings.opendev.org/meetings/cyborg/2026/cyborg.2026-04-28-14.03.html | |
| 14:35:18 | opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/cyborg/2026/cyborg.2026-04-28-14.03.txt | |
| 14:35:18 | opendevmeet | Log: https://meetings.opendev.org/meetings/cyborg/2026/cyborg.2026-04-28-14.03.log.html | |
| 14:36:44 | sean-k-mooney | chandankumar: jgilaber i dont know if we want to continue to discuss the upgrade condieration or not right now or if ye want to reflect on my comments | |
| 14:37:31 | sean-k-mooney | i think before we proceed with reproting the owner triat (that shoudl be safe) we need to do a deeper review of how that currently interact with t hegeenrat request groups in the arq | |
| 14:38:01 | sean-k-mooney | and look at the exact mechanis when we have up to n-2 compute-agent runnign with a fully upgraded api/conductor | |
| 14:38:49 | chandankumar | +1 to do a deep review on interaction with placement and cyborg on owner traits | |
| 14:39:10 | chandankumar | We might discover more bugs in that. | |
| 14:39:32 | sean-k-mooney | openstack --os-cloud devstack-admin resource provider trait list 45593a00-c178-3159-9687-6d7efa55f64d | |
| 14:39:34 | sean-k-mooney | +----------------------------+ | |
| 14:39:36 | sean-k-mooney | | name | | |
| 14:39:38 | sean-k-mooney | +----------------------------+ | |
| 14:39:40 | sean-k-mooney | | CUSTOM_PCI_INTEL | | |
| 14:39:41 | chandankumar | and then revisit this again in another meeting. | |
| 14:39:42 | sean-k-mooney | | CUSTOM_PCI_PRODUCT_ID_154C | | |
| 14:39:44 | sean-k-mooney | +----------------------------+ | |
| 14:39:46 | sean-k-mooney | i can confirm that the trait is indeed missing | |
| 14:40:09 | sean-k-mooney | so its not stricly a bug | |
| 14:40:19 | sean-k-mooney | the suprpot for this was part of a new spec | |
| 14:40:25 | sean-k-mooney | that was not fully impmented | |
| 14:40:44 | sean-k-mooney | i.e. owner traits were only intoduced a long time after cybrog was creted | |
| 14:41:01 | sean-k-mooney | as part fo a feature that spaned nova and cybrog that was never impmented on teh nova side | |
| 14:41:07 | sean-k-mooney | and only partly doen in cyborg | |
| 14:41:29 | sean-k-mooney | so im wondering if this shoudl really be a bug at all given it was orgially a feature qruieding a spec | |
| 14:41:44 | chandankumar | https://github.com/search?q=repo%3Aopenstack%2Fnova-specs%20OWNER_CYBORG&type=code there are two specs | |
| 14:41:51 | sean-k-mooney | and given the nvova part was never done (its part of the nova spec for vgpus) | |
| 14:42:09 | chandankumar | this one https://github.com/openstack/nova-specs/blob/e53cb185c680e479a44a94ef0cfb7e1f52c762d6/specs/wallaby/approved/cyborg-vgpu-support.rst#L125 ? | |
| 14:42:29 | sean-k-mooney | yep both are appvoed not implemtned | |
| 14:42:41 | sean-k-mooney | so the nvida driver was partly updated in cybrog because of this | |
| 14:42:54 | sean-k-mooney | but it was never completed and the canges for the other cybrog driver were not done | |
| 14:43:54 | sean-k-mooney | https://github.com/openstack/cyborg-specs/blob/master/specs/wallaby/approved/vgpu-driver-proposal.rst | |
| 14:44:06 | sean-k-mooney | was the cyborg spec for this | |
| 14:45:02 | sean-k-mooney | that was finally implemtned in 2023.1 https://github.com/openstack/cyborg-specs/blob/master/specs/2023.1/implemented/vgpu-driver-proposal.rst | |
| 14:45:28 | sean-k-mooney | without the nova half and without updatign the other drivers to also use the cybrog owner trait | |
| 14:47:07 | chandankumar | when you say nova half, how much work is still left to implement to make it work properly? | |
| 14:47:27 | chandankumar | https://review.opendev.org/q/topic:%22bp/cyborg-vgpu-support%22 | |
| 14:47:47 | chandankumar | https://review.opendev.org/c/openstack/nova/+/812247: vGPU support by Cyborg is an open patch | |
| 14:47:48 | sean-k-mooney | none of the nova change merged | |
| 14:48:03 | sean-k-mooney | so on the nova side there are 2 related parts | |
| 14:48:34 | sean-k-mooney | first nova need to report all its resouce provider with OWNER_COMPUTE or OWNER_NOVA i dont recall | |
| 14:49:00 | sean-k-mooney | it thne need to have a prefilter to include that for its device requests that are not cyborg related | |
| 14:51:46 | chandankumar | Since it is not properly implemented. | |
| 14:52:04 | chandankumar | How do you suggest here next course of action? | |
| 14:52:19 | chandankumar | I am going to do a deep review of how placement nova and cyborg works on traits | |
| 14:52:49 | chandankumar | or we want to table this discussion for some other day? | |
| 14:53:38 | jgilaber | once we implement the owner trait in nova, after upgrading will the trait be added to an already existing resource class or does the operator need to do some action? | |
| 14:55:34 | sean-k-mooney | nova shoudl do it | |
| 14:55:39 | sean-k-mooney | for its resouce classes | |
| 14:55:56 | sean-k-mooney | but the requesting of it need to be gated by a min compute service version check | |
| 14:56:42 | chandankumar | can you share one example of min service version check? | |
| 14:56:47 | chandankumar | from other project | |
| 14:58:08 | sean-k-mooney | sure but i need to check if cybrog has the same concpet today | |
| 14:58:43 | sean-k-mooney | so in the service object | |
| 14:58:45 | sean-k-mooney | https://github.com/openstack/nova/blob/master/nova/objects/service.py#L40 | |
| 14:59:04 | sean-k-mooney | we have an integer constant that we increment when we add specific feautres | |
| 14:59:28 | sean-k-mooney | eacg compute serivce looks at there local constant and report that value to the db | |
| 14:59:43 | sean-k-mooney | nova maintian a history of what those are https://github.com/openstack/nova/blob/master/nova/objects/service.py#L65 | |
| 15:00:08 | sean-k-mooney | we use them both for feature discoverya and detecting nodes that are tooo old | |
| 15:01:31 | sean-k-mooney | in the compute api among other places we can then check the compute service version and block actions | |
| 15:01:33 | sean-k-mooney | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L373-L394 | |
| 15:01:48 | sean-k-mooney | so operation that dont work with cybrog are decorated with block_accelerators | |
| 15:01:56 | sean-k-mooney | which acppate a verion | |
| 15:02:20 | sean-k-mooney | if the min version fo any compute agent cant supprot it | |
| 15:02:22 | sean-k-mooney | https://github.com/openstack/nova/blob/master/nova/objects/service.py#L621-L686 | |
| 15:02:35 | sean-k-mooney | we block teh request | |
| 15:03:43 | sean-k-mooney | we also have other hleper https://github.com/openstack/nova/blob/master/nova/objects/service.py#L495-L518 | |
| 15:04:12 | sean-k-mooney | but the simple version of it is nova will refuse to starte some serices if a compute agent is older then it can supprot | |