| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2026-10-21 | |||
| 15:03:37 | melwitt | it looks like indeed openstacksdk is already negotiating the latest API version but yet it seemed to not be the case when func tests ran in CI ... so I think I must not understanding something somewhere. still investigating | |
| 15:04:03 | sean-k-mooney | o/ | |
| 15:04:10 | sean-k-mooney | is ^ related to cyborg? | |
| 15:04:43 | stephenfin | melwitt: I assumed the failues were because the Device resource was missing the `_max_microversion` field? | |
| 15:04:50 | stephenfin | sean-k-mooney: yarp | |
| 15:05:23 | sean-k-mooney | so i had to fix microversion in cyborg last cycle because they were entirly broken | |
| 15:05:51 | melwitt | stephenfin: ohhh ... I dunno. I mean, why/how could it work installed in a devstack locally but then not work in upstream func test job if that's the case? | |
| 15:06:06 | sean-k-mooney | https://github.com/openstack/cyborg/commit/7faf80e098f189e6220ccbe5bfc9d179f482fa76 | |
| 15:06:24 | melwitt | I could also be totally wrong and remembering it wrong | |
| 15:06:24 | sean-k-mooney | well not entirly but they were not conforment to the spec | |
| 15:06:30 | stephenfin | yeah, that bit's the suprising bit. Have you a link to a failing build? | |
| 15:07:16 | melwitt | not to hand. there might be a failed build somewhere in the history of the command migration patches that may have showed it. I'll look around | |
| 15:07:41 | sean-k-mooney | is this in relateion to a particly patch | |
| 15:07:49 | sean-k-mooney | i can take a look as well | |
| 15:07:51 | stephenfin | sean-k-mooney: https://review.opendev.org/c/openstack/python-openstackclient/+/1011568 | |
| 15:08:18 | stephenfin | though I think melwitt has it in hand. It's an SDK/OSC interaction thing as opposed to something cyborg specific we think | |
| 15:08:21 | sean-k-mooney | i think i have a deployment with cyborg i can test that on one sec | |
| 15:08:53 | stephenfin | sean-k-mooney: before you do, see my comments on same | |
| 15:09:44 | sean-k-mooney | ah | |
| 15:09:46 | sean-k-mooney | i see | |
| 15:10:02 | sean-k-mooney | yes that is plasuaible | |
| 15:10:03 | stephenfin | cardoe: jfyi I left some comments on https://review.opendev.org/c/openstack/openstacksdk/+/1011547/ so Ade has respun. I'm holding off +W'ing in case you want to take another spin through it | |
| 15:10:26 | sean-k-mooney | melwitt: i assuem you dont have teh max api version set in your clouds.yaml or openrc | |
| 15:10:42 | sean-k-mooney | i dont think the cybrog plugin does that but good to check in yoru devstack | |
| 15:11:01 | melwitt | ohhhh I do have it set | |
| 15:11:14 | stephenfin | yeah, that'd do it :) | |
| 15:11:18 | melwitt | that must be why. argh I don't remember doing that sorry | |
| 15:12:20 | sean-k-mooney | ok i jsut checked my devstack and they are not set by default | |
| 15:12:29 | melwitt | ok ok. good. but yeah, I am working on this bc I was going through cyborgclient plugin <=> OSC parity and realizing that is missing still while i was working on making the docs work | |
| 15:13:16 | melwitt | thanks sean-k-mooney you saved me a lot of time haha | |
| 15:13:47 | sean-k-mooney | i just went throught the list of things that bit me in the past | |
| 15:13:53 | sean-k-mooney | and it happend to be the first one | |
| 15:14:00 | melwitt | :D | |
| 15:28:58 | melwitt | stephenfin: ok, so what is the right way to bump the version if we want to do it for all resources under cyborg, not only Device? | |
| 15:30:09 | melwitt | unless it's to set _max_microversion on all of them separately. I assumed that would not be the way | |
| 15:31:53 | stephenfin | the expectation is that each Resource gets _max_microversion set to the latest version that changed something about that resource | |
| 15:32:12 | stephenfin | is there a reason you want to use >= 2.0 for resources != Device? | |
| 15:32:16 | melwitt | ok, I see | |
| 15:32:22 | stephenfin | err, > 2.0 | |
| 15:33:00 | melwitt | yes for device profile we have some capability under 2.2 for handling both name and uuid | |
| 15:33:22 | stephenfin | right, so DeviceProfile would get 2.2 | |
| 15:33:44 | melwitt | so I would just set that one to 2.2 according to this pattern | |
| 15:34:06 | stephenfin | yeah | |
| 15:34:15 | melwitt | ok got it | |
| 15:38:13 | stephenfin | This should be documented. If it's not done, I'd be happy to review some (brief, if LLM-written) contributor docs on same | |
| 15:38:25 | stephenfin | Only if you want to, obviously 🤷 | |
| 15:38:43 | melwitt | sure. sorry if I missed the doc | |
| 15:51:15 | opendevreview | Merged openstack/cliff master: Move dependencies to pyproject.toml https://review.opendev.org/c/openstack/cliff/+/1010701 | |
| 15:51:17 | opendevreview | Merged openstack/cliff master: Add release notes build https://review.opendev.org/c/openstack/cliff/+/1010702 | |
| 15:55:43 | opendevreview | Merged openstack/cliff master: Add ShowOneWithDetails for field-oriented tables https://review.opendev.org/c/openstack/cliff/+/1010703 | |
| 16:30:30 | opendevreview | Merged openstack/python-openstackclient master: Fix the block storage v3 index to say v3 instead of v2 https://review.opendev.org/c/openstack/python-openstackclient/+/1011354 | |
| 22:19:25 | cardoe | stephenfin: done going through stuff. | |
| 22:20:26 | cardoe | stephenfin: do we have a type for UUID-like? We pass around a UUID that's really string-ified. We also need a UUID | "default" type. | |
| #openstack-sdks - 2026-10-22 | |||
| 08:30:52 | opendevreview | Merged openstack/ansible-collections-openstack master: Fix dns zone update issues in CI https://review.opendev.org/c/openstack/ansible-collections-openstack/+/1011348 | |
| 08:37:39 | stephenfin | cardoe: in what context? | |
| 08:37:51 | stephenfin | (the UUID-like type) | |
| 09:57:07 | opendevreview | Jed Preist proposed openstack/ansible-collections-openstack master: Add target_provision_state option to baremetal_node https://review.opendev.org/c/openstack/ansible-collections-openstack/+/1010852 | |
| 10:06:25 | opendevreview | Merged openstack/openstacksdk master: Fix Barbican key-manager microversion negotiation https://review.opendev.org/c/openstack/openstacksdk/+/1011547 | |
| 10:06:28 | opendevreview | Merged openstack/openstacksdk master: Add Barbican secret store migrate helper https://review.opendev.org/c/openstack/openstacksdk/+/1002400 | |
| 10:08:59 | opendevreview | Jed Preist proposed openstack/ansible-collections-openstack master: Add target_provision_state option to baremetal_node https://review.opendev.org/c/openstack/ansible-collections-openstack/+/1010852 | |
| 10:20:04 | opendevreview | Moritz Tanner proposed openstack/ansible-collections-openstack master: image: do a web-download import when uri is given https://review.opendev.org/c/openstack/ansible-collections-openstack/+/1006899 | |
| 11:50:13 | opendevreview | Merged openstack/ansible-collections-openstack master: Handle the cloud parameter as a dict in application_credential module https://review.opendev.org/c/openstack/ansible-collections-openstack/+/955695 | |
| 11:59:59 | opendevreview | Merged openstack/ansible-collections-openstack master: Use **-operator to unpack dict in volume_type_access https://review.opendev.org/c/openstack/ansible-collections-openstack/+/1006027 | |
| 12:00:02 | opendevreview | Merged openstack/ansible-collections-openstack master: Fix: Include target_project_id in RBAC creation https://review.opendev.org/c/openstack/ansible-collections-openstack/+/964684 | |