| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2025-02-20 | |||
| 19:24:19 | mikal | Yeah, I am not sure I see how my tempest test would affect ironic at all. | |
| 19:24:24 | mikal | Thank you for chasing it though. | |
| 19:25:25 | sean-k-mooney | most of us are off tomorrow but im goign to t check in perodicaly over the weekend | |
| 19:26:07 | sean-k-mooney | if the tempest fix is merged and my job fix merges hten ill trigger one recheck on the top patch and if that passes ill approve the 3 pathces for spice direct | |
| 19:27:20 | sean-k-mooney | im actully goign to finish now so ill be online intermitntly later to check in | |
| 19:37:45 | mikal | sean-k-mooney: no worries. Thanks for keeping an eye on it. | |
| 19:39:40 | MengyangZhang[m] | sean-k-mooney: I saw your comment about the min check yesterday. Just wondering if we still want to have it. I looked into the API code a bit and I don't think it's trivial to implement it. | |
| 20:16:39 | opendevreview | Merged openstack/nova master: move nova-ovs-hybrid-plug to deploy with spice and fix qxl default https://review.opendev.org/c/openstack/nova/+/940873 | |
| 20:16:47 | opendevreview | Merged openstack/nova master: FUP for reno issues https://review.opendev.org/c/openstack/nova/+/942386 | |
| 21:14:31 | opendevreview | Masahito Muroi proposed openstack/nova master: Use dict object for request_specs_dict in the _list_view https://review.opendev.org/c/openstack/nova/+/939658 | |
| 21:33:40 | opendevreview | melanie witt proposed openstack/nova master: libvirt: Fix regression of listDevices() return type https://review.opendev.org/c/openstack/nova/+/942334 | |
| 21:33:41 | opendevreview | melanie witt proposed openstack/nova master: Reproducer for bug 2098892 https://review.opendev.org/c/openstack/nova/+/942402 | |
| 21:40:05 | opendevreview | melanie witt proposed openstack/nova master: libvirt: Fix regression of listDevices() return type https://review.opendev.org/c/openstack/nova/+/942334 | |
| 22:51:19 | opendevreview | melanie witt proposed openstack/nova master: Reproducer for bug 2098892 https://review.opendev.org/c/openstack/nova/+/942402 | |
| 22:51:19 | opendevreview | melanie witt proposed openstack/nova master: libvirt: Fix regression of listDevices() return type https://review.opendev.org/c/openstack/nova/+/942334 | |
| 23:41:21 | melwitt | weird new-to-me CI failure in nova.tests.unit.db.main.test_migrations.TestMigrationsWalkPostgreSQL.test_single_head_revision: sqlalchemy.exc.OperationalError: (psycopg2.errors.ObjectInUse) source database "template1" is being accessed by other users | |
| 23:50:56 | sean-k-mooney | ... werid | |
| #openstack-nova - 2025-02-21 | |||
| 00:15:00 | sean-k-mooney | melwitt: +2 on your repodcuer and fix for the listDevices thing | |
| 00:15:23 | melwitt | sean-k-mooney: thanks :) | |
| 03:11:18 | melwitt | sean-k-mooney: hm ... the top patch is still failing the spice test after merging the tempest change. seems to be the same error about "additional schema properties" | |
| 03:12:32 | sean-k-mooney | did the tempet one merge? | |
| 03:12:41 | melwitt | yes. it merged and then I rechecked it | |
| 03:13:12 | melwitt | not sure if I'm missing something | |
| 03:13:22 | sean-k-mooney | let me check the results | |
| 03:13:49 | sean-k-mooney | were you watching it live | |
| 03:13:54 | sean-k-mooney | i dont see zuul reproting yet | |
| 03:14:18 | melwitt | yeah watching it live. as soon as it turned red | |
| 03:14:34 | melwitt | not sure if this will work https://zuul.openstack.org/stream/85ee800351e444bcbf37a907f0e9ac50?logfile=console.log | |
| 03:14:44 | sean-k-mooney | i have it open :) | |
| 03:14:56 | melwitt | heh | |
| 03:15:33 | sean-k-mooney | i think this is a diffent error | |
| 03:15:36 | melwitt | my first thought is I wonder if it somehow has to do with the image properties patch merged, if some schema update was needed for that? | |
| 03:15:48 | sean-k-mooney | yep that what im thinking | |
| 03:15:54 | melwitt | bleh | |
| 03:16:19 | melwitt | we really need a better way (or a way :P) to make sure we sync tempest with api changes | |
| 03:16:20 | sean-k-mooney | so we need ot also update the tempest schema validate with the 2.98 stuff | |
| 03:16:45 | sean-k-mooney | well normlay we merge the tempets change after the api change and have a dnm on nova to show its working | |
| 03:17:15 | sean-k-mooney | so nova(api) <- tempest <- nova (DNM) | |
| 03:18:10 | melwitt | I see | |
| 03:18:33 | sean-k-mooney | so this isnt a problem with any of mikal patches | |
| 03:18:43 | sean-k-mooney | its because tempest is not aware of 2.98 | |
| 03:18:59 | melwitt | ok. that makes sense then | |
| 03:19:00 | sean-k-mooney | and any test that uses 2.99 is going to fail if it does a server show i think | |
| 03:19:26 | melwitt | yeah I think you're right | |
| 03:21:28 | sean-k-mooney | given it 3:20 i shoudl probaly either go to bed or get a glass of wiskey | |
| 03:22:29 | sean-k-mooney | we could push a patch to tempest and then put a patch on top of the speice direct one to verify it fixes it | |
| 03:22:51 | melwitt | those are some choices lol | |
| 03:23:16 | melwitt | yeah. I could probably do that. I have done a tempest schema patch once before | |
| 03:23:16 | sean-k-mooney | i mean it clearly to late? early? for coffee | |
| 03:23:29 | melwitt | yeah. no coffee | |
| 03:29:12 | sean-k-mooney | ok so it kind of an issue with mikals tempest patch but only because the schemav299 tehy defiend does not have the image stuff | |
| 03:29:46 | sean-k-mooney | so we need to add a 2_98 and update https://review.opendev.org/c/openstack/tempest/+/942023/3/tempest/lib/api_schema/response/compute/v2_99/servers.py#15 | |
| 03:29:51 | sean-k-mooney | to import that instead | |
| 03:30:32 | melwitt | ah ok | |
| 03:30:39 | sean-k-mooney | also when we hit 2.100 can we defien a 3.0 that is just 2.100 and move on withour lives | |
| 03:30:57 | melwitt | lol. that's ... a good point | |
| 03:31:23 | sean-k-mooney | it wouldnt be the first tiem we had a 3.0 ... | |
| 03:31:34 | melwitt | hah. yeah | |
| 03:32:19 | masahito | hi nova folks, please review the API bug fix for v2.96 and later version if you have time. https://review.opendev.org/c/openstack/nova/+/939658 | |
| 03:33:05 | sean-k-mooney | oht he pined az | |
| 03:34:07 | masahito | yes :-) The show scheduler hint feature needs to follow similar change because it checks request_spec. | |
| 03:36:24 | sean-k-mooney | im currently ruminating on https://review.opendev.org/c/openstack/nova/+/939658/6/nova/api/openstack/compute/views/servers.py#222 | |
| 03:36:39 | sean-k-mooney | syliticlly this annoys me but im tryign to decied if there is actully an issue | |
| 03:37:43 | sean-k-mooney | if i read that if cascade | |
| 03:38:14 | sean-k-mooney | it imples provided_az is either the litral boolean False or its a sting contiaing the az name | |
| 03:40:41 | sean-k-mooney | so the data type of provided_az is union(NONE|bool|str) | |
| 03:40:58 | masahito | yes, I also wondering it follows Nova's nice coding manner. The _get_pinned_az needs to identify 3 patterns as commented. | |
| 03:42:45 | sean-k-mooney | so None is only passe if show is invoed and prived_az is not passed | |
| 03:42:48 | sean-k-mooney | https://review.opendev.org/c/openstack/nova/+/939658/6/nova/api/openstack/compute/views/servers.py#244 | |
| 03:42:54 | sean-k-mooney | correct? | |
| 03:43:04 | sean-k-mooney | in what case is false passed? | |
| 03:43:35 | masahito | right. | |
| 03:44:43 | masahito | fales is passed if list is invoked and no request_specs object to an instance | |
| 03:44:57 | sean-k-mooney | ah | |
| 03:45:13 | sean-k-mooney | it woudl be nice if None was passed then | |
| 03:45:31 | sean-k-mooney | thats form here right https://review.opendev.org/c/openstack/nova/+/939658/6/nova/api/openstack/compute/views/servers.py#539 | |
| 03:46:08 | sean-k-mooney | if we change req_specs_dict.get(server.uuid, False) to req_specs_dict.get(server.uuid, None) | |
| 03:47:27 | sean-k-mooney | im trying to decide if that allows use to simplfy the if | |
| 03:48:06 | sean-k-mooney | it would allow use toe get rid of the first two brances and jsut replace it with pinned_az = provided_az | |
| 03:48:29 | sean-k-mooney | but the else woudl have to become if pinned_az is None: | |
| 03:48:34 | masahito | but in that case, _get_pinned_az can't identify pre-fetheched None in list or no input in show. | |
| 03:48:58 | masahito | s/no input in show/no input None in show/ | |
| 03:48:58 | sean-k-mooney | sure but im trying to decied if that matters | |
| 03:49:46 | sean-k-mooney | if we dont have a request spec then it cant be pinned | |
| 03:49:52 | masahito | i see. | |
| 03:50:41 | sean-k-mooney | im not sayign you should change this | |
| 03:51:00 | sean-k-mooney | im trying to undersand the desing chooices that lead to this current patch | |
| 03:51:29 | sean-k-mooney | "if provided_az is False:" is very specific and odd | |
| 03:51:30 | masahito | thanks for the follow up. my reply is one previous your message. | |
| 03:51:39 | sean-k-mooney | you are using False as a sential | |
| 03:51:51 | sean-k-mooney | but it might be better to actuly create a new sentail object instead | |
| 03:51:55 | sean-k-mooney | and use that instead | |
| 03:52:09 | sean-k-mooney | so its clear what its role is | |
| 03:52:40 | masahito | like a empty request_spec object instead of False? | |
| 03:52:56 | sean-k-mooney | no | |
| 03:53:07 | sean-k-mooney | so at teh module level define sentinel = object() | |
| 03:53:40 | sean-k-mooney | then do req_specs_dict.get(server.uuid, sentinel) here | |
| 03:53:42 | sean-k-mooney | https://review.opendev.org/c/openstack/nova/+/939658/6/nova/api/openstack/compute/views/servers.py#539 | |
| 03:54:03 | sean-k-mooney | and if provided_az is sentinel: | |
| 03:54:05 | sean-k-mooney | https://review.opendev.org/c/openstack/nova/+/939658/6/nova/api/openstack/compute/views/servers.py#222 | |