| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-07-31 | |||
| 16:37:38 | ganso | Hi Nova folks! I have a quick question about Nova+Cyborg in Ussuri. From looking at the code, it seems that if Cyborg is not present and/or no "accel:device_profile" is specified in the flavor, Nova will work just fine behaving the same way as it was in Train (not having Cyborg), however, I am not 100% sure about its capabilities of using vGPUs in Ussuri without Cyborg, does it still work in Ussuri with vGPUs as it worked in Train without Cyborg? | |
| 16:38:40 | ganso | or is it strictly necessary to have Cyborg to be able to use vGPUs at all in Ussuri? | |
| 17:23:24 | artom | ganso, I believe GPUs work in U as they did in Train | |
| 17:23:30 | artom | IIRC support was first introduced in Queens | |
| 17:23:33 | artom | Without Cybord | |
| 17:24:07 | ganso | artom: thanks for the clarification! =) | |
| 17:29:20 | openstackgerrit | Merged openstack/nova master: Provider Config File: YAML file loading and schema validation https://review.opendev.org/673341 | |
| 18:28:59 | sean-k-mooney | ganso: cyborg has limited support for phsyical gpu passthough with the nvidia driver but that has no real benifit over using pci passthough in nova in ussuri | |
| 18:29:16 | sean-k-mooney | ganso: cyborg currently has no support for vGPUs | |
| 18:30:16 | sean-k-mooney | vgpus are only supported vai nova, and i think artom is right that the suppor in nove is more or less unchange in ussuri | |
| 18:30:34 | sean-k-mooney | they may have been some bug fixes but i dont think we added new features | |
| 18:30:45 | artom | Oh, I think we added move operations, no? | |
| 18:31:56 | artom | But nothing has fundamentally changed in how you have to set it up, and it's completely without Cyborg | |
| 18:42:58 | sean-k-mooney | i know bauzas was working on cold migrate/resize | |
| 18:43:10 | sean-k-mooney | but i dont know if that landed in ussuri or not | |
| 18:43:26 | sean-k-mooney | the other change was multiple mdev types per host but i think that was in train | |
| 19:14:12 | jroll | current status: patching out tools/check-cherry-picks.sh so that our patched branches don't fail pep8, thanks a lot :P | |
| 19:31:19 | artom | jroll, oh damn, I finally get to blame dansmith for something! | |
| 22:58:51 | openstackgerrit | Ghanshyam Mann proposed openstack/nova master: migrate testing to ubuntu focal https://review.opendev.org/738126 | |
| #openstack-nova - 2020-08-01 | |||
| 00:26:59 | openstackgerrit | Merged openstack/nova stable/queens: ironic: add instance_uuid before any other spawn activity https://review.opendev.org/743493 | |
| 06:27:26 | openstackgerrit | tony yan proposed openstack/nova master: add notification info when server attach or detach interface https://review.opendev.org/744273 | |
| 08:38:58 | openstackgerrit | Brin Zhang proposed openstack/nova master: Cyborg evacuate support https://review.opendev.org/715326 | |
| 08:47:49 | openstackgerrit | Brin Zhang proposed openstack/nova master: [Trivial] Remove wrong format_message() conversion https://review.opendev.org/744280 | |
| #openstack-nova - 2020-08-02 | |||
| 00:41:06 | openstackgerrit | Ghanshyam Mann proposed openstack/os-resource-classes master: DNM: testing gate on ubuntu focal https://review.opendev.org/744359 | |
| 00:41:09 | openstackgerrit | Ghanshyam Mann proposed openstack/os-traits master: DNM: testing gate on ubuntu focal https://review.opendev.org/744360 | |
| 08:04:57 | openstackgerrit | Jie Li proposed openstack/nova-specs master: add volume backed server rebuild spec https://review.opendev.org/739349 | |
| 08:31:11 | openstackgerrit | Jie Li proposed openstack/nova-specs master: add volume backed server rebuild spec https://review.opendev.org/739349 | |
| 14:33:11 | openstackgerrit | tony yan proposed openstack/nova master: add notification info when server attach or detach interface https://review.opendev.org/744273 | |
| 20:00:48 | openstack | Launchpad bug 1890057 in OpenStack Compute (nova) "EC2 instance_id_mappings are never deleted" [Undecided,New] | |
| 20:00:48 | mnaser | found this nugget we have to figure out tomororw hopefully -- https://bugs.launchpad.net/nova/+bug/1890057 | |
| #openstack-nova - 2020-08-03 | |||
| 00:41:07 | openstackgerrit | Brin Zhang proposed openstack/nova master: Fix CI failture: Introduce decorator>=3.4.0 to test-requirements.txt https://review.opendev.org/744385 | |
| 00:42:11 | openstackgerrit | Brin Zhang proposed openstack/nova master: [Trivial] Remove wrong format_message() conversion https://review.opendev.org/744280 | |
| 01:34:37 | openstackgerrit | Tony Su proposed openstack/nova master: Provider Config File: Function to further validate and retrieve configs https://review.opendev.org/676029 | |
| 01:34:38 | openstackgerrit | Tony Su proposed openstack/nova master: Provider Config File: Functions to merge provider configs to provider tree https://review.opendev.org/676522 | |
| 02:34:09 | openstackgerrit | Tony Su proposed openstack/nova master: Provider Config File: Enable loading and merging of provider configs https://review.opendev.org/693460 | |
| 03:05:40 | openstackgerrit | Brin Zhang proposed openstack/nova master: Cyborg evacuate support https://review.opendev.org/715326 | |
| 06:35:55 | devfaz_ | hello, just tried to evacuate an instance with ServerGroupAffinity - this failed in sheduler. Is this a known issue or a bug? | |
| 06:41:47 | openstackgerrit | renminmin proposed openstack/nova master: SRIOV: Fix instance with macvtap port live migration failed https://review.opendev.org/738432 | |
| 07:00:46 | devfaz_ | answering myself - yes, known issue - workaround could be SoftServerGroupAffinity or admin-api (force) during evacuate | |
| 07:03:43 | openstackgerrit | Tony Su proposed openstack/nova master: Provider Config File: Enable loading and merging of provider configs https://review.opendev.org/693460 | |
| 07:12:03 | gibi | good morning | |
| 07:43:34 | openstackgerrit | Tony Su proposed openstack/nova master: Provider Config File: Functions to merge provider configs to provider tree https://review.opendev.org/676522 | |
| 07:45:39 | nightmare_unreal | hello nova, I am trying to write a testcase for the patch. I am not able to resolve this error : " TypeError: '<' not supported between instances of 'Mock' and 'MagicMock'" . Has someone experienced this or how did you fixed it ? | |
| 07:45:52 | nightmare_unreal | patch : https://review.opendev.org/#/c/742210/ | |
| 08:04:45 | gibi | nightmare_unreal: after a quick glance. Did you mocked too many things there? e.g. you set the api version in the setUp and then mock the api_version.APIVersion in the test case | |
| 08:06:48 | nightmare_unreal | gibi yes , I was trying to mock api version . I am mocking 2 things the migration list methods and api_version methods. | |
| 08:07:40 | nightmare_unreal | if I don't provide @mock.patch , Then i get this error : ' TypeError: '<Mock name='mock.api_version' id='140452243441928'>' should be an instance of '<class 'novaclient.api_versions.APIVersion'>' | |
| 08:07:41 | gibi | I'm not at all familiar with the openstackclient unit tests but seems strange to set api version here https://review.opendev.org/#/c/742210/7/openstackclient/tests/unit/compute/v2/test_migration_list.py@65 and then mock it also here https://review.opendev.org/#/c/742210/7/openstackclient/tests/unit/compute/v2/test_migration_list.py@68 | |
| 08:08:24 | nightmare_unreal | true. | |
| 08:28:03 | nightmare_unreal | Solved gibi :) . Thanks | |
| 08:28:46 | gibi | cool ;0 | |
| 08:28:48 | gibi | ;) | |
| 08:32:56 | mlycka | Hullo | |
| 09:06:31 | lyarwood | Morning all | |
| 09:06:33 | lyarwood | gibi: welcome back | |
| 09:06:52 | lyarwood | is anyone working on the lower-constraints job failure on master? | |
| 09:07:40 | lyarwood | https://zuul.opendev.org/t/openstack/builds?job_name=openstack-tox-lower-constraints&project=openstack%2Fnova&branch=master | |
| 09:07:48 | lyarwood | if not I'll take a swing at it now | |
| 09:27:19 | openstackgerrit | Lee Yarwood proposed openstack/nova master: WIP lower-constraints: Bump dogpile.cache to 1.0.1 and decorator to 4.0.0 https://review.opendev.org/744426 | |
| 09:27:19 | openstackgerrit | Lee Yarwood proposed openstack/nova master: compute: Don't delete the original attachment during pre LM rollback https://review.opendev.org/743319 | |
| 09:41:05 | luyao | stephenfin: I replied your comments on https://review.opendev.org/#/c/714653/5, please let me know if you still have question about it. | |
| 09:41:59 | lyarwood | luyao: FYI stephenfin is out today | |
| 09:44:59 | kashyap | It's a holiday in Ireland, IIRC | |
| 09:47:02 | luyao | lyarwood: OK, thanks | |
| 10:16:22 | openstack | bug 1890064 in OpenStack Compute (nova) "CI failture: ImportError: cannot import decorate" [High,In progress] https://launchpad.net/bugs/1890064 - Assigned to Brin Zhang (zhangbailin) | |
| 10:16:22 | brinzhang_ | lyarwood: the CI report a bug of importError, see bug 1890064 | |
| 10:16:58 | brinzhang_ | lyarwood: I submit a patch, but is seems not right, pls review https://review.opendev.org/#/c/744385/ | |
| 10:26:32 | lyarwood | brinzhang_: yeah that's isn't right | |
| 10:26:45 | lyarwood | brinzhang_: https://review.opendev.org/#/c/744426/ - the issue is in lower-constraints | |
| 10:27:07 | lyarwood | brinzhang_: for some reason in the job we are pulling in a newer version of dogpile.cache | |
| 10:27:21 | lyarwood | brinzhang_: I couldn't reproduce it on Fedora, trying again on bionic now | |
| 10:30:45 | brinzhang_ | lyarwood: ok, left comment in the patch | |
| 10:43:01 | gibi | lyarwood: o/ | |
| 10:49:18 | lyarwood | does anyone know the version of setuptools used on our jobs? | |
| 10:49:31 | lyarwood | in the lower-constraint 18.04 based jobs | |
| 10:51:06 | lyarwood | with the distro provided 39.0.1-2 version lower-constraints fails on MarkupSafe==1.0 | |
| 10:52:34 | lyarwood | ah wait lower-constraints is pulling it in | |
| 10:55:40 | lyarwood | what a total shower | |
| 10:55:59 | lyarwood | so we can't reproduce these jobs on a freshly installed 18.04 because all of the workarounds in the image | |
| 10:56:02 | lyarwood | gah! | |
| 11:45:57 | openstackgerrit | Elod Illes proposed openstack/nova stable/pike: Fix os-simple-tenant-usage result order https://review.opendev.org/709600 | |
| 11:58:08 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Delete non tag related device metadata during detach https://review.opendev.org/741436 | |
| 12:40:38 | openstackgerrit | Lee Yarwood proposed openstack/nova master: WIP lower-constraints: Pin to python3.6 https://review.opendev.org/744458 | |
| 12:40:38 | openstackgerrit | Lee Yarwood proposed openstack/nova master: WIP lower-constraints: Update and address cyclic dependencies https://review.opendev.org/744426 | |
| 12:57:56 | openstackgerrit | Lee Yarwood proposed openstack/nova master: WIP lower-constraints: Update and address cyclic dependencies https://review.opendev.org/744426 | |
| 12:57:58 | lyarwood | gibi: ^ if you have time btw, I need to drop for a few hours in ~30mins but will pick this back up this evening if it's still broken. | |
| 12:58:16 | gibi | lyarwood: thanks | |
| 12:58:21 | gibi | I will try to look | |
| 13:01:44 | gibi | lyarwood: is there a way to reproduce the dependency problem locally? | |
| 13:02:47 | lyarwood | gibi: tox -e lower-constraints --notests | |
| 13:02:53 | gibi | ack | |
| 13:02:54 | gibi | thanks | |
| 13:03:11 | lyarwood | gibi: ^ but even then I had to update lower-contraints.txt for it to even build that env for me locally | |
| 13:03:29 | lyarwood | gibi: to use MarkupSafe==1.1.0 | |
| 13:03:55 | lyarwood | gibi: as somthing here is pulling in the latest version of setuptools | |
| 13:05:14 | gibi | I do not promise to get to it but it sounds juicy | |
| 13:07:10 | lyarwood | gibi: yup cool no issues, I'll work on it this evening either way | |
| 13:37:24 | openstackgerrit | Kevin Zhao proposed openstack/nova master: [WIP] CI: add tempest-integrated-compute-aarch64 job https://review.opendev.org/714439 | |
| 14:01:32 | smcginnis | lyarwood: Responded on the ML, but I'd hold off on playing whack a mole with packages. The job itself appears to be broken somehow. | |
| 14:02:14 | smcginnis | lyarwood: Would be interesting to know what version you have locally if you are able to reproduce this (.tox/lower-constraints/bin/pip --version) | |