| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-02-23 | |||
| 19:47:15 | openstackgerrit | Merged openstack/python-novaclient master: Remove six https://review.opendev.org/708831 | |
| 19:47:16 | openstackgerrit | Merged openstack/python-novaclient master: Bump to hacking 2.x https://review.opendev.org/708832 | |
| 19:47:17 | openstackgerrit | Merged openstack/python-novaclient master: Random cleanups https://review.opendev.org/709081 | |
| #openstack-nova - 2020-02-24 | |||
| 01:13:39 | brinzhang_ | gmann: refresh default policy of os-instance-action API meet a problem, could you please review this patch [1]? I have a SPEC need this change. [1]https://review.opendev.org/#/c/707777/2/nova/tests/unit/policies/test_instance_actions.py@131 | |
| 01:21:35 | brinzhang_ | gmann, johnthetubaguy: There is an issue is the context.can(xx,xx, fatal=false), since fatal=False the unauthorized role cannot get the expect raise exception, so I would like to get your suggestion, and than fast complete this change | |
| 01:24:46 | brinzhang_ | gmann, johnthetubaguy: and another change is the destroy-instance-with-datavolume SPEC, and I will refresh the os-volumes API's default policy, because of I will add a new PATCH API to the os-volumes API. In order to use the new policy, I had to do this work in advance, please unstand me do this. | |
| 01:26:23 | brinzhang_ | gmann, johnthetubaguy: s/os-volumes API/os-volumes-attachments API | |
| 02:00:21 | openstackgerrit | Brin Zhang proposed openstack/nova master: Introduce scope_types in os-volumes-attachments policy https://review.opendev.org/709388 | |
| 08:55:24 | openstackgerrit | Fan Zhang proposed openstack/nova master: Fix hypervisors paginted collection_name. https://review.opendev.org/709431 | |
| 10:19:51 | openstackgerrit | Mark Goddard proposed openstack/nova master: Prevent deletion of a compute node belonging to another host https://review.opendev.org/694802 | |
| 10:31:18 | bauzas | ergh, fuckingly impacted by https://review.opendev.org/#/c/709389/ | |
| 10:34:06 | bauzas | frickler: would appreciate a +2 on ^ in order to fix my issues | |
| 10:34:13 | bauzas | (I just recreated my tox venv :() | |
| 10:40:52 | frickler | bauzas: humm, usually I don't vote on reqs. but I guess in this case I can reapply the +3 by prometheanfire and smcginnis | |
| 10:41:08 | bauzas | as you can | |
| 10:46:45 | frickler | guess I also need to promote it in gate, everything else is currently failing, right? | |
| 10:51:34 | frickler | infra-root: wanting to promote the reqs fix https://review.opendev.org/#/c/709389/ , zuul command is now broken by the thing this patch is fixing, too http://paste.openstack.org/show/789920/ | |
| 10:51:59 | frickler | ehm, wrong chan | |
| 10:59:56 | bauzas | sean-k-mooney: do you remember my issue with stestr last week ? | |
| 11:00:17 | bauzas | sean-k-mooney: I fully cleaned all my local eggs but I still had the issue | |
| 11:00:51 | bauzas | sean-k-mooney: and I finally was able to fix it by asking tox to use sdist (ie. removing use_develop in tox.ini) | |
| 11:01:30 | bauzas | weirod | |
| 11:01:33 | bauzas | weirdo | |
| 11:03:54 | bauzas | stephenfin: you're a pbr expert (heh), any idea why https://tox.readthedocs.io/en/latest/example/general.html#avoiding-expensive-sdist doesn't work for me and why by not asking use_develop=True works ? | |
| 11:04:00 | bauzas | for the nova repo | |
| 11:12:50 | bauzas | stephenfin: context is http://paste.openstack.org/show/789922/ and looks like I can't use use_develop, but I rather need to create a packaged install | |
| 11:32:32 | stephenfin | bauzas: Not sure, but I don't think that's the issue. I think the issue is with the latest major version of oslo.config | |
| 11:32:49 | stephenfin | We've been seeing issues with other projects like oslo.policy because of it | |
| 11:34:22 | stephenfin | bauzas: What do you mean it doesn't work though? | |
| 11:34:43 | bauzas | stephenfin: like I said, I get http://paste.openstack.org/show/789922/ by not modifying tox.ini | |
| 11:35:15 | bauzas | stephenfin: but if i ask to not using the development mode (ie. https://github.com/openstack/nova/blob/master/tox.ini#L11) then it works | |
| 11:35:30 | stephenfin | and if you remove that, things work? Can you give me the output until 'py37 run-test:' ? | |
| 11:36:32 | bauzas | stephenfin: yes, but WDYM by py36 run-test N | |
| 11:36:36 | bauzas | whoops | |
| 11:36:41 | bauzas | py37 run-test | |
| 11:36:57 | stephenfin | line 7 in the paste you provided | |
| 11:37:11 | stephenfin | I want to see what the difference in the commands is if you remove usedevelop | |
| 11:37:32 | bauzas | ah ah ok | |
| 11:38:32 | bauzas | stephenfin: http://paste.openstack.org/show/789923/ | |
| 11:39:14 | bauzas | stephenfin: also see the git diff tox.ini http://paste.openstack.org/show/789924/ | |
| 11:40:25 | stephenfin | you're also ignoring upper constraints too. Why's that necessary? | |
| 11:47:15 | sean-k-mooney | that is strange | |
| 11:47:23 | sean-k-mooney | bauzas: did your env break again | |
| 11:54:57 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Make image fixtures simpler to reference https://review.opendev.org/690403 | |
| 13:11:14 | openstackgerrit | Kevin Zhao proposed openstack/nova master: Add default cpu model for aarch64 https://review.opendev.org/709494 | |
| 13:26:52 | bauzas | stephenfin: because of https://review.opendev.org/#/c/709389/5 that wasn't merged yet | |
| 13:30:15 | openstackgerrit | Alexandre arents proposed openstack/nova master: Avoid allocation leak when deleting instance stuck in BUILD https://review.opendev.org/702368 | |
| 13:54:55 | bauzas | ERROR: Could not find a version that satisfies the requirement gitdb2===4.0.1 (from -c https://releases.openstack.org/constraints/upper/master (line 777)) (from versions: 2.0.0, 2.0.1, 2.0.2, 2.0.3, 2.0.4, 2.0.5, 2.0.6, 3.0.0, 3.0.1, 3.0.2, 3.0.3, 3.0.3.post1, 4.0.2) | |
| 13:55:36 | bauzas | but https://review.opendev.org/#/c/709389/ is merged and I can see gitdb==4.0.2 in https://opendev.org/openstack/requirements/raw/branch/master/upper-constraints.txt | |
| 13:56:17 | aarents | bauzas: got the same issue since this morning | |
| 13:56:31 | bauzas | aarents: was due to some bug, see the change | |
| 13:57:31 | aarents | bauzas: ok | |
| 14:02:23 | bauzas | mmm, removed my local pip cache, the URL was cached | |
| 14:07:10 | stephenfin | merged now, thankfully | |
| 14:11:13 | bauzas | okay, still puzzled by stestr not knowing the run command | |
| 14:11:30 | bauzas | stephenfin: looks like 'run' is a pbr entrypoint | |
| 14:11:40 | bauzas | https://github.com/mtreinish/stestr/blob/master/setup.cfg#L40 | |
| 14:11:48 | bauzas | and stestr relies on cliff | |
| 14:12:10 | bauzas | so I suspect something messed up with entrypoints when I use develop | |
| 14:12:19 | stephenfin | bauzas: '--remote_debug-port' is a nova config option though | |
| 14:12:50 | bauzas | stephenfin: sorry, where ? | |
| 14:12:56 | stephenfin | http://paste.openstack.org/show/789922/ | |
| 14:12:59 | stephenfin | from that | |
| 14:13:15 | stephenfin | what seems to be happening is arguments from some command are being registered against others | |
| 14:13:30 | stephenfin | it's like argparse is maintaining some kind of global state | |
| 14:14:31 | stephenfin | bauzas: As proof, try this: virtualenv .venv; source .venv; pip install stestr; stestr -h | |
| 14:14:35 | bauzas | oh, you're right | |
| 14:14:42 | stephenfin | you'll get a totally different output | |
| 14:14:47 | bauzas | I can't see any of the usual stestr options | |
| 14:14:51 | stephenfin | exactly | |
| 14:15:01 | stephenfin | I just haven't figured out why | |
| 14:15:41 | bauzas | I'm just creating a venv as a tox target | |
| 14:15:43 | stephenfin | We worked around it in glance with some patch that I'll need to go root for, but the ultimate issue is that there is some kind of global state being maintained by cliff or oslo.config that there shouldn't be | |
| 14:15:49 | bauzas | to see whether stestr is messed up too | |
| 14:16:39 | stephenfin | bauzas: I assume you're not seeing this issue on master? | |
| 14:16:54 | bauzas | I'm seeing it on any branches | |
| 14:16:55 | bauzas | but, | |
| 14:17:18 | bauzas | if I reclone nova to another directory, it then works | |
| 14:17:31 | bauzas | so the issue is only on the main git repo | |
| 14:17:34 | stephenfin | Backup anything you want and 'git clean -xfd' ? | |
| 14:17:56 | stephenfin | Maybe you have a hangover pyc file that the interpreter is loading | |
| 14:18:04 | stephenfin | I've seen that cause other issues in the past | |
| 14:18:24 | bauzas | maybe | |
| 14:18:41 | stephenfin | ('git clean -xfd' will wipe all untracked files including those ignored by '.gitgnore' files, btw) | |
| 14:18:56 | stephenfin | (I know you probably know that already) | |
| 14:21:31 | bauzas | indeed, but haven't thought about it | |
| 14:23:15 | bauzas | it does fix the problem | |
| 14:23:36 | bauzas | have no f*** idea what the issue was, but now stestr works with develop mode | |
| 14:23:57 | bauzas | (which is understandable, since I asked to delete all pyc's) | |
| 14:36:09 | efried | I'm going to take at least the morning, possibly the day. Email if anything needs my attention. o/ | |
| 14:36:20 | gibi | efried_pto: o/ | |
| 15:06:58 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Unplug VIFs as part of cleanup of networks https://review.opendev.org/663382 | |
| 15:06:58 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Functional test for UnexpectedDeletingTaskStateError https://review.opendev.org/689278 | |
| 15:06:59 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Fix incorrect vm and task state after build failure race https://review.opendev.org/689388 | |
| 15:08:04 | stephenfin | gibi: Think you could take a look at ^ this week? | |
| 15:10:04 | elod | lyarwood: hi, do we need to wait for any patch in Rocky (for final release before Extended Maintenance transition?) | |
| 15:11:02 | lyarwood | elod: there's nothing from my side that can't wait until after EM | |
| 15:11:24 | lyarwood | elod: RH isn't shipping Rocky anymore FWIW | |
| 15:12:22 | elod | lyarwood: i see, good. Then I guess we could create a release patch | |
| 15:12:31 | gibi | stephenfin: queued | |