| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-12-12 | |||
| 13:32:57 | mriedem | and as discussed with dan in here last night it's also a hack | |
| 13:34:22 | mriedem | efried: while you're about, and i know you've got a full plate, but this is a 7 LOC FUP that if you will approve will help unblock several other approved changes in the series https://review.opendev.org/#/c/698028/ | |
| 13:36:11 | efried | mriedem: done. | |
| 13:36:15 | mriedem | thanks | |
| 13:37:07 | openstackgerrit | Merged openstack/nova master: functional: Remove 'api' parameter https://review.opendev.org/697694 | |
| 13:42:57 | openstackgerrit | Matt Riedemann proposed openstack/nova master: DNM: debug cross-cell resize https://review.opendev.org/698304 | |
| 13:53:52 | efried | Nova meeting in 7 minutes in -meeting | |
| 13:54:26 | shilpasd | efried: (y) | |
| 14:05:58 | openstack | Launchpad bug 1855493 in OpenStack Compute (nova) "py38 unit test failure - test_validate_patternProperties_fails" [Undecided,New] | |
| 14:05:58 | mriedem | stephenfin: do you have a py38 system? https://bugs.launchpad.net/nova/+bug/1855493 | |
| 14:06:06 | mriedem | i know you love bleeding edge fedora | |
| 14:07:52 | sean-k-mooney | kasyap like bleeding edge fedora | |
| 14:08:52 | sean-k-mooney | mriedem: stephenfin and i tend to use an alternitive operating system for dev although i think stephenfin still runs fedora on his laptop | |
| 14:09:40 | mriedem | alternative, like early 90s R.E.M.? | |
| 14:10:22 | sean-k-mooney | that is one way do describe ubuntu 18.04 | |
| 14:10:53 | sean-k-mooney | which happens to have python 3.8 in its packages becasue i just installed it | |
| 14:11:31 | sean-k-mooney | ill run the test and see what happens | |
| 14:13:34 | sean-k-mooney | from the bug report it like like this a byte string vs text string issue if im reading that right | |
| 14:15:26 | shilpasd | stephenfin: where i can get team photo during shanghai PTG? | |
| 14:15:59 | sean-k-mooney | mriedem: so yes it fails as reported | |
| 14:16:06 | sean-k-mooney | there are some other warning too | |
| 14:16:40 | mriedem | shilpasd: there was a link in the ML | |
| 14:17:14 | shilpasd | mriedem: ok, sure will check, thanks | |
| 14:18:54 | mriedem | shilpasd: https://www.dropbox.com/sh/1my6wdtuc1hf58o/AACU49pjWxzFNzcZJgjLG8n1a?dl=0 | |
| 14:19:51 | shilpasd | mriedem: thanks | |
| 14:22:43 | luyao | stephenfin: Are you around? | |
| 14:26:37 | gmann | mriedem: stephenfin FYI. there is plan to add py38 unit test job in ussuri testing runtime. - https://review.opendev.org/#/c/693401/ | |
| 14:26:51 | gmann | this is hold till V cycle name is decided | |
| 14:27:26 | luyao | sean-k-mooney: could you review vpmem live migration spec if you get time, you are familiar with the vpmem feature, so do stephenfin, but he seems not here :) https://review.opendev.org/#/c/695863/ | |
| 14:32:34 | luyao | stephenfin, sean-k-mooney : FYI. the spec is short, an initial patch based on spec is also submitted, I has verified its function on hardware env, still need further polishing | |
| 14:34:57 | stephenfin | mriedem: I do. Will try reproduce now | |
| 14:34:59 | stephenfin | luyao: I'll try hit that before EOD | |
| 14:35:35 | stephenfin | shilpasd: You should ask Kendall Nelson but I've forgotten her IRC nick /o\ | |
| 14:36:14 | luyao | stephenfin: thanks :D | |
| 14:36:41 | efried | stephenfin: it's diablo_rojo. But mriedem found them for shilpasd. | |
| 14:37:00 | stephenfin | efried: whoops, missed that | |
| 14:37:01 | stephenfin | cool cool | |
| 14:37:29 | efried | stephenfin: diablo rojo is spanish for diabhal dearg | |
| 14:37:39 | efried | (how's your gaelic :P) | |
| 14:37:55 | shilpasd | efried: thanks | |
| 14:38:04 | shilpasd | stephenfin: i got it | |
| 14:38:08 | shilpasd | thanks | |
| 14:39:05 | efried | Is it still a thing? | |
| 14:39:05 | efried | https://review.opendev.org/#/c/270289 | |
| 14:39:05 | efried | lyarwood: this is currently the oldest patch on the review dashboard (http://status.openstack.org/reviews/#nova) | |
| 14:39:32 | efried | oh, I see you restored it recently, so I guess it must be. | |
| 14:39:46 | lyarwood | efried: yeah the spec has landed, I was going to ask if I should use a fresh change-id given this is so old and crusty | |
| 14:40:12 | lyarwood | efried: hopefully pushing things for review later today once I've worked on some issues with the tempest tests | |
| 14:40:16 | openstackgerrit | Guo Jingyu proposed openstack/nova master: Make scheduling more debuggable https://review.opendev.org/698421 | |
| 14:40:25 | efried | lyarwood: is there anything useful in the history? | |
| 14:40:39 | efried | I assume the rebase will be a train wreck | |
| 14:40:44 | lyarwood | efried: not really tbh | |
| 14:40:49 | lyarwood | efried: yeah it was | |
| 14:41:04 | efried | Then meh, use a new change-id but ref the old one in the commit message | |
| 14:41:28 | lyarwood | ack will do | |
| 14:43:44 | dansmith | mriedem: for sure | |
| 14:44:09 | dansmith | mriedem: theoretically you could handle it in rpcapi but definitely should have a bump imho | |
| 14:44:33 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Move _update_pci_request_spec_with_allocated_interface_name https://review.opendev.org/696574 | |
| 14:44:33 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Move _get_request_group_mapping() to RequestSpec https://review.opendev.org/696541 | |
| 14:46:15 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Support live migration with qos ports https://review.opendev.org/695905 | |
| 14:47:41 | huaqiang | stephenfin: morning | |
| 14:48:12 | huaqiang | stephenfin: for https://review.opendev.org/#/c/668656/, can you make a review when you have time | |
| 14:48:27 | gibi | efried, mriedem: fixed up your comments in the qos live migration series | |
| 14:48:40 | stephenfin | mriedem: Yeah. I can reproduce that too. Will try root out a fix, assuming sean-k-mooney hasn't fixed it yet | |
| 14:49:18 | sean-k-mooney | luyao: left a few comments but more or less i think the direction of the spec is good. just grabing something to eat so i wont be around for a the next 30 mins or so | |
| 14:49:39 | mriedem | gibi: thanks, much better | |
| 14:49:55 | sean-k-mooney | stephenfin: nope havent started was going to take a look later today but have a few other thing to get done first | |
| 14:54:05 | luyao | sean-k-mooney: thanks for your comments:) | |
| 14:56:39 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Keep pre-commit inline with hacking and fix whitespace https://review.opendev.org/674057 | |
| 14:57:17 | stephenfin | sean-k-mooney: Cool | |
| 14:57:58 | stephenfin | also, "foo is 'string'" raises a syntax deprecation warning in Python 3, so we should probably merge https://review.opendev.org/#/c/695732/ sooner rather than later | |
| 14:59:01 | huaqiang | :xa | |
| 15:01:15 | efried | stephenfin: +2 | |
| 15:02:28 | stephenfin | ta | |
| 15:02:48 | stephenfin | huaqiang: I've a few other things ahead of it, but I'll try | |
| 15:05:28 | huaqiang | stephenfin: I see. I especially need comments for if we need to create 'mixed' instance through 'resources:PCPU' and 'resources:VCPU' and specifying the count of dedicated vCPU number in flavor/image or using a CPU mask for them | |
| 15:05:57 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Don't hardcore Python versions in test https://review.opendev.org/698732 | |
| 15:05:58 | stephenfin | efried, mriedem, sean-k-mooney: ^ | |
| 15:06:05 | stephenfin | hardcore. FML | |
| 15:06:18 | mriedem | stephenfin: efried: what's wrong with the dict usage for urlencode here? https://review.opendev.org/#/c/695732/2/nova/tests/unit/api/openstack/compute/test_volumes.py | |
| 15:06:18 | efried | love it | |
| 15:06:19 | huaqiang | stephenfin: I know we have had some discussion on these two questions, but there are some other opionions on them | |
| 15:06:35 | efried | mriedem: URLs allow repetition of keys; dicts don't. | |
| 15:06:45 | efried | mriedem: querystrings are also ordered. | |
| 15:06:50 | mriedem | oh guh | |
| 15:06:56 | mriedem | didn't even think about the duplicate keys | |
| 15:06:58 | efried | So old encoders just kinda went with it | |
| 15:07:09 | efried | but good code should be explicit | |
| 15:07:44 | huaqiang | stephenfin: so I want to reach consitency on these two questions. | |
| 15:08:52 | KeithMnemonic | Hello mridem. Do you think there will be any movement on the queens/pike cherry picks of your "marker" fix? https://review.opendev.org/#/c/690725/2. I know those are EOL but since the patches were created was not sure if you thought they could still get in? | |
| 15:09:08 | KeithMnemonic | argh mriedem: | |
| 15:09:34 | awalende | Anyone knows what nova-compute wants to tell me when trying to attach a volume to a new instance on a freshly deployed ubuntu 18.04 compute node?: unable to execute QEMU command 'device_add': Property 'virtio-blk-device.drive' can't find value 'drive-virtio-disk1' | |
| 15:11:13 | mriedem | KeithMnemonic: you have to ask stable cores | |
| 15:11:31 | KeithMnemonic | ok thanks | |
| 15:11:38 | mriedem | those branches aren't EOL either, | |
| 15:11:42 | mriedem | they are EM = extended maintenance | |
| 15:11:53 | mriedem | https://review.opendev.org/#/admin/groups/540,members | |
| 15:15:52 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Don't hardcore Python versions in test https://review.opendev.org/698732 | |
| 15:15:58 | KeithMnemonic | mriedem is there a preferred way to ping the "stable cores" | |
| 15:16:08 | efried | stephenfin: I thought you were going to fix that commit title? | |