| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-09-23 | |||
| 15:02:41 | mriedem | but it's hard to say without just proposing | |
| 15:02:57 | mriedem | efried: and we'd be running nova-next with 4 test workers rather than 2 | |
| 15:03:35 | sean-k-mooney | i did a multinoe tempest full fun for one of the cpu pinning jobs | |
| 15:03:42 | sean-k-mooney | it was just aboud 2 hours | |
| 15:04:01 | mriedem | tempest-integrated-compute SUCCESS in 1h 38m 29s | |
| 15:04:17 | mriedem | i'd say if we can stay within a reasonable comparison time-wise to tempest-integrated-compute it's a win | |
| 15:04:29 | mriedem | so make nova-next multinode and drop tempest-slow-py3 from nova runs | |
| 15:05:42 | sean-k-mooney | oh sorry tempest-full with concurrance:1 is just about 2 hours | |
| 15:06:04 | sean-k-mooney | so with concurrance:4 it should be closer to 1.5h as you suggested | |
| 15:08:41 | mriedem | stephenfin: was there any reason why you didn't remove this cellsv1 mention here? https://github.com/openstack/nova/blob/5a1c2d4ffa0815e874f373a87eb38b1833d03b24/nova/conductor/manager.py#L567 | |
| 15:09:50 | efried | sean-k-mooney: do you understand mriedem's suggestion enough to propose... whatever change(s) are necessary? | |
| 15:10:10 | stephenfin | mriedem: I started doing it but got stuck because of highly coupled tests https://review.opendev.org/#/c/651316/2/nova/conductor/manager.py I'd prefer to leave it until that patch is finished | |
| 15:10:27 | stephenfin | Or at least put in a TODO to remove all the cells v1 stuff | |
| 15:11:24 | mriedem | stephenfin: that patch is wrong anyway | |
| 15:11:32 | mriedem | you can't just drop compat code w/o a major rpc version bump | |
| 15:13:51 | stephenfin | Yeah, I don't get why we do that. I get that we can't change the signature but can't we start crapping out if a too-old client calls us? | |
| 15:14:15 | stephenfin | i.e. 'if request_spec if None: raise Exception('too old')' | |
| 15:14:18 | mriedem | i'll let dansmith answer that one | |
| 15:14:42 | cdent | if cdent: raise Exception('too old') | |
| 15:15:13 | mriedem | i ask b/c i want to drop ocata-era error handling in _populate_instance_mapping which would only be hit in the cells v1 case, which is no longer possible | |
| 15:15:35 | dansmith | stephenfin: mriedem: Of course we *can* but we're breaking the contract/rules of that 5.x means vs 6.x. So yeah, we could go all wild-west and just deprecate minor versions by raising random exceptions in the receiving code, | |
| 15:15:52 | dansmith | but then it becomes hard to reason about when we can remove things and what the impacts will be | |
| 15:16:15 | dansmith | the version numbers and rules are there to make it easier for the humans to know what falls into what bucket, IMHO | |
| 15:17:14 | mriedem | i know, i just pinged you since you're better with the words on this | |
| 15:17:21 | mriedem | "we can but we shouldn't" | |
| 15:17:23 | dansmith | okay | |
| 15:17:26 | mriedem | bad habits etc | |
| 15:17:52 | stephenfin | Would bumping the major version each release be too expensive? | |
| 15:18:01 | mriedem | conductor doesn't need to change that often | |
| 15:18:02 | dansmith | it used to be done almost every release | |
| 15:18:08 | dansmith | but we don't change that much anymore, | |
| 15:18:12 | mriedem | but we are definitely due for a conductor comptue task api 2.0 | |
| 15:18:12 | dansmith | so yeah, I think it's too expensive | |
| 15:18:20 | mriedem | b/c there is a lot of old shit in here | |
| 15:18:31 | stephenfin | maybe that's the solution so | |
| 15:18:37 | mriedem | once we go to 2.0 that _populate_instance_mapping just gets dropped | |
| 15:19:11 | stephenfin | I can't remove that stuff yet but I can go to 2.0 and then drop that stuff | |
| 15:19:23 | openstackgerrit | Merged openstack/os-resource-classes master: Update the constraints url https://review.opendev.org/683872 | |
| 15:19:38 | stephenfin | mriedem: While I have you - can we remove the os-networks entirely when we drop nova-net? | |
| 15:20:08 | mriedem | i had an etherpad with notes about the more complicated apis that involved networks b/c they aren't all nova-net only anymore | |
| 15:20:09 | stephenfin | It seems some of the APIs work with neutron but most don't, and I'm trying to decide if we should selectively 404 them or 404 everything | |
| 15:20:15 | mriedem | that's the issue | |
| 15:20:24 | mriedem | if there are apis that work with neutron we can't just 410 those | |
| 15:20:30 | stephenfin | 410, sorry | |
| 15:20:32 | stephenfin | yeah | |
| 15:22:03 | stephenfin | I was afraid you'd say that :( Time to rework again | |
| 15:22:09 | stephenfin | Any idea where that old etherpad is? | |
| 15:22:10 | mriedem | btw this is the last time we did an rpc api major version bump https://review.opendev.org/#/c/541005/ | |
| 15:22:31 | mriedem | https://etherpad.openstack.org/p/nova-network-removal-rocky | |
| 15:22:39 | stephenfin | I've a rough idea from converting all the API sample functional tests over but maybe I've missed some stuff | |
| 15:25:24 | stephenfin | gibi: Just in case, you haven't tried adding floating IP stuff to NeutronFixture or some subclass, have you? | |
| 15:25:52 | gibi | stephenfin: let me check some notification tests | |
| 15:26:27 | stephenfin | gibi: It would be local if so. We don't have it in tree | |
| 15:26:54 | stephenfin | at least searching for 'create_floatingip' doesn't turn up anything but mocks/the actual call | |
| 15:28:32 | gibi | stephenfin: unfortunately no, the IpPayload only contains fixed ips | |
| 15:28:44 | stephenfin | damn | |
| 15:28:46 | stephenfin | thanks for checking | |
| 15:28:49 | gibi | np | |
| 15:37:36 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Make nova-next multinode and drop tempest-slow-py3 https://review.opendev.org/683988 | |
| 15:37:37 | mriedem | efried: i think this is what you're looking for ^ | |
| 15:37:50 | efried | thanks mriedem | |
| 15:48:22 | dansmith | mriedem: so on your remap sanity check patch, | |
| 15:48:33 | dansmith | mriedem: it looks to me like you're now letting whichever one comes last win, right? | |
| 15:49:10 | dansmith | mriedem: really what we want is to let the non-cell0 one win, if possible | |
| 15:49:49 | mriedem | dansmith: in the bury in cell0 case, if it's already mapped we don't bury it in cell0, | |
| 15:50:00 | mriedem | in the non-cell0 case, we log an error and map it to the cell we just scheduled it to | |
| 15:50:05 | mriedem | which would not be cell0 | |
| 15:50:32 | dansmith | mriedem: oh I see, you created a helper but only call it in the non-cell0 case | |
| 15:50:33 | mriedem | i think those are both letting the non-cell0 ones win | |
| 15:50:37 | mriedem | yes | |
| 15:50:45 | dansmith | I assumed you were calling it from both places | |
| 15:50:45 | dansmith | okay | |
| 15:50:48 | mriedem | b/c i didn't want to have to mock 100 LOC just to test 1 line change in schedule_and_build_instances | |
| 15:50:59 | dansmith | sounds like cheating | |
| 15:51:39 | mriedem | schedule_and_build_instances is a monster | |
| 15:52:18 | mriedem | i also noted that if we drop that InstanceMappingNotFound pre-ocata compat in _bury_in_cell0 the _bury_in_cell0 method can just call the new _map_instance_to_cell method | |
| 15:52:44 | mriedem | we have a bunch of pre-ocata pre-cells v2 compat handlers all over the conductor task manager code which it'd be nice to remove | |
| 15:52:49 | mriedem | since they shouldn't be possible anymore | |
| 15:53:58 | dansmith | meh, I love piles of compat code | |
| 15:54:01 | dansmith | makes me feel nostalgic | |
| 15:54:24 | mriedem | all of the TODO(alaski)s to feel like an old sweater | |
| 15:54:25 | mriedem | i'll admit | |
| 15:54:32 | mriedem | s/to/do/ | |
| 15:57:25 | dansmith | lol | |
| 16:03:06 | stephenfin | I'm almost certain we've discussed this before, but why do we drop support for old microversion in novaclient? | |
| 16:03:16 | stephenfin | Referring to the API part, rather than the CLI | |
| 16:03:24 | stephenfin | Doesn't osc use that? | |
| 16:07:28 | mriedem | gibi: a few comments in https://review.opendev.org/#/c/683947/ but not worth holding it up, | |
| 16:07:36 | mriedem | gibi: but we might want to reconsider the fault message that the user could see | |
| 16:08:01 | mriedem | "oh hi your cloud provider is doing an upgrade and you shouldn't be trying to resize your instance right now bob!" | |
| 16:08:43 | mriedem | stephenfin: where have we dropped support for old microversions in novaclient? | |
| 16:08:52 | gibi | mriedem: the funny thing is that I think we cannot hit that error after the fix | |
| 16:09:04 | gibi | mriedem: so we could even remove it | |
| 16:09:21 | stephenfin | mriedem: e.g. 01fb16533bf562f39fe822bc12b9cc34b8580359 | |
| 16:09:30 | mriedem | gibi: well, technically your fix is in conductor checking it's config but the computes involved in the resize/cold migrate could have different config with different pins | |
| 16:09:53 | mriedem | stephenfin: that one broke osc | |
| 16:09:55 | gibi | mriedem: can we pin our computes differently? how will they talk to each other? | |
| 16:10:08 | mriedem | and they had to fix to avoid using novaclient and hit the api directly | |
| 16:10:26 | mriedem | gibi: well i meant if your conductor was unpinned but the computes were, or not restarted yet or something | |
| 16:10:28 | mriedem | after unpinning them | |