| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-05-25 | |||
| 16:14:03 | cdent | what would that gain? | |
| 16:14:12 | dansmith | jaypipes: nobody wants to claim on compute right? | |
| 16:14:31 | jaypipes | it would mean we would solve the 80% case now. and come up with a solution for the 20% case later. | |
| 16:14:34 | edleafe | jaypipes: yeah, that sounds terrible | |
| 16:15:04 | edleafe | the only thing missing to solve 100% of the claims is the instance UUIDs. | |
| 16:15:20 | edleafe | As long as we pass those in some form to the scheduler, we are good | |
| 16:15:38 | edleafe | the only question is what is the best way to pass them | |
| 16:15:47 | jaypipes | edleafe: right, but what I'm saying is we solve that later and for the first patch(es), we bite off what we can chew for right now. | |
| 16:16:00 | jaypipes | instead of trying to solve everything at once. | |
| 16:16:17 | edleafe | IMO the patches are bite-sized | |
| 16:16:42 | jaypipes | edleafe: the patches might be. the debate isn't. :) | |
| 16:16:50 | mriedem | pass them over rpc | |
| 16:16:54 | mriedem | don't store them in the db | |
| 16:16:58 | mriedem | i thought that was clear? | |
| 16:17:07 | dansmith | mriedem: me too | |
| 16:17:10 | jaypipes | mriedem: nobody has suggested storing them in the DB. | |
| 16:17:13 | edleafe | mriedem: well, if that's the decision, then everything is ready for review | |
| 16:17:18 | dansmith | jaypipes: perhaps we can nail this down quicker in a hangout? | |
| 16:17:22 | dansmith | jaypipes: edleafe did :) | |
| 16:17:28 | jaypipes | he did? | |
| 16:17:31 | mriedem | if you put them in the reqspec they go in the db unless you monkey the object to not persist | |
| 16:17:35 | mriedem | but that seems wrong | |
| 16:17:39 | edleafe | I suggested putting them in the reqspec | |
| 16:17:51 | jaypipes | ok. | |
| 16:17:54 | edleafe | so I guess DB storage is inevitable :) | |
| 16:17:59 | mriedem | i mean sure you can throw any attribute on an object and it won't persist | |
| 16:18:07 | mriedem | if it's not in the model | |
| 16:18:09 | jaypipes | right | |
| 16:18:12 | edleafe | jaypipes: to be fair, you suggested the same thing :) | |
| 16:18:18 | jaypipes | that's what I thought we were going to do.. | |
| 16:18:24 | edleafe | me too | |
| 16:18:40 | jaypipes | edleafe: not store in the DB. I suggested just tacking instance_uuids as a field on the object, since that would avoid an RPC API change. | |
| 16:18:43 | mriedem | it seems very not intuitive | |
| 16:18:51 | dansmith | jaypipes: it doesn't avoid an rpc change | |
| 16:18:55 | mriedem | 2 years from now someone is going to ask, wtf, why are these on this object but not in the db? | |
| 16:18:59 | dansmith | yeah | |
| 16:19:03 | dansmith | it's silly to do that | |
| 16:19:16 | cdent | there's nothing wrong with an rpc api change, and it is semantically clean. let's just do that. | |
| 16:19:17 | jaypipes | fine. call me silly. I don't care. | |
| 16:19:33 | edleafe | I've done it both ways, so I don't care either :) | |
| 16:19:33 | cdent | "that" == add a param | |
| 16:19:34 | dansmith | jaypipes: you know you're just generally silly anyway :) | |
| 16:19:38 | jaypipes | indeed. | |
| 16:19:54 | jaypipes | ok, follow on question... | |
| 16:20:37 | jaypipes | let's say we pass an instance_uuids param to the scheduler. cool, fine by me. But what about the whole "alternate destinations" thing that is designed to solve the no-cell-upcall issue? | |
| 16:20:46 | jaypipes | are we still going for that? | |
| 16:21:10 | dansmith | ... | |
| 16:21:22 | dansmith | have there been other suggestions on how to resolve that other than doing the alternates thing? | |
| 16:22:06 | edleafe | jaypipes: have you seen https://review.openstack.org/#/c/465767/ ? | |
| 16:22:21 | cdent | I gotta run, if there's a summary of this, can someone let me know somehow later this evening? | |
| 16:22:31 | edleafe | cdent: will do | |
| 16:22:36 | cdent | thanks | |
| 16:23:02 | jaypipes | dansmith: not that I'm aware of. was just trying to catch up my thoughts on this series. | |
| 16:23:08 | dansmith | okay | |
| 16:23:23 | edleafe | Adding alternates changes the result of select_destinations() from a list of hosts to a list of a list of hosts | |
| 16:24:12 | jaypipes | edleafe: and you/we are OK with that, yes?> | |
| 16:24:39 | edleafe | It wasn't my preference, no | |
| 16:24:52 | edleafe | But it was a big request from operators, so... | |
| 16:25:11 | openstackgerrit | OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/468089 | |
| 16:25:20 | mriedem | alex_xu: oomichi: we're kind of at an impass with some of the new cinder attachment reviews - john and i are the only reviewers on these but i'm also working on the patches too, so we need a 3rd core to help move some stuff along, can you help review the 3 bottom changes in this series which already have a +2 from john? https://review.openstack.org/#/c/456896/ | |
| 16:25:32 | mriedem | *impasse | |
| 16:29:35 | jaypipes | edleafe: is there a reason you undid Sylvain's changes from patch 10 on https://review.openstack.org/#/c/460177/10..11/nova/compute/resource_tracker.py? | |
| 16:29:57 | jaypipes | edleafe: I'd asked him to change the dict comprehension into a simple tracked_keys = set(self.tracked_instances.keys()) | |
| 16:30:03 | jaypipes | edleafe: and you undid that. | |
| 16:30:26 | jaypipes | edleafe: and changed instance_uuid -> allocation_uuid. | |
| 16:30:31 | edleafe | jaypipes: probably a rebase screw-up | |
| 16:30:34 | jaypipes | edleafe: just a case of rebase screw up? | |
| 16:30:40 | jaypipes | edleafe: jinx :) | |
| 16:30:46 | edleafe | I rebased my series on master, and his is in the series | |
| 16:30:53 | jaypipes | edleafe: could you please push quick update on that? | |
| 16:30:57 | jaypipes | edleafe: and I will quick approve. | |
| 16:30:57 | edleafe | I'll fix that in a bit | |
| 16:31:05 | jaypipes | since dansmith is already +2 on that | |
| 16:31:11 | edleafe | gimme a few | |
| 16:33:45 | dansmith | mriedem: what ovo impact do you think that context change has? | |
| 16:34:14 | mriedem | dansmith: don't know, but i've thought in the past when we talk about changes to context it can have implications for rpc | |
| 16:34:19 | mriedem | so i was waiting for you | |
| 16:34:24 | dansmith | mriedem: you might also want to get a read from tonyb or jamie on it.. there was some upgrade issue when we bumped oslo.context the last time | |
| 16:34:38 | dansmith | mriedem: yeah, rpc, but not ovo specifically that I know of | |
| 16:34:39 | mriedem | waiting for tony isn't going to be an option | |
| 16:34:50 | dansmith | tonyb was the one handling the upgrade issue last time | |
| 16:35:22 | openstackgerrit | Sean Dague proposed openstack/nova-specs master: oslo.middleware request_id spec https://review.openstack.org/468066 | |
| 16:36:33 | openstackgerrit | Sean Dague proposed openstack/nova master: Migrate to oslo request_id middleware - mv 2.46 https://review.openstack.org/467998 | |
| 16:36:46 | sdague | mriedem ok all the cross links and blueprint bits filed | |
| 16:37:30 | sdague | dansmith: how do you imagine it would have an impact? | |
| 16:38:33 | openstackgerrit | Ed Leafe proposed openstack/nova master: Have scheduler claim resources https://review.openstack.org/463993 | |
| 16:38:34 | openstackgerrit | Ed Leafe proposed openstack/nova master: Pass the number of hosts to return explicitly https://review.openstack.org/465767 | |
| 16:38:34 | openstackgerrit | Ed Leafe proposed openstack/nova master: Remove selecting host based on host_subset_size https://review.openstack.org/465174 | |
| 16:38:35 | openstackgerrit | Ed Leafe proposed openstack/nova master: Make the method to put allocations public https://review.openstack.org/464086 | |
| 16:38:36 | openstackgerrit | Ed Leafe proposed openstack/nova master: Pass a list of instance UUIDs to scheduler https://review.openstack.org/465171 | |
| 16:38:36 | openstackgerrit | Ed Leafe proposed openstack/nova master: Return multiple hosts from select_destinations https://review.openstack.org/465768 | |
| 16:38:38 | dansmith | sdague: i don't recall the details, which is why I said, but I think that during one of the bumps, we could end up with say a new compute sending a context to an old compute that it couldn't parse or something.. like maybe it wasn't defensive enough in the original version maybe? | |
| 16:38:39 | edleafe | jaypipes: ^^ should be fixed | |
| 16:39:02 | sdague | dansmith: maybe, the from_dict is pretty defensive | |
| 16:39:09 | dansmith | sdague: if the super'd method is defensive with gets like ours is there, then I guess we're okay for this change, I just don't know | |
| 16:39:10 | sdague | it's all .get() calls | |
| 16:39:25 | mriedem | hmm, | |
| 16:39:31 | dansmith | sdague: yeah, either it wasn't before or there was some really subtle detail | |
| 16:39:31 | mriedem | we do have tests for this | |
| 16:39:33 | sdague | https://github.com/openstack/oslo.context/blob/cce83ad5e12baf2edc0e12c2944c587224e71148/oslo_context/context.py#L358-L382 | |