| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-09-25 | |||
| 21:04:54 | openstackgerrit | Dan Smith proposed openstack/nova master: Pre-create migration object https://review.openstack.org/498950 | |
| 21:04:55 | openstackgerrit | Dan Smith proposed openstack/nova master: Refactor resource tracker to account for migration allocations https://review.openstack.org/506419 | |
| 21:04:55 | mriedem | https://docs.openstack.org/nova/latest/reference/api-microversion-history.html#id14 | |
| 21:04:55 | openstackgerrit | Dan Smith proposed openstack/nova master: Revert allocations by migration uuid https://review.openstack.org/498949 | |
| 21:04:56 | openstackgerrit | Dan Smith proposed openstack/nova master: Make migration uuid hold allocations for migrating instances https://review.openstack.org/506420 | |
| 21:05:05 | dansmith | jaypipes: ready for another flogging I think ^ | |
| 21:05:26 | mriedem | microversion >= 2.16 GET /servers/detail will lookup host status if you're admin | |
| 21:05:34 | mriedem | per instance | |
| 21:05:47 | jaypipes | dansmith: I'll ready my cat-o-nine-tails then | |
| 21:06:22 | dansmith | jaypipes: oooh, you know what I liiiiike | |
| 21:06:35 | jaypipes | lol | |
| 21:11:31 | mriedem | dansmith: 15.73s average to list servers with details at microversion 2.53 | |
| 21:11:46 | mriedem | double that of listing servers with microversion 2.1 | |
| 21:11:57 | dansmith | that sucks | |
| 21:12:03 | dansmith | this is all baseline still right? | |
| 21:12:13 | mriedem | ypu | |
| 21:12:15 | mriedem | *up | |
| 21:12:16 | mriedem | gdi | |
| 21:12:17 | mriedem | YES | |
| 21:12:27 | mriedem | and that's demo tenant | |
| 21:13:19 | mriedem | fuck man, looking at the code for microversion 2.16 if you're an admin, we aren't caching host status | |
| 21:15:38 | mriedem | so i think if you're an admin, listing instances from all tenants, with microversion >= 2.16, we're doing a lazy-load of instance.services on every instance | |
| 21:16:22 | mriedem | and that does a server group is_up call per service if it's not forced_down or disabled | |
| 21:16:23 | mriedem | jesus | |
| 21:17:29 | mriedem | oh nvm, for instance list it's a bit smarter | |
| 21:17:33 | mriedem | whew | |
| 21:26:30 | openstackgerrit | Chris Dent proposed openstack/nova master: DNM: Don't monkey patch eventlet in functional or unit tests https://review.openstack.org/506668 | |
| 21:26:51 | mriedem | ah we don't lazy-load tags https://github.com/openstack/nova/blob/29ef2474d9d6a59ae6859b5b01caad350252c23e/nova/api/openstack/compute/servers.py#L297 | |
| 21:26:55 | mriedem | we join those in the db query | |
| 21:27:01 | mriedem | same for services | |
| 21:27:13 | dansmith | mriedem: I shot you that link right? | |
| 21:27:18 | mriedem | https://github.com/openstack/nova/blob/29ef2474d9d6a59ae6859b5b01caad350252c23e/nova/api/openstack/compute/servers.py#L295 should probably be version constrained | |
| 21:27:24 | mriedem | dansmith: you send me something else | |
| 21:27:36 | mriedem | you sent me https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api.py#L2229-L2229 | |
| 21:27:48 | mriedem | but that's in a conditional block | |
| 21:28:19 | dansmith | oh I see, but actually I guess it's because they end up in columns_to_join_new | |
| 21:28:43 | mriedem | https://review.openstack.org/#/c/355558/ | |
| 21:28:47 | mriedem | ^ doesn't go far enough | |
| 21:29:44 | dansmith | that's services | |
| 21:29:49 | mriedem | i know, | |
| 21:29:53 | dansmith | but yeah, that won't do anything I think, | |
| 21:30:01 | dansmith | because it's not in the _expected_cols() thing in instance | |
| 21:32:10 | mriedem | it worked because it was in _INSTANCE_OPTIONAL_JOINED_FIELDS | |
| 21:32:12 | mriedem | for services | |
| 21:32:19 | mriedem | fault wasn't before queens | |
| 21:32:30 | dansmith | services is in optional? | |
| 21:32:33 | mriedem | yup | |
| 21:33:30 | dansmith | oh we actually have a .services on instance | |
| 21:33:37 | dansmith | I didn't remember | |
| 21:35:55 | mriedem | https://bugs.launchpad.net/nova/+bug/1719460 | |
| 21:35:55 | mriedem | yeah | |
| 21:35:56 | openstack | Launchpad bug 1719460 in OpenStack Compute (nova) "instance.services loading when listing instances regardless of microversion" [Medium,Triaged] | |
| 21:35:58 | mriedem | i think ^ sums it up | |
| 21:36:09 | openstackgerrit | Ed Leafe proposed openstack/nova-specs master: Re-propose nested resource providers spec https://review.openstack.org/505209 | |
| 21:37:14 | mriedem | btw, deleting 1000 instances via the API takes awhile | |
| 21:49:07 | openstackgerrit | Merged openstack/nova master: [placement] correct error on bad resource class in allocation https://review.openstack.org/502205 | |
| 21:52:47 | efried | mriedem Can I get some love on https://review.openstack.org/#/c/488137/ so I can get the rest of the series moving? | |
| 21:53:15 | mriedem | how about tough love | |
| 21:53:32 | efried | I expect nothing less from you | |
| 22:01:36 | mriedem | efried: your unit tests failed | |
| 22:02:15 | efried | sonofa... Okay, I think I know what happened. | |
| 22:03:51 | efried | Okay, since I'm going to need an edit anyway, I guess I might as well rebase... | |
| 22:04:10 | openstackgerrit | Eric Fried proposed openstack/nova master: nova.utils.get_ksa_adapter() https://review.openstack.org/488137 | |
| 22:06:12 | openstackgerrit | Eric Fried proposed openstack/nova master: nova.utils.get_ksa_adapter() https://review.openstack.org/488137 | |
| 22:06:21 | openstackgerrit | Eric Fried proposed openstack/nova master: Get auth from context for glance endpoint https://review.openstack.org/490057 | |
| 22:06:32 | efried | mriedem That ought to fix it ^ | |
| 22:08:16 | mriedem | well, meanwhile +2 on https://review.openstack.org/#/c/506370/ again | |
| 22:08:55 | efried | whee! | |
| 22:09:01 | mriedem | good god it takes a long time to delete 1000 vms | |
| 22:10:53 | dansmith | mriedem: is it actually deleting things or hung? | |
| 22:11:05 | mriedem | down to 512 left | |
| 22:15:01 | openstackgerrit | Eric Fried proposed openstack/nova master: Use ksa adapter for placement conf & requests https://review.openstack.org/492247 | |
| 22:15:26 | openstackgerrit | Eric Fried proposed openstack/nova master: Don't fix protocol-less glance api_servers anymore https://review.openstack.org/505317 | |
| 22:17:39 | jaypipes | mriedem: but don't you know? each launch of a VM does 400 calls to the database... | |
| 22:23:56 | jaypipes | dansmith: easy one from cdent: https://review.openstack.org/#/c/495913/ | |
| 22:32:28 | dansmith | jaypipes: I'm pretty meh on stuff like that in o.vo | |
| 22:32:47 | dansmith | doesn't save anything having that in the library and makes it harder to trace why you have these two fields in your object | |
| 22:32:49 | dansmith | but whatever | |
| 22:32:58 | jaypipes | dansmith: yeah? seems like it's generally useful, no? | |
| 22:33:36 | dansmith | no, it "saves" two lines from a single object in a project, or four for multiples, but makes chasing down the definitions a cross-repo venture | |
| 22:33:43 | jaypipes | dansmith: I had asked cdent to submit that to ovo, so blame me :) | |
| 22:33:49 | dansmith | I blame you. | |
| 22:33:53 | jaypipes | :) | |
| 22:40:49 | mriedem | jaypipes: this is delete | |
| 22:41:08 | mriedem | your callback joke doesn't fly with me | |
| 22:42:12 | jaypipes | mriedem: :) | |
| 22:43:13 | mriedem | ooo cpu pinning failure in the ML | |
| 22:44:57 | mriedem | how many db calls are made when we create a server? | |
| 22:45:02 | mriedem | 375? | |
| 22:46:01 | dansmith | 375, 376, however many it takes | |
| 22:49:09 | mriedem | 7.5 deletes per minute, just counted | |
| 22:49:11 | mriedem | blazing | |
| 22:49:34 | dansmith | you could restack quicker | |
| 22:49:56 | mriedem | probably | |
| 22:50:04 | mriedem | do unstack/stack work? or restack? | |
| 22:53:39 | openstackgerrit | Merged openstack/nova master: Add get_node_uuid() helper to ResourceTracker https://review.openstack.org/506730 | |
| 22:53:40 | dansmith | reboot, stack works :) | |
| 22:54:00 | dansmith | I don't know what still works for teardown after the systemd thing | |
| 22:54:28 | cfriesen | mriedem: do you know where the cost is? cuz that seems higher than it should be. | |
| 22:58:28 | jaypipes | edleafe: hopefully I explained the whole server-side PUT /allocations/{consumer_uuid} API microversion mismatch scenario properly in that last comment on the alternate hosts spec. | |