| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-09-22 | |||
| 12:08:46 | liuyulong | sdague, Kevin_Zheng, I'm glad to. so here is the patch for that spec, https://review.openstack.org/#/c/379128/. I don't think there has any conflicts. | |
| 12:09:55 | liuyulong | correct me, if i missing something. | |
| 12:10:58 | liuyulong | Idividual contributor has a small chance to attend ptgs. haha | |
| 12:12:58 | liuyulong | Kevin_Zheng, feel free to directly update the spec: https://review.openstack.org/#/c/375221/, since the PTG etherpad linked this. | |
| 12:15:35 | Kevin_Zheng | I already had one up, but sure I can do that | |
| 12:17:47 | liuyulong | https://review.openstack.org/#/c/506552/, this one? nice work. | |
| 12:19:14 | Kevin_Zheng | Yeah I can merge these two | |
| 12:24:20 | liuyulong | I don't think this two spec can be merged. Your spec is going to add a new API, mine updating rebuild API. Basically they don't affect each other. | |
| 12:24:48 | Kevin_Zheng | Ok | |
| 12:38:01 | openstackgerrit | Dinesh Bhor proposed openstack/nova master: Fix 500 if list servers called with empty regex pattern https://review.openstack.org/506585 | |
| 12:46:01 | liuyulong | mriedem | |
| 12:47:10 | openstackgerrit | Elod Illes proposed openstack/nova master: Add instance.interface_detach notification https://review.openstack.org/506284 | |
| 12:57:33 | openstackgerrit | OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/506640 | |
| 12:59:46 | openstackgerrit | OpenStack Proposal Bot proposed openstack/os-traits master: Updated from global requirements https://review.openstack.org/503646 | |
| 13:02:40 | openstackgerrit | Elod Illes proposed openstack/nova master: Add error notification for instance.interface_attach https://review.openstack.org/506643 | |
| 13:13:09 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Remove allocations when unshelve fails on host https://review.openstack.org/506458 | |
| 13:23:14 | openstackgerrit | Chris Dent proposed openstack/nova master: Optional separate database for placement API https://review.openstack.org/362766 | |
| 13:42:01 | openstackgerrit | Rodolfo Alonso Hernandez proposed openstack/os-vif master: Add abstract OVSDB API https://review.openstack.org/476612 | |
| 13:42:25 | openstackgerrit | Rodolfo Alonso Hernandez proposed openstack/os-vif master: Add native implementation OVSDB API https://review.openstack.org/482226 | |
| 13:43:19 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add _wait_for_action_fail_completion to InstanceHelperMixin https://review.openstack.org/506660 | |
| 13:43:28 | mriedem | gibi: ^ is where i was going with the refactor | |
| 13:54:25 | mriedem | bauzas: you've got a race in the test here https://review.openstack.org/#/c/506093/ | |
| 13:54:30 | mriedem | need to wait for the migration to be done | |
| 13:54:55 | bauzas | mriedem: yeah I know | |
| 13:55:01 | bauzas | mriedem: I saw that this morning | |
| 13:55:07 | bauzas | mriedem: I have a f*** agenda today | |
| 13:55:16 | mriedem | agenda?! | |
| 13:55:16 | bauzas | but I'll try to look at it soon | |
| 13:55:22 | bauzas | schedule | |
| 13:55:35 | mriedem | oh right, red hat friday meetings and bug smash day | |
| 13:55:36 | bauzas | meetings day | |
| 13:55:50 | mriedem | i love how you guys load all of the worst stuff to the end of the week | |
| 13:55:51 | bauzas | even more than that - spoiling some very specific French stuff | |
| 13:56:16 | mriedem | very specific french stuff? like a smoking mime eating a croissant? | |
| 13:58:20 | stephenfin | mriedem, bauzas: Can one of you merge this so it stops flooding my inbox with Gerrit mails? https://review.openstack.org/#/c/503646/ | |
| 13:59:14 | bauzas | stephenfin: Wipped | |
| 13:59:20 | openstackgerrit | Evgeny Antyshev proposed openstack/nova master: Vzstorage: synchronize volume connect https://review.openstack.org/505708 | |
| 14:00:58 | mriedem | alex_xu: you might like to see this one https://review.openstack.org/#/c/506585/ | |
| 14:01:15 | mriedem | stephenfin: your gerrit emails go to your inbox? | |
| 14:01:21 | mriedem | that's your mistake | |
| 14:01:47 | mriedem | gotta setup a filter rule, especially for the CI emails | |
| 14:02:14 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix 500 if list servers called with empty regex pattern https://review.openstack.org/506585 | |
| 14:02:20 | openstackgerrit | Chris Dent proposed openstack/nova master: DNM: Don't monkey patch eventlet in functional tests https://review.openstack.org/506668 | |
| 14:05:07 | liuyulong | hi mriedem, https://review.openstack.org/#/c/375221/, I've updated the spec, could you please review it again? | |
| 14:05:39 | openstackgerrit | Merged openstack/os-traits master: Updated from global requirements https://review.openstack.org/503646 | |
| 14:07:08 | mriedem | liuyulong: Kevin_Zheng was going to update that spec since we thought it was abandoned, | |
| 14:07:13 | liuyulong | mriedem, there are some things that need to be clarified is that this spec can be under parallel development with Kenvin_zheng's os-keypair API. | |
| 14:07:19 | mriedem | liuyulong: we discussed this at the PTG last week, and want a different direction | |
| 14:07:27 | mriedem | right | |
| 14:07:49 | mriedem | we said we'd provide a way to update the keypair for a specific server, and then the user can reboot or rebuild the instance | |
| 14:08:00 | liuyulong | it's not the same. One adding API, one updating rebuild API. | |
| 14:08:01 | mriedem | because they might not need to rebuild the guest depending on how cloud-init is configured | |
| 14:08:10 | mriedem | the goal is the same | |
| 14:08:26 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'keymap' options https://review.openstack.org/483994 | |
| 14:10:24 | liuyulong | One API can done for updating keypair. But the new API may need two steps. It's waste. | |
| 14:10:55 | liuyulong | but the new API can be used for a running injection. | |
| 14:11:27 | liuyulong | So I'd like to say, both can be added to nova. | |
| 14:12:06 | mriedem | liuyulong: by that logic, why don't we also add keypair to the reboot API so you can specify a new keypair on reboot? | |
| 14:12:13 | mriedem | it starts to become a slippery slope | |
| 14:12:33 | liuyulong | New API can result an inconsistent situtation when the vm is not reboot. | |
| 14:12:48 | mriedem | does Kevin_Zheng have a spec up for review yet? | |
| 14:12:54 | liuyulong | yes | |
| 14:13:00 | mriedem | inconsistent how? | |
| 14:13:06 | liuyulong | https://review.openstack.org/#/c/506552 | |
| 14:13:17 | mriedem | yes the keypair is different, because the user changed it | |
| 14:13:28 | mriedem | and it's different from the guest until the guest is rebooted or rebuilt | |
| 14:14:11 | mriedem | mikal: cburgess: keypair update API spec https://review.openstack.org/#/c/506552 | |
| 14:14:13 | mriedem | since you wanted to review | |
| 14:14:49 | liuyulong | That makes could user confused. But rebuild API naturally with a restart. | |
| 14:17:11 | mriedem | well, we'd be documenting this API | |
| 14:17:23 | mriedem | if you update the keypair on the instance, you have to reboot or rebuild after that | |
| 14:17:26 | liuyulong | IMHO, both specs are OK for now. One step for rebuild API make more reasonable. | |
| 14:17:26 | mriedem | for it to take effect | |
| 14:17:46 | mriedem | sorry but we don't really need 3 ways to do this same thing | |
| 14:18:05 | leakypipes | mriedem: +3 | |
| 14:18:06 | mriedem | 3 = keypair update + user reboot/rebuild, keypair during rebuild, keypair during reboot | |
| 14:18:29 | mriedem | because then people will also say, "why can't i update the keypair when i unshelve my offloaded instance too?!" | |
| 14:18:38 | mriedem | or migrate it | |
| 14:20:56 | leakypipes | superdan: morning. | |
| 14:21:14 | superdan | leakypipes: I can confirm that it is indeed morning. | |
| 14:21:20 | liuyulong | reboot API can also update the keypair? | |
| 14:21:33 | leakypipes | superdan: has Jax attempted homocide yet? | |
| 14:22:01 | superdan | leakypipes: I think you mean homicide, and not yet this morning | |
| 14:22:33 | superdan | but. it's early. | |
| 14:22:33 | mriedem | ha | |
| 14:22:40 | leakypipes | superdan: ha, indeed. | |
| 14:22:46 | mriedem | liuyulong: no, but that's my point, | |
| 14:23:11 | mriedem | if we have one api to update the keypair just on the instance, and one api to rebuild with a new keypair, then people would also want the reboot api to take a keypair, and possibly other APIs, like migrate and unshelve | |
| 14:23:32 | mriedem | and we don't want an explosion of API changes to provide a new keypair for every operation | |
| 14:25:16 | finucannot | leakypipes: Finally settled on a Friday nick that doubles as a learning opportunity (how to pronounce my last name) | |
| 14:26:01 | openstackgerrit | Stephen Finucane proposed openstack/nova master: placement: add ProviderTree.is_inventory_empty https://review.openstack.org/480957 | |
| 14:26:07 | leakypipes | finucannot: well played, sir. | |
| 14:26:52 | Kevin_Zheng | https://blueprints.launchpad.net/nova/+spec/use-constrain-for-microversion-values does anyone find this useful? | |
| 14:27:08 | liuyulong | `rebuild` means recreate a instance, so for custom data, they should have ways to change it. Like name, adminPass, image etc. | |
| 14:27:33 | liuyulong | Its not a explosion. | |
| 14:27:44 | liuyulong | Just a expanding of rebuilding API | |
| 14:27:57 | mriedem | liuyulong: so do you think we should also allow passing in new volumes and ports when rebuilding an instance? | |
| 14:28:53 | openstackgerrit | Stephen Finucane proposed openstack/nova master: libvirt: Make 'get_domain' private https://review.openstack.org/417378 | |
| 14:29:29 | artom | finucannot, I guess apostrophes aren't allowed in IRC nicks, eh? | |
| 14:29:32 | finucannot | leakypipes, sahid: That patch has been open for a long time as is a pretty trivial TODO resolution. Could ye take a look at some point | |