| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-02-25 | |||
| 15:30:08 | kashyap | It doesn't follow the general "scheme" most of the human brains are used to: describe the problem, then tell the solution. | |
| 15:30:10 | sean-k-mooney | kashyap: there are 2 thing i need to fix | |
| 15:30:26 | kashyap | Describe the problem. | |
| 15:30:26 | kashyap | [One line summary -- in imperative mood] | |
| 15:30:26 | kashyap | Lastly, this is my Git commit message 'scheme': | |
| 15:30:26 | sean-k-mooney | kashyap: so leave a comment and ill adress them shortly | |
| 15:30:28 | kashyap | Describe your solution. And more importantly, tell *why*. | |
| 15:31:18 | kashyap | (You do that, though. But all of them in bullets :-)) Anyway, don't want to belabor on this. | |
| 15:32:08 | stephenfin | bauzas: Fancy doing me the honour? https://review.openstack.org/#/c/445436/ | |
| 15:32:59 | bauzas | stephenfin: with pleasure | |
| 15:33:43 | bauzas | stephenfin: oh wait, sec. the option is *already* deprecated ? | |
| 15:33:52 | bauzas | stephenfin: if so, it should have a reno note | |
| 15:33:56 | bauzas | if not, we need it | |
| 15:34:03 | stephenfin | that's a good point | |
| 15:34:09 | stephenfin | lemme fix that | |
| 15:35:36 | bauzas | stephenfin: will update the commit msg to stop the zuul check then | |
| 15:35:52 | stephenfin | If you remove the -W, it'll take it out of the gate | |
| 15:35:56 | openstackgerrit | Sylvain Bauza proposed openstack/nova master: conf: Improve documentation for defer_iptables_apply https://review.openstack.org/445436 | |
| 15:37:35 | bauzas | stephenfin: nope, unless zuul changed | |
| 15:37:58 | bauzas | for pushing the change out of the gate pipeline, we need to have a new revision | |
| 15:38:05 | bauzas | anyway, it's done | |
| 15:46:03 | openstackgerrit | Lajos Katona proposed openstack/python-novaclient master: Add support for microversion v2.70 https://review.openstack.org/637234 | |
| 15:54:43 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecated 'defer_iptables_apply' https://review.openstack.org/445436 | |
| 15:54:46 | stephenfin | bauzas, jaypipes: That should be better ^ | |
| 15:55:27 | stephenfin | sean-k-mooney tells me what I'm doing is likely correct and I'm going to b̶l̶a̶m̶e̶ ̶h̶i̶m̶ ̶i̶f̶ ̶I̶'̶m̶ ̶w̶r̶o̶n̶g̶ trust him :) | |
| 15:56:18 | jaypipes | heh | |
| 15:59:23 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add unit tests for missing VirtualInterface in 2.70 os-interface https://review.openstack.org/639141 | |
| 16:00:02 | mriedem | can i get another core on the 'expose virtual device tags' api change? https://review.openstack.org/#/c/631948/ alex +2ed it and it's pretty straight-forward. last day in the runway queue for this one. | |
| 16:00:14 | stephenfin | sean-k-mooney: My docs are "nice to have". I feel cheated, Good Sir ;) | |
| 16:00:20 | stephenfin | mriedem: I can take a gawk | |
| 16:00:25 | mriedem | thanks | |
| 16:01:37 | yonglihe | this one in same situation: https://review.openstack.org/#/c/621474/22 | |
| 16:01:38 | sean-k-mooney | stephenfin: docs chagnes are non fuctional and can can be merged after the client lib freeze. since we generally just use the latest version of the docs people should still see them but they also can be backported to stable branches more easily so nice to have | |
| 16:02:08 | sean-k-mooney | stephenfin: hehe but i also think they will be straigt forward to merge so im going to go review them shortly anyway | |
| 16:02:16 | mriedem | yonglihe: i'll try to get to that one | |
| 16:03:23 | yonglihe | thanks. | |
| 16:04:26 | yonglihe | i suppose all stuff piled up at end of dev cycle, sorry for that. | |
| 16:12:37 | mriedem | np, it always happens | |
| 16:20:50 | bauzas | stephenfin: +Waboom | |
| 16:25:33 | openstackgerrit | Merged openstack/nova master: Refactor "networks" processing in ServersController.create https://review.openstack.org/633594 | |
| 16:27:31 | stephenfin | bauzas: Tank u | |
| 16:28:23 | stephenfin | (O_o_o_o_o_O) | |
| 16:28:23 | stephenfin | .-='=='==-, " | |
| 16:28:23 | stephenfin | .--._____, | |
| 16:33:29 | yonglihe | -:) | |
| 16:40:09 | openstackgerrit | Stephen Finucane proposed openstack/nova-specs master: Add 'flavor-extra-spec-image-property-validation' spec https://review.openstack.org/638734 | |
| 16:42:36 | openstackgerrit | Matt Riedemann proposed openstack/nova master: [Doc] Best practices for effectively tolerating down cells https://review.openstack.org/638173 | |
| 16:43:11 | tssurya | thanks mriedem ^ | |
| 16:43:15 | mriedem | np | |
| 16:57:49 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Fup for the bandwidth series https://review.openstack.org/639159 | |
| 16:58:10 | gibi | mriedem: your comments for the bandwidth series are fixed in ^^ | |
| 16:58:54 | gibi | jaypipes, efried: thanks for the comments in the bandwidth series, I will update https://review.openstack.org/639159 with your comments as well, possibly tomorrow | |
| 17:00:44 | artom | Object equality in tests in a PITA | |
| 17:00:48 | artom | *is | |
| 17:01:11 | artom | Expect call: Flavor(<some stuff>), Actual call: Flavor(<exact same stuff>) | |
| 17:03:49 | artom | *facepalm* | |
| 17:03:54 | artom | No, that's not i :( | |
| 17:03:58 | artom | *it | |
| 17:04:46 | sean-k-mooney | stephenfin: left some comments on you os-vif docs changes most are minor | |
| 17:07:24 | sean-k-mooney | artom: object equality check pending change so you need to reset changes on both object before comparing them | |
| 17:08:06 | sean-k-mooney | otherwise you get X != X issues | |
| 17:10:23 | jaypipes | artom: what sean-k-mooney said is almost always the problem with that. | |
| 17:10:54 | sean-k-mooney | jaypipes: artom we proably should just create a function in the base tescae for comparing objects | |
| 17:11:13 | jaypipes | sean-k-mooney: there was one somewhere I think... maybe dansmith can remember :) | |
| 17:11:22 | sean-k-mooney | e.g. self.assertObjEquals | |
| 17:12:30 | sean-k-mooney | jaypipes: would it break the world if we changed __eq__ in base ovo to ignore the changed state of fields? | |
| 17:12:46 | sean-k-mooney | i assume yes since we have not done so before | |
| 17:12:50 | artom | sean-k-mooney, jaypipes, yeah, so it this case it was dumber than that - my main problem was I had an - and an _ in my fake values | |
| 17:13:02 | sean-k-mooney | oh :) | |
| 17:13:11 | artom | But once that hurdle was over, there was still an object I had to replace with a fake string. | |
| 17:13:45 | artom | But yeah, a more intelligent way of comparing objects would be super | |
| 17:13:46 | jaypipes | hah :) | |
| 17:14:18 | artom | It wouldn't even be that hard - implement __eq__ in the base fields, and then recursively compare fields | |
| 17:15:13 | sean-k-mooney | artom: yes but we dont know if anyting depens on the fact that object comparisons current check the changed fields state of the objects | |
| 17:15:34 | sean-k-mooney | os its not that its hard to do but would it break anything | |
| 17:17:03 | sean-k-mooney | artom: https://github.com/openstack/nova/blob/eb5bdd33052166e4375f924456438f11be03310a/nova/test.py#L704 we have this by the way | |
| 17:17:37 | artom | sean-k-mooney, that's not useful when you're asseting call param tho | |
| 17:17:50 | artom | Anyways, it's a pain, but all things considered a minor one | |
| 17:33:22 | mriedem | artom: i've got some object equality test utils in my cross-cell resize series, sec | |
| 17:36:51 | mriedem | artom: https://review.openstack.org/#/c/627892/15/nova/tests/unit/conductor/tasks/test_cross_cell_migrate.py@193 | |
| 17:44:03 | mriedem | dansmith: random question, when detaching the root volume of a server and attaching a new root volume, would you expect the device name on that bdm to change? or remain vda or whatever? | |
| 17:44:12 | mriedem | i would expect it to *not* change | |
| 17:44:18 | mriedem | since the boot_index is still 0 | |
| 17:44:32 | mriedem | and the disk_bus and device_type can't change | |
| 17:49:51 | sean-k-mooney | mriedem: i think the only time we would expect that it could change would be a rebuild with a different image or perhaps a volume retype operation. so if your question was related to the cross cell resize i think that is a safe assumtion to make | |
| 17:50:20 | mriedem | it's not | |
| 17:50:32 | mriedem | it's for Kevin_Zheng's root bdm attach/detach series | |
| 17:50:45 | sean-k-mooney | oh ok | |
| 17:51:21 | sean-k-mooney | am well if you detach attach anoth volume and then attach the root again i guess it could change | |
| 17:51:42 | sean-k-mooney | i dont know that code that well however | |
| 17:52:15 | mriedem | https://review.openstack.org/#/c/614750/34/nova/compute/manager.py | |
| 17:52:57 | mriedem | if i boot from volume and get vda, then attach a data volume which is vdb, then detach the root volume and attach another root volume, would i expect to have that as vda or vdc? | |
| 17:53:10 | mriedem | i would expect vda because the root volume being higher than the data volume seems wrong | |
| 17:54:02 | dansmith | mriedem: yeah, expect the name to remain stable | |
| 17:54:25 | sean-k-mooney | mriedem: that might depend on the os and the udev rules. but i would expect it to stay the same. i dont know if it actully would | |
| 17:56:42 | mriedem | well, we also don't guarantee the device name the user requests is honored by the hypervisor anyway | |
| 18:01:04 | artom | mriedem, interesting, but I feel like that's specific to what you're doing with them (which is fine!) | |
| 18:02:19 | sean-k-mooney | mriedem: for the detach attach root volume spec | |
| 18:02:34 | sean-k-mooney | mriedem: is it the same volume or can it be any volume that is reattached | |
| 18:02:37 | artom | And as far as I can tell _assertEqualObjects doens't handle nested objects | |