| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-12-04 | |||
| 20:44:37 | mriedem | passive aggressive works too | |
| 20:44:57 | efried | Like this: dansmith I think it's great how much time you have for cells stuff. | |
| 20:45:05 | mriedem | no | |
| 20:45:09 | efried | damn | |
| 20:45:13 | efried | I really suck at this. | |
| 20:45:16 | dansmith | yeah, that just makes me mad :) | |
| 20:45:21 | mriedem | heh, yeah :) | |
| 20:45:24 | dansmith | lol | |
| 20:45:35 | mriedem | you need to be a true midwesterner to master passive aggressive | |
| 20:45:58 | efried | My in-laws are from Wisconsin. That's as close as I can come. | |
| 20:46:41 | mriedem | psh, wisconsin | |
| 20:46:43 | efried | If I had my mother in law asking, it would be like, "dansmith, it's okay if you don't want to review my patches; I'll be dead soon and then it won't matter". | |
| 20:46:45 | mriedem | cheese heads | |
| 20:46:48 | mriedem | and serial killers | |
| 20:46:56 | mriedem | efried: that's more like it | |
| 20:47:02 | dansmith | yeah, getting there | |
| 20:54:22 | cfriesen | is it a bug that "nova quota-show" has "injected_file_path_bytes" but "nova limits" doesn't have an equivalent entry? | |
| 20:54:58 | mriedem | cfriesen: meh | |
| 20:55:12 | mriedem | isn't that maxPersonalitySize? | |
| 20:56:03 | cfriesen | So I see "Personality" and "Personality Size" in limits (in Newton) but for quotas I see injected_files/injected_file_content_bytes/injected_file_path_bytes | |
| 20:57:09 | clutchbeyers | looking to achieve some type of multi tenancy exclusive isolation on host aggregates, I noticed there were some blueprints that were abandoned. Is anyone aware of any work being done or other mechanism to exclude tenants to an aggregate or global filter that wont provision a tenant to an aggregate that doesnt have such filter meta set? | |
| 20:58:31 | melwitt | cfriesen: did it use to show it in 'nova limits' but doesn't anymore? if so, that's a bug | |
| 20:59:06 | melwitt | oh, newton. then not a bug. quota rewriting didn't happen until pike | |
| 21:00:07 | melwitt | or rather, not a regression. one could argue whether it's a bug in general | |
| 21:00:21 | cfriesen | melwitt: okay, will be upgrading soon. | |
| 21:00:38 | cfriesen | why do we have both APIs anyway? | |
| 21:00:49 | openstackgerrit | Eric Fried proposed openstack/nova stable/ocata: Proper error handling by _ensure_resource_provider https://review.openstack.org/525309 | |
| 21:01:02 | cfriesen | qutoa and limits I mean | |
| 21:01:03 | efried | mriedem ^ There's that cherry-pick. That was a pretty fun merge. | |
| 21:01:04 | melwitt | that, I don't know | |
| 21:01:25 | cfriesen | okay, thanks | |
| 21:02:19 | mriedem | cfriesen: i'd bet it was likely due to v2.0 api extension stuff | |
| 21:04:35 | mriedem | clutchbeyers: https://review.openstack.org/#/c/381912/ | |
| 21:05:04 | mriedem | clutchbeyers: no one is working on that | |
| 21:05:21 | mriedem | melwitt looked at something related recently, saying maybe it was a regression that we needed this anyway | |
| 21:05:59 | melwitt | I think that was the lack of counting keypairs for the used limits API. it always returns 0 | |
| 21:09:14 | melwitt | this is the redhat bug where it's demonstrated to be a regression in mitaka https://bugzilla.redhat.com/show_bug.cgi?id=1435597 | |
| 21:09:14 | openstack | bugzilla.redhat.com bug 1435597 in openstack-nova "AggegateImagePropertiesIsolatefiler not working" [High,New] - Assigned to sbauza | |
| 21:11:03 | melwitt | given that, I was a bit lost about how it got turned into a blueprint-required feature | |
| 21:11:43 | melwitt | I want to dig into it more to find out what actually happened back then | |
| 21:14:20 | melwitt | i.e. find what changed that made it stop working and whether it was intentional | |
| 21:15:57 | melwitt | I had done some looking while at the summit but didn't find anything yet | |
| 21:25:24 | mriedem | huh, how does this ever work? | |
| 21:25:25 | mriedem | https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L6595 | |
| 21:25:46 | mriedem | ah oh nvm | |
| 21:39:57 | mikal | mriedem: I commented on that bug, but mostly to say "you people confuse me" | |
| 21:40:06 | mikal | In other news, Amazon shopping just launched in my country | |
| 21:40:54 | mriedem | mikal: the perf one? | |
| 21:46:13 | mikal | Yep, https://bugs.launchpad.net/nova/+bug/1735663 | |
| 21:46:15 | openstack | Launchpad bug 1735663 in OpenStack Compute (nova) "privsep.daemon takes 9 seconds during taking volume attaching operation" [Undecided,New] | |
| 21:47:44 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add regression test for rebuilding a volume-backed server https://review.openstack.org/521200 | |
| 21:47:44 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Get original image_id from volume for volume-backed instance rebuild https://review.openstack.org/521391 | |
| 21:47:45 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Fail fast if changing image on a volume-backed server rebuild https://review.openstack.org/520660 | |
| 21:48:04 | mriedem | dansmith: think i addressed your comments in ^ | |
| 21:48:33 | dansmith | slright | |
| 22:07:33 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: [placement] Add aggregate link note in API ref https://review.openstack.org/521502 | |
| 22:07:57 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: [placement] Add functional tests for traits API https://review.openstack.org/524094 | |
| 22:08:20 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: [placement] Add 'Location' parameters in API ref https://review.openstack.org/521541 | |
| 22:08:38 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: [cellv2] Improve getting BDMs in multiple cells https://review.openstack.org/521400 | |
| 22:08:57 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: Add a warning in 'nova-manage cell_v2 delete_cell' https://review.openstack.org/513771 | |
| 22:09:39 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: [placement] Add functional tests for resource class API https://review.openstack.org/524506 | |
| 22:10:37 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: [placement] Fix foreign key constraint error https://review.openstack.org/516880 | |
| 22:11:09 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: [placement] Fix getting placement request ID https://review.openstack.org/523606 | |
| 22:11:26 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: Fix missing marker functions https://review.openstack.org/514579 | |
| 22:12:56 | openstackgerrit | Hongbin Lu proposed openstack/nova master: [WIP] Handle RPC timeout on get_console_output https://review.openstack.org/525335 | |
| 22:21:35 | efried | jaypipes Been staring at this for a bit, can't figure out where 1.10 is coming from: http://logs.openstack.org/93/385693/83/check/legacy-grenade-dsvm-neutron/b513223/logs/screen-n-cpu.txt.gz#_Dec_04_15_47_11_988730 | |
| 22:31:31 | gryf | efried, looks like you run pike version of placement against latest source | |
| 22:31:50 | efried | gryf Wow, that would be... bad. How can you tell? | |
| 22:32:12 | gryf | 1.10 was the max microversion in pike | |
| 22:33:22 | efried | cdent If you're still around -- how does legacy-grenade-dsvm-neutron spin up the placement API? Is there some chance we could be using the pike version in latest master? | |
| 22:33:33 | efried | cdent And wouldn't lots of other stuff have broken by now if that were the case? | |
| 22:33:57 | cdent | efried: I assume you’re talking about the thing that’s claiming 1.10 is the max version? | |
| 22:34:05 | efried | cdent yeah | |
| 22:34:12 | cdent | was just reading that in my email | |
| 22:35:27 | cdent | it could be that a new compute is talking to an older placement, which I think is something we _may_ do intentionally in some tests | |
| 22:35:48 | efried | oy vay, that could be... complicated to fix. | |
| 22:36:03 | cdent | and for the most part we’ve tried to make the report client be smart about that | |
| 22:37:53 | cdent | efried: hmmm. yeah, that code doesn’t have a fall back, but I’m not sure we want that or not. jaypipes, mriedem or dansmith will be able to say more on whether we need to be able to work with new compute/old placement | |
| 22:38:13 | efried | cdent Okay, I gotcha. | |
| 22:38:16 | cdent | If we don’t need to support that, the other chance is simply that the new placement hasn’t manage to get restarted yet | |
| 22:38:19 | dansmith | old compute needs to work with new placement | |
| 22:38:32 | cdent | dansmith: yes, but vice versa? | |
| 22:39:28 | dansmith | not vice versa | |
| 22:39:47 | efried | cdent "hasn't managed to get restarted yet" - so actually a recheck may clear it? | |
| 22:40:41 | cdent | efried: probably, but it may be that the process mgt in grenade is insufficiently robust and could be robustified | |
| 22:40:56 | efried | BTW, this was just exposed because of https://review.openstack.org/#/c/524263/ -- I think otherwise we would have blown right by this error and not noticed. So it may be a long-standing latent bug. | |
| 22:41:23 | cdent | yup, I dig | |
| 22:41:55 | cdent | efried: if you haven’t got time to think about grenade right now (or ever), I’m happy to go digging instead, but wouldn’t be until tomorrow | |
| 22:42:01 | efried | Well, I guess I'll try a recheck, but it seems like something that ought to be fixed *somewhere* | |
| 22:42:10 | efried | cdent I've never heard of grenade. | |
| 22:42:21 | efried | And it's jaypipes' patch anyway :P | |
| 22:43:01 | cdent | code ownership is, like, _so_ 1990, but yeah, my curiosity is piqued | |
| 22:43:16 | cdent | grenade is the sort of framework used to manage upgrade testing | |
| 22:43:32 | cdent | a devstack in the previous version is started, verified a bit | |
| 22:44:18 | cdent | then new code is installed, but old configruations kept | |
| 22:50:07 | openstackgerrit | Ed Leafe proposed openstack/nova master: Return Selection objects from the scheduler driver https://review.openstack.org/495854 | |
| 22:50:07 | openstackgerrit | Ed Leafe proposed openstack/nova master: Move the to_dict() method to the Selection object https://review.openstack.org/523492 | |
| 22:50:07 | openstackgerrit | Ed Leafe proposed openstack/nova master: Modify select_destinations() to return objects and alts https://review.openstack.org/510159 | |
| 22:50:43 | edleafe | mriedem: ^^ addressed your comments | |
| 22:51:02 | edleafe | mriedem: will start on https://review.openstack.org/#/c/516707/ tomorrow | |