| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-04-17 | |||
| 14:58:40 | efried | so I poked at a couple of ways of doing that | |
| 14:59:27 | efried | and it's about time we did *something* to break our pattern of "set up a whole KSA adapter just to grab the endpoint out of it and send that into *client which uses it to create a whole nother Adapter to do its work" | |
| 14:59:38 | efried | which we do (in different ways!) for glance, ironic, neutron... | |
| 15:00:45 | mriedem | the ksa stuff in nova still feels new to me, but i'm slow and old | |
| 15:00:53 | mriedem | and was a big improvement over the per-service config and all that | |
| 15:01:02 | mriedem | cinder excluded since we haven't fixed that one yet | |
| 15:01:46 | gmann | mriedem: updated this with review comments. - https://review.openstack.org/#/c/603969/ | |
| 15:04:42 | efried | mriedem: It was an improvement in the sense that operators can now use same-looking conf values for talking to the services. But internally we didn't improve much: we're still doing the dance I mentioned above --^ | |
| 15:05:30 | efried | And, wasn't the ksa cutover in Queens? That's a pretty long time ago. Maybe just seems that way for me as it's pretty much when I first got involved in nova proper. | |
| 15:07:56 | mriedem | queens is still very new to me | |
| 15:08:03 | mriedem | hell pike is as well | |
| 15:08:17 | mriedem | if i'm still fixing bugs in ocata/pike etc it's new to me | |
| 15:16:52 | mriedem | vdrok: just some small updates to https://review.openstack.org/#/c/571410/ | |
| 15:24:04 | mriedem | dansmith: can you hit this backport to keep things moving https://review.openstack.org/#/c/651639/2 | |
| 15:24:27 | mriedem | and https://review.openstack.org/#/q/If966eb1161c842ff49aa530e4482dbca87b61a3e+status:open since that's a gate issue hitting stable backports | |
| 15:25:06 | mriedem | melwitt: can you hit these backports? https://review.openstack.org/#/q/topic:bug/1819963+(status:open+OR+status:merged)+branch:stable/rocky | |
| 15:25:09 | cfriesen | mriedem: thanks for triaging that resize regression issue and writing the regression test. looks like Jack and I dropped the ball there. | |
| 15:25:16 | mriedem | cfriesen: i approved the change | |
| 15:25:21 | mriedem | i probably approved all of these busted changes | |
| 15:25:46 | mriedem | cfriesen: please see my comment-to-self on the fix and drop your opinion on decoupling the grossitude | |
| 15:25:59 | dansmith | mriedem: ack | |
| 15:26:03 | cfriesen | mriedem: for sure | |
| 15:26:29 | mriedem | cfriesen: i'll also say that as much as i don't like these regressions, i'm happy there are people in your org somewhere doing real testing | |
| 15:27:17 | mriedem | lyarwood: can you hit this backport? https://review.openstack.org/#/c/647623/ | |
| 15:27:29 | openstackgerrit | Vladyslav Drok proposed openstack/nova master: libvirt: Do not reraise DiskNotFound exceptions during resize https://review.openstack.org/571410 | |
| 15:27:51 | mriedem | gmann: ack | |
| 15:28:26 | mriedem | vdrok: +2 thanks | |
| 15:28:45 | mriedem | would be good to get another core on ^ since that's been around a long time | |
| 15:28:47 | mriedem | and is very latent | |
| 15:28:56 | mriedem | efried: ^ | |
| 15:31:23 | lyarwood | mriedem: ack, I'll also try to get around to that bugfix again later, slightly busy downstream today. | |
| 15:31:45 | mriedem | lyarwood: which bug fix? | |
| 15:32:30 | lyarwood | mriedem: vdrok's above | |
| 15:32:37 | mriedem | oh ok | |
| 15:57:24 | mriedem | cfriesen: i'm going with that alternative - have it locally and it's a much simpler diff | |
| 16:03:02 | openstackgerrit | Merged openstack/os-vif stable/stein: Update UPPER_CONSTRAINTS_FILE for stable/stein https://review.openstack.org/644035 | |
| 16:05:15 | cfriesen | mriedem: so you're moving the last ~10 lines of _validate_flavor_image_nostatus() into another function? | |
| 16:05:55 | mriedem | yup | |
| 16:07:02 | aspiers | efried: run out of time for today, but I've almost finished updating the spec to talk about the new resource class rather than the existing HW_CPU_AMD_SEV trait. I'll submit another patchset tomorrow | |
| 16:07:35 | efried | aspiers: I'm not quite caught up, did you get other cores to ack that shift in direction? | |
| 16:07:59 | efried | mriedem: what am I looking at? https://review.openstack.org/#/c/571410/ ? | |
| 16:09:27 | cfriesen | mriedem: seems like it could work | |
| 16:11:36 | mriedem | efried: hold on that one | |
| 16:11:47 | efried | ack | |
| 16:18:24 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix volume-backed resize with a smaller disk flavor https://review.openstack.org/653273 | |
| 16:18:25 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Update volume-backed comment in _validate_flavor_image_nostatus https://review.openstack.org/653483 | |
| 16:18:25 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Remove FlavorNotFound dead code condition in API.resize https://review.openstack.org/653484 | |
| 16:18:34 | mriedem | cfriesen: https://review.openstack.org/#/c/653273/ | |
| 16:23:40 | openstackgerrit | Eric Fried proposed openstack/nova master: Add ids to sections of flavors guide to allow deep-linking https://review.openstack.org/653458 | |
| 16:30:35 | cfriesen | mriedem: couple comments in the review | |
| 16:33:29 | mriedem | i'll fix those image status docstrings even though i blame you :) | |
| 16:34:22 | cfriesen | my bad entirely. :) | |
| 16:34:57 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix volume-backed resize with a smaller disk flavor https://review.openstack.org/653273 | |
| 16:34:57 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Update volume-backed comment in _validate_flavor_image_nostatus https://review.openstack.org/653483 | |
| 16:34:58 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Remove FlavorNotFound dead code condition in API.resize https://review.openstack.org/653484 | |
| 16:35:39 | cfriesen | mriedem: it's still a bit grody, but the underlying special-casing is complicated so I don't see a nice way to make it cleaner | |
| 17:08:31 | openstackgerrit | Dan Smith proposed openstack/nova master: Make libvirt expose supported image types https://review.openstack.org/653454 | |
| 17:32:03 | artom | Would there be a point in adding multinode resize tests somewhere? | |
| 17:32:23 | artom | AFAICT currently we're only testing in in the allinone tempest-full job | |
| 17:32:30 | artom | Which isn't super realistic... | |
| 17:33:12 | artom | I want to try it regardless, but how I do it will depend. Either as a throwaway DNM hack, or I'll put in some minimal amount of effort | |
| 17:40:26 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: Revert "Wait for network-vif-plugged on resize revert" https://review.openstack.org/639396 | |
| 17:40:26 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: Revert resize: wait for external events in compute manager https://review.openstack.org/644881 | |
| 17:40:27 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: DNM: Run test_resize_server_revert in the live migration job https://review.openstack.org/653498 | |
| 17:40:47 | artom | h4x | |
| 18:01:00 | openstackgerrit | Eric Fried proposed openstack/nova master: DNM: DEMO: Reinstate bug 1819963 https://review.openstack.org/653503 | |
| 18:01:01 | openstack | bug 1819963 in OpenStack Compute (nova) rocky "Reverting a resize does not update the instance.availability_zone value to the source az" [Medium,In progress] https://launchpad.net/bugs/1819963 - Assigned to Matt Riedemann (mriedem) | |
| 18:01:01 | openstackgerrit | Eric Fried proposed openstack/nova master: DEMO: Functional recreate test for bug 1819963 https://review.openstack.org/653504 | |
| 18:01:01 | openstackgerrit | Eric Fried proposed openstack/nova master: DEMO: Update instance.availability_zone on revertResize https://review.openstack.org/653505 | |
| 18:03:13 | efried | dammit, spelling those tags wrong didn't prevent them from posting to lp | |
| 18:03:34 | efried | oh well, guess I might as well spell them correctly then. | |
| 18:04:48 | openstackgerrit | Eric Fried proposed openstack/nova master: DEMO: Functional recreate test for bug 1819963 https://review.openstack.org/653504 | |
| 18:04:49 | openstack | bug 1819963 in OpenStack Compute (nova) rocky "Reverting a resize does not update the instance.availability_zone value to the source az" [Medium,In progress] https://launchpad.net/bugs/1819963 - Assigned to Matt Riedemann (mriedem) | |
| 18:04:49 | openstackgerrit | Eric Fried proposed openstack/nova master: DEMO: Update instance.availability_zone on revertResize https://review.openstack.org/653505 | |
| 18:06:42 | openstackgerrit | sean mooney proposed openstack/nova master: [WIP] only update port binding if there is not an inactive binding https://review.openstack.org/653506 | |
| 18:12:01 | sean-k-mooney | mriedem: i spent the last hour or more fighting with some of the unit test so im takeing a break form it but ^ i think will fix NewBruce issue | |
| 18:12:14 | NewBruce | will test it out shortly | |
| 18:12:29 | NewBruce | Thanks sean-k-mooney | |
| 18:13:31 | sean-k-mooney | ok im going to go have dinner. ill check back later to see what the gate thinks. | |
| 18:21:06 | openstackgerrit | Matt Riedemann proposed openstack/nova master: libvirt: set device address tag only if setting disk unit https://review.openstack.org/611974 | |
| 18:21:58 | mriedem | melwitt: begin backport procedures | |
| 18:34:52 | efried | mriedem: I'm setting up a demo for the summit onboarding session. I want to show adding a regression test for a bug and then fixing the bug. I want to use https://bugs.launchpad.net/nova/+bug/1819963 | |
| 18:34:54 | openstack | Launchpad bug 1819963 in OpenStack Compute (nova) rocky "Reverting a resize does not update the instance.availability_zone value to the source az" [Medium,In progress] - Assigned to Matt Riedemann (mriedem) | |
| 18:34:54 | efried | I've got most of the setup done at this point, just have a question about the bug itself: Given that the API was masking that the instance AZ in the db doesn't match where the instance actually is, for whom is it actually a problem, and why? | |
| 18:38:26 | mriedem | the most obvious one is filtering instances by az | |
| 18:38:34 | mriedem | GET /servers?availability_zone=foo | |
| 18:38:48 | mriedem | when it's really in bar | |
| 18:52:42 | openstackgerrit | melanie witt proposed openstack/nova stable/stein: libvirt: set device address tag only if setting disk unit https://review.openstack.org/653510 | |
| 18:56:27 | mriedem | gmann: done https://review.openstack.org/#/c/603969/ | |
| 18:57:21 | openstackgerrit | melanie witt proposed openstack/nova stable/stein: libvirt: set device address tag only if setting disk unit https://review.openstack.org/653510 | |
| 18:59:45 | openstackgerrit | melanie witt proposed openstack/nova stable/rocky: libvirt: set device address tag only if setting disk unit https://review.openstack.org/653511 | |
| 19:00:35 | openstackgerrit | melanie witt proposed openstack/nova stable/queens: libvirt: set device address tag only if setting disk unit https://review.openstack.org/653512 | |
| 19:05:37 | openstackgerrit | melanie witt proposed openstack/nova stable/pike: libvirt: set device address tag only if setting disk unit https://review.openstack.org/653514 | |
| 19:08:09 | openstackgerrit | Eric Fried proposed openstack/nova master: DEMO: Functional recreate test for bug 1819963 https://review.openstack.org/653504 | |
| 19:08:10 | openstack | bug 1819963 in OpenStack Compute (nova) rocky "Reverting a resize does not update the instance.availability_zone value to the source az" [Medium,In progress] https://launchpad.net/bugs/1819963 - Assigned to Matt Riedemann (mriedem) | |
| 19:08:10 | openstackgerrit | Eric Fried proposed openstack/nova master: DEMO: Update instance.availability_zone on revertResize https://review.openstack.org/653505 | |
| 19:08:35 | efried | mriedem: Thanks. I've done a couple of things to improve that test case ^ - lmk if you think it's worth pulling those out into a real patch. | |
| 19:09:29 | efried | actually, the first thing - checking the db explicitly - would have only applied to the interim recreate patch. So it's just the part where we do that GET /servers?availability_zone=* query that would be added now. | |
| 19:20:49 | openstackgerrit | Ghanshyam Mann proposed openstack/nova-specs master: Spec for API inconsistency cleanup https://review.openstack.org/603969 | |
| 19:21:29 | gmann | mriedem: thanks. ^^ updated. | |