| *** tetsuro_ has joined #openstack-placement | 00:28 | |
| *** tetsuro_ has quit IRC | 00:36 | |
| *** edmondsw has joined #openstack-placement | 00:37 | |
| *** edmondsw has quit IRC | 00:42 | |
| openstackgerrit | melanie witt proposed openstack/nova stable/ocata: [stable only] Handle quota usage during create/delete races https://review.openstack.org/582413 | 00:58 |
|---|---|---|
| openstackgerrit | melanie witt proposed openstack/nova stable/ocata: Add functional regression test for bug 1783613 https://review.openstack.org/588416 | 00:58 |
| openstack | bug 1783613 in OpenStack Compute (nova) ocata "[ocata only] quota usage not decremented during boot/delete race" [Undecided,In progress] https://launchpad.net/bugs/1783613 - Assigned to melanie witt (melwitt) | 00:58 |
| openstackgerrit | Merged openstack/nova master: In Python3.7 async is a keyword [1] https://review.openstack.org/584365 | 00:58 |
| openstackgerrit | melanie witt proposed openstack/nova stable/ocata: [stable only] Add functional regression test for bug 1783613 https://review.openstack.org/588416 | 00:59 |
| openstackgerrit | melanie witt proposed openstack/nova stable/ocata: [stable only] Handle quota usage during create/delete races https://review.openstack.org/582413 | 00:59 |
| openstack | bug 1783613 in OpenStack Compute (nova) ocata "[ocata only] quota usage not decremented during boot/delete race" [Undecided,In progress] https://launchpad.net/bugs/1783613 - Assigned to melanie witt (melwitt) | 00:59 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Remove unused request API sample template https://review.openstack.org/588420 | 01:30 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Remove unused request API sample template https://review.openstack.org/588420 | 01:31 |
| *** mriedem_afk has quit IRC | 01:34 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Remove unused request API sample template https://review.openstack.org/588420 | 01:34 |
| *** tetsuro_ has joined #openstack-placement | 01:36 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Update installation guide to be more clear about cellsv2 https://review.openstack.org/584244 | 01:51 |
| openstackgerrit | zhufl proposed openstack/nova master: Fix none-ascii char in doc https://review.openstack.org/588422 | 01:56 |
| *** tetsuro_ has quit IRC | 02:04 | |
| *** edmondsw has joined #openstack-placement | 02:25 | |
| *** edmondsw has quit IRC | 02:30 | |
| openstackgerrit | Vishakha Agarwal proposed openstack/nova master: No change in field 'updated' in server https://review.openstack.org/586446 | 03:03 |
| *** diga has joined #openstack-placement | 04:14 | |
| openstackgerrit | melanie witt proposed openstack/nova stable/ocata: [stable only] Handle quota usage during create/delete races https://review.openstack.org/582413 | 04:38 |
| *** tetsuro_ has joined #openstack-placement | 04:55 | |
| *** tetsuro_ has quit IRC | 05:02 | |
| *** tetsuro_ has joined #openstack-placement | 05:08 | |
| *** tetsuro_ has quit IRC | 05:10 | |
| *** tetsuro__ has joined #openstack-placement | 05:10 | |
| *** tetsuro__ has quit IRC | 05:46 | |
| *** tetsuro_ has joined #openstack-placement | 05:49 | |
| openstackgerrit | Vishakha Agarwal proposed openstack/nova master: 'Updated_at' is NULL when show aggregate info https://review.openstack.org/580271 | 05:57 |
| *** tetsuro_ has quit IRC | 05:57 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Remove unused request API sample template https://review.openstack.org/588420 | 06:06 |
| *** gibi is now known as giblet | 06:25 | |
| alex_xu | stephenfin: those sample file are delete by this commit https://review.openstack.org/#/c/149129/ | 06:44 |
| *** tetsuro_ has joined #openstack-placement | 06:52 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Remove unused request API sample template https://review.openstack.org/588420 | 07:00 |
| *** tetsuro_ has quit IRC | 07:03 | |
| *** tetsuro_ has joined #openstack-placement | 07:04 | |
| openstackgerrit | Yongli He proposed openstack/nova master: Load expected attr pci_devices while migrate https://review.openstack.org/588455 | 07:23 |
| openstackgerrit | Merged openstack/nova master: Add another up-call to the cells v2 caveats list https://review.openstack.org/581910 | 07:31 |
| *** tetsuro_ has quit IRC | 07:41 | |
| *** tetsuro_ has joined #openstack-placement | 07:44 | |
| *** tetsuro__ has joined #openstack-placement | 07:54 | |
| *** tetsuro_ has quit IRC | 07:55 | |
| *** tetsuro__ has quit IRC | 08:07 | |
| *** tetsuro_ has joined #openstack-placement | 08:27 | |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Fix nits in resource_provider.py https://review.openstack.org/588470 | 08:28 |
| *** tetsuro_ has quit IRC | 08:40 | |
| *** tetsuro_ has joined #openstack-placement | 08:41 | |
| *** cdent has joined #openstack-placement | 09:09 | |
| openstackgerrit | zhufl proposed openstack/nova master: xx_instance_type_id in list_migrations should be integer https://review.openstack.org/588481 | 09:12 |
| *** deepak_mourya has quit IRC | 09:25 | |
| *** avolkov has joined #openstack-placement | 09:31 | |
| *** tetsuro_ has quit IRC | 09:31 | |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Fix nits in resource_provider.py https://review.openstack.org/588470 | 09:55 |
| *** fanzhang has quit IRC | 10:18 | |
| *** cdent has quit IRC | 10:43 | |
| *** diga has quit IRC | 10:54 | |
| openstackgerrit | Chen proposed openstack/nova master: Revert task_state to none for LM failure due to invalid dest https://review.openstack.org/588512 | 11:01 |
| *** e0ne has joined #openstack-placement | 11:01 | |
| *** e0ne has quit IRC | 11:01 | |
| *** e0ne has joined #openstack-placement | 11:02 | |
| *** e0ne has quit IRC | 11:02 | |
| *** tssurya has joined #openstack-placement | 11:17 | |
| *** e0ne has joined #openstack-placement | 11:17 | |
| *** e0ne has quit IRC | 11:17 | |
| *** cdent has joined #openstack-placement | 11:26 | |
| openstackgerrit | Merged openstack/nova master: Remove unused request API sample template https://review.openstack.org/588420 | 11:29 |
| cdent | jaypipes: I'm eagerly anticipating your blog post, how much longer will I have to wait? | 12:02 |
| openstackgerrit | Liam Young proposed openstack/nova master: Target metadata requests at the correct cell. https://review.openstack.org/588520 | 12:06 |
| openstackgerrit | Merged openstack/nova master: Docs: Add Placement to Nova system architecture https://review.openstack.org/584338 | 12:13 |
| openstackgerrit | Liam Young proposed openstack/nova master: Remove Neutron MetaAPIProxy from cellsv2-layout https://review.openstack.org/588525 | 12:20 |
| openstackgerrit | Surya Seetharaman proposed openstack/nova master: Cleanup comp_node, res_prov, services, aggregate_hosts during cell deletion https://review.openstack.org/546660 | 12:30 |
| *** mriedem has joined #openstack-placement | 12:39 | |
| *** efried is now known as fried_rice | 12:48 | |
| fried_rice | Guten morgen cdent, wie geht's? | 12:49 |
| cdent | tired. vaguely stupid. writing a pupdate. | 12:50 |
| cdent | you? | 12:50 |
| fried_rice | Feeling similar, not quite enough sleep last night. Meant to sleep in, but failed. | 12:51 |
| fried_rice | Anything I should attend to for you today? | 12:52 |
| fried_rice | Haven't gotten through email yet, but I was also curious whether you feel the/a use case for placement debug info has been described if not fully, then at least adequately to justify some action. | 12:54 |
| *** nicolasbock has joined #openstack-placement | 13:00 | |
| cdent | a) nothing in particular I need, I'd say reshaper is the main digs | 13:00 |
| *** edmondsw_ has joined #openstack-placement | 13:00 | |
| fried_rice | I updated the reshaper series last night | 13:02 |
| fried_rice | didn't address nits on +X'd patches, just on -1'd | 13:02 |
| fried_rice | figured the rest for a cumulative fup patch | 13:02 |
| cdent | b) alex raised a good point on the placement debug thing (in response to your patch), which is the disjunction between host and resource provider when "counting things" in placement results. I think to get closer to good action, we need to think that through a bit more. I also don't think we should do anything more, code wise, beyond what you've already done, with placement debugging until after we've wiped other stuff that nova isn't doing yet off | 13:03 |
| cdent | slate | 13:03 |
| fried_rice | noted | 13:05 |
| openstackgerrit | Merged openstack/nova master: Scrub hw:cpu_model from API samples https://review.openstack.org/588371 | 13:19 |
| *** mriedem is now known as mriedem_afk | 13:25 | |
| *** stephenfin is now known as finucannot | 13:28 | |
| fried_rice | cdent: You have/had (a) patch(es) somewhere to ensure traits and resource class cache only on startup, right? | 13:38 |
| fried_rice | cdent: Which I guess haven't merged yet, because I'm still seeing that happen multiple times in a request. | 13:38 |
| cdent | fried_rice: yeah, they were above my os-resource-class changes, because with a sync was added | 13:39 |
| cdent | https://review.openstack.org/#/c/584086/ | 13:39 |
| cdent | it's probably movable | 13:39 |
| fried_rice | no biggie, just wanted to make sure I wasn't seeing this behavior *after* those merged. | 13:40 |
| cdent | depending on how the near future goes I might pull that one out to it's own thing | 13:40 |
| openstackgerrit | Chris Dent proposed openstack/nova stable/queens: [placement] Retry allocation writes server side https://review.openstack.org/588569 | 13:41 |
| openstackgerrit | Eric Fried proposed openstack/nova master: [placement] Debug log per granular request group https://review.openstack.org/588350 | 13:41 |
| openstackgerrit | Stephen Finucane proposed openstack/nova stable/queens: Don't filter out sibling sets with one core https://review.openstack.org/588570 | 13:42 |
| openstackgerrit | Stephen Finucane proposed openstack/nova stable/queens: Ensure emulator threads are always calculated https://review.openstack.org/588571 | 13:42 |
| openstackgerrit | Stephen Finucane proposed openstack/nova stable/queens: Always pass 'NUMACell.siblings' to _pack_instance_onto_cores' https://review.openstack.org/588572 | 13:42 |
| openstackgerrit | Stephen Finucane proposed openstack/nova stable/queens: trivialfix: cleanup _pack_instance_onto_cores() https://review.openstack.org/588573 | 13:42 |
| openstackgerrit | Stephen Finucane proposed openstack/nova stable/queens: Add unit tests for EmulatorThreadsTestCase https://review.openstack.org/588574 | 13:42 |
| openstackgerrit | Stephen Finucane proposed openstack/nova stable/queens: Not use thread alloc policy for emulator thread https://review.openstack.org/588575 | 13:42 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Use placement 1.28 in scheduler report client https://review.openstack.org/583667 | 13:44 |
| jaypipes | cdent: should be done tomorrow. it's a two parter. | 13:45 |
| *** jaypipes is now known as leakypipes | 13:46 | |
| openstackgerrit | Chris Dent proposed openstack/nova master: [placement] Move resource_class_cache into placement hierarchy https://review.openstack.org/584085 | 14:02 |
| openstackgerrit | Chris Dent proposed openstack/nova master: [placement] ensure_rc_cache only at start of process https://review.openstack.org/584086 | 14:02 |
| cdent | fried_rice: there that pull it out from under os-resource-classes | 14:03 |
| fried_rice | cdent: There was one for traits too? | 14:07 |
| cdent | fried_rice: traits merged long ago | 14:08 |
| fried_rice | hm | 14:08 |
| *** alex_xu has quit IRC | 14:08 | |
| cdent | if you're seeing it multiple times now it is because it happens each time a new wsgi-intercept is created | 14:08 |
| cdent | which will be for each yaml file | 14:08 |
| *** edmondsw_ is now known as edmondsw | 14:08 | |
| fried_rice | okay, yes, I see now it was just the rc one that was repeating. | 14:08 |
| cdent | and in the non-gabbi tests, for each tests | 14:08 |
| cdent | (it wil be tried, but the global will prevent it) | 14:08 |
| fried_rice | cdent: Is it still appropriate to remove this comment in light of having pulled this out of the os-resource-classes stack? https://review.openstack.org/#/c/584085/2/nova/api/openstack/placement/resource_class_cache.py | 14:10 |
| cdent | it's been moved | 14:11 |
| fried_rice | I'm not really awake enough to suss that out for certain on my own :( | 14:11 |
| cdent | so it is no longer problematic | 14:11 |
| cdent | it could stop being problematic in two ways | 14:11 |
| cdent | and so far we've done one of them, which is sufficient | 14:11 |
| fried_rice | oh, "this file" meaning the file we're in, not the one referenced by the import. | 14:11 |
| fried_rice | Weird place for that comment to have been. Cool. | 14:12 |
| cdent | yeah, something went awry in spacing somewhere in the history | 14:12 |
| cdent | you've bee bcc'd | 14:15 |
| cdent | bbl | 14:22 |
| *** cdent has quit IRC | 14:23 | |
| *** mriedem_afk is now known as mriedem | 14:43 | |
| *** cdent has joined #openstack-placement | 15:01 | |
| * fried_rice gives up trying to make sure he is profound, complete, and accurate and just hits the f'ing send button. | 15:12 | |
| cdent | good message | 15:13 |
| *** cfriesen has joined #openstack-placement | 15:17 | |
| leakypipes | cfriesen: it's not an exception to return 0 results though. just like, see my email response to the Oracle guy, it's not an exception for a SELECT statement to return 0 rows. | 15:17 |
| fried_rice | cfriesen: I agree with you that it would be nice if blazar (or whatever) could have captured something from nova, which could have captured something from placement, that it could ultimately bubble up to the user, that would let the user know why we denied their request. Easier said than done, by a long shot. | 15:18 |
| cfriesen | I totally get that it's not an exceptional case for placement itself, and I agree | 15:18 |
| fried_rice | which is why we've had the NoValidHost-and-you-have-to-go-digging-in-logs situation forever. | 15:18 |
| cdent | there's a middle ground here. It's not exceptional that 0 results were returned, but clearly people have no faith that things worked as expectred so want to see where the gap was | 15:18 |
| cfriesen | cdent: it's not even necessarly that they have no faith, but that they want to know what went wrong so they can reconfigure the cluster | 15:19 |
| fried_rice | cdent: The working hypothesis is that the middle ground is having more detailed logs in the placement service, yah? | 15:19 |
| cdent | fried_rice: I guess, yeah | 15:19 |
| fried_rice | I mean, surely we're not talking about ?verbose=true where the response includes EXPLAIN output or whatever, are we? | 15:20 |
| cdent | i hope not | 15:20 |
| cdent | cfriesen: does it have to be placement where people look | 15:20 |
| cfriesen | I wouldn't think so. but right now some logic is hidden away inside placement | 15:21 |
| fried_rice | yes, lots of it, no argument there. Monolithic sql statement gives 0 results, no telling which combination of ON/WHERE subclauses were responsible. | 15:22 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Avoid live migrate to same host https://review.openstack.org/542689 | 15:24 |
| fried_rice | The obvious response, as has been stated by leakypipes and others, is to break that monolithic sql statement up and do more of the JOINing, as it were, in python, with helpful debug logs like http://paste.openstack.org/raw/727229/ in between each step. | 15:24 |
| melwitt | fwiw, we don't actually want the end user to know the details of why the request failed (hide details of the cloud especially in public cloud settings) but this is more about information for the operator/admin/support people | 15:24 |
| leakypipes | cdent: we need to combat the lack of faith with good ol' fashioned missionary work, my friend. | 15:25 |
| cdent | praise be | 15:26 |
| leakypipes | lord have mercy | 15:26 |
| fried_rice | I think we're not doing that work (splitting monolithic sql) immediately because a) we want to better understand the problem space to make sure that's the/a right thing to do; and b) we have big feature work that we should be concentrating on finishing out, rather than getting distracted (again). | 15:26 |
| cdent | under his eye | 15:26 |
| cdent | fried_rice++ | 15:27 |
| openstackgerrit | Chris Dent proposed openstack/nova master: [placement] ensure_rc_cache only at start of process https://review.openstack.org/584086 | 15:28 |
| sean-k-mooney | fried_rice: the perfromce hit to that is why we have never gone that way | 15:28 |
| melwitt | also fwiw, I wasn't seeking feature work, just at a high level was curious what's possible because until that thread, I didn't know it was possible at all | 15:28 |
| melwitt | for me, the thread was informational and not in any way "let's do something about this right now" | 15:29 |
| fried_rice | sean-k-mooney: Whereas that's a concern, I don't think it's the main reason. We did monolithic sql originally for perf reasons (and so leakypipes could show off :) but having started that way, we haven't tried splitting it up because it'd be a bunch of work, and we're not clear on the value. | 15:29 |
| cfriesen | right, I'm in no way saying it needs to be done immediately, but rather something to keep in mind so we don't paint ourselves into a corner | 15:29 |
| leakypipes | melwitt: I think we're saying this isn't feature work but rather refactoring/cleanup work that has been *delayed* because of feature work (on things like reshaper, which is necessary for nested/NUMA/CPU pinning/GPU/FPGA jazz) | 15:30 |
| fried_rice | I can see where a contributor like cfriesen could construe his position (in openstack and employer terms) in such a way that it makes sense for him to start working on sql-split-and-log as opposed to other feature work :) Of course, reviewer attention would remain a factor. | 15:30 |
| sean-k-mooney | fried_rice: perhaps be there is also the argument that is what sql is designed to do so why reinvent it in python our seleves | 15:30 |
| melwitt | and I'm not saying it needs to be done at all. just wanted to talk about the issue and share ideas at this stage | 15:31 |
| sean-k-mooney | that said i understand the debug aspect | 15:31 |
| leakypipes | sean-k-mooney: bingo. that was the original argument. | 15:31 |
| fried_rice | sean-k-mooney: True story. I think we're coming to the conclusion that the debug aspect makes it worth reinventing that wheel to some extent. | 15:31 |
| fried_rice | debug and even just code readability. | 15:31 |
| melwitt | I could also see us coming up with a CLI tool that does some placement queries to narrow things and show the caller likely resource culprits | 15:32 |
| melwitt | based on passing in a flavor | 15:32 |
| sean-k-mooney | im not sure leakypipes has reached that conclutions but i have not read teh email thread yet | 15:32 |
| leakypipes | sean-k-mooney: that said, I do believe there's a good middle ground that can be achieved between monster query (and complete lack of visibility) and repeated smaller queries (for each resource class or request group) to increase visibility for the operator. | 15:32 |
| leakypipes | sean-k-mooney: which is what I said in my ML response(s) | 15:32 |
| leakypipes | which reminds me, I didn't respond directly to cdent's ML response on that thread, which I should have. | 15:33 |
| sean-k-mooney | ya thats a reasonably stance in my book | 15:33 |
| leakypipes | sorry about the cdent. :( | 15:33 |
| cdent | it is what it is ;) | 15:33 |
| leakypipes | sean-k-mooney: or even a reasonable stance, as it were. | 15:33 |
| sean-k-mooney | :) | 15:33 |
| * leakypipes morphs into sean-k-mooney's grammar and spell-check bot | 15:33 | |
| cdent | I think diagnostic tools that throw changing queries into placement would be useful | 15:34 |
| cdent | more useful that log digging | 15:34 |
| cdent | s/that/than/ | 15:34 |
| leakypipes | has the OP (bnemec I think?) responded to which approach (split queries vs. a diagnostic-y tool) would be more in-line with his needs? | 15:35 |
| melwitt | yeah, I don't know why I didn't think of it before but since leakypipes mentioned doing multiple queries and narrowing things down, I realized we could automate that if we take a flavor as an arg | 15:35 |
| melwitt | as a CLI tool | 15:35 |
| fried_rice | leakypipes: https://review.openstack.org/#/c/584085/ seems worthy of landing at this stage (post FF, pre release). | 15:36 |
| sean-k-mooney | well as an on looker im getting the feeling as placement is strarting to mature we are starting to get to the point where the operational aspects such as tooling are becomeing more of a fornt vs raw feature dev/creap | 15:36 |
| cdent | very true sean-k-mooney | 15:36 |
| fried_rice | leakypipes: Noticed when I was working on the logging thing that the rc_cache was hitting frequently, be nice to get rid of that. | 15:36 |
| *** cfriesen is now known as cfriesen_afk | 15:36 | |
| fried_rice | leakypipes: (both patches in that series, tbc) | 15:37 |
| leakypipes | ack, on it. | 15:38 |
| melwitt | fwiw, our support people have experience a lot of bewilderment when "no allocation candidates" happens and would be grateful for anything that helps them figure out which resource failed. so far, they have learned how to use osc-placement to look at resource providers and such | 15:38 |
| melwitt | *experienced | 15:38 |
| cdent | pre placement what would they do? | 15:39 |
| cdent | for sake of comparison | 15:39 |
| melwitt | and by "anything" that includes docs like a troubleshooting guide | 15:39 |
| melwitt | they'd look at scheduler debug logs | 15:39 |
| melwitt | as I described in the thread | 15:39 |
| leakypipes | fried_rice, cdent: both +W'd. nice stuff. | 15:39 |
| melwitt | so fwiw I don't think anyone (at least not our support people) are asking for features, they just want to learn what's the new way to troubleshoot | 15:39 |
| cdent | rad, thanks | 15:39 |
| melwitt | and I had been thinking since we have one query for allocation candidates, I thought there was nothing we could do in the debug log department | 15:40 |
| cdent | melwitt: I reckon the big diffrence between the old scheduler logs and now is that the old logs made it clear which type of thing was the killer: "hey there are no hosts with vcpus" | 15:41 |
| melwitt | cdent: yes, exactly | 15:41 |
| cdent | there's not an easy map for that in placement | 15:41 |
| cdent | not at the level of resource class | 15:41 |
| melwitt | right | 15:41 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Deprecate upgrade_levels options for deprecated/removed services https://review.openstack.org/588607 | 15:41 |
| melwitt | last night I was thinking about writing up a doc for our docs with some troubleshooting tips. and maybe I could come up with a CLI tool for nova/tools/ that can do some queries and dump some info using osc-placement underneath based on a flavor passed in | 15:43 |
| cdent | at least exploring those things will lead somewhere | 15:45 |
| cdent | leakypipes, fried_rice in case neither of you made it all the way to the endu of the pupdate: I'm taking a vacation | 15:45 |
| cdent | (from the pupdate itself) | 15:45 |
| fried_rice | cdent: Yup, saw that. Haven't decided whether to act on it. | 15:46 |
| cdent | I don't reckon immediate action is required, because of the current position in the cycle | 15:46 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Deprecate upgrade_levels options for deprecated/removed services https://review.openstack.org/588607 | 15:53 |
| leakypipes | cdent: I appreciate all the hard work you put into the pupdates. I can't commit to taking it on, but I will see what I can do about continuing at least a partial update for folks on the ML. | 15:53 |
| cdent | Once stein starts up, I might pick it up again. I worry that without doing it, I'll have no idea what's going on | 15:54 |
| cdent | we'll see. | 15:54 |
| leakypipes | cdent: well, we can discuss it in September. for now, just know we appreciate your efforts in doing this. | 15:55 |
| cdent | thanks | 15:55 |
| fried_rice | ++ | 15:56 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Reno for notification-transformation-rocky https://review.openstack.org/588403 | 15:57 |
| *** mriedem is now known as hansmoleman | 15:57 | |
| -openstackstatus- NOTICE: The infra team is renaming projects in Gerrit. There will be a short ~10 minute Gerrit downtime in a few minutes as a result. | 16:03 | |
| *** tssurya has quit IRC | 16:05 | |
| cdent | leakypipes, fried_rice: can I safely abandon https://review.openstack.org/#/c/587050/ in favor of some the other exploration that has happened since then or is it (one of the simpler options) still an option? | 16:09 |
| fried_rice | cdent: I think it's worth keeping open for now. | 16:10 |
| cdent | okee | 16:10 |
| fried_rice | cdent: It's *almost* the same as https://review.openstack.org/#/c/588091/ -- except in reverse, kinda. | 16:11 |
| * cdent nods | 16:11 | |
| fried_rice | You're removing the second one; I'm removing the first one. | 16:12 |
| cdent | It would make me a lot happier if something, somewhere, would break when we makes these kinds of changes | 16:13 |
| *** melwitt is now known as jgwentworth | 16:14 | |
| leakypipes | cdent: gerrit's down. can't see those patches just now. | 16:15 |
| fried_rice | cdent: You'll break tests on top of the reshaper stack for sure. It adds tests that actually check the number of calls. | 16:17 |
| fried_rice | which is why mine are failing. | 16:17 |
| cdent | that's a bit artificial. I mean functionality changing | 16:18 |
| fried_rice | well, yes. Unless it's truly not necessary. Which is kind of what we've been hoping, the best-case scenario, nicht war? | 16:18 |
| fried_rice | s/war/wahr/ | 16:19 |
| *** fried_rice is now known as fried_rolls | 16:19 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Remove unused flavor_delete_info() method https://review.openstack.org/588621 | 16:45 |
| *** openstackgerrit has quit IRC | 16:49 | |
| *** openstackgerrit has joined #openstack-placement | 16:56 | |
| openstackgerrit | Lee Yarwood proposed openstack/nova master: fixtures: Track volume attachments within CinderFixtureNewAttachFlow https://review.openstack.org/587013 | 16:56 |
| openstackgerrit | Lee Yarwood proposed openstack/nova master: Add regression test for bug#1784353 https://review.openstack.org/587014 | 16:56 |
| openstackgerrit | Lee Yarwood proposed openstack/nova master: compute: Recreate volume attachments during a reschedule https://review.openstack.org/587071 | 16:56 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix host validity check for live-migration https://review.openstack.org/401009 | 17:05 |
| *** cfriesen_afk is now known as cfriesen | 17:25 | |
| -openstackstatus- NOTICE: Project renames and review.openstack.org downtime are complete without any major issue. | 17:29 | |
| *** e0ne has joined #openstack-placement | 18:15 | |
| *** openstackgerrit has quit IRC | 18:19 | |
| *** openstackgerrit has joined #openstack-placement | 18:49 | |
| openstackgerrit | Merged openstack/nova master: Reno for notification-transformation-rocky https://review.openstack.org/588403 | 18:49 |
| openstackgerrit | Chris Dent proposed openstack/nova master: [placement] ensure_rc_cache only at start of process https://review.openstack.org/584086 | 18:55 |
| cdent | fried_rolls, leakypipes : that fixes the failing unit test (which I hadn't thought about because "unit" but they aren't really unit tests) | 18:56 |
| leakypipes | ack | 19:10 |
| openstackgerrit | Merged openstack/nova master: [placement] Move resource_class_cache into placement hierarchy https://review.openstack.org/584085 | 19:17 |
| *** fried_rolls is now known as fried_rice | 19:18 | |
| leakypipes | cdent: Wallaby'd. | 19:26 |
| cdent | huzzah | 19:26 |
| *** cdent has quit IRC | 19:35 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Remove old check_attach version check in API https://review.openstack.org/588348 | 20:05 |
| *** e0ne has quit IRC | 20:40 | |
| *** e0ne has joined #openstack-placement | 20:41 | |
| *** e0ne has quit IRC | 20:45 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Fix image-defined numa claims during evacuate https://review.openstack.org/588657 | 20:59 |
| openstackgerrit | Eric Fried proposed openstack/nova master: Remove redundant _update()s https://review.openstack.org/588091 | 21:22 |
| fried_rice | cdent, leakypipes: ^ I think this one ought to pass all tests at this point. What are we going need to do to achieve confidence that any of these forays into update deduplication are actually viable for mergitude? | 21:23 |
| *** avolkov has quit IRC | 21:26 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Optimize AZ lookup during schedule_and_build_instances https://review.openstack.org/588665 | 21:29 |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Use CONF.long_rpc_timeout in post_live_migration_at_destination https://review.openstack.org/588668 | 21:47 |
| *** nicolasbock has quit IRC | 22:43 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova master: Update RequestSpec.flavor on resize_revert https://review.openstack.org/588689 | 23:14 |
| openstackgerrit | Merged openstack/nova master: Remove unused flavor_delete_info() method https://review.openstack.org/588621 | 23:18 |
| *** tetsuro_ has joined #openstack-placement | 23:29 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!