| openstackgerrit | fupingxie proposed openstack/nova master: Add an example to add more pci devices in nova.conf https://review.openstack.org/592243 | 00:42 |
|---|---|---|
| openstackgerrit | Merged openstack/nova master: Move str to six.string_types https://review.openstack.org/599493 | 00:49 |
| openstackgerrit | Merged openstack/nova master: Fix a failure to format config sample https://review.openstack.org/597986 | 00:50 |
| openstackgerrit | fupingxie proposed openstack/nova master: Delete allocations for instances that have been moved to another node https://review.openstack.org/582899 | 01:23 |
| *** lei-zh has joined #openstack-placement | 02:13 | |
| *** nicolasbock has quit IRC | 02:18 | |
| openstackgerrit | Brin Zhang proposed openstack/nova-specs master: Resource retrieving: add changes-before filter https://review.openstack.org/591976 | 03:33 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/placement master: Rename PlacementPolicyFixture to PolicyFixture https://review.openstack.org/600326 | 03:47 |
| *** lei-zh has quit IRC | 04:53 | |
| openstackgerrit | Merged openstack/nova stable/rocky: Don't use '_TransactionContextManager._async' https://review.openstack.org/597421 | 04:56 |
| *** lei-zh has joined #openstack-placement | 05:17 | |
| openstackgerrit | Merged openstack/nova master: Doc: PowerVM does support shelve https://review.openstack.org/593052 | 05:47 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova stable/rocky: Fix a failure to format config sample https://review.openstack.org/600336 | 06:10 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/placement master: Unify utils.py and util.py https://review.openstack.org/600353 | 06:54 |
| *** e0ne has joined #openstack-placement | 06:58 | |
| *** tetsuro has joined #openstack-placement | 08:01 | |
| openstackgerrit | Brin Zhang proposed openstack/nova master: Resource retrieving: add changes-before filter https://review.openstack.org/599276 | 08:10 |
| *** cdent has joined #openstack-placement | 08:19 | |
| openstackgerrit | Elod Illes proposed openstack/nova stable/queens: Remove noisy DEBUG log https://review.openstack.org/600372 | 08:29 |
| *** tetsuro has quit IRC | 08:37 | |
| *** tetsuro has joined #openstack-placement | 08:46 | |
| *** lei-zh has quit IRC | 09:01 | |
| *** tetsuro has quit IRC | 09:46 | |
| *** tetsuro has joined #openstack-placement | 09:48 | |
| *** e0ne has quit IRC | 10:15 | |
| *** nicolasbock has joined #openstack-placement | 10:31 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: de-novify wsgi application to expect placement config https://review.openstack.org/600157 | 10:35 |
| *** e0ne has joined #openstack-placement | 10:39 | |
| *** cdent has quit IRC | 11:02 | |
| openstackgerrit | Merged openstack/nova master: Fix nits: Compute: Handle reshaped provider trees https://review.openstack.org/598387 | 11:02 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/placement master: Rename PlacementPolicyFixture to PolicyFixture https://review.openstack.org/600326 | 11:12 |
| openstackgerrit | Claudiu Belu proposed openstack/nova master: tests: autospecs all the mock.patch usages https://review.openstack.org/470775 | 11:17 |
| *** tetsuro has quit IRC | 11:20 | |
| *** cdent has joined #openstack-placement | 11:33 | |
| cdent | update placedock to use openstack/placement, it's happy, placecat is happy, as long as https://review.openstack.org/#/c/600157/ is in there | 11:36 |
| jroll | cdent: holy cow those are some fast tests, much applause | 11:59 |
| cdent | ain't it fun? | 11:59 |
| openstackgerrit | Vladyslav Drok proposed openstack/nova master: Fix resource tracker updates during instance evacuation https://review.openstack.org/600425 | 12:15 |
| *** tetsuro has joined #openstack-placement | 12:23 | |
| *** rubasov has joined #openstack-placement | 12:24 | |
| *** cdent has quit IRC | 12:34 | |
| *** tetsuro has quit IRC | 12:34 | |
| *** tetsuro has joined #openstack-placement | 12:38 | |
| *** e0ne has quit IRC | 12:41 | |
| jaypipes | jroll: which tests did you think were fast? | 12:47 |
| *** e0ne has joined #openstack-placement | 12:59 | |
| *** mriedem has joined #openstack-placement | 13:08 | |
| jroll | jaypipes: just over a minute for -epy27,functional,pep8,py36,functional-py36, from scrollback yesterday | 13:15 |
| jroll | that's pretty dang fast (especially in openstack) | 13:16 |
| efried | stephenfin: Do you know how to add doc jobs? https://review.openstack.org/#/c/600268/ | 13:23 |
| stephenfin | efried: zuul job? If so, AJaeger would be your best bet | 13:24 |
| efried | Thanks. | 13:24 |
| jaypipes | jroll: oh... I thought you were referring to cdent's message about referencing https://review.openstack.org/#/c/600157/ | 13:28 |
| jaypipes | jroll: and I see 5+ minute times for those... | 13:28 |
| jroll | jaypipes: ah, gotcha | 13:29 |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova stable/rocky: placement: use simple code paths when possible https://review.openstack.org/600447 | 13:30 |
| efried | apparently that time is all spend setting up the node/env | 13:30 |
| efried | spent | 13:30 |
| efried | though I haven't looked. | 13:31 |
| edleafe | efried: http://logs.openstack.org/57/600157/2/check/openstack-tox-functional/b0513f8/job-output.txt.gz#_2018-09-06_11_17_05_071371 | 13:33 |
| edleafe | Functional tests took just under 53 seconds to run | 13:33 |
| efried | noyce | 13:33 |
| *** takashin has joined #openstack-placement | 13:34 | |
| efried | There's a thing stephenfin was going to fix in tox so it didn't do some kind of heavyweight check up front, which ought to trim the local times down too. | 13:34 |
| stephenfin | efried: https://github.com/tox-dev/tox/commit/95f5476e31fd20a93b70027efac1f1864b6ed25c | 13:35 |
| efried | stephenfin: What's with the quote changes? | 13:36 |
| stephenfin | efried: I had it in multiple patches and they squashed them | 13:37 |
| efried | ight. This in a release yet? | 13:37 |
| stephenfin | efried: i.e. blame them | 13:37 |
| efried | heh | 13:37 |
| stephenfin | efried: Since 3.1.3, apparently | 13:38 |
| efried | neat | 13:38 |
| stephenfin | (If the GitHub UI is to be believed) | 13:38 |
| efried | I so don't know how to use the github ui. | 13:39 |
| efried | apparently 3.1.3 isn't available for my distro, or something. | 13:40 |
| openstackgerrit | Eric Fried proposed openstack/nova master: Add an example to add more pci devices in nova.conf https://review.openstack.org/592243 | 13:41 |
| *** cdent has joined #openstack-placement | 13:45 | |
| *** takashin has quit IRC | 13:58 | |
| *** takashin has joined #openstack-placement | 14:00 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova master: hyper-v: autospec classes before they are instantiated https://review.openstack.org/342211 | 14:04 |
| openstackgerrit | Eric Fried proposed openstack/placement master: Make docs build https://review.openstack.org/600268 | 14:21 |
| openstackgerrit | Vladyslav Drok proposed openstack/nova master: Fix resource tracker updates during instance evacuation https://review.openstack.org/600425 | 14:22 |
| openstackgerrit | sean mooney proposed openstack/nova master: add caching to _build_regex_range https://review.openstack.org/599071 | 14:38 |
| efried | cdent: FYI AJaeger says he's going to propose a patch to cut us over to zuul templates | 14:40 |
| efried | cdent: in case that was on your soon-list. | 14:41 |
| cdent | efried: awesome, thanks. it was sort of in the back of mind, now that we've got all those jobs | 14:41 |
| cdent | but he wants to do it, awesome | 14:41 |
| efried | yeah, your comment said "once we're pretty sure these work" which I think is now. | 14:41 |
| efried | I like that people are getting involved. Cool to see those patches from kero | 14:42 |
| cdent | yes, very very cool | 14:42 |
| openstackgerrit | Andreas Jaeger proposed openstack/placement master: Use templates in .zuul.yaml https://review.openstack.org/600459 | 14:45 |
| openstackgerrit | Andreas Jaeger proposed openstack/placement master: Add lower-constraints job https://review.openstack.org/600460 | 14:47 |
| *** takashin has left #openstack-placement | 14:53 | |
| *** tetsuro has quit IRC | 14:53 | |
| cdent | yay! | 14:54 |
| edleafe | cdent: efried: are we divvying up the tasks at the bottom of https://etherpad.openstack.org/p/placement-extract-stein-3 yet? | 15:00 |
| edleafe | Or can I just grab one and mark it as being worked on? | 15:00 |
| cdent | edleafe: I think the process is: look at the pending reviews (there aren't many) and start something. Marking it as "taken" is a good idea | 15:01 |
| edleafe | cdent: that was my inclination, but just wanted to check | 15:02 |
| cdent | you happy with that ^ efried ? | 15:02 |
| cdent | edleafe: some of the things listed there already have things in progress but have no marks | 15:03 |
| edleafe | Yeah, I noticed that | 15:03 |
| cdent | if you're feeling really enthusiastic...fix it | 15:04 |
| efried | yes, sounds like a good process | 15:04 |
| cdent | I also added a line in the more general section above about the question of CD, brought on by discussion with jaypipes on https://review.openstack.org/#/c/600157/ | 15:06 |
| openstackgerrit | Matt Riedemann proposed openstack/nova stable/rocky: Configure placement DB context manager for nova-manage/status https://review.openstack.org/600464 | 15:07 |
| cdent | man, I really miss those noop gate jobs | 15:08 |
| edleafe | You know you are a bad programmer when your patch fails the noop jobs | 15:09 |
| efried | edleafe: cdent: I think we should send out an email to the dev ML describing the process in the etherpad and encouraging contributors to pick (and mark) things to work on. | 15:12 |
| efried | As I mentioned earlier, I'm encouraged by the contributions I've seen from random people - we should foster that and get some momentum going. | 15:12 |
| cdent | i was gonna do a pupdate tomorrow. either somebody else do it before then, or I'll put it in there | 15:13 |
| edleafe | I'll write something for the ML | 15:13 |
| efried | thanks ed | 15:15 |
| edleafe | sent | 15:22 |
| *** e0ne has quit IRC | 15:42 | |
| *** vdrok has joined #openstack-placement | 15:55 | |
| openstackgerrit | Eric Fried proposed openstack/nova master: fup: Fix import order and test nit https://review.openstack.org/600474 | 16:16 |
| *** mriedem is now known as mriedem_away | 16:19 | |
| efried | cdent, edleafe: Is https://etherpad.openstack.org/p/placement-extraction-file-notes obsolete at this point? And/or can it be merged into ...-3? | 16:27 |
| cdent | efried: obsolete and/or done | 16:28 |
| efried | cool | 16:28 |
| efried | cleaning up tabs :) | 16:29 |
| efried | I have... an unreasonable number of them open. | 16:29 |
| edleafe | efried: yeah, it's outlived its usefulness | 16:29 |
| efried | I prefer "served its purpose" | 16:29 |
| openstackgerrit | Eric Fried proposed openstack/placement master: Make docs build https://review.openstack.org/600268 | 16:42 |
| openstackgerrit | Eric Fried proposed openstack/placement master: Use templates in .zuul.yaml https://review.openstack.org/600459 | 16:42 |
| openstackgerrit | Eric Fried proposed openstack/placement master: Add lower-constraints job https://review.openstack.org/600460 | 16:42 |
| openstackgerrit | Eric Fried proposed openstack/placement master: Add api-ref job https://review.openstack.org/599524 | 16:45 |
| openstackgerrit | Eric Fried proposed openstack/placement master: Make docs build https://review.openstack.org/600268 | 16:45 |
| openstackgerrit | Eric Fried proposed openstack/placement master: Use templates in .zuul.yaml https://review.openstack.org/600459 | 16:45 |
| openstackgerrit | Eric Fried proposed openstack/placement master: Add lower-constraints job https://review.openstack.org/600460 | 16:45 |
| cdent | efried, edleafe, jaypipes: in case you've missed it jay of cinder has posted a doodle poll for a time to meet with cinder about their thinking about placement | 16:54 |
| efried | ml? | 16:55 |
| cdent | email at: http://lists.openstack.org/pipermail/openstack-dev/2018-September/134371.html | 16:55 |
| efried | ack | 16:55 |
| smcginnis | Jay of Cinder, son of Peter. | 16:56 |
| edleafe | cdent: We have the Blazar meeting on Tuesday at 10 | 17:04 |
| edleafe | You have it checked on the doodle | 17:05 |
| cdent | oh, I misread that | 17:05 |
| cdent | I thouht that was the evening | 17:05 |
| * cdent tries to fix | 17:05 | |
| cdent | edleafe: it _is_ pm | 17:05 |
| cdent | oh wait | 17:05 |
| cdent | timezones | 17:06 |
| * cdent sighs deeply | 17:07 | |
| *** sean-k-mooney has quit IRC | 17:09 | |
| *** ChanServ sets mode: -rf | 17:14 | |
| *** e0ne has joined #openstack-placement | 17:34 | |
| *** mriedem_away is now known as mriedem | 17:41 | |
| cdent | jaypipes, dansmith : I put a response to you both on https://review.openstack.org/#/c/600157/ with a sort of proposal that I'd like to see dismissed quickly if it unrealistic. I don't really know. mriedem and melwitt you probably interested too | 17:46 |
| melwitt | ack | 17:47 |
| melwitt | thanks | 17:48 |
| dansmith | cdent: I was basically agreeing with you, if it wasn't clear | 17:48 |
| cdent | dansmith: I got that you were agreeing with me on the change in that patch, but wasn't clear on the bigger picture/plan | 17:50 |
| * mriedem just got back, needs to read this wall of commentage | 17:51 | |
| dansmith | cdent: the bigger plan I'm referring to is that we're going to make nova in stein not have its own placement, which I think is the consensus | 17:51 |
| mriedem | i need to shat some stuff into jay's ownership transfer spec before i forget too | 17:51 |
| dansmith | that's not how it went for things like the nova-volume split | 17:51 |
| dansmith | which is my only point there | 17:52 |
| dansmith | I'm trying to say that since we're foregoing the (what I consider) normal upgrade-to-a-split scenario, the conf thing is rather moot | 17:52 |
| cdent | does the nova-placement-api vs placement-api thing make any sense? in both senses: did I explain it well enough and is it sensible | 17:52 |
| cdent | because if we do that people can choose to carry on as before if they like | 17:53 |
| dansmith | um, I dunno | 17:53 |
| dansmith | I think it's unnecessary | 17:53 |
| dansmith | in that I don't think it addresses the nothing-changed situation sufficiently to be worth the potential confusion | 17:54 |
| cdent | I guess to me it is nothing changed: that placement is on a different import path doesn't change that it is the same as it ever was. I can see how that would not feel like that to everyone but me, however. | 17:55 |
| dansmith | it won't pass grenade, for example | 17:55 |
| dansmith | well, pip might cheat that for us, I'm not positive | 17:56 |
| cdent | it would if we configured grenader to use nova-placement-api | 17:56 |
| cdent | and make placement a dependency of nova | 17:56 |
| dansmith | if that upgrade pulls in the new placement via pip yeah | 17:56 |
| dansmith | but to be clear, I'm not going to be satisfied with that scenario ticking the grenade box :) | 17:57 |
| * cdent nods | 17:57 | |
| cdent | I dont' think it is good for grenade but it _might_ be good for some users and that's what I'm trying to fiture out | 17:57 |
| dansmith | *shrug* | 17:57 |
| dansmith | it'd be cheating the db schema too, | 17:57 |
| dansmith | which I also think is problematic | 17:57 |
| dansmith | because if we renumber the migrations, which I think we shoudl do, | 17:58 |
| dansmith | you couldn't run placement-manage against the nova-api db without issues | 17:58 |
| * dansmith has to jump on a call in 70 seconds | 17:58 | |
| cdent | is it fair to say "don't do that"? | 17:59 |
| cdent | (in this usage scenario) | 17:59 |
| dansmith | not imho | 17:59 |
| cdent | presumably this scenario is only for people who have "reasons" | 17:59 |
| dansmith | I'd say if someone wants to take that risk, they should cook up their own wsgi glue | 17:59 |
| cdent | well that would certainly be easy enough | 18:00 |
| *** e0ne has quit IRC | 18:01 | |
| *** openstackgerrit has quit IRC | 18:05 | |
| *** e0ne has joined #openstack-placement | 18:06 | |
| *** e0ne has quit IRC | 18:08 | |
| *** melwitt is now known as melwitt_awaysick | 18:18 | |
| *** openstackgerrit has joined #openstack-placement | 18:21 | |
| openstackgerrit | Ed Leafe proposed openstack/placement master: Remove multiple database scaffolding https://review.openstack.org/600497 | 18:21 |
| mriedem | cdent: jaypipes: dansmith: replied on https://review.openstack.org/#/c/600157/ | 18:29 |
| cdent | thanks | 18:29 |
| mriedem | tl;dr let's ask mnaser to weigh in and hold off on merging that until we have an idea of what the grenade change looks like | 18:29 |
| openstackgerrit | Chris Dent proposed openstack/placement master: Use both unit and functional for coverage testing https://review.openstack.org/600501 | 18:36 |
| openstackgerrit | Chris Dent proposed openstack/placement master: Add a zuul check job for coverage https://review.openstack.org/600502 | 18:36 |
| mriedem | so besides needing to write placement.conf in grenade before upgrading to stein and starting placement in devstack on the *new* side, | 18:38 |
| mriedem | the other major thing is dealing with the db migration right? | 18:38 |
| cdent | yes, assuming those are the constraits we're setting | 18:38 |
| cdent | i agree with dansmith that a bash script of mysqldump is likely the fastest thing | 18:39 |
| mriedem | i don't think we want to just copy/rename nova.conf to placement.conf and point placement at nova_api db and say grenade is done | 18:39 |
| mriedem | cdent: yeah agree there too | 18:39 |
| mriedem | i'm just not sure where that script would live | 18:39 |
| mriedem | since it doesn't exist in nova in rocky | 18:39 |
| mriedem | we could script it into grenade itself, but then it's not really re-usable by other deployment projects that want to do the same thing | 18:40 |
| cdent | either in grenade on the "new" side, as part of placement | 18:40 |
| cdent | or perhaps in placement's tools? | 18:40 |
| mriedem | i'm not sure if we have placement installed at that point | 18:40 |
| cdent | which point are you thinkign? | 18:40 |
| mriedem | by the time the from_rocky script runs in grenade | 18:40 |
| cdent | grenade's stages are hard to remember | 18:40 |
| mriedem | yeah, it's in the docs though i think, looking | 18:41 |
| mriedem | https://docs.openstack.org/grenade/latest/plugins.html | 18:41 |
| cdent | I suppose I should remember this, I think I wrote one of the first plugins, but damn that was a long time ago | 18:42 |
| mriedem | item 4 in https://docs.openstack.org/grenade/latest/readme.html#theory-of-upgrade mentions the from-<release> scripts but not when they are run, i mean clearly during the upgrade phase, but | 18:44 |
| * mriedem digs into code | 18:44 | |
| cdent | I got scolded recently by my lovely wife for working too much of late, so I'm going to avoid getting sucked in now, mriedem, but if you are able to put your findings somewhere that would be teh ossum | 18:45 |
| openstackgerrit | Ed Leafe proposed openstack/placement master: Rename files to remove 'placement' https://review.openstack.org/600506 | 18:45 |
| mriedem | sure, maybe i'll push a grenade docs patch :) | 18:46 |
| mriedem | so uh, i think grenade is broken on master | 18:56 |
| mriedem | http://logs.openstack.org/75/585475/12/check/neutron-grenade/254e48b/logs/grenade_localrc.txt.gz | 18:57 |
| mriedem | grenade on master is running from queens | 18:57 |
| mriedem | rather than rocky | 18:57 |
| mriedem | yikes https://review.openstack.org/#/c/591594/ | 19:01 |
| cdent | hmm | 19:05 |
| cdent | time moves slowly | 19:05 |
| cdent | but doens't this mean skip level upgrades are working? | 19:05 |
| *** e0ne has joined #openstack-placement | 19:05 | |
| cdent | oh you said that over in nova :) | 19:05 |
| mriedem | i'll shout it from the rooftops | 19:10 |
| mriedem | we can all go home now, job done | 19:10 |
| cdent | yay | 19:11 |
| * cdent goes home | 19:18 | |
| cdent | goodnight all | 19:18 |
| *** cdent has quit IRC | 19:19 | |
| dansmith | mriedem: the script should be in placement I think | 19:35 |
| dansmith | tools/bootstrap-from-nova.sh | 19:35 |
| dansmith | make sure we're using the thing that people will be given, | 19:35 |
| dansmith | not require them to go fetch it out of grenade | 19:35 |
| mriedem | dansmith: ideally yes, i'm just trying to figure out if we even have placement pip installed by the time we run the from-rocky script | 19:36 |
| dansmith | not pip installed, | 19:36 |
| dansmith | but we should have the tree in /opt/stack/new yeah | 19:37 |
| dansmith | even still, | 19:37 |
| dansmith | you could make it a placement upgrade step, not a nova one | 19:37 |
| mriedem | ok so it looks like we do install from the new side before calling the per-release from-<release> scripts | 19:44 |
| mriedem | so if we put the upgrade scripts in placement, that should work | 19:44 |
| dansmith | well, even still, it's in /opt/stack if we want to go grab it | 19:45 |
| dansmith | but yeah | 19:45 |
| *** e0ne has quit IRC | 20:02 | |
| mriedem | so uh, i'm thinking it might be good to maybe recap in the ML some of what came out of today? | 20:33 |
| mriedem | maybe new thread specific to grenade? | 20:33 |
| mriedem | shall i? | 20:33 |
| mriedem | efried: ^ | 20:33 |
| efried | mriedem: Yes please. I still don't understand upgrade/grenade well enough to speak about it. | 20:34 |
| *** nicolasbock has quit IRC | 20:34 | |
| *** nicolasbock has joined #openstack-placement | 20:54 | |
| *** dims has quit IRC | 20:55 | |
| mriedem | efried: done | 20:58 |
| efried | thanks dood | 20:58 |
| *** dims has joined #openstack-placement | 21:11 | |
| openstackgerrit | Eric Fried proposed openstack/placement master: De-nova-ify and reformat contributor guide https://review.openstack.org/599864 | 21:20 |
| *** mriedem is now known as mriedem_away | 22:02 | |
| *** mriedem_away has quit IRC | 22:02 | |
| openstackgerrit | Merged openstack/nova-specs master: Add serial numbers for local disks https://review.openstack.org/595247 | 22:05 |
| *** rubasov has quit IRC | 23:42 | |
| *** rubasov has joined #openstack-placement | 23:47 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!