| *** s10 has quit IRC | 00:06 | |
| *** s10 has joined #openstack-placement | 00:06 | |
| *** s10 has quit IRC | 00:07 | |
| *** s10 has joined #openstack-placement | 00:07 | |
| *** s10 has quit IRC | 00:08 | |
| *** s10 has joined #openstack-placement | 00:08 | |
| *** s10 has quit IRC | 00:08 | |
| *** s10 has joined #openstack-placement | 00:09 | |
| *** s10 has quit IRC | 00:09 | |
| *** s10 has joined #openstack-placement | 00:10 | |
| *** s10 has quit IRC | 00:10 | |
| *** s10 has joined #openstack-placement | 00:10 | |
| *** s10 has quit IRC | 00:11 | |
| *** s10 has joined #openstack-placement | 00:11 | |
| *** s10 has quit IRC | 00:12 | |
| *** takashin has joined #openstack-placement | 00:12 | |
| *** s10 has joined #openstack-placement | 00:12 | |
| *** s10 has quit IRC | 00:12 | |
| *** s10 has joined #openstack-placement | 00:13 | |
| *** s10 has quit IRC | 00:13 | |
| *** s10 has joined #openstack-placement | 00:13 | |
| *** s10 has quit IRC | 00:14 | |
| *** s10 has joined #openstack-placement | 00:14 | |
| *** s10 has quit IRC | 00:15 | |
| *** s10 has joined #openstack-placement | 00:15 | |
| *** s10 has quit IRC | 00:15 | |
| *** s10 has joined #openstack-placement | 00:16 | |
| *** s10 has quit IRC | 00:16 | |
| *** s10 has joined #openstack-placement | 00:17 | |
| *** s10 has quit IRC | 00:17 | |
| *** s10 has joined #openstack-placement | 00:17 | |
| *** s10 has quit IRC | 00:18 | |
| *** s10 has joined #openstack-placement | 00:18 | |
| *** s10 has quit IRC | 00:18 | |
| *** s10 has joined #openstack-placement | 00:19 | |
| *** s10 has quit IRC | 00:19 | |
| *** s10 has joined #openstack-placement | 00:20 | |
| *** s10 has quit IRC | 00:20 | |
| *** s10 has joined #openstack-placement | 00:20 | |
| *** s10 has quit IRC | 00:21 | |
| *** s10 has joined #openstack-placement | 00:21 | |
| *** s10 has quit IRC | 00:22 | |
| *** s10 has joined #openstack-placement | 00:22 | |
| *** s10 has quit IRC | 00:22 | |
| *** s10 has joined #openstack-placement | 00:23 | |
| *** s10 has quit IRC | 00:23 | |
| *** s10 has joined #openstack-placement | 00:23 | |
| *** s10 has quit IRC | 00:24 | |
| *** s10 has joined #openstack-placement | 00:24 | |
| *** s10 has quit IRC | 00:25 | |
| *** s10 has joined #openstack-placement | 00:25 | |
| *** s10 has quit IRC | 00:25 | |
| *** s10 has joined #openstack-placement | 00:26 | |
| *** s10 has quit IRC | 00:26 | |
| *** s10 has joined #openstack-placement | 00:27 | |
| *** s10 has quit IRC | 00:27 | |
| *** s10 has joined #openstack-placement | 00:27 | |
| *** s10 has quit IRC | 00:28 | |
| *** s10 has joined #openstack-placement | 00:28 | |
| *** s10 has quit IRC | 00:29 | |
| *** s10 has joined #openstack-placement | 00:29 | |
| *** s10 has quit IRC | 00:29 | |
| *** s10 has joined #openstack-placement | 00:30 | |
| *** s10 has quit IRC | 00:30 | |
| *** s10 has joined #openstack-placement | 00:30 | |
| *** s10 has quit IRC | 00:31 | |
| *** s10 has joined #openstack-placement | 00:31 | |
| *** s10 has quit IRC | 00:32 | |
| *** s10 has joined #openstack-placement | 00:32 | |
| *** s10 has quit IRC | 00:32 | |
| *** s10 has joined #openstack-placement | 00:33 | |
| *** s10 has quit IRC | 00:33 | |
| *** s10 has joined #openstack-placement | 00:34 | |
| *** s10 has quit IRC | 00:34 | |
| *** s10 has joined #openstack-placement | 00:34 | |
| *** s10 has quit IRC | 00:35 | |
| *** s10 has joined #openstack-placement | 00:35 | |
| *** s10 has quit IRC | 00:36 | |
| *** s10 has joined #openstack-placement | 00:36 | |
| *** s10 has quit IRC | 00:36 | |
| *** s10 has joined #openstack-placement | 00:37 | |
| *** s10 has quit IRC | 00:37 | |
| *** s10 has joined #openstack-placement | 00:37 | |
| *** s10 has quit IRC | 00:38 | |
| *** s10 has joined #openstack-placement | 00:38 | |
| *** s10 has quit IRC | 00:39 | |
| *** s10 has joined #openstack-placement | 00:39 | |
| *** s10 has quit IRC | 00:39 | |
| *** s10 has joined #openstack-placement | 00:40 | |
| *** s10 has quit IRC | 00:40 | |
| *** s10 has joined #openstack-placement | 00:41 | |
| *** s10 has quit IRC | 00:41 | |
| *** s10 has joined #openstack-placement | 00:41 | |
| *** s10 has quit IRC | 00:42 | |
| *** s10 has joined #openstack-placement | 00:42 | |
| *** s10 has quit IRC | 00:42 | |
| *** s10 has joined #openstack-placement | 00:43 | |
| *** s10 has quit IRC | 00:43 | |
| *** s10 has joined #openstack-placement | 00:44 | |
| *** s10 has quit IRC | 00:44 | |
| *** s10 has joined #openstack-placement | 00:44 | |
| *** s10 has quit IRC | 00:45 | |
| *** s10 has joined #openstack-placement | 00:45 | |
| *** s10 has quit IRC | 00:46 | |
| *** s10 has joined #openstack-placement | 00:46 | |
| *** s10 has quit IRC | 00:46 | |
| *** s10 has joined #openstack-placement | 00:47 | |
| *** s10 has quit IRC | 00:47 | |
| *** s10 has joined #openstack-placement | 00:47 | |
| *** s10 has quit IRC | 00:48 | |
| *** s10 has joined #openstack-placement | 00:48 | |
| *** s10 has quit IRC | 00:49 | |
| *** tetsuro has joined #openstack-placement | 01:37 | |
| *** lei-zh has joined #openstack-placement | 02:06 | |
| *** lei-zh has quit IRC | 02:31 | |
| *** lei-zh has joined #openstack-placement | 02:31 | |
| *** tetsuro has quit IRC | 02:59 | |
| *** tetsuro has joined #openstack-placement | 03:02 | |
| *** lei-zh has quit IRC | 03:32 | |
| *** tetsuro has quit IRC | 04:03 | |
| *** tetsuro has joined #openstack-placement | 04:05 | |
| *** tetsuro has quit IRC | 04:33 | |
| *** s10 has joined #openstack-placement | 05:18 | |
| *** s10 has quit IRC | 06:19 | |
| *** lei-zh has joined #openstack-placement | 06:20 | |
| *** e0ne has joined #openstack-placement | 06:37 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs master: Detach and attach boot volumes - Stein https://review.openstack.org/600628 | 07:17 |
|---|---|---|
| *** helenafm has joined #openstack-placement | 07:33 | |
| *** tssurya has joined #openstack-placement | 07:49 | |
| *** rubasov has quit IRC | 07:55 | |
| *** rubasov has joined #openstack-placement | 07:55 | |
| *** takashin has left #openstack-placement | 08:00 | |
| *** e0ne has quit IRC | 08:03 | |
| *** tetsuro has joined #openstack-placement | 08:27 | |
| *** s10 has joined #openstack-placement | 08:34 | |
| *** lei-zh has quit IRC | 08:58 | |
| *** helenafm has quit IRC | 09:01 | |
| *** e0ne has joined #openstack-placement | 09:09 | |
| *** ttsiouts has joined #openstack-placement | 09:12 | |
| *** s10 has quit IRC | 09:15 | |
| *** helenafm has joined #openstack-placement | 09:21 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: Clean up and clarify tox.ini https://review.openstack.org/611719 | 09:39 |
| *** s10 has joined #openstack-placement | 09:50 | |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova-specs master: Spec: Support filtering by forbidden aggregate https://review.openstack.org/603352 | 10:13 |
| *** tetsuro has quit IRC | 10:20 | |
| *** helenafm has quit IRC | 10:23 | |
| *** tetsuro has joined #openstack-placement | 10:31 | |
| *** cdent has joined #openstack-placement | 10:35 | |
| *** helenafm has joined #openstack-placement | 10:45 | |
| *** tetsuro has quit IRC | 10:58 | |
| *** e0ne has quit IRC | 11:01 | |
| *** e0ne has joined #openstack-placement | 11:52 | |
| *** mriedem has joined #openstack-placement | 12:19 | |
| *** cdent has quit IRC | 12:24 | |
| *** dansmith is now known as SteelyDan | 12:45 | |
| *** mdbooth has joined #openstack-placement | 12:49 | |
| *** stephenfin is now known as finucannot | 12:57 | |
| finucannot | I'm trying to run the tests for osc-placement but everything's failing with 'Missing value auth-url required for auth plugin password'. What do I need to do? | 13:32 |
| *** rubasov has quit IRC | 14:05 | |
| *** mriedem has quit IRC | 14:06 | |
| *** mriedem has joined #openstack-placement | 14:09 | |
| *** s10 has quit IRC | 14:15 | |
| openstackgerrit | Stephen Finucane proposed openstack/osc-placement master: Enforce key-value'ness for 'allocation candidate list --resource' https://review.openstack.org/611883 | 14:18 |
| openstackgerrit | Stephen Finucane proposed openstack/osc-placement master: tox: Hide deprecation warnings from stdlib https://review.openstack.org/611884 | 14:18 |
| *** s10 has joined #openstack-placement | 14:21 | |
| *** efried is now known as efried_pto | 14:26 | |
| *** ttsiouts has quit IRC | 14:50 | |
| *** e0ne has quit IRC | 14:52 | |
| sean-k-mooney | finucannot: have you sourced your openrc file in devstack? | 14:55 |
| sean-k-mooney | by the way do people know if cdent has left for the weekend or just for lunch? | 14:56 |
| *** s10 has quit IRC | 15:01 | |
| *** ttsiouts has joined #openstack-placement | 15:01 | |
| finucannot | sean-k-mooney: Do I need DevStack deployed? They're functional tests, no integration tests | 15:03 |
| finucannot | *not | 15:03 |
| sean-k-mooney | finucannot: well if they are fucntional test osc-placement is not going to help | 15:03 |
| finucannot | sean-k-mooney: I just want to run the tests I added above in https://review.openstack.org/#/c/611883/ | 15:04 |
| finucannot | but 'tox -e functional osc_placement.tests.functional.test_allocation_candidate' moans with that error | 15:04 |
| sean-k-mooney | oh ok | 15:04 |
| finucannot | So I'm obviously doing _something_ wrong but it's not documented what | 15:04 |
| *** helenafm has quit IRC | 15:05 | |
| sean-k-mooney | let me try and run that locally | 15:05 |
| finucannot | sean-k-mooney: No rush. It's in the CI now so it's either good or I'll either have to rework based on that results | 15:06 |
| finucannot | *the | 15:06 |
| sean-k-mooney | i get the same error | 15:08 |
| finucannot | Yeah, we must have to do something so that's not documented | 15:09 |
| * finucannot looks as zuul config | 15:09 | |
| sean-k-mooney | same issue on master too | 15:09 |
| finucannot | Yeah, the functional tests are using DevStack | 15:09 |
| finucannot | Meaning these are actually integration tests. Lovely. | 15:10 |
| sean-k-mooney | finucannot: right so they are like the neutron ones | 15:10 |
| finucannot | Yup, and not like the nova ones | 15:10 |
| finucannot | Aight, I'm leaving that to that gate so :) | 15:10 |
| sean-k-mooney | finucannot: not exactly they are only testing placement so they are not full integration testes they are still functional | 15:10 |
| sean-k-mooney | am i think cdent has a docker container he uses or a gabby server for placement testing | 15:11 |
| *** tssurya has quit IRC | 15:11 | |
| sean-k-mooney | not sure if the osc-plugin expects those | 15:11 |
| finucannot | My understanding of functional tests was that they should test interactions between various sub-services of that same service | 15:11 |
| finucannot | Mocking out all external dependencies | 15:11 |
| *** ttsiouts has quit IRC | 15:12 | |
| finucannot | But I know we've differed there in the past :) (libvirt stuff) | 15:12 |
| sean-k-mooney | finucannot: yes but useing the placement api would be resonably for the placement api client | 15:12 |
| finucannot | Not if all I want to test is the CLI parsing | 15:12 |
| sean-k-mooney | neutron has openvswithc functional test the require openvswtich to be installed | 15:12 |
| finucannot | You could still stub out openvswitch | 15:13 |
| sean-k-mooney | finucannot: then use unit test not fuctional tests | 15:13 |
| sean-k-mooney | finucannot: i think they have a fake api servece somewhere however that you can use | 15:13 |
| finucannot | Eh, the test I've written should work. I just can't run it locally atm | 15:13 |
| finucannot | Good enough for me | 15:13 |
| sean-k-mooney | finucannot: this was the dev enve thing i was refering too that cdent created https://github.com/cdent/placedock | 15:18 |
| *** mriedem is now known as mriedem_afk | 15:27 | |
| sean-k-mooney | finucannot: https://github.com/openstack/python-openstackclient/blob/master/doc/source/contributor/developing.rst#running-functional-tests so the osc docs mention that they functional tests expect a devstack env just like neutrons dsvm-functional jobs | 15:34 |
| finucannot | sean-k-mooney: Indeed. Good spot. I guess we should add that doc to osc-placement too | 15:35 |
| finucannot | Or at least link to it | 15:35 |
| sean-k-mooney | ya probably a good idea. | 15:36 |
| *** ttsiouts has joined #openstack-placement | 15:37 | |
| *** openstackgerrit has quit IRC | 16:24 | |
| *** ttsiouts has quit IRC | 16:30 | |
| *** e0ne has joined #openstack-placement | 16:35 | |
| *** e0ne has quit IRC | 16:37 | |
| *** mdbooth has quit IRC | 17:06 | |
| *** cdent has joined #openstack-placement | 17:20 | |
| *** e0ne has joined #openstack-placement | 17:53 | |
| *** openstackgerrit has joined #openstack-placement | 17:55 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: Use gabbi-tempest job from gabbi-tempest plugin https://review.openstack.org/607508 | 17:56 |
| cdent | efried_pto, finucannot : sorry for rushing that tox change without checking it properly. made some adjustments and doing that checking now | 17:57 |
| *** e0ne has quit IRC | 18:00 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: Fix genpolicy tox job https://review.openstack.org/611927 | 18:01 |
| openstackgerrit | Chris Dent proposed openstack/placement master: Clean up and clarify tox.ini https://review.openstack.org/611719 | 18:06 |
| *** e0ne has joined #openstack-placement | 18:08 | |
| *** gibi is now known as gibi_off | 18:12 | |
| *** mriedem_afk is now known as mriedem | 18:41 | |
| cdent | sean-k-mooney: lemme find the thing | 18:53 |
| sean-k-mooney | cdent: ya jay mentioned you were fixing some things in this area | 18:53 |
| cdent | ah, I guess this is a new one, becuse it was lee who fixed the first one: "so we've worked past one issue and are now hitting another" | 18:54 |
| sean-k-mooney | yes | 18:55 |
| sean-k-mooney | lee is on PTO so he asked me to contiue working on it | 18:55 |
| sean-k-mooney | so the issue as far as i can tell is that since the placement_context_manager = enginefacade.transaction_context() | 18:56 |
| sean-k-mooney | is initalised when we first load the module. if we reload the wsgi app due to a failure it gets loaded back into the same python interperater process and its not reinitallised | 18:56 |
| sean-k-mooney | specificlally https://github.com/openstack/placement/blob/master/placement/db_api.py#L21 | 18:57 |
| sean-k-mooney | so intailly i was going to catch the exception and skip the configeration the second tiem but mdbooth suggestion just recreating the transation_context in the configure function | 18:58 |
| cdent | you're running under mod_wsgi, yeah? | 18:59 |
| cdent | i suppose it would be unfair to say "don't do that" :) ? | 18:59 |
| cdent | also, you're seeing these problems in placement, there's some chance you'll get the same problem with running nova-api under mod wsgi, if you do that | 19:00 |
| cdent | anyway, reading the code | 19:00 |
| sean-k-mooney | am im not sure i would have to check what tripplo does. there two questions i wanted to ask. 1 do you have a perference/other suggestion and second what is the process for submiting patchs to both nova and neutron. | 19:00 |
| melwitt | hm, that looks similar to what we have in nova, so do we have the same problem in nova too? https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api.py#L80 | 19:01 |
| melwitt | which I now see cdent has already said | 19:01 |
| cdent | sean-k-mooney: do you mean both nova and placement? | 19:02 |
| sean-k-mooney | that is a good question. so this is partly being uncovered because there is a race in how tripplo starts nova and placement. and how trippleo doe the db sync/migration of the same | 19:02 |
| sean-k-mooney | cdent: yes i ment nova and placement | 19:02 |
| sean-k-mooney | cdent: i belive both patches should have the same change id | 19:03 |
| sean-k-mooney | cdent: is there an odering i should follow? | 19:03 |
| cdent | we haven't really established a pattern. it would certainly be nice to have the same change id. Since they won't really be a cherry-pick I don't it matters what order you use. | 19:03 |
| cdent | When mriedem recently did something similar for the consumer_id grouping bug, the code started in nova and went to placement. | 19:04 |
| cdent | I'd prefer it go the other way, but I don't think it actively matters _now_. Later it will. | 19:04 |
| melwitt | +1 change-id should be the same for both nova and placement, that's how we've been doing it | 19:04 |
| sean-k-mooney | ok that is quite simple to do | 19:05 |
| sean-k-mooney | as for will this happen with nova api i guess its possible | 19:05 |
| cdent | by the way, the change-id for lee's earlier change is at I41b5c7990d4d62a3a397f1686261f3fb7dc1a0be | 19:07 |
| sean-k-mooney | cdent: so yes it looks like the docker contaienr command is httpd so mod_wsgi https://github.com/openstack/tripleo-heat-templates/blob/master/docker/services/nova-placement.yaml#L107 | 19:07 |
| cdent | that's what I thought. mod-wsgi and uwsgi have different approaches to how they deal with a wsgi application that has an exception during initialization | 19:07 |
| sean-k-mooney | ya so it looks like the code is not expecting for the apllication to be reloaded inside the same interperter process on failure | 19:09 |
| sean-k-mooney | which is what i assume mod-wsgi is doing | 19:09 |
| cdent | yes | 19:10 |
| sean-k-mooney | and for uwsgi i assume if it reloads the app it receretate the python iterperter | 19:10 |
| cdent | yup | 19:10 |
| sean-k-mooney | ya that is a pain | 19:10 |
| sean-k-mooney | so do you have a perference. | 19:11 |
| sean-k-mooney | i can recreate the context in configure or i can catch the excpetion | 19:11 |
| sean-k-mooney | both would work but we likely need to make the same change in the nova-api in that case | 19:11 |
| sean-k-mooney | melwitt: cdent by the way should i also associate this with lee's bug https://bugs.launchpad.net/tripleo/+bug/1784155 or open a new one? | 19:12 |
| openstack | Launchpad bug 1784155 in tripleo "nova_placement service start not coordinated with api db sync on multiple controllers" [High,Triaged] - Assigned to Martin Schuppert (mschuppert) | 19:12 |
| sean-k-mooney | from a down stream perspecive its all related to the same issue but its technically a different failure | 19:13 |
| melwitt | sean-k-mooney: normally you would just add a project to the existing bug but that one already shows a Fix Released for nova. is it that there was a fix but it didn't solve the problem? | 19:14 |
| melwitt | if so, maybe just open a new bug to describe the current situation and use that | 19:14 |
| cdent | sean-k-mooney: I think this is pretty much a second issue, so agree with melwitt | 19:14 |
| sean-k-mooney | cool | 19:14 |
| sean-k-mooney | melwitt: the first fix exposed a second issue | 19:15 |
| melwitt | aye | 19:15 |
| cdent | I think I'd prefer the context in configure fix, but that might lead to needing to do some more cleanups elsewhere, I'm leaving some comments about that on the review. | 19:15 |
| sean-k-mooney | cdent: ok cool. ill make whatever changes are asked for on monday and file the bug ectra | 19:16 |
| sean-k-mooney | cdent: i dont hink anything shoudl have been using the context directly form the module import | 19:16 |
| cdent | probably not, but that's the pattern that's been established in nova for a long long time | 19:17 |
| sean-k-mooney | ya the placemnet change im not expecting to break anything(partly because of the lack of any unit tests for that file) if i don the same transformaion in the nova api im less certin | 19:18 |
| sean-k-mooney | cdent: anyway thanks. its getting late for both of us so have a nice weekend. thanks for weighing in on what to do to adress this | 19:20 |
| cdent | if you were to straight up remove the module-level context, all the functional tests will fail | 19:20 |
| cdent | it needs to exist. but it doesn't need to be initialized at import time | 19:20 |
| sean-k-mooney | cdent: right i was going to set it to non but initalsed it in configure | 19:20 |
| sean-k-mooney | *None | 19:20 |
| cdent | yeah, exactly, said the sqme thing in my comments that I'm nearly done with | 19:21 |
| sean-k-mooney | cool that was mdbooth suggestion too. | 19:21 |
| sean-k-mooney | melwitt: it would be a different patch obviously but would you like me to look at doing a similar change for the nova api | 19:22 |
| cdent | my comments are submitted now | 19:22 |
| sean-k-mooney | cdent: thanks :) | 19:23 |
| melwitt | sean-k-mooney: it's up to you. if you could just add nova to the new launchpad bug so we don't lose track of it, that is good enough | 19:23 |
| sean-k-mooney | melwitt: sure. i need to submit a patch to both placemnet and nova in both cases so ill add both projects | 19:25 |
| melwitt | sean-k-mooney: coolness, thanks | 19:25 |
| cdent | placement doesn't have a project in launchpad yet (last I checked), still a tag in nova | 19:26 |
| cdent | operational bugs are the best kind of bugs | 19:26 |
| sean-k-mooney | oh ok ill just file the bug against nova so and add the placement tag | 19:28 |
| sean-k-mooney | ill also mention the triplo bug as a related bug so that can track it too. | 19:28 |
| * sean-k-mooney somtimes i feel like i spend more times tracking work then doing it | 19:29 | |
| *** e0ne has quit IRC | 19:34 | |
| *** e0ne has joined #openstack-placement | 19:34 | |
| *** e0ne has quit IRC | 19:36 | |
| * cdent waves goodnight | 20:12 | |
| *** cdent has quit IRC | 20:12 | |
| *** e0ne has joined #openstack-placement | 20:28 | |
| *** e0ne has quit IRC | 20:33 | |
| *** e0ne has joined #openstack-placement | 20:44 | |
| *** e0ne has quit IRC | 20:48 | |
| *** mriedem has quit IRC | 21:40 | |
| *** sean-k-mooney has quit IRC | 22:03 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!