| *** tetsuro has quit IRC | 00:07 | |
| *** tetsuro has joined #openstack-placement | 00:10 | |
| *** mriedem has quit IRC | 00:17 | |
| openstackgerrit | Tetsuro Nakamura proposed openstack/placement master: WIP: Set root_provider_id in database https://review.openstack.org/619126 | 01:17 |
|---|---|---|
| openstackgerrit | Merged openstack/placement master: Add recreate test for bug 1799892 https://review.openstack.org/617031 | 04:14 |
| openstack | bug 1799892 in OpenStack Compute (nova) rocky "Placement API crashes with 500s in Rocky upgrade with downed compute nodes" [Medium,In progress] https://launchpad.net/bugs/1799892 - Assigned to Matt Riedemann (mriedem) | 04:14 |
| openstackgerrit | Merged openstack/placement master: Consider root id is None in the database case https://review.openstack.org/619041 | 04:17 |
| *** diga has quit IRC | 06:00 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs master: Amend the detach-boot-volume design https://review.openstack.org/619161 | 07:37 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs master: Make scheduling weight more granular https://review.openstack.org/599308 | 07:38 |
| openstackgerrit | Tetsuro Nakamura proposed openstack/placement master: Set root_provider_id in the database https://review.openstack.org/619126 | 07:57 |
| *** sean-k-mooney has quit IRC | 08:06 | |
| *** helenafm has joined #openstack-placement | 08:21 | |
| *** takashin has left #openstack-placement | 08:30 | |
| *** dtantsur|afk is now known as dtantsur | 08:42 | |
| openstackgerrit | Tetsuro Nakamura proposed openstack/placement master: Set root_provider_id in the database https://review.openstack.org/619126 | 08:51 |
| *** e0ne has joined #openstack-placement | 09:03 | |
| *** tetsuro has quit IRC | 09:12 | |
| *** tssurya has joined #openstack-placement | 09:32 | |
| *** cdent has joined #openstack-placement | 09:37 | |
| *** gibi has quit IRC | 10:01 | |
| *** gibi has joined #openstack-placement | 10:08 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: Initial integration test using gabbits https://review.openstack.org/601614 | 10:41 |
| openstackgerrit | Chris Dent proposed openstack/placement master: Use gabbi-tempest job from gabbi-tempest plugin https://review.openstack.org/607508 | 10:41 |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Add placeload to integration test https://review.openstack.org/602484 | 10:41 |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Add a perfload job. https://review.openstack.org/602484 | 11:01 |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Add a perfload job. https://review.openstack.org/602484 | 11:04 |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Add a perfload job. https://review.openstack.org/602484 | 11:31 |
| cdent | getting so close to goodness | 11:32 |
| cdent | gibi: yesterday I abandoned some of the other work related using placement from nova and the fixtures. Some of my stuff, some of your stuff. Seemed like noise in the face of the thing that we seem to have settled on. | 11:44 |
| *** sean-k-mooney has joined #openstack-placement | 11:46 | |
| cdent | mornin' sean-k-mooney | 11:51 |
| sean-k-mooney | o/ | 11:52 |
| sean-k-mooney | for some reason my router disconnected at 8am so getting my vpn working and reconnecting to irc | 11:53 |
| sean-k-mooney | how are things in placement land :) | 11:54 |
| *** tssurya has quit IRC | 11:55 | |
| cdent | we merge a lot of code, so now there's a placement-manage and placement-status command | 12:01 |
| cdent | which means that the devstack change can merge soon | 12:01 |
| cdent | at which point it will be time to write a lot of documentation | 12:01 |
| sean-k-mooney | ah yes i saw there was a ML thread on placement-manage | 12:01 |
| sean-k-mooney | did not have a chance to read it | 12:01 |
| sean-k-mooney | cdent: and update all the other installers to use split out placement | 12:02 |
| cdent | at least tripleo and osa are making some pretty good headway on that already | 12:02 |
| sean-k-mooney | ya. kolla should proably change its containers to refect that placement is a seperate service form nova | 12:03 |
| gibi | cdent: thanks for the cleanup. | 12:07 |
| openstackgerrit | Jan Gutter proposed openstack/nova-specs master: Spec to implement os-vif generic datapath offloads https://review.openstack.org/607610 | 12:10 |
| cdent | sean-k-mooney: yes, true | 12:10 |
| cdent | I wish they would stop using apache as well :) | 12:10 |
| sean-k-mooney | ya | 12:11 |
| sean-k-mooney | they are open to it | 12:11 |
| sean-k-mooney | would the wsgi config need to change or can the binary just be swapped | 12:11 |
| dtantsur | cdent: hey, any plans for microversion negotiation in the placement OSC plugin? | 12:13 |
| sean-k-mooney | dtantsur: microvirsion in osc is a topic that goes beyond placement | 12:14 |
| cdent | dtantsur: I've not had much involvement with the plugin, but yeah, what sean-k-mooney said | 12:14 |
| dtantsur | sean-k-mooney: well, ironic does it :) | 12:14 |
| dtantsur | essentially, you need to pick the latest version that is known to both client and server | 12:14 |
| cdent | oh, that kind of negotiation | 12:14 |
| dtantsur | I can show anyone interested the relevant ironicclient bits | 12:14 |
| cdent | you don't mean choosing | 12:14 |
| dtantsur | well, it's good enough to not confuse people with Operation or argument is not supported with version 1.0 | 12:15 |
| sean-k-mooney | dtantsur: plugins can do it but osc does not allow the intree service to do the negociation | 12:15 |
| dtantsur | sean-k-mooney: who should I talk to about changing this? :) | 12:15 |
| dtantsur | this is really, REALLY bad for user experience | 12:15 |
| sean-k-mooney | that will likely be change after osc moves to openstacksdk | 12:15 |
| dtantsur | very likely indeed | 12:16 |
| sean-k-mooney | what i ment is we discussed it at the PTG and they decided that they would change it after that point if people wanted it still | 12:16 |
| cdent | dtantsur: the "who" is a good question. I don't know. I've made a point of avoiding involvement in client-side stuff because my preference is to just use requests :) | 12:16 |
| cdent | or curl even | 12:16 |
| dtantsur | you don't even support "latest", cmon folks :) | 12:16 |
| sean-k-mooney | dtantsur: again ironic should not be supporting this technically | 12:17 |
| cdent | dtantsur: who is "you"? | 12:17 |
| dtantsur | well some abstract "you" that claims that "latest" is not a supported version | 12:17 |
| dtantsur | which now I remember may be OSC itself. we had to fight with it in ironicclient | 12:17 |
| * dtantsur can show ironicclient as a collection of OSC-fighting hacks :D | 12:18 | |
| sean-k-mooney | its not that it cant be supported it was not allowed to be for intree services and was not ment to be in plugins for consitancy | 12:18 |
| dtantsur | well, practicality beats purity. we did it and we do not regret it at all. | 12:18 |
| cdent | dtantsur: I get the impression you're trying to do things with the plugin in a way that the plugin is not oriented towards | 12:18 |
| dtantsur | if by that you mean "try to provide decent user experience with microversions", then yes :) | 12:19 |
| sean-k-mooney | dtantsur: i am one of the people that wants microversion support in osc by the way | 12:19 |
| cdent | dtantsure: I mean what info are you trying to use/have that you aren't able to do now? | 12:19 |
| dtantsur | sean-k-mooney: you have a passionate ally here :) | 12:19 |
| sean-k-mooney | dtantsur: consitent user expericance is required for that. if ironic is the only service that negociates that is not a good user experiance | 12:20 |
| dtantsur | cdent: back to placement: I tried openstack resource provider list --resource CUSTOM_BAREMETAL=1 | 12:20 |
| dtantsur | I naively expected it to just work | 12:20 |
| dtantsur | sean-k-mooney: well, ironic is one of few services that do microversions at all. for other services new features just work. | 12:20 |
| sean-k-mooney | dtantsur: it may but what did you want it to do | 12:20 |
| sean-k-mooney | did you want it to list all RP with that Resouce class or all with that resouce class and free inventory | 12:21 |
| dtantsur | sean-k-mooney: I expected it NOT to return Operation or argument is not supported with version 1.0 | 12:21 |
| sean-k-mooney | dtantsur: ah ok you just wanted the mircroverios to work | 12:22 |
| dtantsur | overall I'm just playing with Placement, trying to understand how it works. I don't have any specific goal here. | 12:22 |
| dtantsur | yep | 12:22 |
| dtantsur | now, with the correct microversions explicitly specified, the result is still kind of surprising to me. but now I think it's because I don't understand Placement :) | 12:23 |
| sean-k-mooney | dtantsur: the current behavior is it will show RP that have a resouce of that type free | 12:23 |
| cdent | dtantsur: presumably you followed that with a '--os-placement-api-version 1.4' and it did work. I agree that not being able to say 'latest' there is pain. However, if your goal is to "understand how placement works" I strongly suggest you avoid using a client. Use something that lets you do http and see it: curl, httppie, gabbi | 12:23 |
| sean-k-mooney | dtantsur: cdent: has a spec for allowing different query to list all resource providers with a resource i think | 12:24 |
| dtantsur | I care more about free resources, but thanks | 12:25 |
| cdent | dtantsur: I'm not trying to be (at least not intentionally) difficult here, just trying to understand what you personally are trying to do, and what has helped me in similar cases | 12:25 |
| dtantsur | anyway, I think I got what I wanted, and I complained about microversions :) thanks all! | 12:26 |
| cdent | and for me, at least, clients _always_ get in the way | 12:26 |
| dtantsur | heh | 12:26 |
| dtantsur | I have to deal with field/QE ppl using OSC stuff. they don't want curl, and they don't want to do --os-placement-api-version :) | 12:26 |
| cdent | sure, but that's not the same as "I'm just playing with Placement, trying to understand how it works" | 12:27 |
| dtantsur | right, right. two parallel things: "I'm learning placement" and "Oh, btw, you don't do microversion negotiation, that's bad" | 12:27 |
| cdent | please raise a bug on osc-placement if you have the energy to do so. the main person who was doing that may have gone on to other pastures, so I'm not sure who/how that is going to go | 12:28 |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Add a perfload job. https://review.openstack.org/602484 | 12:28 |
| dtantsur | cdent: is it storyboard or lp? | 12:28 |
| cdent | I'm not willing for it to be me, because of the aforementioned meh about clients, but also because if I say yes to everything I'll die and nobody else will do anything | 12:29 |
| cdent | I believe https://launchpad.net/placement-osc-plugin | 12:29 |
| dtantsur | lol, will do | 12:30 |
| *** e0ne has quit IRC | 12:45 | |
| *** dtantsur is now known as dtantsur|brb | 12:47 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Add a perfload job. https://review.openstack.org/602484 | 12:51 |
| *** tssurya has joined #openstack-placement | 13:02 | |
| *** e0ne has joined #openstack-placement | 13:02 | |
| *** e0ne has quit IRC | 13:03 | |
| *** e0ne has joined #openstack-placement | 13:03 | |
| cdent | jaypipes: here's an entirely unrealistic bug you might find amusing: https://bugs.launchpad.net/nova/+bug/1804453 | 13:33 |
| openstack | Launchpad bug 1804453 in OpenStack Compute (nova) "maximum recursion possible while setting aggregates in placement" [Undecided,New] | 13:33 |
| jaypipes | cdent: hmm. | 13:34 |
| cdent | the log file helps to make it a big more clear about what's going on | 13:35 |
| *** mriedem has joined #openstack-placement | 13:49 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 14:01 |
| *** dtantsur|brb is now known as dtantsur | 14:07 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 14:24 |
| cdent | try try again | 14:24 |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 14:28 |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 15:10 |
| *** lyarwood has joined #openstack-placement | 15:21 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: Documentation cleanup: front page https://review.openstack.org/619273 | 15:36 |
| *** tssurya has quit IRC | 15:50 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 16:12 |
| openstackgerrit | Chris Dent proposed openstack/placement master: Remove keystoneauth1 opts from placement config group https://review.openstack.org/619299 | 16:19 |
| openstackgerrit | Chris Dent proposed openstack/placement master: Documentation cleanup: front page https://review.openstack.org/619273 | 16:33 |
| dtantsur | cdent: https://bugs.launchpad.net/placement-osc-plugin/+bug/1804489 | 16:45 |
| openstack | Launchpad bug 1804489 in placement-osc-plugin "Microversion negotiation support" [Undecided,New] | 16:45 |
| cdent | thanks | 16:46 |
| *** efried_back_mon has quit IRC | 16:51 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 17:00 |
| *** helenafm has quit IRC | 17:02 | |
| *** e0ne has quit IRC | 17:02 | |
| *** alex_xu has quit IRC | 17:17 | |
| *** alex_xu has joined #openstack-placement | 17:19 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 17:30 |
| *** dtantsur is now known as dtantsur|afk | 17:49 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 17:50 |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 17:58 |
| *** mriedem has quit IRC | 18:26 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: Add a perfload job. https://review.openstack.org/602484 | 18:29 |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 18:29 |
| *** mriedem has joined #openstack-placement | 18:37 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 18:52 |
| openstackgerrit | Chris Dent proposed openstack/placement master: WIP: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 19:08 |
| openstackgerrit | Chris Dent proposed openstack/placement master: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 19:30 |
| *** e0ne has joined #openstack-placement | 19:33 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: Remove keystoneauth1 opts from placement config group https://review.openstack.org/619299 | 19:36 |
| openstackgerrit | Chris Dent proposed openstack/placement master: Use a smaller base job for the perfload run https://review.openstack.org/619248 | 19:58 |
| cdent | jaypipes: I updated that bug https://bugs.launchpad.net/nova/+bug/1804453 with a bit more info. Turns out that it had an impact while trying to make a better placeload perf job: https://review.openstack.org/#/c/619248/ | 20:16 |
| openstack | Launchpad bug 1804453 in OpenStack Compute (nova) "maximum recursion possible while setting aggregates in placement" [Undecided,New] | 20:16 |
| cdent | even when it doesn't reach recursion it causes the server to get very congested and the database to get overloaded | 20:16 |
| cdent | and that's the end of my too long coding day. goodnight all. happy thanksgiving to those of you doing that. | 20:18 |
| *** cdent has quit IRC | 20:18 | |
| *** e0ne has quit IRC | 21:32 | |
| openstackgerrit | melanie witt proposed openstack/nova-specs master: Propose counting quota usage from placement and API database https://review.openstack.org/509042 | 21:44 |
| *** mriedem has quit IRC | 21:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!