Wednesday, 2019-08-07

*** betherly has quit IRC00:01
*** mmethot_ has joined #openstack-nova00:09
*** logan- has quit IRC00:12
*** mmethot has quit IRC00:13
*** logan- has joined #openstack-nova00:13
*** betherly has joined #openstack-nova00:17
*** ivve has quit IRC00:21
*** betherly has quit IRC00:22
*** markvoelker has joined #openstack-nova00:23
*** tetsuro has joined #openstack-nova00:32
*** hongbin has joined #openstack-nova00:34
*** tetsuro has quit IRC00:57
*** ricolin has joined #openstack-nova01:02
*** tetsuro has joined #openstack-nova01:02
*** mkrai has joined #openstack-nova01:12
*** igordc has quit IRC01:13
*** igordc has joined #openstack-nova01:15
*** ricolin_ has joined #openstack-nova01:17
*** betherly has joined #openstack-nova01:19
*** tbachman has quit IRC01:19
*** ricolin has quit IRC01:20
*** igordc has quit IRC01:23
*** betherly has quit IRC01:24
*** pcaruana has quit IRC01:26
*** tetsuro has quit IRC01:28
*** gyee has quit IRC01:33
*** markvoelker has quit IRC01:39
*** markvoelker has joined #openstack-nova01:40
*** betherly has joined #openstack-nova01:48
*** betherly has quit IRC01:53
openstackgerritLi Liu proposed openstack/nova master: WIP: Create and bind Cyborg ARQs.  https://review.opendev.org/63124402:09
openstackgerritEric Fried proposed openstack/nova master: trivial: Remove unused '_instance_to_allocations_dict' function  https://review.opendev.org/67489302:11
*** mkrai has quit IRC02:19
*** _mlavalle_1 has joined #openstack-nova02:24
*** mkrai has joined #openstack-nova02:25
*** mlavalle has quit IRC02:26
*** ricolin_ is now known as ricolin02:27
*** huaqiang has joined #openstack-nova02:27
*** betherly has joined #openstack-nova02:29
*** _mlavalle_2 has joined #openstack-nova02:32
*** betherly has quit IRC02:34
*** _mlavalle_1 has quit IRC02:35
*** BjoernT has quit IRC02:43
*** _mlavalle3 has joined #openstack-nova02:43
*** _mlavalle_2 has quit IRC02:44
*** igordc has joined #openstack-nova02:49
*** betherly has joined #openstack-nova02:50
*** BjoernT has joined #openstack-nova02:52
*** ircuser-1 has quit IRC02:52
*** huaqiang has quit IRC02:52
*** betherly has quit IRC02:55
fungii directly reenqueued 674821,1 back into the gate after that last failure, so no need to recheck it02:56
fungii'd promote it but the only thing ahead is a cinder change which is eta 28 minutes to merge anyway02:56
openstackgerritZhengMa proposed openstack/nova master: Check container_format on instance creation  https://review.opendev.org/67340702:56
*** tetsuro has joined #openstack-nova02:58
openstackgerritZhengMa proposed openstack/nova master: Check container_format on instance creation  https://review.opendev.org/67340702:59
*** _mlavalle3 has quit IRC02:59
*** _mlavalle3 has joined #openstack-nova03:00
fungithe failure was a TimeoutException in test_walk_versions on py36, for the record03:01
openstackgerritLi Liu proposed openstack/nova master: WIP: Create and bind Cyborg ARQs.  https://review.opendev.org/63124403:12
*** markvoelker has quit IRC03:14
*** huaqiang has joined #openstack-nova03:17
*** whoami-rajat has joined #openstack-nova03:19
*** tetsuro has quit IRC03:29
*** psachin has joined #openstack-nova03:36
openstackgerritYAMAMOTO Takashi proposed openstack/nova master: Revert "Revert resize: wait for events according to hybrid plug"  https://review.opendev.org/67502103:36
*** markvoelker has joined #openstack-nova03:38
*** betherly has joined #openstack-nova03:42
*** markvoelker has quit IRC03:43
*** pcaruana has joined #openstack-nova03:44
*** betherly has quit IRC03:47
*** hongbin has quit IRC03:49
openstackgerritLi Liu proposed openstack/nova master: WIP: Create and bind Cyborg ARQs.  https://review.opendev.org/63124403:53
*** tetsuro has joined #openstack-nova04:03
*** udesale has joined #openstack-nova04:04
*** ratailor has joined #openstack-nova04:08
openstackgerritLi Liu proposed openstack/nova master: WIP: Added Cyborg External Events  https://review.opendev.org/67502204:11
*** BjoernT has quit IRC04:12
openstackgerritBrin Zhang proposed openstack/python-novaclient master: Microversion 2.76: Add user_id/project_id to migration-list API  https://review.opendev.org/67502304:14
openstackgerritAlex Xu proposed openstack/nova-specs master: Update the virtual persistent memory spec based on discussion  https://review.opendev.org/67502404:15
openstackgerritJing Zhang proposed openstack/nova master: Remove the current host from cold migration target host list  https://review.opendev.org/67502504:17
*** bhagyashris has joined #openstack-nova04:18
*** jaosorior has quit IRC04:22
*** markvoelker has joined #openstack-nova04:24
*** markvoelker has quit IRC04:38
*** betherly has joined #openstack-nova05:05
*** abaindur has joined #openstack-nova05:07
*** icarusfactor has joined #openstack-nova05:08
*** factor has quit IRC05:08
*** betherly has quit IRC05:10
*** abaindur has quit IRC05:11
*** abaindur has joined #openstack-nova05:11
*** Luzi has joined #openstack-nova05:19
openstackgerritMerged openstack/nova master: Replace non-nova server fault message  https://review.opendev.org/67482105:20
*** betherly has joined #openstack-nova05:26
*** tetsuro has quit IRC05:30
*** igordc has quit IRC05:31
*** betherly has quit IRC05:31
openstackgerritZhengMa proposed openstack/nova master: Check container_format on instance creation  https://review.opendev.org/67340705:40
*** takashin has quit IRC05:47
*** ociuhandu has joined #openstack-nova05:49
*** ociuhandu has quit IRC05:53
*** dpawlik has joined #openstack-nova05:56
*** icarusfactor has quit IRC06:01
*** dpawlik has quit IRC06:01
*** factor has joined #openstack-nova06:02
*** markvoelker has joined #openstack-nova06:18
*** betherly has joined #openstack-nova06:18
*** markvoelker has quit IRC06:22
*** betherly has quit IRC06:23
*** tetsuro has joined #openstack-nova06:24
*** udesale has quit IRC06:24
*** udesale has joined #openstack-nova06:24
*** maciejjozefczyk has joined #openstack-nova06:25
*** udesale has quit IRC06:27
*** udesale has joined #openstack-nova06:27
*** jaosorior has joined #openstack-nova06:28
*** ccamacho has joined #openstack-nova06:45
*** dpawlik has joined #openstack-nova06:46
*** slaweq has joined #openstack-nova06:48
*** betherly has joined #openstack-nova06:50
*** betherly has quit IRC06:54
openstackgerritMerged openstack/nova master: trivial: Remove unused '_instance_to_allocations_dict' function  https://review.opendev.org/67489307:01
*** rcernin has quit IRC07:03
*** rpittau|afk is now known as rpittau07:07
*** tssurya has joined #openstack-nova07:08
*** huaqiang has quit IRC07:10
*** huaqiang has joined #openstack-nova07:10
*** udesale has quit IRC07:10
openstackgerritTakashi NATSUME proposed openstack/nova master: Fix non-existent method of Mock  https://review.opendev.org/67504107:14
*** takashin has joined #openstack-nova07:18
*** tesseract has joined #openstack-nova07:26
*** ralonsoh has joined #openstack-nova07:29
*** ivve has joined #openstack-nova07:34
*** xek has joined #openstack-nova07:34
*** lpetrut has joined #openstack-nova07:38
*** jistr is now known as jistr|afk07:42
*** helenafm has joined #openstack-nova07:48
*** spsurya has joined #openstack-nova07:50
*** ralonsoh has quit IRC07:54
brinzhangtakashi: Why abandon this patch https://review.opendev.org/#/c/450119?07:55
*** huaqiang has quit IRC07:56
brinzhangtakashi: I have a bp, it's codes need to add parameters to support for migration-list command, but there isnot the ``openstack server migration-list`` CLI07:57
*** udesale has joined #openstack-nova08:00
brinzhangtakashin: :)08:01
takashinbrinzhang: Hi08:02
brinzhang https://review.opendev.org/#/c/450119 this patch, looks good, but why are you abandon?08:02
takashinbrinzhang: There was a subsequnt BP. It was the function to abort cold migration.08:03
*** cgoncalves has quit IRC08:04
takashinbrinzhang: But in community discussin, it is not necessary to abort cold migration. So https://review.opendev.org/#/c/450119  was abondaned.08:04
*** cgoncalves has joined #openstack-nova08:04
brinzhangtakashin: Now in the openstackclient, there is not has the 'opesntack server migration-list', I was reviewed this patch, found you added it08:05
*** abaindur has quit IRC08:05
*** jangutter has joined #openstack-nova08:05
brinzhangtakashin: so I asked for that, if so, I will push a new patch for the migration-list08:06
takashinbrinzhang: You can take over it.08:06
*** threestrands has quit IRC08:06
brinzhangtakashin: cool, I will have a try, and remove the cold migration deal, and add my requirement.08:08
*** mauro|call has quit IRC08:13
*** mauro|call has joined #openstack-nova08:14
*** panda has quit IRC08:22
openstackgerritLee Yarwood proposed openstack/nova master: Use os-brick locking for volume attach and detach  https://review.opendev.org/61419008:23
*** panda has joined #openstack-nova08:23
*** cdent has joined #openstack-nova08:23
*** ralonsoh has joined #openstack-nova08:26
*** derekh has joined #openstack-nova08:29
*** takashin has left #openstack-nova08:30
*** tetsuro has quit IRC08:35
*** ralonsoh has quit IRC08:35
*** tetsuro has joined #openstack-nova08:36
*** huaqiang has joined #openstack-nova08:37
*** betherly has joined #openstack-nova08:45
*** ralonsoh has joined #openstack-nova08:47
*** markvoelker has joined #openstack-nova08:50
*** markvoelker has quit IRC08:55
*** mauro|call has quit IRC08:58
*** mauro|call has joined #openstack-nova09:00
*** tkajinam has quit IRC09:02
*** ociuhandu has joined #openstack-nova09:03
*** ociuhandu has quit IRC09:13
*** mauro|call is now known as takamatsu09:14
*** ociuhandu has joined #openstack-nova09:16
*** tetsuro has quit IRC09:19
openstackgerritChris Dent proposed openstack/nova master: Add a document that describes profiling eventlet services  https://review.opendev.org/67486509:22
*** ociuhandu has quit IRC09:25
bhagyashrisstephenfin: Hi , Good morning :)09:28
stephenfinbhagyashris: Evening o/09:29
*** ociuhandu has joined #openstack-nova09:29
bhagyashrisstephenfin: I saw you have uploaded patches including the upgrade changes, But I found few issues after applying all the patches. #link: http://paste.openstack.org/show/755606/09:29
bhagyashrisstephenfin: noted issues in above #link path09:30
*** mkrai has quit IRC09:33
bhagyashrisstephenfin: As per discussion and mentioned in spec we are not going to support both the PCPU and VCPU in single boot request so IMO we will need to add check some where to reject such request with proper error message to user.09:34
stephenfinbhagyashris: Yeah, I should have a WIP flag on that. There's a few corner cases I still need to cover including that one09:34
stephenfinYou're correct about the pinning though. I guess that's because we need to set the 'cpu_policy' field on the 'InstanceNUMATopology' object09:35
*** mkrai has joined #openstack-nova09:36
bhagyashrisstephenfin:yeah09:36
bhagyashrisstephenfin: And also I would like to know about are we going tp supporting new syntax of flavor extra specs like resources:PCPU and resources:VCPU in Train?09:37
bhagyashrisS/tp/to09:38
stephenfinThat would require changes to the InstanceNUMATopology object, but like we discussed previously, they don't need to be done yet09:39
bhagyashrisstephenfin: if yes then in that case we will also need to take care the first issue that I have mentioned above ... gust xml will give wrong result if user will create the instance using flavor extra specs resources:PCPU=109:40
stephenfinYup. Indeed. That needs to be done regardless.09:41
bhagyashrisstephenfin: OK09:42
bhagyashrisstephenfin: and one more thing I would like to ask about the translation of flavor extra specs and image meta data .. are you going to use my change https://review.opendev.org/#/c/672223/1 or you are going to continue on your change?09:42
stephenfinI think I'm going to keep going with https://review.opendev.org/#/c/671801/. I don't like the idea of rewriting the flavor behind the users back and this change doesn't require that. I might change my mind again, but I'll go with this for now09:44
bhagyashrisstephenfin: OK No problem.09:45
stephenfinbhagyashris: Thanks for testing this stuff, btw. Really good to have as many eyes on it as possible, as we're seeing :)09:45
bhagyashrisstephenfin: If you are OK?  then I will take up adding the functional test cases and will fix the existing test that you have added for https://review.opendev.org/#/c/674895/209:47
stephenfinbhagyashris: Yes, please. I kept the Author the same on that. I'm going to be working on the previous two patches and adding support for reporting the 'HW_CPU_HYPERTHREADING' trait today09:48
bhagyashrisstephenfin: OK then I will take up this task and will upload the patch soon09:48
*** jistr|afk is now known as jistr09:54
*** sapd1_x has joined #openstack-nova09:54
*** ricolin_ has joined #openstack-nova10:01
bhagyashrisstephenfin: Thank you :)10:01
*** bhagyashris has quit IRC10:03
*** ricolin has quit IRC10:03
*** spatel has joined #openstack-nova10:27
*** helenafm has quit IRC10:29
*** udesale has quit IRC10:31
*** spatel has quit IRC10:31
*** mvkr has joined #openstack-nova10:36
*** sean-k-mooney has quit IRC10:52
*** jangutter_ has joined #openstack-nova10:55
*** jangutter has quit IRC10:58
*** ociuhandu has quit IRC11:07
*** mkrai has quit IRC11:08
*** jangutter_ has quit IRC11:08
*** sean-k-mooney has joined #openstack-nova11:09
*** jangutter has joined #openstack-nova11:12
*** jangutter has quit IRC11:14
*** jangutter has joined #openstack-nova11:38
*** jangutter has quit IRC11:40
sean-k-mooneystephenfin: gibi alex_xu  https://review.opendev.org/#/c/621646/23 has been around for a while and it hasnt gotten any core review. i have review this several times at this point, would ye be able to take a look?11:40
*** boxiang has quit IRC11:42
*** boxiang has joined #openstack-nova11:42
*** jangutter has joined #openstack-nova11:43
*** jangutter has quit IRC11:45
gibisean-k-mooney: I opened it now on a tab but I cannot promies to reach there today11:48
sean-k-mooneygibi: no worries i was pingged to review it yesterday so i did it this morning11:48
*** markvoelker has joined #openstack-nova11:49
sean-k-mooneyi just wanted to flag it to some people since it was opened back in december and they have been rebasing it since11:49
*** mkrai has joined #openstack-nova11:58
*** mgariepy has joined #openstack-nova11:58
*** jangutter has joined #openstack-nova11:59
alex_xusean-k-mooney: I added it to my review list, will reach it when I do review work this week12:03
sean-k-mooneyalex_xu: cool thanks12:05
*** ratailor has quit IRC12:05
*** shilpasd has joined #openstack-nova12:06
stephenfinsean-k-mooney: sure12:07
*** ociuhandu has joined #openstack-nova12:09
sean-k-mooneyspeaking of reveiw where are we at with https://review.opendev.org/#/c/668656/7/specs/train/approved/use-pcpu-vcpu-in-one-instance.rst12:10
sean-k-mooneyoh wrong version12:11
*** markvoelker has quit IRC12:11
sean-k-mooneyall the discusstion seams to be on 7 but the patch is on version 1112:11
*** ociuhandu has quit IRC12:14
sean-k-mooneyi think the table for the setting in the image vs flavor is wrong by the way12:14
*** markvoelker has joined #openstack-nova12:16
*** ociuhandu has joined #openstack-nova12:19
*** ociuhandu has quit IRC12:22
*** ociuhandu has joined #openstack-nova12:23
openstackgerritAlex Xu proposed openstack/nova master: Add functional tests for resize with VGPU  https://review.opendev.org/67510612:26
openstackgerritStephen Finucane proposed openstack/nova master: scheduler: Flatten 'ResourceRequest.from_extra_specs', 'from_image_props'  https://review.opendev.org/67489412:27
openstackgerritStephen Finucane proposed openstack/nova master: Add support for translating CPU policy extra specs, image meta  https://review.opendev.org/67180112:27
openstackgerritStephen Finucane proposed openstack/nova master: Add reshaper for PCPU  https://review.opendev.org/67489512:27
*** kaisers has joined #openstack-nova12:28
*** jistr is now known as jistr|call12:37
*** ociuhandu has quit IRC12:38
*** ociuhandu has joined #openstack-nova12:39
alex_xusean-k-mooney: dansmith also has a point about whether we should let the user specific which vcpu is dedicated12:40
*** helenafm has joined #openstack-nova12:42
openstackgerritShilpa Devharakar proposed openstack/nova master: Nova object changes for forbidden aggregates request filter  https://review.opendev.org/67107212:43
openstackgerritShilpa Devharakar proposed openstack/nova master: DB API changes to get non-matching aggregates from metadata  https://review.opendev.org/67107412:45
*** zbr has quit IRC12:46
openstackgerritShilpa Devharakar proposed openstack/nova master: Add a new request filter to forbidden aggregates  https://review.opendev.org/67107512:46
*** ociuhandu has quit IRC12:46
sean-k-mooneyalex_xu: dansmith  the reason for being able to define which cpus are dedicated is the same for why we allow it for realtime cpus. we want to be able to say that core x is low performace and can be used for the os/manament interfaces vs core y is high perfomace and can be used for the application12:47
openstackgerritShilpa Devharakar proposed openstack/nova master: Enable request filter forbidden_aggregates  https://review.opendev.org/66795212:48
*** ociuhandu has joined #openstack-nova12:48
sean-k-mooneyalex_xu: dansmith: so yes we do need to allow the user to pass a mask in my opipion and i dont think nova should do any magic to try and calualte it.12:48
*** ricolin_ is now known as ricolin12:48
alex_xusean-k-mooney: hint by you, if I want the realtime cpu to be dedicated, then I still need a way to define that.12:48
sean-k-mooneyif the user has multiple numa nodes and/or a specific cpu toplogy then they shoudl understand that mapping12:49
sean-k-mooneywell that is simple. use the same mask for both12:49
alex_xuyea, otherwise, we will have a magic to mapping realtime vcpu to dedicated vcpu inside nova12:49
sean-k-mooneybut really i dont think we should allow realtime core to not be dedicated12:49
alex_xuah, that make sense. but it still will be a magic12:50
sean-k-mooneyif you are declaring a core as realtime it does not make sense for it to be multiplex to a shared host core that other vms can execute on as that will violate any realtiem logic12:51
sean-k-mooneywell we just and the bitmasks12:51
sean-k-mooneyactuuly xor the bits and check the result is 012:51
sean-k-mooneyactully no that is also wrong12:52
sean-k-mooneybut you get the idea we just need to validate each bit set in the realtime mask is set in the dedicated mask12:53
alex_xuyea12:53
alex_xui got it12:53
sean-k-mooneyto dansmith point however i do belive we discusssed the idea of not exposing a mask in the past and instead doing what he say and just having tally counts of how many of each type you want12:55
sean-k-mooneythen evenly distubting them over the numa nodes and ensureing within a numa node the shared cpus are first followed by the pinned ones12:55
sean-k-mooneythat fits the paradime where core 0 is often reserved for the os12:56
sean-k-mooneyand in a multi numa works the first core on each numa node is often reserved for the os12:56
sean-k-mooneybut that does seam like magic that would easily break12:57
alex_xugood to know13:00
* alex_xu has too much fun with functional test and vgpu and resize, can't open the eye13:01
sean-k-mooneyalex_xu: i saw the irc ping for the patch. what tempeted you to work on that13:01
sean-k-mooneyalso what time zone are you in. i assume its eitehr quite late or very early for you13:02
*** zbr has joined #openstack-nova13:03
alex_xusean-k-mooney: hah13:05
*** mriedem has joined #openstack-nova13:10
*** udesale has joined #openstack-nova13:19
openstackgerritShilpa Devharakar proposed openstack/nova master: Enable request filter isolate_aggregates  https://review.opendev.org/66795213:19
openstackgerritMatt Riedemann proposed openstack/python-novaclient master: docs: clarify nova migration-list --host option  https://review.opendev.org/67495013:23
openstackgerritMatt Riedemann proposed openstack/python-novaclient master: WIP: Add --migration-type and --source-compute to migration-list  https://review.opendev.org/67511713:23
mriedemlyarwood: as you're probably aware once these land https://review.opendev.org/#/q/I5e0a43ec59341c9ac62f89105ddf82c4a014df81 we'll be doing stable branch releases13:28
sean-k-mooneymriedem: those are the cve backports right13:29
mriedemyes13:29
*** mkrai has quit IRC13:29
mriedemdansmith: can you get this this stein backport? https://review.opendev.org/#/c/672723/13:29
*** mkrai has joined #openstack-nova13:29
mriedemmelwitt: can you get these stein backports for a series you're familiar with https://review.opendev.org/#/q/topic:bug/1764556+(status:open+OR+status:merged)+branch:stable/stein13:30
lyarwoodmriedem: ack yup13:30
dansmithmriedem: jes13:30
mriedemthanks13:31
*** mkrai has quit IRC13:35
*** BjoernT_ has joined #openstack-nova13:36
mnasercan i get a second pair of eyes13:42
mnaseri think there is a nova bug here that we uncovered in OSA13:42
mnaserhttps://logs.opendev.org/33/660333/10/check/openstack-ansible-deploy-aio_metal-ubuntu-bionic/97d8bc3/logs/host/nova-compute.service.journal-23-20-40.log.txt.gz#_Aug_06_22_51_2513:42
mnasernova-compute magically trying to access db?13:42
mnaser(and raising an exception because of that)13:43
*** jaosorior has quit IRC13:44
*** factor has quit IRC13:44
*** factor has joined #openstack-nova13:45
yonglihe<sean-k-mooney:  Hope you have time take a look the Orphan clean up patch, which i revised, but busy on another patch last week.  thanks. https://review.opendev.org/#/c/627765/13:45
sean-k-mooneyam i can make time. i saw melwitt had a question when i looked last week about shutdown_unkown. was that resolved13:47
mnaserim trying to do git blame on github and its timing out13:47
mnaserboo13:47
mriedemmnaser: hmm, yeah compute is calling compute.API.stop to stop the instance, which is hitting some code in the compute rpcapi client router to determine the minimum compute version across all cells which it shouldn't need to be doing, b/c you're in a cell13:47
mnaserthats what i figured based on the traceback but i havent done my share of git blame yet to see when this might be started13:48
mnaserhttps://github.com/openstack/nova/blame/master/nova/compute/manager.py crashes github13:48
*** ociuhandu has quit IRC13:48
mriedemmnaser: https://review.opendev.org/#/c/649197/13:48
sean-k-mooneymnaser: ya that and the libvirt dirver are two of the files better blamed on the terminal13:48
*** ociuhandu has joined #openstack-nova13:49
mriedemi want to say we should be hitting that when compute starts up though, or used to?13:49
mriedemmnaser: do you have the api db configured in nova.conf that you're computes are using?13:49
mnasermriedem: in this case i believe its an aio so yes it will be there13:50
mnaserin an actual deployment, osa doesnt do this13:50
mriedemeven if it's a single node you can run nova-compute with a different config file13:50
mriedemw/o the db creds13:50
mriedemthat's what devstack does13:50
dansmithmnaser: is the api db connection set?13:50
dansmithI think that's how it determines if it should use the all-cells or this-cell method13:50
mnaseryeah.. it would be a very 'CI' specific thing if we used 2 different configs, unless we always used 2 configs, but yeah13:51
mnaserif we have to roll out nova-compute.conf and nova.conf -- that's fine, we'll fix that13:51
sean-k-mooneymnaser: well always using two config makes sense in a way13:51
dansmithmnaser: was that aimed at me?13:51
sean-k-mooneyit is nice to be able to seperate compute only things for contoler only things13:52
mnaserdansmith: just in general, im wondering if we're doing things wrong (i.e. nova operates with the assumption that api db _should not_ be set in nova-compute)13:52
*** mkrai has joined #openstack-nova13:52
mriedemi've got to run my kid somewhere, bbiab13:52
dansmithmnaser: the assumption is that computes should not be talking to any db, and thus shouldn't have creds set at all13:52
dansmithmnaser: like, the creds shouldn't even need to be on those systems13:52
sean-k-mooneythe compute talkign directly to a db was a legacy feature correect13:53
sean-k-mooneyits inteded to go via the conductor13:53
mnaserright, in a 'production' deployment of OSA (where compute != control), the value is not set, but in our aio, the logic is compute is inside the controller group so it gets added13:53
mnaseri agree with you that it should not be set13:53
dansmithmnaser: okay, but it should have a different config which matches what a compute looks like non-aio13:53
mnaserhttps://github.com/openstack/openstack-ansible-os_nova/blob/master/templates/nova.conf.j2#L196-L21213:54
mnaserso we only set these if we're running against a host that is not part of nova-compute group, but in this case in aio, every host is so it gets included13:54
dansmiththis is obviously failing because of a somewhat fragile assumption about what having api db creds set means, but.. this is why it's good to use realistic configs even in an aio situation13:54
sean-k-mooneymnaser: do ye share config files between services?13:55
mnasersean-k-mooney: yes, but its templated to be different depending where it ends up getting deployed13:55
sean-k-mooneye.g. i know kolla ansible generate a different nova.conf per nova service even on the same host13:55
mnaseryes we do that too, but they all end up at /etc/nova/nova.conf13:56
sean-k-mooneymnaser: right that is different then kolla ansible13:56
sean-k-mooneyright within the contianer13:56
sean-k-mooneybut different paths in the host right13:56
mnasernah its always at /etc/nova/nova.conf here13:56
*** tesseract has quit IRC13:56
sean-k-mooneyso its alway one file on the host?13:56
mnaseryep13:57
mnaseri mean for me its fine, if from a nova perspective, the assumption that api_db shouldnt be defined, i can fix it inside osa, but if you think that assumption isn't right, i can also fix it inside nova13:57
*** tesseract has joined #openstack-nova13:57
sean-k-mooneyach well that explains why this does not work then13:57
*** whoami-rajat has quit IRC13:57
sean-k-mooneywell the assumtion is no db will be defiend in the compute agent's nova.conf13:57
sean-k-mooneyall db acess form the comptue agent is expected to go via the cell conductor13:58
mnaseryep, i agree with that13:58
dansmithmnaser: I think you should change OSA so that AIO and non-AIO are effectively the same by using separate nova confs per service13:59
alex_xusean-k-mooney: can we have hw_cpu_policy as mixed, then have hw:cpu_dedicated_mask?14:00
*** tesseract has quit IRC14:00
mnaserdansmith: makes sense, i'll have us roll out config to nova-$service.conf inside OSA and i think that should clear this up14:00
mnaserthanks for the help and heads up :>?14:00
dansmithmnaser: cool14:00
*** tbachman has joined #openstack-nova14:00
sean-k-mooneyfor what its worth kolla-ansible still use one template but generates multiple files and this is how they guard the db section14:01
sean-k-mooneyhttps://github.com/openstack/kolla-ansible/blob/master/ansible/roles/nova/templates/nova.conf.j2#L156-L16614:01
*** tesseract has joined #openstack-nova14:01
mnaseryeah im thinking that probably is in line of what we'll end u pdoing14:02
sean-k-mooneyit makes the condional simpler as most are based on the service name instaead of the group you are in or if a service is enabled14:03
mnaseryeah that does sound easier than a bunch of complicated jinja214:03
sean-k-mooneybut it requires the template to be evaulated and generated seperatly per service14:03
mnaseryeah that does increase the # of times we have to do it but eh14:03
*** ociuhandu has quit IRC14:03
mnaserif its the right thing(tm) then i rather do that than a few less tasks14:04
sean-k-mooneyya it synergies really nicely with kollas config overide mechanisum14:04
sean-k-mooneyperhaps it could be broken out at somepoint and shared between kolla-ansible, tripleo and osa14:05
yonglihesean-k-mooney:  i reply  melwitt but not sure is that what he concern. but thanks your  review.14:05
*** ociuhandu has joined #openstack-nova14:06
*** slaweq has quit IRC14:07
*** slaweq has joined #openstack-nova14:08
*** noonedeadpunk has joined #openstack-nova14:09
*** betherly has quit IRC14:10
*** ociuhandu has quit IRC14:11
*** mvkr has quit IRC14:12
*** dpawlik has quit IRC14:15
dansmithsean-k-mooney: you did not just say "it synergies" ...14:20
*** ociuhandu has joined #openstack-nova14:20
*** spatel has joined #openstack-nova14:21
sean-k-mooneyfor once im using it an example of its usage in a non marketing bs context14:21
dansmithheh14:22
sean-k-mooneythe config file per service per host design works realy well with the config override system kolla-ansible has. i think it was the config override desigin goals that acttuly motivated that approch initally so i guess its not suprising they work well together14:24
*** ociuhandu has quit IRC14:25
*** mkrai has quit IRC14:26
*** betherly has joined #openstack-nova14:29
*** Luzi has quit IRC14:30
noonedeadpunkAnd does that make sense to create nova-conductor.conf and place db credentials only there?14:36
*** jistr|call is now known as jistr14:36
dansmithnoonedeadpunk: devstack does a nova-cell1.conf for all the non-compute services that need to talk to the database, but yeah14:36
dansmithnova.conf has everything, and is used by the control services, api, etc. nova-cell1 is no api, just cell db creds, and nova-compute.conf is the bare minimum for compute14:37
stephenfinefried, gibi: What's the expected behavior here? http://paste.openstack.org/show/755623/14:37
stephenfinMaybe dansmith and sean-k-mooney too, idk ^14:37
noonedeadpunkdansmith: thanks14:37
sean-k-mooneystephenfin: for legacy reasons that shoudl be 214:38
*** tbachman has quit IRC14:38
sean-k-mooneye.g. we have alway allowed that14:38
sean-k-mooneybut i dont like that you can change it14:38
*** betherly has quit IRC14:39
sean-k-mooneyif we block it it could break existing flavors14:39
sean-k-mooneyso there would be an upgrade impact14:39
dansmithsean-k-mooney: why for legacy reasons?14:39
mriedemtechnically devstack has several conf files for nova because of the superconductor mode14:39
sean-k-mooneydansmith: we decided to allow operators to specify resouce request in the flavor extra specs14:40
sean-k-mooneythat is the sole legacy use of this14:40
dansmithsean-k-mooney: I'm not sure what you mean14:40
dansmithresource requests take precedence over the base flavor values14:40
dansmiththat's new behavior14:40
dansmithand critical for things like ironic14:40
sean-k-mooneydansmith: no its not14:40
mriedemyes...it is14:41
dansmithwhere we "display" a value to the user via the flavor, but ask for something different from placement based on the overrides14:41
mriedemas of pike14:41
sean-k-mooneyresouce request have always had precedece over the flavor value14:41
dansmithsean-k-mooney: not what?14:41
mriedemsean-k-mooney: you just said the same thing as dan14:41
stephenfinthere's a lot of drivers using the flavor.vcpus field14:41
dansmithsean-k-mooney: right, which is the new behavior14:41
sean-k-mooneydansmith: i was saying its not new it has been that way since we added resouces: in the flavor extra spec14:42
mriedemanything new in nova in <3 years is "new"14:42
mriedem:)14:42
sean-k-mooneyhehe :)14:42
dansmithand new since the introduction of that thing14:42
stephenfinincluding the libvirt driver14:42
mriedemstephenfin: yes, and it's going to be a big problem for that pcpu stuff14:42
stephenfinI mean, couldn't you request 'resources:VCPU=1' and 'flavor.vcpus=1000' ?14:43
sean-k-mooneymriedem: not really14:43
sean-k-mooneyunless you ask for less14:43
stephenfinand placement would be screwwwwed14:43
mriedemyou could add a property to the Flavor object that proxies the extra spec override when getting Flavor.vcpus14:43
sean-k-mooneystephenfin: we allow it but i agree its not ideal14:43
stephenfinespecially if placement is thing that enforces proxies now14:43
sean-k-mooneyone of the origial usecase was to allow yo to do flavor.vcpu=4 resouce:vcpu=0 for ironic but we ended up going a different route14:44
stephenfinsean-k-mooney: It feels slightly DDoS'y, at least if placement does quotas. I guess the thing that would prevent that is that flavor creation is admin-only but still14:44
*** ociuhandu has joined #openstack-nova14:45
sean-k-mooneye.g. so we could discribe the resouce an ironic flavor would have and then set the value to 0 and use the custome_resouce_class=1 internally14:45
*** altlogbot_3 has quit IRC14:46
efried        # If any of the three standard resources above was explicitly given in14:46
efried        # the extra_specs - in any group - we need to replace it, or delete it14:46
efried        # if it was given as zero.14:46
dansmithsean-k-mooney: what "different route" did we go?14:46
mriedemstephenfin: read this for context https://docs.openstack.org/ironic/stein/install/configure-nova-flavors.html14:46
efriedregardless of the reason, we definitely coded it so that extra specs would override base amounts.14:46
sean-k-mooneydansmith: we went with setting the flavor vlaues to 0 and just useing the custom resouce class14:46
dansmithsean-k-mooney: no we didn't14:46
dansmithsean-k-mooney: see that doc mriedem just posted14:47
mriedemi think sean-k-mooney means the flavor values via the extra spec,14:47
mriedemnot the actual top-level flavor properties14:47
*** altlogbot_2 has joined #openstack-nova14:47
dansmithI don't think he does :)14:47
*** betherly has joined #openstack-nova14:47
stephenfinsean-k-mooney: We don't touch the flavor properties. We just ignore them14:47
dansmithbecause he described what we *are* doing as the original plan that we abandoned14:47
*** mkrai has joined #openstack-nova14:47
mriedemi'm not sure you can even create a flavor with 0 vcpus or ram14:47
* stephenfin wonders what hell hole he's wandered into with this PCPU stuff...14:48
mriedemhttps://github.com/openstack/nova/blob/master/nova/api/openstack/compute/schemas/flavor_manage.py#L3314:48
sean-k-mooney no https://docs.openstack.org/ironic/stein/install/configure-nova-flavors.html is exactly the use case i though this was for but i though we ended up not doing it14:48
sean-k-mooneyok good to know that actully was done14:48
mriedemstephenfin: you didn't realize this 6 months ago?14:48
efriedWhat we need is to start over with a flavorsv214:48
efriedHow long could it possibly take to deprecate and remove cellsv1, I mean flavorsv1?14:49
dansmithefried: not helping14:49
stephenfinmriedem: I knew it was bad but not this bad /o\ I'd estimate a solid 80% of the code I've written has been tech debt-related14:49
mriedemthe base flavor values for vcpu/ram/disk are *everywhere* and in *everything* in nova, as far as resource accounting is concerned14:50
mriedemso changing the meaning of one of those has huge implications14:50
dansmithyup14:50
sean-k-mooneyya14:50
mriedemthe only one that was ever sort of special was disk for bfv14:50
mriedemwhich dansmith finally fixed in i think rocky14:50
mriedemby fixed i mean don't claim disk usage if you're bfv14:51
mriedemand don't report to the hypervisors API that we're consuming disk for a bfv instance14:51
mriedemec14:51
mriedem*etc14:51
sean-k-mooneyi personally am not sure why we still allow bfv with root_gb !=0 i get the upgrade impact of that but it would be nice to fully lock that down14:51
sean-k-mooneybut that is a differnet hill to die on14:52
sean-k-mooneyanyway for the vcpu stuff we never generate xml from the resouce:* values14:52
cdentso many hills. so many deaths.14:52
sean-k-mooneywe only use thos for placmenet14:52
mriedemunless i'm mistaken,14:53
mriedemuntil this pcpu hullabaloo,14:53
mriedemthe only time you'd ever override vcpu/ram/disk with extra specs is for ironic flavors14:53
mriedemfor which the driver doesn't look at those anymore anyway14:53
sean-k-mooneyno wiht the pcpu stuff we still dont plan to generate the xml form resouce: we plan to use the extra_specs14:53
mriedemresource: *is* extra specs14:53
stephenfinYeah, my life has been made significantly harder by having to care about 'resources:VCPU' and 'resources:PCPU'14:54
sean-k-mooneybut yes i think you are right the only time you would override them is ironic14:54
* mriedem remembers the ironic driver is still busted a bit on stable...14:54
stephenfinI'm seriously considering a big "this isn't doing anything unless you're ironic" warning and moving on14:54
stephenfinefried: ^14:54
mriedemhttps://review.opendev.org/#/c/637217/14:54
stephenfinThere's edge case after edge case14:55
sean-k-mooneymriedem: it is but i mean we have hw:cpu_policy to determin if we translate flavor.vcpu to resouce:vcpu or resouces:pcpu14:55
sean-k-mooneystephenfin: the proably is how do you know its an ironic flavor14:55
mriedemsean-k-mooney: if you're talking about things in the plan in the spec, i haven't read it so don't know14:55
efriedstephenfin: what specifically? resources:VCPU/PCPU in extra specs?14:56
sean-k-mooneymriedem: we knew trying to supprot resouces: for influncing the xml generation and assignment woudl be a paid soe we are not supporting it14:56
stephenfinmriedem: tl;dr: if you specify 'hw:cpu_policy=dedicated', you now get 'PCPU' inventory, but we need to care about 'resources:PCPU' in addition14:56
stephenfinefried: yup14:56
sean-k-mooney*pain14:56
sean-k-mooneyefried: i would expand it to all resocues generated form flavor.* but the problem is we have no way to know if its an ironic flavor or not14:57
mriedemumm, couldn't you just bork if you have hw:cpu_policy=dedicated and resources:VCPU?14:57
sean-k-mooneyso we cant14:57
*** factor has quit IRC14:57
sean-k-mooneywe would still request resocue:PCU=flavor.vcpus in addtion to resources:VCPU14:57
mriedemtechnically any ironic flavor is hw:cpu_policy=dedicated by default yeah?14:57
stephenfinmriedem: Yup, and that's what I'm doing (ditto for explicit PCPU too). The top issue is flavor.vcpus != resources:PCPU14:58
mriedemi mean you do'nt specify that, it's just implied b/c it's baremetal14:58
sean-k-mooneystephenfin: there is one case where its not equal by design14:58
mriedemyou know, mogan wouldn't have this problem....14:58
* mriedem ducks14:58
*** factor has joined #openstack-nova14:58
sean-k-mooneywhich hw:emulator_threads_policy=isolate14:58
stephenfinAlso, 'hw:emulator_threads_policy=isolate' means "consume an extra core and put emulator threads there", so there it's (flavor.vcpus + 1) != resources:PCPU14:59
stephenfinsean-k-mooney: jinx14:59
sean-k-mooneymriedem: for ironic all cpus will still be reported using VCPU i think since we dont plan to update the driver14:59
sean-k-mooneyat least not unless someone asks14:59
sean-k-mooneystephenfin: yes15:00
mriedemmnaser: dansmith: looking at this trace again https://logs.opendev.org/33/660333/10/check/openstack-ansible-deploy-aio_metal-ubuntu-bionic/97d8bc3/logs/host/nova-compute.service.journal-23-20-40.log.txt.gz#_Aug_06_22_51_25 and where we hit it https://github.com/openstack/nova/blob/master/nova/compute/rpcapi.py#L422 ,15:00
stephenfinSo this is all a massive PITA. TBH, I'd rather just say "don't use 'hw:cpu_policy' or any of those extra specs with 'resources:(P|V)CPU' and kick that can down the road15:00
stephenfinefried: Could we do that for now? ^15:00
sean-k-mooneywhich if we kill that option entrily and just always map the emulator thread to the shared set in the futre will make your life simpeler15:00
mriedemwhat do you think about adding a try/except for DBNotAllowed to just log something useful and re-raise, like "this service does not allow direct database access, you should run without the [api_database]/connection configured"15:00
stephenfinsean-k-mooney: I'd be happy to do that but I need a cycle to do it15:01
stephenfinBecause we break everything otherwise, assuming the operator doesn't have cpu_shared_set configured everywhere15:01
sean-k-mooneyya15:01
mriedemsean-k-mooney: reported....where?15:01
mriedemsean-k-mooney: for ironic the only inventory on the node should be the custom resource class reported by the driver's update_provider_tree method, which gets the resource class off the ironic node resource15:02
sean-k-mooneymriedem: in placement15:02
sean-k-mooneyoh right15:02
mriedemand the flavor should only consume, in placement, those custom resource class allocatoins15:02
sean-k-mooneyya i forgot15:02
mriedemthat's why the vcpu/ram/disk are all zero'ed out15:02
mriedemthe values for those in the flavor are just cosmetic15:02
sean-k-mooneyso no change for ironic and hw:cpu_policy will have no effect15:02
sean-k-mooneyit should not be used but its meaning less15:03
mriedemmnaser: if you have a bug for that osa issue i could tack on a nova logging patch to maybe make that more clear if someone hits that misconfig15:03
*** factor has quit IRC15:03
*** tbachman has joined #openstack-nova15:03
melwittmriedem: will do15:04
mnasermriedem: i think noonedeadpunk will likely look into fixing it so might be good if he created a bug to track it ^15:05
efriedstephenfin: I would be fine limiting the use of resources:(P|V)CPU as strictly as possible without breaking backward compat.15:05
efriedbut resources$S:* will be a different story.15:05
stephenfinefried: What would the reason for 'resources$S:PCPU' be?15:05
stephenfinI mean, without NUMA at least15:05
stephenfin*NUMA in placement15:06
dansmithmriedem: sure15:06
efriedstephenfin: Exactly NUMA in placement.15:07
efried"will be" future tense intended15:07
sean-k-mooneyefried: why not block  resources$S:* for pcus and vcpus15:07
efriedfor now, yes, we certainly can ^15:08
stephenfinWhy not forever?15:08
sean-k-mooneywe shoudl only support using the hw:numa_cpu.0 values15:08
sean-k-mooneyefried: e.g. we already have syntax for specify the cpu to numa mappings15:08
efriedstephenfin: pretty sure there will be affinity use cases where we'll need that level of control.15:08
stephenfinGiven that we have ways to request all these things with extra specs15:08
efriedincluding things like accelerators?15:09
efriedand VGPUs?15:09
efriedAnd bandwidth?15:09
efriedesp. that last one15:09
efriedbecause we need to be able to do affinity by subtree matching15:09
sean-k-mooneyif we also support mixed cpus in instance we will jsut combien the mask with the hw:numa_cpu.X=Y status15:09
sean-k-mooney*spec15:09
noonedeadpunkmnaser: I guess writing a patch might be quicker that filing in a bug...15:09
efried...when the requests come from different thingies15:09
huaqiangsean-k-mooney: morning15:10
stephenfinefried: 'hw:numa_affinity:$RESOURCE_CLASS:$POLICY' or something similar?15:10
stephenfine.g. 'hw:numa_affinity:VCPU:prefer'15:11
stephenfin*VGPU15:11
gibiI don't think we ever want to request bandwidth via flavor exta_spec15:11
efriedexactly gibi15:11
efriedbut we still need to be able to express (NUMA) affinity15:12
stephenfinWe're not. We're just saying what the affinity of whatever the bandwidth resource class is15:12
stephenfinAssuming we can know that ahead of time (idk)15:12
efriedby tying together the port bw request group suffix and the NUMA request group suffix15:12
efriedthere's going to be a coupling problem for sure15:13
efriedlike what we have for group_policy, only much worse.15:13
*** jaosorior has joined #openstack-nova15:14
gibiif nova start using the port uuid as the port suffix for the bandwidth request then a single use flavor can be created where the port group can be referenced for affinity. But this sounds pretty horrific15:17
gibis/port suffix/group suffix/15:17
efriedagree, I don't think we want the suffix to be a port UUID, do we?15:18
gibiefried: currently each port's resource request is understood as a separate group15:18
gibiefried: so a natural suffix for the group could be the port uuid15:18
sean-k-mooneystephenfin: why woudl we need hw:numa_affinity:VCPU:prefer15:19
*** ircuser-1 has joined #openstack-nova15:19
stephenfinsean-k-mooney: *VGPU15:19
sean-k-mooneyoh for graphics ya15:19
stephenfinJust some way to define NUMA affinity for individual resource class types15:20
sean-k-mooneycpus and memory are are affinites by adding hw:numa_node or one of the option that imples it so i was confused15:20
stephenfinIt doesn't allow you to define the same for individual instances, but that's slightly insane15:20
stephenfinYup, my mistake15:20
sean-k-mooneywell you would be surprised what we have approved in the past and never implmeneted.15:21
sean-k-mooneystephenfin: https://github.com/openstack/nova-specs/blob/master/specs/newton/approved/virt-image-props-boot-override.rst15:22
efriedgibi: I guess it's not really relevant until we can tie in whatever piece of hardware is associated with the NUMA node. RP for a NIC PF, kind of thing. Cause today you don't request network VF resources or whatever, right?15:23
gibiefried: right. we don't request VF resource from neutron via the resource request.15:24
gibiefried: you are also correct that when VF will be requested from neturon along with the bandwidth then expressing NUMA affinity will be an issue to be solved15:25
efriedgibi: Won't the majority of affinity use cases involve only one NUMA node?15:25
sean-k-mooneywe request vf internally in nova currently based on the vif type15:26
efriedsean-k-mooney: yes but VF not tracked by placement yet15:26
efriedis my point15:26
sean-k-mooneybut sriov vf will be owned and created by nova in placmenet in the future15:26
sean-k-mooneynot by neutron15:26
efriedright15:26
efriedoh15:26
efriedoh?15:26
efriedThat will make it tougher15:27
*** slaweq_ has joined #openstack-nova15:27
sean-k-mooneyyes all pci decice handeling is currently done by nova15:27
efriedto affine the bw resources with the vf resources15:27
sean-k-mooneyneutron does not have a whitelist of device that can be used15:27
efriedwhat if it was a VIF resource?15:27
efriedAnyway, if we had a one-NUMA-node use case, we could handle this fairly well. Even multiple NUMA nodes if they're symmetrical.15:28
sean-k-mooneywe discussed adding vif ressource that would be owned by the neutron backend too15:28
efriedA case where I want $this specific VF/bw associated with $big_numa and $that specific VF/bw associated with $small_numa that things get really hairy.15:28
*** huaqiang has quit IRC15:28
*** slaweq has quit IRC15:29
sean-k-mooneywell nomally its not symetric15:29
*** huaqiang has joined #openstack-nova15:29
sean-k-mooneyidealy it would be15:29
sean-k-mooneybut its quite common to only have nics attached to numa 015:29
efriedIt's not? I thought that was the point of saying hw:numa_nodes=215:29
sean-k-mooneyno15:29
efriedSorry, I mean if the CPU/mem resources in the NUMA nodes are symmetric15:29
efriedI don't mean "same number of VFs per NUMA node"15:30
sean-k-mooneyhw:numa_nodes=2 is refering to the guest numa toplogy15:30
*** factor has joined #openstack-nova15:30
sean-k-mooneyoh by default yest15:30
sean-k-mooney*yes15:30
efriedwhen we do that we're dividing evenly, yah?15:30
sean-k-mooneyyou can change that with hw:numa_mem.0=512 hw:numa_mem.1=153615:31
huaqiangsean-k-mooney: hi15:31
efriedsean-k-mooney: Yes, exactly, that's my point: those cases are rarer, right?15:31
sean-k-mooneyefried: yes if you jsut do hw:numa_nodes it devices cpus and ram evenly15:31
sean-k-mooneywe also check its evenly deivisable15:31
efriedsean-k-mooney: Do we currently support VF NUMA affinity?15:31
sean-k-mooneyefried: yes15:31
sean-k-mooneyfor 4+ years15:32
sean-k-mooneywe require it in the case of neutron sriov which is what im trying to chagne15:32
sean-k-mooneywith https://review.opendev.org/#/c/674072/15:32
sean-k-mooneygibi: ^ i would like your input on that by the way15:33
sean-k-mooneyhuaqiang: hi15:33
efriedsean-k-mooney: even in the case of lopsided NUMA nodes? You can say which one you want your VF associated with?15:33
sean-k-mooneyefried: no15:33
sean-k-mooneyyou cannot specify that at all15:33
efriedokay, then we're fine.15:34
efriedwe continue to say you can't specify that.15:34
sean-k-mooneywe guarentee it will come form one of the numa ndoes the vm is mapped too15:34
sean-k-mooneybut we give no garuntee it on a specific guest numa node15:34
*** _mlavalle3 has quit IRC15:34
gibiefried: if nova keeps handling the VFs then nova can do a VF-NUMA affinity. Bandwidth is already tied to VF today in case of SRIOV.15:34
*** mlavalle has joined #openstack-nova15:35
sean-k-mooneycurrently its always preseneted as if its attache to guest numa node 0 regradelss of what host numa node it came from15:35
efriedand we can give the requests for NUMA nodes and VFs generic known suffixes, and use same_subtree to tie them together.15:35
efriedperfect15:35
sean-k-mooneyi also want to chagne that but not in a way tha will impact placmenet or shcudleing in any way15:35
efriedYeah, let's not give any more granular control over that15:35
efriedcause then we would have to support it via placement :)15:35
efriedOkay, it's the middle of the night for huaqiang, let's wrap up and yield the floor...15:36
sean-k-mooneyefried: what i want to add i the use of a livirt feature that allows mupliple pci roots in the guest15:36
huaqiangsean-k-mooney: got the comments for the 'using PCPU VCPU in one instance', thanks15:36
sean-k-mooneythat will allwo us to attach the pci device to the virtual numa node that maps to the host numa node the deive is on15:36
huaqiangbut I checked the existing logic for 'hw:cpu_policy=dedicated' and 'hw_cpu_policy=shared', the final policy is 'dedicated'15:37
sean-k-mooneycan you show me that in code15:38
huaqiangsure15:38
sean-k-mooneyits this https://github.com/openstack/nova/blob/master/nova/virt/hardware.py#L1327-L133815:39
sean-k-mooneyok we do perfer the flavor over the image in that case15:39
huaqiangyou find it!15:40
sean-k-mooneyya i knew what code shoudl be enforcing it15:40
sean-k-mooneyits incositent with how we handel hw:cpu_threads_policy15:40
sean-k-mooneyhttps://github.com/openstack/nova/blob/master/nova/virt/hardware.py#L1370-L137715:40
sean-k-mooneyand numa topligies https://github.com/openstack/nova/blob/master/nova/virt/hardware.py#L129015:41
sean-k-mooneyhad thought we had made it an error for all  conflitc at this point15:41
sean-k-mooneywe can chagne that without at least 1 cycle to deprecate the old behavior so i guess we have to stick with the current behavior15:42
*** sapd1_x has quit IRC15:42
sean-k-mooneyhuaqiang: it would be good to add a warning here https://github.com/openstack/nova/blob/master/nova/virt/hardware.py#L1328 as i dont think that is what people expect15:42
huaqiangso you mean the flavor first behavior15:43
sean-k-mooneyyes15:43
sean-k-mooneywe started moving to raising exception if they conflict for most of these value a whiel ago15:43
sean-k-mooneyat least when adding new values15:43
sean-k-mooneyi guess we did not change older values proably due to upgrade concerns15:44
huaqiangso for the new 'mixed' policy, we need consider conflict as invalid15:45
*** helenafm has quit IRC15:45
huaqiangbut for 'dedicated' and 'shared' conflicts we follow the rule that flavor first15:45
sean-k-mooneywell we have a choice15:45
sean-k-mooneymy perference would be to consider all conflcit of hw:cpu_policy and hw_cpu_policy to be invalid and raise an exception15:46
sean-k-mooneybut since we cant do that htis cycle15:46
sean-k-mooneyits better to be consistent i think15:46
sean-k-mooneyit would be nice to issue a warning  if we detect a conflict however15:47
huaqiangI am good for both15:47
huaqiangok15:47
huaqianganother thing i don't know if you noticed that15:48
huaqianghw:cpu_realtime exists15:48
huaqiangbut there is no 'hw_cpu_realtime' in image properties15:48
huaqiangthis is different with the mask for dedicated cpus15:48
huaqiangAlex mentioned this difference, and cpu realtime mask has a more loose rules15:49
*** icarusfactor has joined #openstack-nova15:50
sean-k-mooneyyes we did not want to allow users to request realtime guest via the image15:50
sean-k-mooneybut we did allow the mask i belive15:50
huaqiangmask is allowed from both image and flavor15:50
sean-k-mooneyso you can only enabel realtime cpus via the flavor but both the flavor and image can set the mask15:51
huaqiangthat is the truth15:51
sean-k-mooneywe have allowed image to enable cpu pinning provided the the flavor has not forbiding it by setting hw:cpu_policy=shared15:52
*** factor has quit IRC15:52
sean-k-mooneyso i think its fine wto allow hw_cpu_policy=mixed give hw_cpu_policy=dedicated is alredy allowed15:52
sean-k-mooneysimilarly i think having the mask in the image and flavor is ok15:52
mkraiHi Nova team, I am working on implementing the resize API for baremetals and need some help in the design. I tried to put an initial design here https://etherpad.openstack.org/p/nova-resize-ironic-api15:53
sean-k-mooneybut im sure someone else has an opipion on that ^ its what the spec currently say however15:53
gibisean-k-mooney: left comment in https://review.opendev.org/#/c/674072 about the notification tests15:54
huaqiangyes, dansmith is questioned on the masks15:54
sean-k-mooneygibi: thanks i can never rememebr where that is15:54
*** icarusfactor has quit IRC15:55
mkraiefried, sean-k-mooney mriedem alex_xu could you please take a look?15:55
sean-k-mooneyim fixing other typos but was about to spend the next hour finding it15:55
efriedmkrai: Hi there. I recall mriedem had some ideas about that design. dansmith could probably provide some good input as well. To be clear, this is going to be a U release thing, you're just looking for initial design help?15:55
*** whoami-rajat has joined #openstack-nova15:55
sean-k-mooneymkrai: yes i have it open15:55
gibisean-k-mooney: we might improve the test to dump the file path when the assert fails15:55
*** mkrai_ has joined #openstack-nova15:55
mkraiefried, Yes!15:55
mkraisean-k-mooney, Thanks!15:56
huaqiangdansmith: hello15:56
*** ralonsoh has quit IRC15:57
*** ralonsoh has joined #openstack-nova15:57
sean-k-mooneyefried: mkrai could start on any ironic internal prep work that was need this cycle but the ironic driver changes would be next cycle i guess15:58
*** tesseract has quit IRC15:58
mriedemi love being continually distracted about asking to review specs and plans and code for things that do not even have a chance for train15:59
mkraisean-k-mooney, I have already posted spec for the ironic reconfigure API that will be needed by the resize API15:59
sean-k-mooneygibi: are you ok with https://review.opendev.org/#/c/674072 adressing bug: #179520 as a bugfix to the original spec15:59
openstackbug 179468 in python-cups (Ubuntu) "duplicate for #179520 applet.py crashed with TypeError in on_job_hold_activate()" [High,Fix released] https://launchpad.net/bugs/179468 - Assigned to Pascal De Vuyst (pascal-devuyst)15:59
mriedemmeanwhile my mega cross-cell resize series is not getting review but has been up for 2 cycles15:59
mkraisean-k-mooney, https://review.opendev.org/#/c/672252/16:00
mkraiI will add this link to the etherpad16:00
efriedmriedem: I'm going out on a limb and saying distraction is not the most significant factor in lack of reviews on cross-cell resize.16:00
efriedAt least for me that's true16:00
gibisean-k-mooney: yeah I'm OK with that being a bugfix. Maybe a link to the original spec / bp in the commit message would help16:01
sean-k-mooneymriedem: ^ that question is also to use as well are you ok with https://review.opendev.org/#/c/674072 adressing the feature gap in the orginal spec.16:01
efried"I'm askeird" is the major reason16:01
mriedemefried: i'm not saying it's a walk in the park16:01
sean-k-mooneygibi: sure ill add one16:01
efriedfwiw, I have been intending to wade in16:01
mriedemthat's why i had integration testing of cross-cell resize working before putting out a review guide16:01
efriedbut procrastinating successfully so far16:01
mriedemwhich is more than i can say for a lot of the features we merge...16:02
efriedyeah yeah16:02
mriedemi'm also a bit disappointed that cern hasn't been more involved16:02
mriedembut they seem...distracted lately as well16:02
*** tesseract has joined #openstack-nova16:02
mriedemsean-k-mooney: idk even know what https://review.opendev.org/#/c/674072 is about16:03
mriedembut it looks like it's not fun16:03
sean-k-mooneywe only allow you to specify the numa affinity policy for pci device in teh pci alias. the spec origially had flavor and image properites for it16:04
sean-k-mooneyneutron sriov dose not use pci alias16:04
mriedemwhich original spec16:04
mriedem?16:04
sean-k-mooneyso we entirely broke the original neuton port usecase.16:04
sean-k-mooneyone sec ill get it16:04
mriedem"lack of a feature is not a bug"16:05
sean-k-mooneyhttps://github.com/openstack/nova-specs/blob/master/specs/queens/implemented/share-pci-between-numa-nodes.rst16:05
sean-k-mooneyhttps://github.com/openstack/nova-specs/commit/74b47ac2f968b13b7d15987063d0f52ecba02a7e#diff-b5d95cd01ee920f26b55b45f2777b7aa is wher whe chaned it16:05
*** gyee has joined #openstack-nova16:05
mriedemok i'm pretty sure i didn't review any of this https://review.opendev.org/#/q/topic:bp/share-pci-between-numa-nodes+(status:open+OR+status:merged)16:06
mriedemlooks like it was jay, sylvain and gibi16:06
mriedemi do, however, always enjoy the late "oh btw this thing we landed 3 releases ago didn't do what it was supposed to but our QE team just got around to testing it"16:06
sean-k-mooneymriedem: that is true which is why i want non redhat core to basicaly decide if this is valid to adress as a bugfix as redhat folk have a conflict of interest16:06
sean-k-mooneywell we knew it was broken for a while but planned to fix it in U16:07
*** igordc has joined #openstack-nova16:07
sean-k-mooneywe have a custom request to support this that came in last week e.g. after spec freeze16:07
mriedemso you want to slide it in as a bug fix?16:08
sean-k-mooneyso this is why im asking is this something we can address upstream in Train16:08
mriedemyou know i'm not going to let you backport this right?16:08
*** maciejjozefczyk has quit IRC16:09
sean-k-mooneywell we do actully have a bug for it but i also added the wrong one16:09
sean-k-mooneymriedem: and yes i know16:09
mriedemthis smells like a blueprint to me, not a bug16:09
*** tssurya has quit IRC16:10
sean-k-mooneygiven we are pass m2 is this something that people would be willing to consider of train as a blueprint16:10
sean-k-mooneyi am happy to file one and ask for a blueprint exception16:10
mriedemchrist16:11
mriedemtell you what, if you can lean on dansmith and he can lean on me then we can talk16:11
mriedembut i just -1ed this as needing a blueprint16:11
openstackgerritMerged openstack/nova stable/stein: Replace non-nova server fault message  https://review.opendev.org/67482816:11
sean-k-mooneymriedem: i am also ok with punting it but that just forces me to consider downstrema only way to adres this.16:11
dansmithwut16:11
openstackgerritMerged openstack/nova stable/stein: Revert "[libvirt] Filter hypervisor_type by virt_type"  https://review.opendev.org/67272316:11
*** mkrai_ has quit IRC16:11
mriedemdansmith: "we forgot to do a thing in queens" https://review.opendev.org/#/c/674072/16:12
*** efried is now known as efried_rollin16:12
dansmithmriedem: if a config flag to disable certain processor flags for the biggest SEV in decades is a feature, then ffs, this surely is16:13
sean-k-mooneythis is the orignal bug report by the way https://bugs.launchpad.net/nova/+bug/179592016:13
openstackLaunchpad bug 1795920 in OpenStack Compute (nova) "SR-IOV shared PCI numa not working " [Undecided,Confirmed]16:13
melwittsince everybody's in a good mood this morning, think it's a good time to ask for reviews on multi-cell archive_deleted_rows https://review.opendev.org/50748616:13
*** shilpasd has quit IRC16:15
*** betherly has quit IRC16:18
mriedemRAGE16:20
stephenfinTerrible game, that16:20
mriedemoh right you wanted to backport that too... :)16:21
*** maciejjozefczyk has joined #openstack-nova16:22
*** ociuhandu has quit IRC16:22
*** cdent has quit IRC16:22
*** ociuhandu has joined #openstack-nova16:22
*** ricolin_ has joined #openstack-nova16:23
*** ivve has quit IRC16:23
huaqiangdansmith: about the evenly distributing PCPUs/VCPUs in instance, I have replied in v716:23
dansmithhuaqiang: yeah, I know, and so have others16:24
huaqiangI need your comment16:24
dansmithhuaqiang: I don't like that we're just letting the user define the orientation of every last electron instead of being abstract16:24
dansmithbut that's just my opinion16:24
dansmithit's clear that others feel it's very important to be able to arrange all the electrons16:24
sean-k-mooneydansmith: i dont know if you saw my conversation with alex_xu this morning16:26
*** ricolin has quit IRC16:26
huaqiangI just think if the cpu maks is provided, administrator can have more choice16:26
dansmithI *saw* it, but didn't pay close attention16:26
sean-k-mooneyyou are right that we prviosuly discussed an alternitive16:26
melwittlol arranging electrons? *slowly backs away*16:26
dansmithhuaqiang: yeah I know.. everyone wants full control of everything16:26
sean-k-mooneye.g. a tally count of how many pcpu and vcpu16:26
sean-k-mooneyand then a conention that per numa node we place the numa local vcpu first followed by the numa local pcus16:27
dansmithsean-k-mooney: seems like arranging the dedicated cpus first and the non-dedicates ones after would still give you all the control you need, since you can already control numa nodes, memory assignment, etc16:27
sean-k-mooneythis make the core 0 is for the os eveythring else if for the app case just work16:27
dansmithbut I know I should just forget about this ever trying to be anything other than manual bitmasks for everything16:27
sean-k-mooneydansmith: you want the non dedicate ones first16:27
sean-k-mooneythere are some things in the kernel you cant move off core 016:28
*** betherly has joined #openstack-nova16:28
sean-k-mooneyat least not eaisly16:28
melwittmriedem: I think I asked about backporting --before, not this :) but yeah, both are things that would help with operability for queens *emoji with eyelashes*16:28
dansmithwhatever16:28
sean-k-mooneydansmith: well it is a valid alternitve im not totally against16:28
*** ociuhandu has quit IRC16:28
*** markvoelker has quit IRC16:28
sean-k-mooneybut if we were to go that route i think we would need time to think about it and spec out16:28
sean-k-mooneymeaning this defintly would be moved to U16:29
sean-k-mooneythe mask is consitent with realtiem cpus btu i get the concern over complexity16:29
*** ricolin_ is now known as ricolin16:29
*** psachin has quit IRC16:30
openstackgerritMatt Riedemann proposed openstack/nova master: Add useful error log when _determine_version_cap raises DBNotAllowed  https://review.opendev.org/67514816:31
mriedemmnaser: noonedeadpunk: ^ did an OSA nova bug get filed?16:31
mriedemif so i can tack it on there16:31
huaqiangdansmith: got. thanks16:32
mriedemmelwitt: oh right yeah16:32
mriedemis queens the major RHOSP release all of the customers are on now because that seems to be where all of the recent customer stuff is coming from16:32
*** betherly has quit IRC16:32
sean-k-mooneymriedem: yes16:33
mriedemnon-LTS RHOSP i should say16:33
melwittthat is correct, heh16:33
sean-k-mooneyqueens is osp 13. osp 10(netwon) is EOL in december so everyone is moving to 13.16:33
melwittwe're maybe probably gonna backport both of these I think. because RDO cloud and/or our CI runs on queens and not having these tools has been pain for those running the clusters16:34
melwitt*backport downstream, at least16:34
*** factor has joined #openstack-nova16:35
mnasermriedem: i think noonedeadpunk did not do one so perhaps OSA instead can tack on16:35
mriedemsuper. i'm now tasked internally with helping to deploy and operate a distribution based on mitaka, so things are really moving forward in the world.16:35
mnaserlololol16:35
melwittoooof, damn16:36
mriedemand the only red hat recruiter emails i get are for consulting gigs :/ \o/16:36
mriedemneed to fire up the old beat farmer dream16:37
mriedemat least that way i can get in on some sweet sweet government subsidies16:37
melwittlol xD16:37
*** markvoelker has joined #openstack-nova16:38
* mriedem starts stable/stein release procedures16:39
*** tesseract has quit IRC16:39
*** ricolin has quit IRC16:40
*** lpetrut has quit IRC16:41
*** markvoelker has quit IRC16:42
*** markvoelker has joined #openstack-nova16:44
*** ociuhandu has joined #openstack-nova16:44
mriedemstein release wip proposed https://review.opendev.org/#/c/675152/16:46
mriedemartom: sean-k-mooney: ^ has the ovs hybrid plug revert resize thing in it too16:47
noonedeadpunkmriedem: not yet:(16:47
noonedeadpunkI'll fill in then right now16:47
*** mkrai has quit IRC16:48
mriedemmelwitt: thanks for hitting those backports16:48
melwittdeps on route to the gate16:48
melwittnp16:48
mriedemyou can bug me about the all cells archive thing, i've been meaning to get to it, just sidetracked16:48
melwitt*en route (how embarrassing)16:48
melwittyesss, thanks16:49
sean-k-mooneyoh ok cool for tansparancey there is this revert fo that https://review.opendev.org/#/c/675021/ but artom and i both belive the issue that they were trying to fix is unrelated16:49
*** ociuhandu has quit IRC16:50
artommriedem, it's already downstream, but thanks :)16:50
artomsean-k-mooney, yeah, he's barking up the wrong tree with that as far as I'm concerned16:50
*** igordc has quit IRC16:51
sean-k-mooneythe timelines dont match up as you said16:51
sean-k-mooneywe could ad special case support for midonet trivially but there is somthing else going on there16:52
artomTheir errors are the mm executable returning an error, nothing to do with events16:52
sean-k-mooneyah ok i didnt look at the compute log until now16:53
noonedeadpunkmriedem: ok, seems that you've already linked patch to some bug16:53
artomsean-k-mooney, he gave me a good night scare though, I got the gerrit review email right before bed16:55
*** rpittau is now known as rpittau|afk16:56
sean-k-mooneyya im sure the heart rate spike didnt help you drifting off to sleep16:56
mriedemnoonedeadpunk: yeah and old bug16:57
mriedem*an16:57
mriedemsean-k-mooney: were you going to update this? https://review.opendev.org/#/c/647735/16:57
*** maciejjozefczyk has quit IRC16:57
mriedemi'd like to make sure we don't lose track of documenting the features we add16:58
sean-k-mooneyyes16:58
sean-k-mooneyi also have to document the vpmu image property16:58
mriedemoh i know16:58
mriedemwas just going through https://review.opendev.org/#/c/671338/16:58
*** derekh has quit IRC16:59
sean-k-mooneyah right but yes ill update both. im almost finish fixing some other typos in an os-vif patch so i might as well update the glance patchs before i start anything else16:59
sean-k-mooneywant me to add you to them17:00
mriedemumm17:00
mriedemyou can if you want17:01
*** ociuhandu has joined #openstack-nova17:01
sean-k-mooneyi can just update the nova patches with the link17:01
*** betherly has joined #openstack-nova17:03
noonedeadpunkmriedem: https://bugs.launchpad.net/nova/+bug/183936017:07
openstackLaunchpad bug 1839360 in OpenStack Compute (nova) "nova-compute fails with DBNotAllowed error" [Undecided,New]17:07
*** betherly has quit IRC17:08
*** igordc has joined #openstack-nova17:10
*** artom has quit IRC17:13
openstackgerritMatt Riedemann proposed openstack/nova master: Add useful error log when _determine_version_cap raises DBNotAllowed  https://review.opendev.org/67514817:13
mriedemnoonedeadpunk: done thanks17:13
*** udesale has quit IRC17:14
* mriedem gets lunch17:15
openstackgerritStephen Finucane proposed openstack/nova master: Add support for translating CPU policy extra specs, image meta  https://review.opendev.org/67180117:16
openstackgerritStephen Finucane proposed openstack/nova master: Add reshaper for PCPU  https://review.opendev.org/67489517:16
* stephenfin goes home17:16
*** brinzhang_ has joined #openstack-nova17:16
stephenfinsean-k-mooney: If you've a chance tomorrow, I think https://review.opendev.org/671801 is mostly in good shape now. That was a rough one17:16
sean-k-mooneycool how is the series as a whole17:16
stephenfinI need to test it on real hardware but I think it's just the reshape that still needs some work17:17
sean-k-mooneyi need to resinstall my sriov systems i think tomorowo but is your stuff ready to test after i do that17:17
stephenfinFor a new deployment, yes. Upgrades are still WIP due to reshape17:17
*** altlogbot_2 has quit IRC17:17
sean-k-mooneywell its not jsut upgrades right17:18
stephenfinHow so?17:18
sean-k-mooneyits going form not defiing the config to defining them17:18
stephenfinWhat is?17:18
sean-k-mooneythe reshape17:18
stephenfin(my head is pretty much fried at this point)17:19
stephenfinAh. Um, ask me tomorrow? :)17:19
*** brinzhang has quit IRC17:19
sean-k-mooneyok17:19
sean-k-mooneygo clear your head. im going to start dinner. ill have the os-vif patch done soone by the way just on the release note. thanks for taking a look yesterday17:20
stephenfinno problem17:22
* stephenfin really goes home17:23
*** altlogbot_1 has joined #openstack-nova17:23
*** betherly has joined #openstack-nova17:24
*** betherly has quit IRC17:29
*** mgariepy has quit IRC17:29
*** ociuhandu has quit IRC17:29
*** mgariepy has joined #openstack-nova17:30
*** altlogbot_1 has quit IRC17:31
*** altlogbot_2 has joined #openstack-nova17:35
*** ociuhandu has joined #openstack-nova17:39
*** ociuhandu has quit IRC17:43
*** betherly has joined #openstack-nova17:44
*** betherly has quit IRC17:49
*** dpawlik has joined #openstack-nova17:50
*** spatel has quit IRC17:57
openstackgerritsean mooney proposed openstack/os-vif master: only disable mac ageing for ovs hybrid plug  https://review.opendev.org/67283418:03
*** betherly has joined #openstack-nova18:04
*** betherly has quit IRC18:09
melwitthm, looks like hyperv ci is failing with 'sudo: /usr/sbin/mke2fs: command not found' all the time as of late18:16
melwitthttp://cloudbase-ci.com/nova/674828/1/tempest/subunit.html.gz18:17
*** artom has joined #openstack-nova18:17
mriedemmelwitt: https://github.com/openstack/tempest/commit/887f99a288c1d2bef45356c916b317d37828e6ac18:20
melwittahh18:20
mriedemlooks like hyperv ci needs to blacklist that test18:20
mriedemlikely need to email their ci group18:20
dansmithdo they not use a cirros gues?18:20
dansmith*guest18:20
mriedemno idea18:21
dansmithI would think they did, and that this test should be fine18:21
sean-k-mooneythey might be using a windows nano server image?18:21
dansmiththere's also a failure to resolve thing in that error log18:21
melwittyeah, I didn't understand that part bc I don't see that "hostname" as part of the command line18:22
*** spsurya has quit IRC18:22
dansmithyeah18:22
sean-k-mooneylooking at the logs melwitt linked they apper to be useing ubuntu as the ssh user name18:22
dansmithmaybe a bug in some fake sudo wrapper they use or something18:23
sean-k-mooneyso i assume its a ubuntu guest image18:23
mriedemit's ubuntu 1618:23
mriedemname             | ubuntu-1618:23
mriedemopenstack --os-cloud=devstack-admin --os-region-name=RegionOne image create ubuntu-16 --public --container-format=bare --disk-format vhdx18:23
mriedemwhatever this is ubuntu-16.04-minimal-cloudimg-amd64-sync.vhdx18:23
dansmithwonder why that and not cirros18:24
mriedemnot sure18:24
dansmithso it's possible that their image is missing mke2fs I guess18:24
mriedemthey have their own cloud-init thing18:24
mriedemfor hyperv18:24
sean-k-mooneydansmith: perhaps there is a dirver issue with hyperv18:24
dansmithfor windows guests, right?18:24
sean-k-mooneythat said cirros18:25
sean-k-mooneyuses teh ubunut kernel18:25
sean-k-mooneyso proably not18:25
mriedemanywho, here is the email info https://wiki.openstack.org/wiki/ThirdPartySystems/Cloudbase_Compute_Hyper-V_CI18:25
*** betherly has joined #openstack-nova18:25
sean-k-mooneymriedem: cloudbase-init18:25
sean-k-mooneyi think that is only for windows guests18:25
dansmithsean-k-mooney: maybe, but there's not much reason for that (i.e. perf for tests) unless it's physically required for disk access or something18:25
sean-k-mooneyhyperv uses parvitualised devices18:26
dansmithI guess maybe they only expose PV devices and need a real kernel18:26
dansmithokay18:26
sean-k-mooneyso if you do lspci by default you will have no devices18:26
sean-k-mooneyits weird18:26
dansmithokay, well, that's why probably18:26
dansmithbut that opens them up to this kind of thing18:26
dansmithwould think it'd be worth just working on getting that stuff into cirros or building their own cirros,18:27
dansmithbut alas18:27
sean-k-mooneybut what i dont get is cirros is built with the ubuntu kernel18:27
sean-k-mooneyso i dont think it buys them much18:27
sean-k-mooneymaybe they could use cirros 4.x18:27
dansmithit doesn't have a full complement of modules thjough, AFAIK18:27
dansmithanyway, this is clearly a thing the hyperv people can fix if they want :)18:28
sean-k-mooneyya that is true. i have hyperv on my windows install but i have never tried cirros with it18:28
sean-k-mooneywell in this case tehy just need to install mke2fs18:28
sean-k-mooneyin the guest image18:29
*** betherly has quit IRC18:30
melwittI'll try to email their list https://wiki.openstack.org/wiki/ThirdPartySystems/Hyper-V_CI18:30
melwittalso looks like it's ok on the master branch. I'm seeing that failure on older branches18:38
*** betherly has joined #openstack-nova18:45
*** mvkr has joined #openstack-nova18:47
*** BjoernT has joined #openstack-nova18:48
melwittemail bounced :(18:49
*** betherly has quit IRC18:50
*** BjoernT_ has quit IRC18:50
melwittI forwarded to ociuhandu in case he is involved with the CI18:58
*** tbachman has quit IRC18:58
*** tbachman has joined #openstack-nova19:00
*** abaindur has joined #openstack-nova19:00
*** mgariepy has quit IRC19:13
*** mgariepy has joined #openstack-nova19:13
*** mgariepy has quit IRC19:14
*** mgariepy has joined #openstack-nova19:15
*** BjoernT_ has joined #openstack-nova19:15
*** mgariepy has quit IRC19:16
*** BjoernT has quit IRC19:17
*** mgariepy has joined #openstack-nova19:18
*** mgariepy has quit IRC19:19
*** mgariepy has joined #openstack-nova19:20
*** rouk has joined #openstack-nova19:22
*** mgariepy has quit IRC19:33
*** dave-mccowan has joined #openstack-nova19:33
*** mgariepy has joined #openstack-nova19:34
*** lpetrut has joined #openstack-nova19:35
*** vesper11 has quit IRC19:37
*** tbachman has quit IRC19:45
*** betherly has joined #openstack-nova19:46
*** betherly has quit IRC19:51
*** dpawlik has quit IRC19:53
*** betherly has joined #openstack-nova19:55
*** efried_rollin is now known as efried19:59
*** betherly has quit IRC20:00
*** markvoelker has quit IRC20:01
*** lpetrut has quit IRC20:01
*** factor has quit IRC20:03
*** markvoelker has joined #openstack-nova20:03
*** factor has joined #openstack-nova20:03
*** factor has quit IRC20:05
*** tbachman has joined #openstack-nova20:05
*** factor has joined #openstack-nova20:05
*** ivve has joined #openstack-nova20:15
*** betherly has joined #openstack-nova20:16
*** whoami-rajat has quit IRC20:17
*** betherly has quit IRC20:21
openstackgerritSundar Nadathur proposed openstack/nova master: ksa auth conf and client for Cyborg access  https://review.opendev.org/63124220:26
openstackgerritSundar Nadathur proposed openstack/nova master: Refactor some methods for reuse by Cyborg-related code.  https://review.opendev.org/67373420:26
openstackgerritSundar Nadathur proposed openstack/nova master: WIP: Add Cyborg device profile groups to request spec.  https://review.opendev.org/63124320:26
openstackgerritSundar Nadathur proposed openstack/nova master: WIP: Create and bind Cyborg ARQs.  https://review.opendev.org/63124420:26
openstackgerritSundar Nadathur proposed openstack/nova master: WIP: Get resolved Cyborg ARQs and add PCI BDFs to VM's domain XML.  https://review.opendev.org/63124520:26
openstackgerritSundar Nadathur proposed openstack/nova master: Delete ARQs for an instance when the instance is deleted.  https://review.opendev.org/67373520:26
*** factor has quit IRC20:32
*** factor has joined #openstack-nova20:32
*** factor has quit IRC20:35
*** factor has joined #openstack-nova20:36
openstackgerritMerged openstack/nova stable/stein: Add functional recreate test for bug 1764556  https://review.opendev.org/67381220:39
openstackbug 1764556 in OpenStack Compute (nova) stein ""nova list" fails with exception.ServiceNotFound if service is deleted and has no UUID" [Medium,In progress] https://launchpad.net/bugs/1764556 - Assigned to Matt Riedemann (mriedem)20:39
openstackgerritMerged openstack/nova stable/stein: Add functional regression test for bug 1778305  https://review.opendev.org/67381320:39
openstackbug 1778305 in OpenStack Compute (nova) stein "Nova may erronously look up service version of a deleted service, when hostname have been reused" [Medium,In progress] https://launchpad.net/bugs/1778305 - Assigned to Matt Riedemann (mriedem)20:39
openstackgerritMerged openstack/nova master: tox: Keeping going with docs  https://review.opendev.org/67033220:39
*** ralonsoh has quit IRC20:58
melwittmriedem: merge conflict :( https://review.opendev.org/67381420:59
openstackgerritGhanshyam Mann proposed openstack/nova master: Add new default roles in os-services API policies  https://review.opendev.org/64848021:03
openstackgerritMatt Riedemann proposed openstack/nova stable/stein: Don't generate service UUID for deleted services  https://review.opendev.org/67381421:04
mriedemwith what? rebase was simple21:04
*** slaweq_ has quit IRC21:06
*** BjoernT_ has quit IRC21:08
melwittI dunno, I didn't pull it down21:10
*** slaweq_ has joined #openstack-nova21:11
mriedemmelwitt: https://review.opendev.org/#/c/507486/6521:14
mriedemalso had some comments in the nova-next patch on top, which dansmith might have thoughts on21:15
melwittthanks!21:16
*** slaweq_ has quit IRC21:17
mriedemmelwitt: there are a few things in there i commented on which are separate issues, which i can push some patches to address so this doesn't get more cluttered up21:20
melwittok, that sounds great if you're up for it. else I can break things out21:21
melwittI see there's a lot to go through, I didn't look in detail yet21:22
*** xek has quit IRC21:22
*** markvoelker has quit IRC21:23
mriedemnothing really big or complicated,21:26
mriedemi think the functional test could be simplified,21:26
mriedemand if separated from ServersTestBase it could be less dependent on some of the weirdness in that class21:26
*** vesper11 has joined #openstack-nova21:26
* melwitt looks wistfully at ServersTestBase21:27
*** macz has joined #openstack-nova21:28
*** tjgresha has quit IRC21:28
openstackgerritMatt Riedemann proposed openstack/nova master: Don't mention CONF.api_database.connection in user-facing messages/docs  https://review.opendev.org/67520521:31
*** tjgresha has joined #openstack-nova21:31
*** tjgresha has quit IRC21:32
*** factor has quit IRC21:35
*** factor has joined #openstack-nova21:35
*** markvoelker has joined #openstack-nova21:38
*** tbachman has quit IRC21:42
openstackgerritMatt Riedemann proposed openstack/nova master: Clean up docstrings for archive_deleted_rows  https://review.opendev.org/67520921:43
*** mchlumsky has quit IRC21:44
openstackgerritMatt Riedemann proposed openstack/nova master: Re-use DB engine connection during archive_deleted_rows  https://review.opendev.org/67521821:49
mriedemmelwitt: ok i think that's all the separate nit stuff21:49
melwittthanx21:49
*** markvoelker has quit IRC21:50
*** tbachman has joined #openstack-nova22:08
*** slaweq_ has joined #openstack-nova22:11
*** mlavalle has quit IRC22:13
*** rcernin has joined #openstack-nova22:15
*** slaweq_ has quit IRC22:16
*** markvoelker has joined #openstack-nova22:19
mriedemduck22:22
mriedemand22:22
mriedemcover22:22
openstackgerritMatt Riedemann proposed openstack/nova master: Add InstanceAction/Event create() method  https://review.opendev.org/61403622:23
openstackgerritMatt Riedemann proposed openstack/nova master: Add Instance.hidden field  https://review.opendev.org/63112322:23
openstackgerritMatt Riedemann proposed openstack/nova master: Add TargetDBSetupTask  https://review.opendev.org/62789222:23
openstackgerritMatt Riedemann proposed openstack/nova master: Add CrossCellMigrationTask  https://review.opendev.org/63158122:23
openstackgerritMatt Riedemann proposed openstack/nova master: Execute TargetDBSetupTask  https://review.opendev.org/63385322:23
openstackgerritMatt Riedemann proposed openstack/nova master: Add prep_snapshot_based_resize_at_dest compute method  https://review.opendev.org/63329322:23
openstackgerritMatt Riedemann proposed openstack/nova master: Add PrepResizeAtDestTask  https://review.opendev.org/62789022:23
openstackgerritMatt Riedemann proposed openstack/nova master: Add prep_snapshot_based_resize_at_source compute method  https://review.opendev.org/63483222:24
openstackgerritMatt Riedemann proposed openstack/nova master: Add nova.compute.utils.delete_image  https://review.opendev.org/63760522:24
openstackgerritMatt Riedemann proposed openstack/nova master: Add PrepResizeAtSourceTask  https://review.opendev.org/62789122:24
openstackgerritMatt Riedemann proposed openstack/nova master: Refactor ComputeManager.remove_volume_connection  https://review.opendev.org/64218322:24
openstackgerritMatt Riedemann proposed openstack/nova master: Add power_on kwarg to ComputeDriver.spawn() method  https://review.opendev.org/64259022:24
openstackgerritMatt Riedemann proposed openstack/nova master: Add finish_snapshot_based_resize_at_dest compute method  https://review.opendev.org/63508022:24
openstackgerritMatt Riedemann proposed openstack/nova master: Add FinishResizeAtDestTask  https://review.opendev.org/63564622:24
openstackgerritMatt Riedemann proposed openstack/nova master: Add Destination.allow_cross_cell_move field  https://review.opendev.org/61403522:24
openstackgerritMatt Riedemann proposed openstack/nova master: Execute CrossCellMigrationTask from MigrationTask  https://review.opendev.org/63566822:24
openstackgerritMatt Riedemann proposed openstack/nova master: Plumb allow_cross_cell_resize into compute API resize()  https://review.opendev.org/63568422:24
openstackgerritMatt Riedemann proposed openstack/nova master: Filter duplicates from compute API get_migrations_sorted()  https://review.opendev.org/63622422:24
openstackgerritMatt Riedemann proposed openstack/nova master: Add functional test for anti-affinity cross-cell migration  https://review.opendev.org/66185922:24
openstackgerritMatt Riedemann proposed openstack/nova master: Support cross-cell moves in external_instance_event  https://review.opendev.org/65847822:24
openstackgerritMatt Riedemann proposed openstack/nova master: Add cross-cell resize policy rule and enable in API  https://review.opendev.org/63826922:24
openstackgerritMatt Riedemann proposed openstack/nova master: WIP: Enable cross-cell resize in the nova-multi-cell job  https://review.opendev.org/65665622:24
*** mriedem has quit IRC22:24
*** brinzhang has joined #openstack-nova22:31
*** brinzhang_ has quit IRC22:34
*** markvoelker has quit IRC22:38
*** ivve has quit IRC22:45
*** markvoelker has joined #openstack-nova22:53
*** macz has quit IRC22:53
openstackgerritEric Fried proposed openstack/nova master: Use :oslo.config:* in nova-manage doc  https://review.opendev.org/67522422:58
*** markvoelker has quit IRC23:08
openstackgerritDustin Cowles proposed openstack/nova master: Provider config file schema and loader  https://review.opendev.org/67334123:09
*** slaweq_ has joined #openstack-nova23:11
*** betherly has joined #openstack-nova23:15
*** slaweq_ has quit IRC23:16
*** betherly has quit IRC23:20
*** hoonetorg has quit IRC23:26
*** takashin has joined #openstack-nova23:32
*** igordc has quit IRC23:35
*** betherly has joined #openstack-nova23:37
*** markvoelker has joined #openstack-nova23:38
*** hoonetorg has joined #openstack-nova23:40
openstackgerritTakashi NATSUME proposed openstack/nova master: Fix non-existent method of Mock  https://review.opendev.org/67504123:40
*** markvoelker has quit IRC23:43
*** betherly has quit IRC23:45

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!