Thursday, 2020-08-27

*** artom has quit IRC00:21
*** _erlon_ has quit IRC00:25
*** brinzhang_ has joined #openstack-nova00:32
*** brinzhang0 has quit IRC00:36
*** bnemec has quit IRC00:41
*** mgoddard has quit IRC00:41
*** bnemec has joined #openstack-nova00:46
*** jamesdenton has joined #openstack-nova00:47
*** mgoddard has joined #openstack-nova00:47
*** k_mouza has joined #openstack-nova01:08
*** k_mouza has quit IRC01:13
*** kberger_ has quit IRC01:13
*** mgoddard has quit IRC01:17
*** Liang__ has joined #openstack-nova01:21
*** mgoddard has joined #openstack-nova01:24
*** brinzhang_ has quit IRC01:31
*** brinzhang_ has joined #openstack-nova01:31
*** bnemec has quit IRC01:34
*** zzzeek has quit IRC01:36
*** zzzeek has joined #openstack-nova01:38
*** bnemec has joined #openstack-nova01:39
*** xiaolin has joined #openstack-nova01:59
openstackgerritWenping Song proposed openstack/nova master: Process exception of delete resource provider  https://review.opendev.org/74833902:03
*** jamesdenton has quit IRC02:25
*** rcernin has quit IRC02:29
*** links has joined #openstack-nova02:29
*** jamesdenton has joined #openstack-nova02:31
openstackgerritBrin Zhang proposed openstack/nova master: Cyborg evacuate support  https://review.opendev.org/71532602:40
*** boxiang_ has quit IRC02:41
*** boxiang has joined #openstack-nova02:42
*** rcernin has joined #openstack-nova02:51
*** Liang__ has quit IRC02:53
*** Liang__ has joined #openstack-nova02:54
*** rcernin has quit IRC02:56
*** rcernin has joined #openstack-nova02:56
*** efried has quit IRC02:57
*** dave-mccowan has quit IRC03:00
*** efried has joined #openstack-nova03:01
*** sapd1_x has joined #openstack-nova03:01
*** zzzeek has quit IRC03:02
*** zzzeek has joined #openstack-nova03:05
*** k_mouza has joined #openstack-nova03:10
*** k_mouza has quit IRC03:14
*** sapd1_x has quit IRC03:31
*** psachin has joined #openstack-nova03:39
*** sapd1_x has joined #openstack-nova03:42
openstackgerritHarshavardhan Metla proposed openstack/nova master: Raising a deprecated warning in case of 403  https://review.opendev.org/73506803:48
*** ratailor has joined #openstack-nova04:31
*** evrardjp has quit IRC04:33
*** evrardjp has joined #openstack-nova04:33
*** mkrai has joined #openstack-nova04:36
*** sapd1_x has quit IRC04:49
*** belmoreira has joined #openstack-nova04:53
*** suryasingh has joined #openstack-nova05:01
*** bhagyashris|away is now known as bhagyashris05:02
*** vishalmanchanda has joined #openstack-nova05:10
openstackgerritMerged openstack/nova master: Provider Config File: Enable loading and merging of provider configs  https://review.opendev.org/69346005:10
*** k_mouza has joined #openstack-nova05:10
*** k_mouza has quit IRC05:14
*** belmoreira has quit IRC05:26
*** belmoreira has joined #openstack-nova05:34
*** zzzeek has quit IRC05:37
*** zzzeek has joined #openstack-nova05:37
*** gyee has quit IRC05:39
*** ratailor has quit IRC05:47
*** ratailor has joined #openstack-nova05:48
openstackgerritJohannes Kulik proposed openstack/nova master: vmware: Handle exception in destroy with attached volumes  https://review.opendev.org/73841605:54
*** jsuchome has joined #openstack-nova05:59
*** ralonsoh has joined #openstack-nova06:14
*** sapd1_x has joined #openstack-nova06:16
*** zzzeek has quit IRC06:16
*** zzzeek has joined #openstack-nova06:17
*** ratailor has quit IRC06:24
*** slaweq has joined #openstack-nova06:32
*** zzzeek has quit IRC06:47
*** hoonetorg has quit IRC06:48
*** zzzeek has joined #openstack-nova06:50
*** mkrai has quit IRC06:57
*** mkrai has joined #openstack-nova06:57
*** hoonetorg has joined #openstack-nova07:01
*** ratailor has joined #openstack-nova07:03
*** tesseract has joined #openstack-nova07:04
*** k_mouza has joined #openstack-nova07:11
*** damien_r has joined #openstack-nova07:14
*** k_mouza has quit IRC07:16
*** damien_r has quit IRC07:18
*** rcernin has quit IRC07:28
*** sapd1_x has quit IRC07:31
*** brinzhang_ has quit IRC07:31
*** brinzhang_ has joined #openstack-nova07:32
*** yoctozepto has quit IRC07:34
*** dtantsur|afk is now known as dtantsur07:40
*** tosky has joined #openstack-nova07:47
*** derekh has joined #openstack-nova07:52
openstackgerritLee Yarwood proposed openstack/nova stable/ussuri: Ensure source compute is up when confirming a resize  https://review.opendev.org/74836908:08
*** brinzhang_ has quit IRC08:10
*** brinzhang_ has joined #openstack-nova08:11
*** sapd1_x has joined #openstack-nova08:17
*** chenhaw has quit IRC08:25
openstackgerritStephen Finucane proposed openstack/nova master: Add support for resize and cold migration of emulated TPM files  https://review.opendev.org/63993408:32
stephenfingibi: Respun that to address the failing test /o\ ^08:33
stephenfinso dumb :(08:33
gibino worries, I put my +2 back08:36
stephenfinthanks08:37
bauzasgibi: I want to provide the implementation for the routed networks spec, can I supersed https://review.opendev.org/#/c/656885/7 ?08:39
bauzasI mean, not providing a new revision, rather a new change08:39
gibibauzas: sure08:43
bauzasthanks08:44
gibisean-k-mooney: do you have ideas what happenening here https://bugs.launchpad.net/nova/+bug/1892361 ?08:44
openstackLaunchpad bug 1892361 in OpenStack Compute (nova) "SRIOV instance gets type-PF interface, libvirt kvm fails" [Undecided,New]08:44
*** k_mouza has joined #openstack-nova08:46
gibiit seems libvirt driver generates wrong xml for a PF dev08:46
stephenfingibi: btw, finally got around to this https://review.opendev.org/#/c/748377/08:51
stephenfinIt's the backport-candidate label change for Gerrit. I think that's how one does it. I've asked hberaud (who did it for oslo) for advice08:52
openstackgerritLee Yarwood proposed openstack/nova master: WIP compute: Only call _local_delete_cleanup when scheduled instance is found  https://review.opendev.org/74838008:52
lyarwoodnow to work out how to exercise that in a regression test08:53
*** yoctozepto has joined #openstack-nova08:55
gibistephenfin: ack, thanks08:56
*** sapd1_x has quit IRC08:58
gibiartom, bauzas: there is a vGPU related bugs in the tracker https://bugs.launchpad.net/nova/+bug/188738009:02
openstackLaunchpad bug 1887380 in OpenStack Compute (nova) "Attaching virtual GPU devices to guests in nova" [Undecided,New]09:02
bauzasgibi: ack09:03
openstackgerritVlad Gusev proposed openstack/nova stable/ussuri: docs: Resolve issue with deprecated extra specs  https://review.opendev.org/74838609:21
openstackgerritVlad Gusev proposed openstack/nova stable/ussuri: replace the "hide_hypervisor_id" to "hw:hide_hypervisor_id"  https://review.opendev.org/74718909:22
*** k_mouza has quit IRC09:22
*** k_mouza has joined #openstack-nova09:23
*** k_mouza has quit IRC09:24
*** k_mouza_ has joined #openstack-nova09:25
openstackgerritAlexandre Arents proposed openstack/nova master: Add a lock to prevent race during detach/attach of interface  https://review.opendev.org/74795709:35
*** ftarasenko has joined #openstack-nova09:41
jsuchomeHi stephenfin, could we get https://review.opendev.org/#/c/574301/ merged? I think your ,ain concerns were adressed09:46
*** zzzeek has quit IRC09:57
*** derekh has quit IRC10:00
stephenfinjsuchome: I think lyarwood might have accidentally regressed your changes from PS37 when updating the patch below it. That's why my -1 is still there10:01
stephenfinthe docs are still wrong here https://review.opendev.org/#/c/574301/39/nova/conf/glance.py10:01
*** zzzeek has joined #openstack-nova10:01
stephenfinjsuchome: Also, would it impact you if I asked for a new boolean config option to turn on rbd direct download instead of reviving 'allowed_direct_url_schemes'? The more I think about it, the less I like it /o\10:03
stephenfinthat would let you drop the following patch and finish this in one patch10:03
gibistephenfin, jsuchome: if it matters I can quickly re-review the patch(s) after such change10:05
lyarwoodstephenfin: did I?10:05
lyarwoodstephenfin: crap my bad sorry10:05
stephenfineasily fixed10:06
stephenfinI really do think the new option is better idea though. Much easier to discover and just better UX in general, IMHO. Looks like it'll have a minimal impact of the patch too10:09
jsuchomeI'm not exactly in favor of that ... I think reusing that old option was described as a way to go in the blueprint10:14
stephenfinjsuchome: Right, but the spec was first proposed in Queens when that option wasn't yet deprecated. I realize the approved spec notes the deprecation and its impact, but is there a clear technical reason for undeprecating the option or was it simply inertia?10:20
jsuchomeI don't think there's technical reason, main reason is that it all takes so much time...10:22
*** xek has joined #openstack-nova10:24
jsuchomeI hear there are some users using this patch already so I would think it's nice to just keep the option the same, but that's just it. Maybe the inconsistency with the bp is not an issue, it just looks ugly in a different way10:24
openstackgerritJiri Suchomel proposed openstack/nova master: Add ability to download Glance images into the libvirt image cache via RBD  https://review.opendev.org/57430110:29
*** Liang__ has quit IRC10:44
openstackgerritStephen Finucane proposed openstack/nova master: Introduce '[glance] enable_rbd_download'  https://review.opendev.org/74841110:55
stephenfinjsuchome, (lyarwood): ^10:55
*** jangutter has joined #openstack-nova11:00
*** jangutter_ has quit IRC11:03
*** k_mouza_ has quit IRC11:03
openstackgerritBalazs Gibizer proposed openstack/nova master: Use absolute path during qemu img rebase  https://review.opendev.org/73924611:10
gibistephenfin, lyarwood: rebased and fixed up ^^11:12
*** jangutter has quit IRC11:15
*** jangutter has joined #openstack-nova11:15
tony_sustephenfin: gibi: thanks for your kind assistance for code review and all patches were merged today.11:18
*** k_mouza has joined #openstack-nova11:19
gibitony_su: thanks for taking care of those patches. nicely done!11:19
tony_suI will continue to take care of any remaining nit comments as a follow-up patch.11:20
gibithanks11:22
*** mkrai has quit IRC11:24
*** sapd1_x has joined #openstack-nova11:25
*** k_mouza has quit IRC11:35
*** ratailor has quit IRC11:36
jsuchomestephenfin: I understand it can be done and that the change is small (altough you see youreself that it is affecting the testsuite) ... it just means back to the step 1 with reviews after months getting it to the current state...11:37
*** xiaolin has quit IRC11:37
*** efried has quit IRC11:45
stephenfinjsuchome: Hardly. The idea is sound. Just a tweak to the config option11:55
stephenfinand the change is to the new test, not to anything existing11:56
stephenfingibi: Done11:56
gibistephenfin: thanks12:05
*** spatel has joined #openstack-nova12:22
brinzhang_stephenfin: hope you can review the cyborg evacuate support patch, it has +2 by gibi, https://review.opendev.org/#/c/715326/12:23
brinzhang_stephenfin, gibi: thanks12:24
*** sapd1_x has quit IRC12:24
*** dklyle has quit IRC12:24
*** artom has joined #openstack-nova12:25
*** spatel has quit IRC12:27
openstackgerritBalazs Gibizer proposed openstack/nova master: Only unplug vif after the device is detached from libvirt  https://review.opendev.org/74493412:27
*** nweinber has joined #openstack-nova12:43
*** psachin has quit IRC13:14
*** jangutter_ has joined #openstack-nova13:20
stephenfinbrinzhang_: Done. I'm -1 but it's just a collection of mostly trivial stuff13:22
*** efried has joined #openstack-nova13:23
*** jangutter has quit IRC13:24
*** jangutter has joined #openstack-nova13:26
*** k_mouza has joined #openstack-nova13:27
*** jangutter_ has quit IRC13:29
brinzhang_stephenfin: thanks, I will update tomorrow13:29
*** jangutter_ has joined #openstack-nova13:29
brinzhang_stephenfin: https://review.opendev.org/#/c/715326/22/nova/accelerator/cyborg.py@313 this function I would not like to change, and Iwant keep it in the cyborg file13:30
brinzhang_and I think it's easy to maintain in the furture13:30
brinzhang_s/furture/future13:31
stephenfinokay, it just seemed unnecessary when it's only one line long and has a single caller. If you're planning to add additional callers, then maybe it makes sense to keep13:31
*** sapd1_x has joined #openstack-nova13:32
brinzhang_stephenfin: yes, the cyborg shelve/unshelve patch also need to use this function13:32
*** jangutter has quit IRC13:32
stephenfinokay, just leave a comment in the patch then (y)13:32
stephenfin*a reply13:33
brinzhang_stephenfin: thanks13:33
stephenfinIs anyone else see "Code Review - Error Server Unavailable" occasionally from Gerrit?13:34
brinzhang_it's the shelve/unshelve patch https://review.opendev.org/#/c/729563/8/nova/compute/api.py@403313:34
* stephenfin wonders if it's his internet or a dodgy Gerrit instance13:34
sean-k-mooneygiven where you are currently working its proably your internet13:35
*** dave-mccowan has joined #openstack-nova13:38
bauzasstephenfin: I do, lots of time13:42
bauzasbut... I can't say my Internet is the better, right?13:42
stephenfinheh, yeah, no13:42
stephenfin:P13:42
* bauzas clicked on the optic fiber deployment page to see whether there was progress13:42
bauzasand the answer is sadly "yes, but not for your zone"13:43
* bauzas is pretty sure he'll have optic fiber for Christmas13:43
bauzasbut I don't know which one13:43
bauzasstephenfin: that being said, this error seems very server-side to me13:44
gibistephenfin: I do get Server Unavailable about 5 times per day from gerrit on an otherwise stable internet connection13:44
bauzasgibi: do you accept a coworker ?13:44
stephenfinI'm getting it every 30 seconds or so at the moment13:44
bauzasyesterday was bad on this now that I remember13:45
bauzasthey're probably exhausting the number of java processes13:45
gibibauzas: I do, but I repeatedly failed to get more headcount on OpenStack upstream13:45
bauzasgibi: well, I just need a seat and a decent connection, a contract is too much :p13:46
* bauzas won't tell jokes on java13:46
gibibauzas: the seating would be a problem. MY 33m^2 flat does not even serve me well on seating13:47
bauzasgibi: I invested on a small Swedish company which I don't remember the name13:48
bauzasgibi: but they are good at providing you decent chairs around 200€ that are worth it13:49
bauzasIke-something13:49
* gibi needs to either move to a bigger place or get rid of some existing furniture13:49
gibibauzas: :)13:49
* bauzas lives 5 mins away from a Ikea store13:50
*** arxcruz|ruck has left #openstack-nova13:50
bauzasstephenfin can speak of how my house is just an Ikea showroom13:50
*** k_mouza has quit IRC13:51
gibiI have no problem with Ikea stuff, my kichen is from there13:51
bauzasyou know that you frequented too much Ikea when you're able to name furniture by their Ikea names13:51
bauzasgibi: worth saying, the Markus chair is very decent for the price13:52
bauzasthis saves my butt 5 times a week, 10 hours per day13:52
gibithanks for the suggestion13:53
gibido you happen to have one of those Ikea Skarsta desk?13:54
bauzasnope but I know them13:54
gibiis it worth the price tag?13:55
bauzaslooks like, yes13:55
bauzassome coworkers have them13:55
*** k_mouza has joined #openstack-nova13:55
gibicool, thanks.13:56
bauzas(I personnally have the Linnmon desk, cheap but large)13:57
bauzasgibi: still homeoffice, I guess ?13:59
gibiyes. It seems this year I don't have to go back to the office if I don't want to13:59
bauzasthis seems reasonable14:02
*** k_mouza has quit IRC14:05
*** k_mouza has joined #openstack-nova14:06
artomlyarwood, https://review.opendev.org/#/c/720769/5 -0.5 I guess?14:07
artomI left it voteless, take it or leave it :P14:07
artomOK, on second thought, -1. Comment left.14:09
*** k_mouza has quit IRC14:10
lyarwoodartom: that's dead until W btw14:11
*** bnemec has quit IRC14:11
*** bnemec has joined #openstack-nova14:12
*** mriedem has joined #openstack-nova14:14
stephenfinaarents: Comment on https://review.opendev.org/#/c/747957/14:19
stephenfinbauzas: Is https://review.opendev.org/#/c/748271/ sufficient to resolve your concerns with https://review.opendev.org/#/c/747744/ ?14:20
bauzasstephenfin: thanks, yes it is14:20
stephenfinhurray14:20
* bauzas is on the implementation side today but can +2 tho14:20
stephenfinGreat. Can rebase and remove the duplication from the vTPM series then (y)14:21
aarentsstephenfin: thks14:23
* bauzas wasn't thinking he would write a StarWars phrase in a review today14:24
*** k_mouza has joined #openstack-nova14:24
sean-k-mooneybauzas i have a few metere of optic cable left over from when they installed mine i can proably post it to you:P eventurlly you can collect enough and run it yourslef probalem solved right14:25
sean-k-mooneybauzas: starlink is ment to be roling out early next year too14:26
sean-k-mooneyif the price is right that might be an option14:26
bauzassean-k-mooney: hold my beer14:26
bauzassean-k-mooney: I personnally like to consider being able to join a free ISP with a specific AS if my ISP sucks14:29
bauzaswhich won't be possible if I go with starlink14:29
bauzasthat, and the fact I don't like how this will pollute space14:29
sean-k-mooneyalso you dont want to have to wait till 2021 or later when it actully start going live14:30
bauzashonestly, I could just pay a local mobile provider and run a LTE connection with 100 gigs fair use14:30
bauzassean-k-mooney: well, about 2021, I become almost sure that starlink would be operational before I get optic fiber at home14:31
sean-k-mooneyya if im at my mother house i get 150 down and about 80 up on my phone while her landline broadband only get 4 down and .25-.6 up on a good day14:34
*** k_mouza has quit IRC14:35
sean-k-mooneyi have suggested just bying a pone pluging it in and paying the 30 euro a month for unlimited broadband with thearhing allowed14:35
*** k_mouza has joined #openstack-nova14:35
sean-k-mooneyit would be a 1TB fair usge limit14:35
lyarwoodhttps://paste.centos.org/view/39276981 <- can anyone think why I could be getting this error in a func test on stable/train using integrated_helpers._IntegratedTestBase and starting two compute services ?14:37
lyarwoodComputeHostNotFound: Compute host 2 could not be found.14:37
lyarwoodeverything seems to start correctly14:37
* lyarwood is missing something obvious14:38
lyarwoodfor context I'm backporting https://review.opendev.org/#/q/Ib9dbc792dc918e7ea45915e2c1dbd96be82ef56214:38
*** dklyle has joined #openstack-nova14:41
aarentsstephenfin: not sure to get this one: https://review.opendev.org/#/c/747957/4/nova/tests/unit/compute/test_compute.py@10296 (I replied)14:41
stephenfinaarents: Fair point. Ignore that comment :)14:50
stephenfin(replied also)14:50
*** dave-mccowan has quit IRC14:50
*** k_mouza has quit IRC14:52
*** links has quit IRC14:53
openstackgerritBalazs Gibizer proposed openstack/nova master: Make PCI claim NUMA aware during live migration  https://review.opendev.org/74845314:56
aarentsstephenfin: oky thks, I will fixup the others14:57
openstackgerritBalazs Gibizer proposed openstack/nova master: [WIP] Support SRIOV interface attach and detach  https://review.opendev.org/74099514:57
openstackgerritBalazs Gibizer proposed openstack/nova master: DNM: tracing the PCI allocation code path  https://review.opendev.org/74845414:59
gibisean-k-mooney: added some logs ^^ to see why the pci device goes to allocated immediately in your env15:01
*** k_mouza has joined #openstack-nova15:01
gibiI mean why it doesn't go to allocated15:01
*** dklyle has quit IRC15:05
*** mkrai has joined #openstack-nova15:10
sean-k-mooneygibi: so i still dont think we should make sriov use claims for what its worth15:12
sean-k-mooneyim hoping eventully claims wil go away and we will use migrate data like we do with sriov migration15:12
sean-k-mooneyoh sorry this is something esle15:13
sean-k-mooneyya ok i see what you have changed15:13
sean-k-mooneygibi: i was getting confused with the TODO to converge numa migration and sriov migration to claim devices in a common way15:13
sean-k-mooneyhttps://review.opendev.org/#/c/748453/1/nova/compute/resource_tracker.py is adress the comment i left about not passign the numa toplogy15:14
sean-k-mooneyartom: this is better then we do today where we dont pass the numa toplogy but its still wrogn correct^15:24
sean-k-mooneyhttps://review.opendev.org/#/c/748453/115:24
gibisean-k-mooney: do you mean we pass the wrong topology during live migration?15:25
sean-k-mooneyit depend on where that is being called15:25
artomgibi, I think sean-k-mooney means adding PCI devices to the _live_migration_claim()...15:25
sean-k-mooneyartom: well i want to delete that15:26
artomsean-k-mooney, delete... the _live_migration_claim()?15:26
sean-k-mooneybut currently its done in check_can_live_migrate_destination15:26
sean-k-mooneyartom: i want to remove claims eventually15:26
artomIn favor of what?15:26
* gibi listens15:26
artomThey're like the basis of all of NUMA live migraiton15:26
sean-k-mooneyyep i know15:27
artomAnd I mean... they work.15:27
sean-k-mooneyin favor of doing it the way we do for sriov live migration15:27
artom... why?15:27
sean-k-mooneyanyway different topic for a different day15:27
gibi:)15:27
sean-k-mooneyport_id_to_pci = self._claim_pci_for_instance_vifs(ctxt, instance)15:27
artomLike yeah, they're hard to grok initially, but they do a bunch of useful things15:27
sean-k-mooneywe do that in check_can_live_migrate_destination15:28
artomAnd they're battle-tested15:28
sean-k-mooneyim not sure which numa toplogy is stored in the instance at that point15:29
artomI mean not really for another day, because gibi's patch has to chose 1 or the other...15:29
sean-k-mooneyis it the souce numa toplogy or the dest15:29
artomsean-k-mooney, the source, until we apply_move_claim()15:29
sean-k-mooneyok so then that is the wrong numa toplogy for 2 reasons15:29
sean-k-mooneyfirst the souce and dest can differ15:29
gibiI have access to the MoveClaim but that also seems to returning the numa topology stored on the instnace15:30
sean-k-mooneyand second if the dest numa toplogy has not been caulated with the numa reuqirement it also wrong15:30
sean-k-mooneywe shoudl be geting the toplogy info form the migate_data object right15:30
gibisean-k-mooney: I did not found it there15:30
artomsean-k-mooney, the whole point of using claims was to calculate the dest numa topology, and store the old and new until we apply the claim...15:31
sean-k-mooneyartom: right but sriov migration merged a few months before any for the numa migration stuff did15:31
gibiartom let me know where is the dest numa topology stored and I will modify my patch15:31
sean-k-mooneyand it was expictly not using move claims15:31
artomsean-k-mooney, I know, we discussed that during review15:32
sean-k-mooneyyes i wanted to move to a case where we claimd resouce in the db and passed the info via migrate data and start using that of all move operations15:32
artomThat's what claims do :)15:32
sean-k-mooneybadly15:33
artomWhy?15:33
sean-k-mooneyfor one they dont actully store them in the db15:33
sean-k-mooneythe are in memroy we are not claiming indivutal resouce on the numa toplogy blob for example15:33
artomStore what? The resource usage? Yes they do, the update the resource tracker, and the migration is consuming the resources15:33
sean-k-mooneythe host one15:33
sean-k-mooneyclaims basically are boolean this is free or it might be used15:34
sean-k-mooneywith pci devces we have a tristate15:34
artomYeah, the eventually call down to rt._update() or w/e it's called15:34
sean-k-mooneyavailable, claimed for an instnace but not used yet and allocated15:35
sean-k-mooneythat is stored in teh db drictly in a table not in memory15:35
artomsean-k-mooney, so can the resource tracker already call down to the pci tracker to udpate those?15:35
sean-k-mooneywhich means if we restart the compute agent that info is not actully lost15:35
sean-k-mooneyyes15:35
artomsean-k-mooney, so what's wrong with using that exisiting plumbing?15:36
sean-k-mooneybut thats kind of missign the point that i dont really think how we do claimis is a good thing. i know othere wanted to move to a model weher we rely more on plament allocation and directl claims or resouce that are not tempory in memory15:36
artomsean-k-mooney, Placement, sure15:36
artomBut let's be honest, NUMA in placement is never happening :P15:37
sean-k-mooneywe basically have two source of info here the live migration claim and the migrate_data15:37
artomsean-k-mooney, the migrate data is just the info for the source to update the XML15:37
sean-k-mooneywell actully we might eventually have to make that deciesion15:37
artomThe claim is for resources consumed15:37
sean-k-mooneythere are feature that have been waitn 4+year for numa in placment15:38
sean-k-mooneyim not really sure we can justify blocking them to much longer but ill try to get it done next cycle15:38
sean-k-mooneyafter wallaby however we really need to consier if we will ever do int plamcent15:38
sean-k-mooneywe did not need placment ot solve the races we have with numa15:39
sean-k-mooneywe just need to not use claims the way they are right now15:39
sean-k-mooneythat is one of my main issue with them15:39
sean-k-mooneybecause the calims are done locally on the compute node it cause the numa races15:39
artomsean-k-mooney, plus, we *already have* PCI stuff in the rt's _move_claim()15:39
sean-k-mooneyfor what15:40
artomMoves :) Cold migrations15:40
sean-k-mooneyoh you put in a hack i remmeber15:40
sean-k-mooneyya for cold migation15:40
sean-k-mooneybut also the hack for live15:40
sean-k-mooneywhere we dont set them or something15:40
sean-k-mooneythis https://github.com/openstack/nova/blob/f521f4dbace0e35bedd089369da6f6969da5ca32/nova/compute/resource_tracker.py#L303-L31415:41
artomgibi, so https://github.com/openstack/nova/blob/master/nova/compute/resource_tracker.py#L223 is the live migration claim creation method, that creates the migration context with the old and new NUMA topologies by calling down to _move_claim()15:42
artomgibi, and yeah, sean-k-mooney found my TODO about converging the 2 (NUMA and SRIOV live migration)15:42
artomgibi, the live migration claim is created on the destination at https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L767015:43
sean-k-mooneyright which i objected too at the time and still do unless we remove the migration_data object15:43
sean-k-mooneyand and a couple of other things15:43
sean-k-mooneylike moving claims for spawn and other vent to the conductor15:44
sean-k-mooneyor ideally earlier15:44
artomgibi, and then we eventually save the dest stuff (like numa topology) here: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L858715:44
sean-k-mooneyartom: if its not clear as far as i am concerned the main reason we have races with numa is due to claims beign done on the compute node. that is what placment and numa in plamcnet was ment to solve15:45
sean-k-mooneyor we could have fixed it years ago if claims were dont in the scuderler of conductor15:46
sean-k-mooneygibi: https://github.com/openstack/nova/blob/f521f4dbace0e35bedd089369da6f6969da5ca32/nova/compute/resource_tracker.py#L32815:46
sean-k-mooneygibi: you need to use the new_numa_topology for the migration context15:46
artomsean-k-mooney, it's not so much *where* they're done, it's that scheduling and claiming happened in 2 steps15:47
sean-k-mooneyartom: yes which is where they are done15:47
artomsean-k-mooney, but yeah, claiming in the scheduler would have fixed that, because we could have made schedule + claim atomic15:47
sean-k-mooneyyes15:47
sean-k-mooneywhich is what i argured for befor placment was created15:47
artomYeah, I could get behind that15:48
artomBut it's moot at this point15:48
artomFor better or worse15:48
mriedemsearch is busted in the docs https://docs.openstack.org/nova/latest/search.html?q=cross_az_attach - click on one of those15:48
sean-k-mooneyoh it is15:48
sean-k-mooneydid we move things15:48
gibisean-k-mooney, artom: thanks I think I see it now15:49
sean-k-mooneyavailability-zonesundefined15:49
artomWhere did the 'undefined' in that URL come from?15:49
artomgibi, that's actually impressive that you do :)15:49
sean-k-mooneyso we are not substituing somthing15:49
gibiartom: I see from where I need to use the topology15:49
gibiartom: honeslty I don't have the brainpower to deeploy follow the argument about claims vs pci_tracker15:50
sean-k-mooneymigration_contex.new_toplogy15:50
mriedemhttps://bugs.launchpad.net/nova/+bug/189323815:50
openstackLaunchpad bug 1893238 in OpenStack Compute (nova) "nova document search result links not working" [Undecided,New]15:50
sean-k-mooneyha we must have chagne some of the shinx setting15:50
gibiohh, nova meeting starts in 9 minutes in #openstack-meeting-315:51
bauzasthis meeting time kills me :(15:53
gibibauzas: how can I help?15:53
bauzasnothing really, I'll attend15:53
bauzasbut please, be short :)15:53
gibiI will try15:54
openstackgerritArtom Lifshitz proposed openstack/nova master: post live migration: don't call Neutron needlessly  https://review.opendev.org/74745115:58
stephenfinartom: got a flake8 issue on https://review.opendev.org/#/c/747451/16:00
artomgdi16:00
openstackgerritArtom Lifshitz proposed openstack/nova master: trivial: Remove useless comment  https://review.opendev.org/74847516:01
artomstephenfin, in the meantime, here's that useless comment removal ^^16:02
stephenfin+216:04
openstackgerritArtom Lifshitz proposed openstack/nova master: post live migration: don't call Neutron needlessly  https://review.opendev.org/74745116:04
*** gyee has joined #openstack-nova16:06
*** JamesBenson has joined #openstack-nova16:06
*** dustinc has quit IRC16:07
*** manuvakery has joined #openstack-nova16:12
*** chenhaw has joined #openstack-nova16:14
*** derekh has joined #openstack-nova16:21
*** bnemec has quit IRC16:22
*** tesseract has quit IRC16:23
*** dtantsur is now known as dtantsur|afk16:23
*** bnemec has joined #openstack-nova16:30
*** martinkennelly has joined #openstack-nova16:33
*** k_mouza has quit IRC16:38
*** k_mouza has joined #openstack-nova16:45
*** tosky has quit IRC16:45
*** derekh has quit IRC16:47
*** mkrai has quit IRC16:49
*** harsha24 has joined #openstack-nova16:49
* bauzas calls it a day, bye folks16:49
*** k_mouza has quit IRC16:49
artomlyarwood, melwitt, wanna take a gander at https://review.opendev.org/#/c/747451/?16:50
* gibi also leaves for today16:50
* stephenfin too16:50
lyarwoodartom: I can try16:51
artomlyarwood, cheers!16:51
*** belmoreira has quit IRC16:52
gibigmann: sorry I totally forget you a second time too16:53
gibigmann: /o\16:53
gibigmann: is there anything important from API side?16:53
*** harsha24 has quit IRC16:55
gmanngibi:  ah yeah me too :). just json to yaml migration, anyways let me get oslo tooling in for that and then we can talk on nova patch - https://review.opendev.org/#/q/topic:bp/policy-json-to-yaml+(status:open+OR+status:merged)16:56
gibigmann: seems like a fairly easy nova change16:57
gibigmann: if the oslo change lands before the non-client library freeze then we can try to land the nova change before m316:58
gmannyeah. i hope at least with upgrade checks, changing the default value would not break operators but dansmith or johnthetubaguy can corret me16:58
gmanngibi: yeah that is the plan. i think we can merge the olso change by tomorrow16:58
gibicool, let's get back to this tomorrow or next week16:59
gibiI have to drop now16:59
gmannyup16:59
gibio/16:59
dansmithgmann: json still works, right?17:00
gmanndansmith: yeah it will work but default value of policy_file will change. https://review.opendev.org/#/c/748059/1/nova/policy.py17:00
dansmithyaml is the default for the generation, right? I wouldn't think any problem on upgrade that wouldn't happen if we overwrote their file with fresh jsson anyway17:01
dansmithoh17:01
gmanni mean is anyone has policy.json then it has to be converted to policy,yaml17:01
dansmithwait, they *have* to convert to yaml?17:01
gmannyeah because of extension name change in default value. json formatted still work though17:03
dansmithyou mean if they take the default17:03
dansmithfile name17:03
gmannyes17:03
sean-k-mooneyyou need to use yaml to use some of the feautures17:03
dansmithsean-k-mooney: yeah that's not what we're talking about17:03
sean-k-mooneyah ok17:03
gmannconverting json formatted to yaml is separate thing and they can do in their own time and we provide tool also for that.17:04
dansmithgmann: ah so, wait, we don't have a policy file conf, is that right?17:04
gmannwe do have17:04
sean-k-mooneygmann: i assume the tool just does yaml.dump(json.loads(file))17:04
sean-k-mooneyor similar17:04
dansmithgmann: where is it?17:04
sean-k-mooneye.g. parse the profile then dump it in the other format17:05
gmannsean-k-mooney: more than that - https://review.opendev.org/#/c/748055/4//COMMIT_MSG@1717:05
dansmithsean-k-mooney: we're talking about breaking existing people, not generation right now17:05
sean-k-mooneywe kind of need to do this in a falback way17:05
sean-k-mooneycheck for policy.json then check for policy.yaml right17:06
sean-k-mooneyand fail if we find both17:06
dansmithsean-k-mooney: exactly, that's what I'm getting at17:06
gmanndansmith: this is per service option - https://github.com/openstack/oslo.policy/blob/a626ad12fe5a3abd49d70e3e5b95589d279ab578/oslo_policy/opts.py#L4817:06
dansmithprefer yaml, but fall back to json17:06
sean-k-mooneyjust prefer or hard error if both17:07
dansmithgmann: ack, so where is that DEFAULT_POLICY_FILE thing used?17:07
dansmithoh jeez, I see17:07
dansmithright under it, sorry :P17:07
dansmithbut yeah, we *have* to fall back IMHO17:07
sean-k-mooneyto fultile on upgrade we dont change config right17:08
sean-k-mooneythat part of the grenade theory of upgrades workflow17:08
gmanndansmith: this will change the default for nova - https://github.com/openstack/oslo.policy/blob/a626ad12fe5a3abd49d70e3e5b95589d279ab578/oslo_policy/opts.py#L12117:08
dansmithsean-k-mooney: right17:08
dansmithgmann: right, I see it now17:08
dansmithgmann: but we can't just change the default and break everyone, even if the workaround is to change the config or copy the file, IMHO17:08
gmanndansmith: but at some point we have to do that right? may be upgrade check in this cycle with warning and next cycle change default ?17:09
dansmiththat passes grenade because we don't override any policy I imagine, but if we did it'd be broken17:09
dansmithgmann: yep, we can do that, warn now switch next17:09
gmannok. and with new tools operator have time for converting to yaml file too other than default one.17:10
sean-k-mooneywe shoudl add a nova status check too17:11
sean-k-mooneybecause tehy should convert on victoria before upgrading idealy17:11
gmannyeah its there but with failure which i can convert to warning for this cycle17:11
gmannhttps://review.opendev.org/#/c/748059/1/nova/cmd/status.py17:12
*** ralonsoh has quit IRC17:12
sean-k-mooneyso i would propose the following. check for the existine of both policy.yaml and policy.json. error if you find both and warn if you finde policy.json but use either policy.json or policy.yaml if you find one but not the other17:13
gmannall except 'error if you find both' s part of this oslo spec.17:14
sean-k-mooneyright didn oslo say we should treat that as an error17:14
gmann'existine of both policy.yaml and policy.json' is not possible as file can be detected via config potion only17:14
gmannwhich one, finding both?17:15
sean-k-mooneyyes17:15
sean-k-mooneydidnt they call that out as an error as we dont know which policy to follow17:15
gmannwe cannot find both as we can only look for conf.oslo_policy.policy_file17:15
sean-k-mooneyso we shoudl hard error17:15
dansmithsean-k-mooney: he's saying that there is only one option and we will hard fail in both cases,17:15
sean-k-mooneywe can do import os; os.path.exist(filename)17:16
dansmithwhich means we have to warn this time before we can switch the default17:16
gmannyeah17:16
dansmithit would be better if we could make nova prefer one over the other, and I think it's possible to do that, but not required17:16
dansmithgmann: nova could check for the file and if not existing, parse the config opt, check for the .json equivalent and reset the default right?17:16
sean-k-mooneyright instead of hardcoding DEFAULT_POLICY_FILE = 'policy.yaml'17:17
dansmithlike if CONF.policy_file == DEFAULT_POLICY_FILE and os.path.exists(CONF.policy_file.replace('yaml', 'json')): cfg.set_defaults(...)17:17
gmannfirst one is not right thing to do as any existing file should be consider as policy file17:18
dansmithsean-k-mooney: no, we have to hardcode it17:18
dansmithbut we can be flexible in our use of it17:18
sean-k-mooneyya what you sugges seam sane to me too17:18
sean-k-mooneyits more or less what i ment17:18
gmannyou mean just change the extension name?17:18
sean-k-mooneyyep17:19
sean-k-mooneycheck if the file exists17:19
sean-k-mooneyif not change extnetion17:19
sean-k-mooneyand check again if found use it17:19
dansmithgmann: yeah my str.replace was just an example, but basically if the config is not overridden, and the file does not exist but the s/yaml/json/ one does, change the default17:19
gmanni think policy.json -> policy.yaml should be converted via new tool which acutally convert the JSON formatted to YAM L formatted.17:20
gmannotherwise policy.yaml in JSON format is confusing17:21
dansmiththey should, but we can't make them17:21
dansmithputting json in a .yaml file is not a good idea, nor what we're suggesting17:21
sean-k-mooneywell this is not assumign yaml in an .json17:21
sean-k-mooneywe are assumign that are on ussuri upgradeing to victoria17:22
sean-k-mooneyso they have a policy.json with json in it now17:22
gmanni was thinking warn now give them time to properly convert the json to yaml and in next cycle or so we just change default17:22
sean-k-mooneyand we want to make sure we can load that in victoria without them modifying it in any way17:22
*** k_mouza has joined #openstack-nova17:23
dansmithgmann: we can do that too, but we're saying it would be better if we can make it default to yaml now, but fall back to json if need b17:23
dansmith*be17:23
gmannand remove fallback in next cycle?17:24
sean-k-mooneyyep17:24
dansmithpotentially..some future cycle17:24
sean-k-mooneybasicaly treat this as a deprecation17:24
sean-k-mooneywe normally remove in a cycle or two17:24
gmannok. yeah if at some point we can remove fallback then it is fine17:24
dansmithgmann: where is policy.init called from the actual code?17:24
dansmithI only see it from cmd/status17:25
gmannfrom manage also and at the first call of context.can(17:25
sean-k-mooneyyou mean in the api wsgi app17:25
dansmithah, context.can runs it if not run yet?17:25
gmannyeah17:26
gmannhttps://github.com/openstack/nova/blob/f521f4dbace0e35bedd089369da6f6969da5ca32/nova/policy.py#L16817:26
gmannwhich is called from context.can()17:27
sean-k-mooneyi was expecting to see it in here https://github.com/openstack/nova/blob/master/nova/api/wsgi.py but ok17:27
*** k_mouza has quit IRC17:27
dansmithgmann: we're suggesting this: https://pastebin.com/WASufpac17:28
dansmithbut less crappy and working.. I just hacked that up like pseudocode17:28
sean-k-mooneyya something like that17:29
dansmithmake sense?17:29
gmannyeah but we should only replace replace('yaml', 'json') if default file, 'policy.json' and not any other name17:31
sean-k-mooneythats why there is CONF.policy_file == DEFAULT_POLICY_FILE17:31
gmannif CONF.policy_file == OLD_DEFAULT17:31
dansmithgmann: no,17:32
openstackgerritLee Yarwood proposed openstack/nova stable/train: WIP Add regression tests for bug #1889108  https://review.opendev.org/74849417:32
openstackbug 1889108 in OpenStack Compute (nova) train "failures during driver.pre_live_migration remove source attachments during rollback" [Undecided,New] https://launchpad.net/bugs/188910817:32
openstackgerritLee Yarwood proposed openstack/nova stable/train: compute: refactor volume bdm rollback error handling  https://review.opendev.org/74849517:32
gmannDEFAULT_POLICY_FILE will be policy.yaml right17:32
openstackgerritLee Yarwood proposed openstack/nova stable/train: compute: Don't delete the original attachment during pre LM rollback  https://review.opendev.org/74849617:32
dansmithgmann: only do this if they haven't touched their config file17:32
sean-k-mooneywe only do hte fall back if the value is the default17:32
dansmithright17:32
sean-k-mooneydansmith: actully isnt there a way to check if a config value has been set or is the default17:32
dansmithbecause if they're relying on the default file being policy.json, only THEN do we fall back to the old default17:32
dansmithsean-k-mooney: yes, I'm just doing the easy thing without having to look that up :)17:32
gmannelse part change overridden policy file also17:32
gmannxyv_policy_nova.json17:33
sean-k-mooneyok cool i think i have used it before17:33
dansmithgmann: if they have overridden it in the config, then we take what they have set17:33
dansmithgmann: we only need to look for the old file if they haven't, but are assuming the default is the old thing17:33
gmann that is what i am saying. which need to change else: -> elif CONF.policy_file == OLD_DEFAULT_POLICY_FILE17:35
*** nweinber has quit IRC17:36
dansmithNO17:36
dansmith:)17:36
dansmithwe don't need to do anything if they have said their CONF.policy_file to anything, including the old default17:36
dansmithbecause of L1317:36
dansmithwait, maybe I'm missing which else you're talking about.. gmann maybe update the pastebin with what you mean?17:37
gmannif they set CONF.policy_file=nova_policy.json17:37
gmannok17:37
sean-k-mooneyif they set it at all we just use what they set17:37
dansmithsean-k-mooney: right17:38
sean-k-mooneyif they dont set it. we check the new default and if the file does not exists check the old default.17:38
dansmithwe never need to check to see if the conf is set to the old default,17:39
sean-k-mooneythat is what dansmith's pastebin does17:39
sean-k-mooneycorrect17:39
dansmithwe only need to do something different if the conf is unchanged, i.e. set to the NEW default17:39
*** nweinber has joined #openstack-nova17:39
dansmithand knowing if it's overridden is better than knowing if it's set to the default17:39
dansmithbecause if it's just set to the default in their config, we should *only* honor that file17:39
gmannhttps://pastebin.com/GN7mz5tE17:40
dansmithgmann: no, that's not what we want17:40
dansmithif it's set to the old value, then we just use the old value, end of story17:41
dansmithif it's not overridden, then we need to check the new value, and if not present, check the old value..that's it17:41
sean-k-mooneygmann: that woudl expand to 'policy.json'.replace('yaml', 'json') on line 1417:41
dansmithright17:42
gmannin case of overridden to any new-name.json also else block is executed right?17:44
dansmithif the value is set to *anything* other than the new default in config, then we just honor that17:44
dansmiththe ONLY thing we need to handle is the case where the value  is not overridden and the operator or packager assumes the filename is the old value17:45
dansmithi.e. "I don't need to set it in my config because I use the default filename"17:45
sean-k-mooneygmann: did you miss the early return on line 11 https://pastebin.com/WASufpac17:45
dansmithwe break those people if we change the default filename17:45
dansmithwe we'd like to change the default filename, but still honor the old one IF-AND-ONLY-IF they're relying on the default and have not specified a value in their config file17:46
sean-k-mooneyim going to go have food o/17:46
gmannyeah but i was thinking to change the existing default policy file instead if cONF default based on what exist17:46
dansmithgmann: I don't understand that last statement17:47
*** sapd1_x has quit IRC17:47
*** eharney has quit IRC17:49
gmannoverall goal i was thinking is to warn with upgrade checks and change default to yaml so that oslo_policy can remove the json format support at some point.17:51
gmannbut having fallback is kind of same situation with what we have currently.17:51
dansmithwe were discussion two options and maybe we confused you, so let me reiterate the two options:17:52
dansmith1. Leave the default as json now, warn in nova status, change default later and break people that didn't notice17:52
gmannyeah this one17:52
dansmith2. Switch default to yaml now, but fall back to json file for a release or two and warn in nova-status17:52
dansmithif we do #2, which is my pastebin, then we can remove json more aggressively because we have been defaulting to yaml for one more release (i.e. default to yaml starts in V)17:53
dansmithsean-k-mooney and I prefer #2, more aggressive, without breaking people, so we can say json goes away sooner17:53
gmannbut operator would not notice the default change as fallback is there.17:54
*** xinranwang__ has quit IRC17:54
dansmithif we do #1, then we will want to wait longer17:54
gmannunless we fail upgrade checks17:54
dansmiththe upgrade checks can be the same in both cases17:54
gmanni was thinking in option1, fail upgrade checks in W17:54
dansmithgmann: we can fail upgrade checks in V if we do #2, we just need to fall back to the old filename because not everyone runs upgrade checks17:55
*** raildo_ has quit IRC17:55
gmannhumm, you mean fail upgrade check but also fallback? does not that make upgrade checks a false info?17:55
*** raildo_ has joined #openstack-nova17:56
gmannupgrade checks should fail at the same time we start rejecting the json file in both otion17:56
gmannoption17:56
dansmiththe upgrade checks are informational anyway.. we have "warning" and "failure" in upgrade checks, we can warn now and fail on the release where we remove it in either case17:56
*** k_mouza has joined #openstack-nova17:57
*** jsuchome has quit IRC17:57
gmannyeah fail later seems more correct and what our code will behave17:57
dansmithif we do #2, then we get all new users to yaml right now, which means we can dump json support sooner17:57
dansmiththat's what we're going for17:57
dansmithit doesn't change the signaling now at all17:57
dansmithit just means we can get all new users on yaml right away,17:58
gmannyeah that is good i think.17:58
dansmithinstead of new users being on yaml later when everyone else if forced to do the same17:58
gmannyeah.  it is better way17:58
*** kevinz has quit IRC17:58
gmannand this warning in doc still valid for option2 also - https://review.opendev.org/#/c/748059/1/doc/source/configuration/policy-concepts.rst18:00
gmannor we should add default value things also here?18:00
dansmithwarning applies to either for sure, I wouldn't add the json fallback here18:01
gmannok18:01
dansmithI would only mention the fallback in the reno18:01
*** eharney has joined #openstack-nova18:01
*** k_mouza has quit IRC18:01
*** k_mouza has joined #openstack-nova18:02
gmann will update patch with option2 once oslo.policy is released. thanks dansmith sean-k-mooney .18:02
*** k_mouza has quit IRC18:06
dansmithcool, thanks gmann !18:08
gmanndansmith: but i will add  "elif CONF.policy_file == OLD_POLICY_FILE:"  for case - policy.json exist but an old, not supposed to be used file and CONF.policy_file =nova_policy.* means overridden file18:08
dansmithgmann: then I will -1 it :)18:09
gmannso that we do not pick not-supposed-to-use but existing policy.json18:09
gmann:)18:09
dansmithif CONF.policy_file is set to anything, then we should do what they ask18:09
*** eharney has quit IRC18:10
dansmithbut go ahead and propose what you want and let sean-k-mooney and I have a chance to -1 with discussion and we can do it there, which might be easier since we can reference actual code and lines18:10
gmannbut where you are checking override file?18:10
gmannok18:11
dansmithbnemec: apologize for asking this again, but.. there's some oslo_config method for determining if the conf option has been set or not right?18:13
mriedemset vs the value coming from a default?18:13
dansmithbnemec: is it just opt.value == opt.default ?18:14
dansmithmriedem: yeah18:14
mriedemthere is some location thing i thought18:14
mriedemwhich says where the value comes from18:14
*** eharney has joined #openstack-nova18:14
mriedemhttps://github.com/openstack/oslo.config/blob/master/oslo_config/cfg.py#L5718:14
bnemecYes18:14
dansmithokay, perfect18:15
bnemecDocs are here: https://docs.openstack.org/oslo.config/latest/reference/locations.html18:15
dansmithgmann: for sean-k-mooney and my desired behavior, the *only* thing you need to check is if CONF.policy_file's location == opt_default :)18:15
gmann+1, that can save any accidental pick of file if both extensions exist18:15
dansmithright18:15
gmann+1.  CONF.policy_file's location in [opt_default, set_default ] as nova will do set_default to change the default.18:16
dansmithack18:18
*** manuvakery has quit IRC18:22
*** suryasingh has quit IRC18:33
*** tosky has joined #openstack-nova18:35
*** k_mouza has joined #openstack-nova18:36
openstackgerritLee Yarwood proposed openstack/nova stable/stein: Improve CinderFixtureNewAttachFlow  https://review.opendev.org/74850918:39
openstackgerritLee Yarwood proposed openstack/nova stable/stein: Robustify attachment tracking in CinderFixtureNewAttachFlow  https://review.opendev.org/74851018:39
openstackgerritLee Yarwood proposed openstack/nova stable/stein: compute: Use source_bdms to reset attachment_ids during LM rollback  https://review.opendev.org/74851118:39
openstackgerritLee Yarwood proposed openstack/nova stable/stein: compute: refactor volume bdm rollback error handling  https://review.opendev.org/74851218:39
openstackgerritLee Yarwood proposed openstack/nova stable/stein: WIP Add regression tests for bug #1889108  https://review.opendev.org/74851318:39
openstackbug 1889108 in OpenStack Compute (nova) train "failures during driver.pre_live_migration remove source attachments during rollback" [Undecided,In progress] https://launchpad.net/bugs/1889108 - Assigned to Lee Yarwood (lyarwood)18:39
openstackgerritLee Yarwood proposed openstack/nova stable/stein: compute: Don't delete the original attachment during pre LM rollback  https://review.opendev.org/74851418:39
*** k_mouza has quit IRC18:51
lyarwoodmelwitt: https://review.opendev.org/#/c/742415/ - can you look at this if you get time, it's holding up the CVE on stable/train18:54
*** belmoreira has joined #openstack-nova18:54
lyarwoodmelwitt: well the release at least18:54
*** rambo_li has joined #openstack-nova19:02
gansohello Nova folks! I just logged bug: https://bugs.launchpad.net/nova/+bug/1893263 after reproducing the issue on master. I am planning to work on the fix myself, however, I am curious to know if anyone here knows the reason why plug_tap() needs to run, as if the instance is ever being rebooted, the tap device goes away and is re-created by libvirt instead (therefore plug_tap() seems to have never needed to run at all)19:03
openstackLaunchpad bug 1893263 in OpenStack Compute (nova) "Cannot create instance with multiqueue image and vif_type=tap (calico)" [Undecided,New]19:03
gansobecause one of the possible fixes is just to remove plug_tap() and be happy19:04
openstackgerritLee Yarwood proposed openstack/nova stable/train: Add regression tests for bug #1889108  https://review.opendev.org/74849419:07
openstackbug 1889108 in OpenStack Compute (nova) train "failures during driver.pre_live_migration remove source attachments during rollback" [Undecided,In progress] https://launchpad.net/bugs/1889108 - Assigned to Lee Yarwood (lyarwood)19:07
openstackgerritLee Yarwood proposed openstack/nova stable/train: compute: refactor volume bdm rollback error handling  https://review.opendev.org/74849519:07
openstackgerritLee Yarwood proposed openstack/nova stable/train: compute: Don't delete the original attachment during pre LM rollback  https://review.opendev.org/74849619:07
openstackgerritLee Yarwood proposed openstack/nova stable/stein: Add regression tests for bug #1889108  https://review.opendev.org/74851319:11
openstackbug 1889108 in OpenStack Compute (nova) train "failures during driver.pre_live_migration remove source attachments during rollback" [Undecided,In progress] https://launchpad.net/bugs/1889108 - Assigned to Lee Yarwood (lyarwood)19:11
openstackgerritLee Yarwood proposed openstack/nova stable/stein: compute: Don't delete the original attachment during pre LM rollback  https://review.opendev.org/74851419:11
*** rambo_li has quit IRC19:12
*** vishalmanchanda has quit IRC19:25
melwittlyarwood: ack will do19:26
*** bbowen has quit IRC19:28
*** bbowen has joined #openstack-nova19:28
lyarwoodmelwitt: thanks! :)19:30
lyarwoodartom: https://review.opendev.org/#/c/747451/ - LGTM, I've got a few questions but I'm not comfortable +2+W'ing networking stuff so even with these answered I'm going to defer to other cores on this one19:31
melwittsean-k-mooney: fyi I added you to review https://review.opendev.org/747451 please. iirc you have said using the cache is fine but it would be good if you could ack it19:31
lyarwoodah there we go \o/19:32
*** nweinber has quit IRC19:56
*** xek has quit IRC20:08
openstackgerritmelanie witt proposed openstack/nova master: Change default num_retries for glance to 3  https://review.opendev.org/74038920:16
openstackgerritRodrigo Barbieri proposed openstack/nova master: Allow tap interface with multiqueue  https://review.opendev.org/74853320:37
*** rambo_li has joined #openstack-nova20:39
*** martinkennelly has quit IRC20:47
*** martinkennelly has joined #openstack-nova20:47
*** rambo_li has quit IRC20:57
*** slaweq has quit IRC21:22
*** raildo_ has quit IRC21:28
*** slaweq has joined #openstack-nova21:29
*** slaweq has quit IRC21:34
*** martinkennelly has quit IRC21:36
*** JamesBenson has quit IRC21:42
*** belmoreira has quit IRC21:56
sean-k-mooneymelwitt: yes, when using multiple port bindings the ports are still active on the source host at this point and when not using multiple portbining we update the neutron port later in post_live_migration_at_destination21:57
sean-k-mooneymelwitt: there are other point in pre live migration where it would also be valid to use the cache for the same reason21:59
gmannstephenfin: you forgot to update this test - https://review.opendev.org/#/c/745906/4/nova/tests/functional/api_sample_tests/test_servers.py@97822:00
sean-k-mooneyi can note as much in the review22:00
melwittsean-k-mooney: yes, it would be ideal if you could review the change as you're expert on the network stuff22:01
*** bnemec has quit IRC22:07
*** mtreinish has quit IRC22:10
*** bnemec has joined #openstack-nova22:12
*** mtreinish has joined #openstack-nova22:16
sean-k-mooneymelwitt: actully if you dont mind ill review artoms change tomorrow. but ill leave it open to remind me.22:17
melwittsure, thanks22:18
sean-k-mooneysorry brain has alreay turned off. we had tunder today so have had a headache all enving. cam back to check something quickly but dont have the brain power to review something properly22:26
melwittnp at all, take some rest22:29
*** eharney has quit IRC22:44
*** rcernin has joined #openstack-nova22:45
openstackgerritGhanshyam Mann proposed openstack/nova master: DNM: testing tempest copy-image test  https://review.opendev.org/74254723:19
*** lbragstad has quit IRC23:27
*** lbragstad has joined #openstack-nova23:28
openstackgerritGhanshyam Mann proposed openstack/nova master: DNM: testing tempest copy-image test  https://review.opendev.org/74254723:30
*** bnemec has quit IRC23:37
openstackgerritmelanie witt proposed openstack/nova master: Default user_id when not specified in check_num_instances_quota  https://review.opendev.org/74855023:39
*** bnemec has joined #openstack-nova23:43
*** jangutter_ has quit IRC23:49
*** jangutter has joined #openstack-nova23:49
*** tosky has quit IRC23:59

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!