Wednesday, 2017-10-04

*** jistr has quit IRC00:00
*** jistr has joined #openstack-nova00:01
*** slaweq_ has joined #openstack-nova00:05
*** chyka has joined #openstack-nova00:06
*** sree has joined #openstack-nova00:08
*** chyka has quit IRC00:10
*** prometheanfire has joined #openstack-nova00:11
prometheanfireso, what's the story with nova-func tests?00:11
*** lnxnut has joined #openstack-nova00:11
*** sree has quit IRC00:12
openstackgerritKen'ichi Ohmichi proposed openstack/nova master: Remove doc todo related to bug/1506667
*** namnh has quit IRC00:17
*** namnh has joined #openstack-nova00:18
*** yamamoto has joined #openstack-nova00:18
*** flanders_ has quit IRC00:19
*** lnxnut has quit IRC00:21
*** namnh has quit IRC00:22
*** yamamoto has quit IRC00:23
*** exarr has quit IRC00:24
*** acormier has joined #openstack-nova00:28
*** ijw has quit IRC00:29
*** ijw has joined #openstack-nova00:30
*** acormier has quit IRC00:31
*** acormier has joined #openstack-nova00:32
*** ijw has quit IRC00:35
*** ijw has joined #openstack-nova00:36
*** gyee has joined #openstack-nova00:36
*** itlinux has joined #openstack-nova00:36
*** slaweq_ has quit IRC00:37
*** slaweq_ has joined #openstack-nova00:41
*** ragiman has joined #openstack-nova00:42
*** ijw has quit IRC00:48
*** TuanLA has joined #openstack-nova00:50
*** hongbin has joined #openstack-nova00:50
*** ijw_ has joined #openstack-nova00:50
*** ijw_ has quit IRC00:53
*** itlinux has quit IRC00:58
*** markvoelker has joined #openstack-nova01:01
*** mingyu has joined #openstack-nova01:07
*** MarginHu has joined #openstack-nova01:11
*** tonygunk has quit IRC01:11
*** slaweq_ has quit IRC01:13
*** lnxnut has joined #openstack-nova01:18
*** slaweq_ has joined #openstack-nova01:18
*** nicolasbock_ has quit IRC01:26
*** Kevin_Zheng has joined #openstack-nova01:27
*** mingyu has quit IRC01:28
*** lnxnut has quit IRC01:29
*** mingyu has joined #openstack-nova01:31
*** sree has joined #openstack-nova01:34
*** itlinux has joined #openstack-nova01:36
*** gjayavelu has quit IRC01:36
*** gouthamr has quit IRC01:37
*** yamahata has quit IRC01:39
*** phuongnh has joined #openstack-nova01:39
*** itlinux has quit IRC01:42
*** sree has quit IRC01:42
*** Tom has joined #openstack-nova01:43
*** gyee has quit IRC01:47
*** slaweq_ has quit IRC01:51
openstackgerritMerged openstack/nova master: Move ploop commands to privsep.
*** slaweq_ has joined #openstack-nova01:54
openstackgerritmelanie witt proposed openstack/nova master: Request zero root disk for boot-from-volume instances
openstackgerritmelanie witt proposed openstack/nova master: Claim and report zero root disk for boot-from-volume instances
*** takashin has quit IRC02:03
*** catintheroof has joined #openstack-nova02:03
*** tbachman has joined #openstack-nova02:10
*** baoli has joined #openstack-nova02:13
*** kenperkins has quit IRC02:13
*** catintheroof has quit IRC02:19
*** mingyu has quit IRC02:22
*** mingyu has joined #openstack-nova02:22
*** lnxnut has joined #openstack-nova02:26
*** slaweq_ has quit IRC02:28
*** yamamoto has joined #openstack-nova02:28
*** catintheroof has joined #openstack-nova02:29
openstackgerritZhenyu Zheng proposed openstack/nova-specs master: Improve the performance of filtering instances by IP.
*** slaweq_ has joined #openstack-nova02:32
*** lnxnut has quit IRC02:36
*** takashin has joined #openstack-nova02:48
*** acormier has quit IRC02:50
*** acormier has joined #openstack-nova02:50
*** Yikun__ has joined #openstack-nova02:52
*** acormier has quit IRC02:55
*** MasterOfBugs has quit IRC02:56
*** erlon has quit IRC03:01
openstackgerritMerged openstack/nova master: Use improved instance_list module in compute API
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove 400 as expected error
openstackgerritTakashi NATSUME proposed openstack/nova master: Remove 400 as expected error
*** catintheroof has quit IRC03:09
openstackgerritMerged openstack/python-novaclient stable/pike: Updated from global requirements
*** itlinux has joined #openstack-nova03:23
*** edand has joined #openstack-nova03:24
*** baoli has quit IRC03:30
*** itlinux has quit IRC03:31
*** takashin has quit IRC03:32
*** itlinux has joined #openstack-nova03:32
*** lnxnut has joined #openstack-nova03:34
*** hongbin has quit IRC03:36
*** slaweq_ has quit IRC03:36
*** Kevin_Zheng has quit IRC03:37
*** edand has quit IRC03:37
*** slaweq_ has joined #openstack-nova03:38
*** itlinux has quit IRC03:41
*** udesale has joined #openstack-nova03:44
*** lnxnut has quit IRC03:44
*** boris_42_ has quit IRC03:44
openstackgerritMerged openstack/nova master: Move allocation manipulation out of drop_move_claim()
*** armax has quit IRC04:05
*** sapcc-bot has quit IRC04:07
*** sapcc-bot has joined #openstack-nova04:07
*** ratailor has joined #openstack-nova04:09
*** slaweq_ has quit IRC04:10
*** mnaser has quit IRC04:15
*** slaweq_ has joined #openstack-nova04:16
*** tristanC has quit IRC04:16
*** tristanC has joined #openstack-nova04:23
*** mnaser has joined #openstack-nova04:28
*** claudiub has joined #openstack-nova04:30
*** tbh_ has joined #openstack-nova04:31
*** yamahata has joined #openstack-nova04:37
*** lnxnut has joined #openstack-nova04:40
*** slaweq_ has quit IRC04:49
openstackgerritMerged openstack/nova stable/pike: Split out the core of the ironic flavor migration
*** slaweq_ has joined #openstack-nova04:53
*** Yikun__ has quit IRC05:01
*** sree has joined #openstack-nova05:04
*** lnxnut has quit IRC05:06
*** psachin has joined #openstack-nova05:15
*** markvoelker has quit IRC05:20
openstackgerritMerged openstack/nova stable/pike: Add ComputeNodeList.get_by_hypervisor_type()
*** takashin has joined #openstack-nova05:24
*** slaweq_ has quit IRC05:26
*** gbarros has joined #openstack-nova05:28
*** sree has quit IRC05:31
*** slaweq_ has joined #openstack-nova05:32
*** sree has joined #openstack-nova05:32
*** lnxnut has joined #openstack-nova05:33
*** sree has quit IRC05:36
*** lnxnut has quit IRC05:37
*** lajoskatona has joined #openstack-nova05:39
*** manasm has joined #openstack-nova05:46
*** markmc has quit IRC05:54
*** kashyap has quit IRC05:55
*** rnoriega has quit IRC05:55
*** jmlowe has quit IRC05:55
*** migi has quit IRC05:55
*** weshay has quit IRC05:55
*** jmlowe has joined #openstack-nova05:56
*** mdnadeem has joined #openstack-nova06:00
*** edand has joined #openstack-nova06:00
*** cfriesen has quit IRC06:04
*** udesale__ has joined #openstack-nova06:05
*** slaweq_ has quit IRC06:05
*** udesale has quit IRC06:06
*** sree has joined #openstack-nova06:07
*** MarginHu has quit IRC06:07
*** Oku_OS-away is now known as Oku_OS06:08
*** pooja_jadhav has joined #openstack-nova06:09
*** slaweq_ has joined #openstack-nova06:11
*** rm_work has quit IRC06:11
*** rm_work has joined #openstack-nova06:14
*** Apoorva has joined #openstack-nova06:15
*** avolkov has joined #openstack-nova06:16
*** Apoorva has quit IRC06:20
*** rnoriega has joined #openstack-nova06:20
*** weshay has joined #openstack-nova06:20
*** kashyap has joined #openstack-nova06:20
*** gbarros has quit IRC06:24
*** sahid has joined #openstack-nova06:24
*** markmc has joined #openstack-nova06:25
*** udesale__ has quit IRC06:25
*** udesale__ has joined #openstack-nova06:25
*** migi has joined #openstack-nova06:27
*** rcernin has joined #openstack-nova06:30
*** sridharg has joined #openstack-nova06:32
*** edmondsw has joined #openstack-nova06:33
*** lnxnut has joined #openstack-nova06:34
*** _pewp_ has quit IRC06:35
*** _pewp_ has joined #openstack-nova06:36
*** edmondsw has quit IRC06:37
*** belmoreira has joined #openstack-nova06:38
openstackgerritMerged openstack/nova stable/pike: Test InstanceNotFound handling in 'nova usage'
*** slaweq_ has quit IRC06:43
*** sshwarts has joined #openstack-nova06:43
*** jgriffit1 has joined #openstack-nova06:44
*** chyka has joined #openstack-nova06:45
*** pcaruana has joined #openstack-nova06:48
*** chyka has quit IRC06:50
*** slaweq_ has joined #openstack-nova06:51
*** udesale has joined #openstack-nova06:51
*** tbh_ has quit IRC06:51
*** chyka has joined #openstack-nova06:52
*** udesale__ has quit IRC06:53
*** spectr has quit IRC06:54
*** andreas_s has joined #openstack-nova06:54
*** spectr has joined #openstack-nova06:55
*** chyka has quit IRC06:56
*** lnxnut has quit IRC06:57
*** gjayavelu has joined #openstack-nova06:57
*** udesale__ has joined #openstack-nova07:02
*** udesale has quit IRC07:04
*** tesseract has joined #openstack-nova07:16
*** sdake_ has joined #openstack-nova07:16
*** sdake_ is now known as Guest5227907:17
*** Guest52279 has quit IRC07:17
*** markvoelker has joined #openstack-nova07:21
*** slaweq_ has quit IRC07:24
*** slaweq_ has joined #openstack-nova07:28
*** flanders_ has joined #openstack-nova07:40
*** jpena|off is now known as jpena07:44
*** gjayavelu has quit IRC07:44
*** manasm has quit IRC07:45
*** ralonsoh has joined #openstack-nova07:47
*** brault has joined #openstack-nova07:47
*** manasm has joined #openstack-nova07:49
*** acormier has joined #openstack-nova07:51
*** markvoelker has quit IRC07:55
*** acormier has quit IRC07:55
*** abalutoiu has joined #openstack-nova08:01
*** udesale has joined #openstack-nova08:02
*** udesale__ has quit IRC08:03
*** takashin has left #openstack-nova08:14
openstackgerritRadoslav Gerganov proposed openstack/nova master: VMware: serial console log (completed)
stephenfindansmith: RE: so what is the deal on this nova-manage spec? we're going to just dump syntax compatibility across one release?08:21
*** edmondsw has joined #openstack-nova08:21
stephenfindansmith: we're going to drop compatibility for a single option - '--version' - for the 'db sync' and 'api_db sync' commands. This was replaced by a positional argument in Pike (I bugged mriedem relentlessly to get it in)08:22
stephenfindansmith: All the commands will otherwise stay the exact same, to the best of my knowledge08:23
stephenfinmelwitt: ^08:23
*** edmondsw has quit IRC08:26
openstackgerritStephen Finucane proposed openstack/nova-specs master: Add 'move-nova-cmds-to-cliff' spec
*** lucas-afk is now known as lucasagomes08:27
*** lnxnut has joined #openstack-nova08:28
*** slaweq_ has quit IRC08:33
*** slaweq_ has joined #openstack-nova08:43
openstackgerritStephen Finucane proposed openstack/nova master: doc: Add documentation for cpu_realtime, cpu_realtime_mask
*** lnxnut has quit IRC08:47
*** jaosorior has quit IRC08:51
*** derekh has joined #openstack-nova08:52
*** markvoelker has joined #openstack-nova08:52
*** jaosorior has joined #openstack-nova08:56
*** yamamoto has quit IRC09:01
*** tssurya has quit IRC09:01
*** sshwarts has quit IRC09:05
*** dgonzalez_ has joined #openstack-nova09:10
*** mkoderer_ has joined #openstack-nova09:10
*** david_ has joined #openstack-nova09:10
*** dgonzalez_ has quit IRC09:12
*** mkoderer_ has quit IRC09:12
*** david_ has quit IRC09:12
openstackgerritMerged openstack/os-vif master: Add Port Profile info to VIF objects OVS plugin
*** tssurya has joined #openstack-nova09:13
openstackgerritBalazs Gibizer proposed openstack/nova stable/pike: Fix race in delete allocation in ServerMovingTests
*** slaweq_ has quit IRC09:16
*** cshen has joined #openstack-nova09:18
*** slaweq_ has joined #openstack-nova09:23
*** abhishekk has joined #openstack-nova09:24
*** markvoelker has quit IRC09:26
*** yamamoto has joined #openstack-nova09:27
*** gszasz has joined #openstack-nova09:33
*** brault has quit IRC09:36
*** yamahata has quit IRC09:38
*** sdague has joined #openstack-nova09:38
*** lnxnut has joined #openstack-nova09:45
*** brault has joined #openstack-nova09:47
*** artom has joined #openstack-nova09:54
*** trinaths has joined #openstack-nova09:54
*** lnxnut has quit IRC09:56
*** slaweq_ has quit IRC09:56
*** abalutoiu has quit IRC10:00
*** artom_ has joined #openstack-nova10:01
*** hshiina has joined #openstack-nova10:04
*** artom has quit IRC10:05
*** TuanLA has quit IRC10:06
*** edmondsw has joined #openstack-nova10:09
*** edmondsw has quit IRC10:13
*** phuongnh has quit IRC10:17
*** markvoelker has joined #openstack-nova10:23
openstackgerritBalazs Gibizer proposed openstack/nova master: use already loaded BDM in instance.<action>
openstackgerritBalazs Gibizer proposed openstack/nova master: use already loaded BDM in instance.<action> (2)
openstackgerritBalazs Gibizer proposed openstack/nova master: use already loaded BDM in instance.create
*** chyka has joined #openstack-nova10:29
*** yamamoto has quit IRC10:30
gmannalex_xu: ping10:31
*** slaweq_ has joined #openstack-nova10:31
*** udesale has quit IRC10:32
*** chyka has quit IRC10:33
*** mingyu has quit IRC10:34
*** andreas_s_ has joined #openstack-nova10:34
*** edand has quit IRC10:36
*** andreas_s has quit IRC10:37
*** tesseract has quit IRC10:40
*** tesseract has joined #openstack-nova10:44
openstackgerritMerged openstack/nova master: Do not monkey patch eventlet in unit tests
*** gszasz has quit IRC10:50
*** sambetts|afk is now known as sambetts10:50
*** gszasz has joined #openstack-nova10:50
*** yamamoto has joined #openstack-nova10:51
*** lucasagomes is now known as lucas-hungry10:51
*** sree has quit IRC10:53
*** sree has joined #openstack-nova10:53
*** lnxnut has joined #openstack-nova10:53
*** cdent has joined #openstack-nova10:54
*** markvoelker has quit IRC10:55
*** sree has quit IRC10:57
*** spectr-RH has joined #openstack-nova10:59
*** dave-mccowan has joined #openstack-nova11:01
*** spectr has quit IRC11:02
*** lnxnut has quit IRC11:03
*** nicolasbock_ has joined #openstack-nova11:03
openstackgerritBalazs Gibizer proposed openstack/nova master: Moving more utils to ServerResourceAllocationTestBase
openstackgerritBalazs Gibizer proposed openstack/nova master: factor out compute service start in ServerMovingTest
*** nicolasbock_ has quit IRC11:08
openstackgerritBalazs Gibizer proposed openstack/nova master: Test resource allocation during soft delete
*** dave-mcc_ has joined #openstack-nova11:11
*** ociuhandu has quit IRC11:11
*** hamzy has joined #openstack-nova11:12
*** dave-mccowan has quit IRC11:13
*** brault has quit IRC11:14
*** tonygunk has joined #openstack-nova11:16
openstackgerritSean Dague proposed openstack/nova master: test_mount_unmount cleanup
*** tonygunk has quit IRC11:20
*** josecastroleon has quit IRC11:21
openstackgerritSean Dague proposed openstack/nova master: test_mount_unmount cleanup
*** dtantsur|afk is now known as dtantsur11:22
*** smatzek has joined #openstack-nova11:25
*** abhishekk has quit IRC11:35
*** ociuhandu has joined #openstack-nova11:37
*** karts is now known as karthiks11:44
*** tonygunk has joined #openstack-nova11:50
*** markvoelker has joined #openstack-nova11:53
cdentsdague: if you get a chance, could you review: You were involved in some of the discussion on the related bug11:55
*** jaypipes has joined #openstack-nova11:56
jaypipesmorning supernovas11:59
*** lnxnut has joined #openstack-nova12:00
* cdent waves meekly12:02
*** ralonsoh has quit IRC12:02
*** lpetrut has joined #openstack-nova12:03
*** lucas-hungry is now known as lucasagomes12:03
cdentlooks like this long lifed spec about cert validation didn’t get much attention in the spec review, it’s a preposal:
*** sapcc-bot has quit IRC12:04
*** sapcc-bot has joined #openstack-nova12:04
*** david_1 has joined #openstack-nova12:05
*** dgonzalez_ has joined #openstack-nova12:05
*** jpena is now known as jpena|lunch12:06
*** dgonzalez_ has quit IRC12:06
*** david_1 has quit IRC12:06
*** Tom has quit IRC12:08
sdaguecdent: +212:09
*** brault has joined #openstack-nova12:09
*** slaweq_ has quit IRC12:09
*** lnxnut has quit IRC12:11
openstackgerritRadoslav Gerganov proposed openstack/nova master: VMware: serial console log (completed)
openstackgerritRadoslav Gerganov proposed openstack/nova master: Move last_bytes into the path module
*** slaweq_ has joined #openstack-nova12:13
*** Tom_____ has joined #openstack-nova12:14
*** edmondsw has joined #openstack-nova12:16
*** edmondsw_ has joined #openstack-nova12:17
*** manasm has quit IRC12:20
*** edmondsw has quit IRC12:21
*** udesale has joined #openstack-nova12:23
*** hemna__ has joined #openstack-nova12:26
*** markvoelker has quit IRC12:26
*** markvoelker has joined #openstack-nova12:27
*** smatzek has quit IRC12:27
*** smatzek has joined #openstack-nova12:28
openstackgerritMerged openstack/nova-specs master: Return Selection Objects
*** trinaths has left #openstack-nova12:32
*** smatzek has quit IRC12:34
openstackgerritMerged openstack/nova-specs master: Re-propose nested resource providers spec
*** namnh has joined #openstack-nova12:37
bauzasjaypipes: around ?12:38
bauzasjaypipes: I discussed with sahid about the vGPU spec12:39
bauzasjaypipes: and we discussed about a specific question12:39
jaypipesbauzas: yep, go ahead12:39
bauzasjaypipes: say the virt driver is providing an inventory for VGPU:812:39
bauzasjaypipes: then the scheduler would allocate a VGPU:112:40
bauzasjaypipes: if so, the compute service would ask to get a vGPU to the virt driver12:40
bauzasjaypipes: so, it would be the virt driver that would look at which MDEV to use ?12:41
jaypipesbauzas: well, since QEMU/KVM is the only one that understands mdev currently, yes.12:42
jaypipesbauzas: but placement has no concerns about that.12:42
bauzasokay, say I have multiple children RPs (because for example of multple GPUs)12:42
*** kylek3h has quit IRC12:43
jaypipesbauzas: placement deals with UUIDs, of course. so it's up to the virt driver (or the generic device manager in the future once efried hurries up and finishes that work please) to map UUIDs of resource providers to internal names/identifiers of actual devices.12:43
jaypipesbauzas: or pGPU *groups* in the case of Xen, but yes.12:43
bauzasjaypipes: okay, so it would be the virt driver who would mapping in between RP UUIDs and specific virt identifiers ?12:44
bauzassay I have two children UUIDs, we would be passing the one that was allocated to the virt driver, so it would use one specific mdev or PCI device ?12:44
jaypipesbauzas: yep, and that's the update_provider_tree() virt driver method that we've been discussing at the PTG and since.12:45
*** namnh has quit IRC12:45
jaypipesbauzas: again, the scheduler doesn't know or care what the PCI device address or mdev identifier is.12:45
*** namnh has joined #openstack-nova12:45
jaypipesbauzas: it will just be passing a UUID in the allocation request.12:45
bauzasjaypipes: yup I know about that, hence my discussion with sahid12:45
jaypipesbauzas: and the virt driver (or generic device manager in future) is responsible for looking up that UUID to a known device.12:46
jaypipesbauzas: I just wish efried would get off his ass and finish the generic device manager already. sheeesh. ;)12:46
*** trinaths has joined #openstack-nova12:46
*** namnh_ has joined #openstack-nova12:46
*** slaweq_ has quit IRC12:46
bauzasjaypipes: okay I'm cool with that, that's just different from the existing, where we have some specific tracker that passes the specific virt device to the virt driver12:47
jaypipesbauzas: yes, understood.12:47
*** kfarr has quit IRC12:47
jaypipesbauzas: same12:48
bauzasjaypipes: looks like something was misundertood in the ML thread, but I'll leave sahid explain it12:48
jaypipesbauzas: again, it's not that we're not responding on the ML. it's that we have priority items we're trying to complete before sahid's patch series.12:48
bauzasfrom what I understand, for Queens, we would just map in the driver between the virt device (here, mdevs) and the RP UUID, that's it12:49
*** derekh has quit IRC12:49
*** namnh has quit IRC12:50
*** sridharg has quit IRC12:50
*** liverpooler has joined #openstack-nova12:51
*** esberglu has joined #openstack-nova12:51
*** hshiina has quit IRC12:53
*** smatzek has joined #openstack-nova12:53
*** hemna__ has quit IRC12:54
*** rodrigods has quit IRC12:54
openstackgerritOpenStack Proposal Bot proposed openstack/os-vif stable/ocata: Updated from global requirements
*** lyan has joined #openstack-nova12:54
*** sree has joined #openstack-nova12:55
*** slaweq_ has joined #openstack-nova12:57
bauzasjaypipes: follow-up thoughts12:58
bauzasjaypipes: since we create allocations by the scheduler, we could imagine an allocation for VGPU:112:58
gmannNova API meeting in 5 min on #openstack-meeting-4 channel12:59
bauzasjaypipes: then, how the virt driver is knowing that we asked for that specific resource and should therefore use an existing mdev ?12:59
*** eharney has joined #openstack-nova12:59
bauzasbecause we're passing the flavor and we introspect it ?12:59
*** spectr-RH has quit IRC13:00
*** takashin has joined #openstack-nova13:00
gmanntakashin: can you check my comment in  -
gmanntakashin: i feel we should return 'type' always13:01
takashingmann: Thank you for your review.13:01
takashingmann: I will check your comment.13:02
gmanntakashin: thanks. i am reviewing tempest patch also, hope we get all these in soon.13:02
*** mriedem has joined #openstack-nova13:02
*** pcaruana has quit IRC13:02
gmanntakashin: got that while checking the schema where you added 'type' as required param13:03
*** smatzek has quit IRC13:04
*** lbragstad has joined #openstack-nova13:05
*** sridharg has joined #openstack-nova13:07
*** ratailor has quit IRC13:07
*** ratailor has joined #openstack-nova13:08
*** kylek3h has joined #openstack-nova13:08
*** lnxnut has joined #openstack-nova13:08
*** flanders_ has quit IRC13:09
*** krtaylor has joined #openstack-nova13:09
*** pchavva has joined #openstack-nova13:09
jaypipesbauzas: sorry, reading back... electricians at my house13:09
*** hemna_ has joined #openstack-nova13:10
*** smatzek has joined #openstack-nova13:11
*** karthiks is now known as karthiks_afk13:11
mriedemstill looking for a final +2 on this pike regression fix
mriedembauzas: ^ is in your wheelhouse13:12
*** ratailor has quit IRC13:12
*** spectr-RH has joined #openstack-nova13:12
bauzasmriedem: oh coolness13:12
*** jpena|lunch is now known as jpena13:13
*** trinaths has left #openstack-nova13:13
openstackgerritBalazs Gibizer proposed openstack/nova master: Moving more utils to ServerResourceAllocationTestBase
openstackgerritBalazs Gibizer proposed openstack/nova master: factor out compute service start in ServerMovingTest
openstackgerritBalazs Gibizer proposed openstack/nova master: Test resource allocation during soft delete
bauzasmriedem: if you remember, I also have that is related13:15
bauzasI need to rebase it13:15
*** pcaruana has joined #openstack-nova13:15
*** smatzek has quit IRC13:15
*** spectr-RH has quit IRC13:16
bauzasjaypipes: np, my 2nd question is more about how you see the virt driver consuming that specific allocation13:17
*** spectr-RH has joined #openstack-nova13:17
*** lnxnut has quit IRC13:17
bauzasjaypipes: that will be set by the scheduler13:17
bauzasjaypipes: for the moment, we can just do things in the virt driver by introspecting the flavor and see that if we have VGPU:1 in the flavor, we need to hook up a mdev device13:18
*** sree has quit IRC13:18
bauzasjaypipes: but that's not a super clear interface13:18
bauzasnote that the problem is identical for Xen13:18
*** sree has joined #openstack-nova13:18
bauzaswith the slight detail it doesn't use mdevs, of course13:18
sahidbauzas, jaypipes: without talking about mdev, how that is going to work for the physical devices, which is on what you are working i think13:19
mriedemmaciejjozefczyk: question about - it was listed as Partial-Bug fix in the commit message but why? is there more to fix in that bug?13:19
jaypipesbauzas: k, so the way we've been talking about that is that on startup, the generic device manager (or virt driver) would go through the host devices it discovers and populate a ProviderTree object supplied to it by the RT. The ProviderTree allows looking up providers by UUID or by name. So, when creating nodes in the ProviderTree, the device manager / virt driver would create the resource provider with a UUID and a unique name. It would then13:19
jaypipeslook up resource providers by UUID or by name later on13:19
bauzasjaypipes: that I understood13:20
bauzasjaypipes: how we consume that ? by passing the allocation down to the compute ?13:21
jaypipesbauzas: yes, the allocation request contains the resource provider UUID(s) of the providers providing resources for an instance. If the scheduler claimed one VGPU resource against a particular device represented by UUID1, the allocation request will contain UUID1: {resources: {VGPU: 1}}}13:22
*** sree has quit IRC13:22
jaypipesbauzas: and it will be up to the virt driver or generic device manager to look up which device corresponds to which UUID.13:23
gmannsdague: mriedem need your eyes and feedback on policy removal list in this spec -
*** lnxnut has joined #openstack-nova13:23
bauzasjaypipes: so that requires
bauzasjaypipes: if we say we're just going to do a quick POC for Queens with vGPUs, we won't have that yet, so permission to just introspect the flavor in the virt driver ?13:24
jaypipesbauzas: not really, no. The compute host looks up allocations for an instance at the moment by doing a GET /allocations/{consumer_uuid} in the RT13:24
openstackgerritBalazs Gibizer proposed openstack/nova master: use already loaded BDM in instance.<action>
openstackgerritBalazs Gibizer proposed openstack/nova master: use already loaded BDM in instance.<action> (2)
openstackgerritBalazs Gibizer proposed openstack/nova master: use already loaded BDM in instance.create
openstackgerritMatt Riedemann proposed openstack/nova stable/pike: doc: fix flavor notes
jaypipesbauzas: the alternate hosts thing is for retry ability.13:24
bauzasjaypipes: oh that's right13:24
*** catintheroof has joined #openstack-nova13:24
bauzaswe don't need to pass them over RPC13:24
bauzasbecause compute is already looking up them13:24
bauzasokay, so the RT knows the allocation, it can then passes the allocation to the virt driver, right?13:25
bauzasjaypipes: ^13:25
openstackgerritMatt Riedemann proposed openstack/nova stable/pike: Account for compute.metrics.update in legacy notification whitelist
openstackgerritMatt Riedemann proposed openstack/nova stable/ocata: Account for compute.metrics.update in legacy notification whitelist
jaypipesbauzas: yes.13:26
openstackgerritMatt Riedemann proposed openstack/nova stable/newton: Account for compute.metrics.update in legacy notification whitelist
bauzasjaypipes: sahid: I think it's enough for us to be able to assign vGPUs for Queens13:26
mriedemsdague: how do you feel about this backport ?13:27
jaypipesbauzas, sahid: if the virt driver (or generic device manager in the future) wants to store that mapping of mdev identifier to resource provider UUID in a DB table (pci_devices?), cool. If it wants to store it in etcd, cool. An inventory.yaml file on the host? also cool, doesn't matter to me :)13:27
bauzassahid: the virt driver reports how many vGPUs it can assign by filling in the get_inventory() method13:27
mriedemsdague: i think it's ok, it's adding the ability to specify certs when talking to keystone for the os-quota-sets and flavor-access APIs13:27
*** yamamoto has quit IRC13:27
bauzassahid: then at the creation time, the RT would pass the allocated claim (I mean the allocation record) to the virt driver so it would get the RP UUID and the amount to consume13:28
sahidjaypipes: it' good point, since currently libvirt is using PciDevice, so it will be a good transition then13:28
bauzasthen the virt driver would allocate from the pool of physical devices it manages13:28
jaypipessahid: yup13:28
jaypipesbauzas: no, not really... the device will have already been picked by the scheduler. all the virt driver would need to do is plumb the specific device to the guest.13:29
*** slaweq_ has quit IRC13:29
bauzasjaypipes: sahid: I guess the most important matter is that whatever the technical persistence is, it's not provided outside of the virt driver13:29
jaypipesbauzas: and that's the point we're trying to get to. the RT and scheduler do the claiming/allocating resources stuff and the virt driver does the guest plumbing.13:30
bauzasjaypipes: if we pass VGPU:8 as a RC13:30
*** alexchadin has joined #openstack-nova13:30
bauzasjaypipes: then the allocation would be against the root RP13:30
bauzasfor queens I mean13:30
jaypipesbauzas: no, not necessarily.13:30
*** baoli has joined #openstack-nova13:30
bauzasI'm all ears :)13:31
jaypipesbauzas: we're aiming to get n-r-p work done in Queens. so, the allocation would be against one or more child providers (in libvirt, those would be pGPUs, in Xen they would be pGPU groups).13:31
*** ralonsoh has joined #openstack-nova13:32
jaypipesbauzas: so all the virt driver would be responsible for doing is a) looking up physical device information by resource provider UUID and b) doing the necessary guest plumbing for the device (in other words, in libvirt's case, writing the XML snippet information for the device, etc)13:32
bauzasI agree, I just thought we said we could try to provide GPU resources as a global resource class for the node in Queens13:33
bauzasif nested-RPs is already there, then of course we would modify that to just lookup the child RP13:33
*** tbachman has quit IRC13:33
dansmithbauzas: yes that's wht we should do13:34
*** tbachman has joined #openstack-nova13:34
dansmithbauzas: we can expose gpu resources for the compute node right now13:34
*** mingyu has joined #openstack-nova13:34
* efried enters the conversation late13:36
sahidefried: please ping me when you have a moment so we can talk about your work on the generic devices management13:36
efriedjaypipes GDM dig accepted13:36
jaypipesefried: lol :)13:36
efriedsahid Now's good, unless I need to catch up on the ML first.13:36
efriedjaypipes BUT - I've actually been thinking along the lines that, once NRP is in place, there will be no need for such a thing as a GDM.13:37
sahidi did not expected to see you respond so quickly :)13:37
maciejjozefczykmriedem: Yes other patch should be applied to rollback migration when delete is called, similiar to:
maciejjozefczykmriedem: I solved an effect of broken migration, not the source13:38
jaypipesefried: oh, there still will be.13:38
jaypipesefried: there's still a need for discovery of hardware on the host.13:38
efriedvirt driver13:38
sahidefried: do you have some pointers of work in progress?13:38
efriedwhitelisting?  virt driver13:39
jaypipesefried: doing so would probably lead to a lot of dup code.13:39
efriedMapping devices to RPs?  virt driver.13:39
*** mingyu has quit IRC13:39
efriedsahid The only "work in progress" is scribbles on etherpads, which we discussed at the PTG.13:39
sahidefried: yes i was not here, if you can give me the link13:40
efriedsahid It has generally been a drive towards understanding how devices are going to be managed once we go full-bore with placement & resource providers13:40
*** gmann is now known as gmann_sleep13:40
*** slaweq_ has joined #openstack-nova13:40
efriedsahid The goal being to get rid of the existing PCI manager code.13:40
mriedemgmann_sleep: doesn't it seem odd that we still have this filtering code when listing instances to be able to filter by metadata and system_metadata?
mriedemi thought that was a 400 in the API now13:41
efriedjaypipes Potentially duplication among the linuxy hypervisors, I suppose.  I would still expect the code to ultimately run under the auspices of the virt driver.13:42
sahidefried: no transition phase as suggested jaypipes by adding a update_from_inventory() method to the PciManager?13:42
efriedjaypipes So I could see having some shared class above the ComputeDriver base class that provides linuxy impls for devicey methods.13:43
efriedsahid Well, the cores have made their position pretty clear: Investment in the existing PCI manager is going to be very limited.13:44
*** cleong has joined #openstack-nova13:44
*** armax has joined #openstack-nova13:45
*** burt has joined #openstack-nova13:45
*** tonygunk has quit IRC13:46
efriedsahid That said, I think it may be possible to get very close to what you want just using NRP with traits and careful modeling, which is happening in Queens.13:46
efriedsahid Fancy use cases like (anti)affinity won't work yet.13:46
jaypipesefried: yeah, we're gonna need to have some sort of transitionary plan anyway...13:46
efriedjaypipes Oh, like sahid is saying with update_from_inventory() - presumably a RT method that peels dev info out of the virt inventory and flushes it back to the PCI manager?13:47
jaypipesefried: chatting with dansmith and bauzas, I'm cool with trying to get single-inventory VGPU support pushed for Queens if we miss nested resource provider targets. That would means no support for multiple GPU types on a single compute node, but would at least get some rudimentary VGPU support13:47
*** sree has joined #openstack-nova13:48
dansmithjaypipes: multiple gpu types can be handled with aggregates until we have richer support, and at least one customer has told me that's fine, FWIW13:49
bauzasefried: I just wanted to clarify how we would translate a specific {'VPGU:1"} request into something very virt-specific13:49
*** crushil has quit IRC13:49
bauzasheh VGPUs even13:49
sahidit seems that XenServer provides an abstraction which makes easy what you want to achieve13:49
sahidbut that does not look reasonable for libvirt13:49
efriedbauzas So the only way we're going to translate an allocation to a specific device is if the RP is modeled as representing that specific device.13:50
sahidXenServer is doing the managmenet ofr the devices, libvirt is using the PciManager13:50
bauzasefried: sahid: dansmith: I just feel we need code so we could chat on the details13:50
bauzasreporting the inventory for those resource classes is easy13:51
efriedThe virt driver will ultimately be responsible for designing that model, maintaining that mapping, and providing the appropriate RPs and inventory to placement (via get_inventory, or update_provider_tree, or whatever it winds up being)13:51
dansmithbauzas: you're asking how the virt driver knows a thing has been requested?13:51
bauzasdansmith: yup, correct13:51
dansmithbauzas: in the simple case it can just look at extra_specs on instance.flavor13:51
bauzasdansmith: jaypipes told me RT looks up the allocation13:52
bauzasdansmith: yeah that was my initial approach13:52
dansmithbauzas: we have a routine that can parse the flavor and give you a merged resource view13:52
bauzasthat would be an easy thing then13:52
bauzasokay, I need coding then13:52
dansmithor look at the allocation, yeah, but plumbing that from rt to virt will take a little work13:52
efriedThe virt driver will be able to see the allocation, no?13:52
bauzasyeah, for a POC, introspecting the flavor seems the quickiest path13:52
dansmithyou will need the allocation once we have n-r-p or traits though13:52
mriedemthese -1s from zuul messing up my dashboard is messing up my life13:53
bauzasefried: it requires a new interface which we don't have yet13:53
dansmithefried: it could fetch it itself, yeah, but better if it didn't I think13:53
*** gouthamr has joined #openstack-nova13:53
bauzasdansmith: yeah, I'm not a fan of the virt driver calling placement13:53
dansmithbauzas: yeah, a good first step would be figuring out the best way to tell the virt driver about the allocation13:53
dansmithbauzas: yep13:53
efriedTalking like something in the Instance object?13:53
bauzasanyway, /me coding then13:53
dansmithefried: no13:53
efriedparam to spawn?13:54
dansmithefried: maybe just a param in spawn13:54
efriedbtw, virt will at some point be calling placement.13:54
efriedNot necessarily in spawn13:54
bauzasdansmith: jaypipes was thinking of a specific interface for nested RPs13:54
bauzasdansmith: something like update_my_provider_tree()13:54
dansmithefried: you're saying that because of virt drivers reporting resource?13:54
efriedbut in init_host, and/or get_inventory, and/or update_provider_tree, whatever - to set up the RPs and whatnot.13:55
efriedYeah, for example, to know whether a RP has been created yet.13:55
dansmithefried: that should be abstracted by the compute manager, not virt calling placement directly13:55
dansmithbauzas: yep makes sense13:55
*** manasm has joined #openstack-nova13:55
efrieddansmith Yeah, I suppose it could be.13:55
bauzasI second dansmith on not having placement calls from the driver13:55
efriedThis means virt is always responsible for producing RP UUIDs.13:56
dansmithefried: I think we should shoot for that goal, and if there's some compelling reason to break that rule, then we can discuss it13:56
efrieddansmith Dig.13:56
bauzasmriedem: (Zuul, Jenkins) is now the tuple to care13:56
jaypipesdansmith: right. RT constructs the known ProviderTree. passes it to the virt driver's update_provider_tree() method, virt driver adds, removes, changes inventory and traits for resource providers in the tree, RT then saves any of those changes to placement.13:56
dansmithjaypipes: yes, that13:56
efriedbauzas What is it you're going off to code now?13:56
bauzasefried: I'll just update libvirt to pass vGPU resources and lookup the flavor extraspecs for plumbing a mdev13:57
efriedbauzas Pass vGPU resources from get_inventory?13:57
*** crushil has joined #openstack-nova13:58
efriedbauzas and look up flavor extra specs from spawn?13:58
bauzasefried: yup13:58
efriedbauzas Cool.  sahid ^13:58
*** felipemonteiro_ has joined #openstack-nova14:00
*** stephenfin has quit IRC14:00
*** Suramya has joined #openstack-nova14:01
*** stephenfin has joined #openstack-nova14:01
*** felipemonteiro__ has joined #openstack-nova14:01
sahidthat seems a bit archaic - get_inventory to retourn ResourceClass.GPU and then ? you are going to hack the virt driver to read a flavor in the spawn phase? add a conditon that a vgpu, (which kind?, what numa?) and then update the XML14:02
*** yamamoto has joined #openstack-nova14:03
*** yamamoto has quit IRC14:03
bauzasthat's basically my intent, yes :)14:03
*** hongbin has joined #openstack-nova14:03
dansmith460 uses of instance.flavor in the virt drivers today14:03
dansmithnot exactly a hack14:03
*** felipemonteiro_ has quit IRC14:05
sahidIt is totally a hack and it's going to provide a very basic support14:05
sahidlibvirt have a pci device manager, and it's a bad idea to just ignore it14:05
dansmithyep, it's a first step to get us basic support, as stated above14:05
sahidbut it's a hack, no need ot RP or anything to provide that basic support14:07
*** edmondsw_ is now known as edmondsw14:07
openstackgerritRodolfo Alonso Hernandez proposed openstack/os-vif master: Add support for Windows network commands
*** archit has joined #openstack-nova14:10
openstackgerritEric Berglund proposed openstack/nova master: WIP(5): PowerVM driver: ovs vif
openstackgerritRodolfo Alonso Hernandez proposed openstack/os-vif master: Add Port Profile info to VIF objects Linux Bridge plugin
*** slaweq_ has quit IRC14:14
*** READ10 has joined #openstack-nova14:14
*** smatzek has joined #openstack-nova14:15
openstackgerritRodolfo Alonso Hernandez proposed openstack/os-vif master: Add VersionedObjectPrintable mixin
*** smatzek has quit IRC14:15
*** smatzek has joined #openstack-nova14:15
*** kenperkins has joined #openstack-nova14:18
openstackgerritTakashi NATSUME proposed openstack/nova master: Fix test_get_volume_config method
*** acormier has joined #openstack-nova14:19
*** takashin has left #openstack-nova14:19
*** slaweq_ has joined #openstack-nova14:20
openstackgerritBalazs Gibizer proposed openstack/nova master: factor out compute service start in ServerMovingTest
openstackgerritBalazs Gibizer proposed openstack/nova master: Test resource allocation during soft delete
*** acormier has quit IRC14:20
openstackgerritMatt Riedemann proposed openstack/nova master: Remove unused get_all_instance_*metadata methods
openstackgerritMatt Riedemann proposed openstack/nova master: Remove old compat code from servers ViewBuilder._get_metadata
openstackgerritMatt Riedemann proposed openstack/nova master: Stop joining on system_metadata when listing instances
openstackgerritMatt Riedemann proposed openstack/nova master: Remove system_metadata loading in Instance._load_flavor
*** kenperkins has quit IRC14:23
dansmithjaypipes: bauzas: stephenfin: easy +W on this cleanup:
jaypipesdansmith: finito14:25
dansmithjaypipes: ${thanks_in_some_fancy_language}14:26
*** mdnadeem has quit IRC14:27
*** eharney has quit IRC14:29
*** gbarros has joined #openstack-nova14:30
*** udesale has quit IRC14:32
*** pcaruana has quit IRC14:33
mriedemwhew, got my aussy visitor visa14:35
dansmithum I believe it's "aussie"14:36
*** cfriesen has joined #openstack-nova14:38
*** spectr-RH has quit IRC14:41
*** smatzek has quit IRC14:42
*** spectr has joined #openstack-nova14:42
*** smatzek has joined #openstack-nova14:43
*** smatzek has quit IRC14:44
*** eharney has joined #openstack-nova14:45
*** pcaruana has joined #openstack-nova14:45
*** sree has quit IRC14:47
*** sree has joined #openstack-nova14:48
*** alexchadin has quit IRC14:50
*** sdague has quit IRC14:50
*** sree has quit IRC14:52
*** slaweq_ has quit IRC14:53
*** tonygunk has joined #openstack-nova14:55
*** Nel1x has joined #openstack-nova14:55
*** smatzek has joined #openstack-nova14:55
cdentanybody able to sail this gabbi test addition, already has jay’s +2:
*** lnxnut_ has joined #openstack-nova14:56
*** awaugama has joined #openstack-nova14:56
*** slaweq_ has joined #openstack-nova14:56
gibicdent: looking...14:57
*** smatzek_ has joined #openstack-nova14:57
gibidansmith was faster14:58
cdentthanks danpawlik14:58
cdentoh noes! thanks dansmith14:58
*** lnxnut has quit IRC14:58
danpawlikcdent: lol14:58
* dansmith bows14:58
danpawlikcdent: I was wondering why you thanks me :D14:59
cdentdanpawlik: I’m sure you’ve done something worth being thanked for? Thanks for existing.14:59
gibidanpawlik: now you have to do someting for cdent :)14:59
*** smatzek has quit IRC15:00
*** smatzek_ is now known as smatzek15:00
danpawlikgibi: In that case, Im going to work!15:00
*** diegows has quit IRC15:00
danpawlikcdent: :D15:00
*** diegows has joined #openstack-nova15:01
*** edleafe- has joined #openstack-nova15:03
*** sahid has quit IRC15:03
*** yamamoto has joined #openstack-nova15:04
*** sahid has joined #openstack-nova15:05
*** edleafe has quit IRC15:06
*** Oku_OS is now known as Oku_OS-away15:08
melwittmriedem: I started looking at the ceph job last night and something weird is happening where keystone can't start. still researching how to fix it15:10
mriedemok, but it's still a smoldering pile despite that15:11
*** manasm has quit IRC15:11
*** yamamoto has quit IRC15:13
*** rcernin has quit IRC15:13
*** sdague has joined #openstack-nova15:14
*** andreas_s_ has quit IRC15:16
*** Suramya has quit IRC15:19
openstackgerritMatt Riedemann proposed openstack/nova master: Modernize set_vm_state_and_notify
openstackgerritMatt Riedemann proposed openstack/nova master: Pass migration from API to conductor for evacuate
*** gbarros has quit IRC15:20
*** brault has quit IRC15:20
*** lajoskatona has left #openstack-nova15:23
*** erlon has joined #openstack-nova15:25
*** psachin has quit IRC15:25
melwittmriedem: I see now, the keystone thing must have been from getting zuul'd. the failures I see now are from a 404 from cinder api test, like you said last time15:26
*** burt has quit IRC15:26
smcginnismelwitt: Anything you need us to look at?15:26
mriedemthere was a long-standing change in cinder for ceph,15:27
mriedemthat jbernard thought would help stabilize some things15:27
*** slaweq_ has quit IRC15:27
*** slaweq_ has joined #openstack-nova15:28
*** pcaruana has quit IRC15:28
mriedemand there was an alternative proposed
*** tesseract has quit IRC15:29
mriedemas far as i can tell, something something locks15:29
*** gyee has joined #openstack-nova15:29
mriedemthere are some newer volume snaphot tests in tempest that are failing at a pretty high rate, globally, so those are probably not helping the situation15:30
melwittsmcginnis: not yet sure. there's a couple of cinder api tempest tests failing on only our ceph job (for a long time) and I'm starting to look at it15:30
mriedemand by "the situation", yes, i mean this guy
*** Sukhdev_ has joined #openstack-nova15:31
openstackgerritMerged openstack/nova-specs master: List/show all server migration types
*** igordc has quit IRC15:32
openstackgerritMatt Riedemann proposed openstack/nova master: What is the meaning of....recreate?
*** igordc has joined #openstack-nova15:33
*** yamahata has joined #openstack-nova15:33
*** belmoreira has quit IRC15:35
mriedemsdague: dansmith: grenade creates a vm and leaves it running through the upgrade so it can ping it on the other side right?15:38
*** smatzek has quit IRC15:38
dansmithmriedem: yeah15:39
mriedemok, i guess nova supports accessible upgrades then!15:39
mriedemyay more tags15:39
*** mdavidson has quit IRC15:39
*** smatzek has joined #openstack-nova15:39
dansmithis that what that tag means?15:39
dansmithbecause there was "no-impact-upgrades" proposed at one point that covered that bit15:40
cdentnobody is ever sure15:40
cdentsmcginnis tried to kill the accessible tag but then someone came along and said, but wait, nova and cinder match that one15:40
*** prometheanfire has left #openstack-nova15:40
mriedemthere are like 5-6 upgrade tags15:40
mriedemi didn't know what a "controlled resource" was in reading the description,15:41
mriedemassumed it was a VM,15:41
mriedembut without examples in the tag descriptions it's hard to know what the author intended15:41
mriedemi know what it means to the ATF...15:42
cdentin nova’s case it does mean that running vm survives and can be worked with (accessed) through an upgrade15:42
mriedemso what is "no-impact-upgrades"?15:42
cdentbut yeah, it’s entirely unclear15:42
cdentdunno! :)15:42
cdentI don’t like tags15:42
mriedemi don't either really, but probably for slightly different reasons15:42
cdenti cannot say, since I dont know why you don’t like15:43
mriedemno one in the dev teams probably thinks about tags15:43
*** smatzek_ has joined #openstack-nova15:43
mriedeme.g. there is no bi-annual audit15:43
mriedemi do'nt know what stick or carrot exists for us to care about tags15:43
*** smatzek has quit IRC15:43
mriedemand i do'nt know if people consuming openstack are really giving them much weight, but maybe they are depending on project,15:44
mriedeme.g. if there are 3 monitoring type projects, you'd care about what tags are applied to those15:44
cdentthe stick or carror aspect is part of my concerns15:44
*** chyka has joined #openstack-nova15:45
*** trinaths has joined #openstack-nova15:45
*** tuanla_fujitsu has joined #openstack-nova15:45
*** smatzek_ has quit IRC15:46
sdaguemriedem: a bunch of them were going to be purged15:46
sdaguemriedem: there is a distinction somewhere about 0 api downtime15:47
sdaguewhich is important for things like keystone15:47
sdaguebecause, if keystone is out, operations fail randomly15:47
*** penick has joined #openstack-nova15:48
*** slaweq_ has quit IRC15:49
*** smatzek has joined #openstack-nova15:49
*** slaweq_ has joined #openstack-nova15:50
*** slaweq_ has quit IRC15:50
*** smatzek_ has joined #openstack-nova15:51
*** slaweq_ has joined #openstack-nova15:51
mriedemi think that's the zero-downtime-upgrade one15:51
mriedemthe zero-impact-upgrade one is about performance during upgrade, as my understanding15:51
mriedemso there are subtle differences15:51
dansmithyeah something15:51
mriedemreplied to the ML thread on this,15:51
mriedembut i'd think zero-impact-upgrade re perf would be something like strain on the network doing live migrations while upgrading compute hosts15:52
*** edleafe- is now known as edleafe15:52
mriedembut i'm not sure how nova would really have anything to do with that, it seems like a deployment topology decision15:52
mriedemlet's put that tag on openstack public clouds :)15:53
*** smatzek has quit IRC15:54
mriedemonto another topic, sounds like cern would like to be able to pass user_data to rebuild...15:54
*** smatzek_ has quit IRC15:54
mriedemsounds like they rely on rebuild pretty heavily15:55
*** slaweq_ has quit IRC15:55
*** smatzek has joined #openstack-nova15:57
*** lucasagomes is now known as lucas-afk15:57
*** penick has quit IRC15:58
*** smatzek_ has joined #openstack-nova16:00
cfriesendoes anyone know why AggregateImagePropertiesIsolation and AggregateInstanceExtraSpecsFilter behave differently?  It's confusing...16:01
*** smatzek has quit IRC16:01
*** smatzek_ has quit IRC16:03
*** smatzek has joined #openstack-nova16:03
*** smatzek has quit IRC16:03
*** smatzek has joined #openstack-nova16:04
*** tuanla_fujitsu has quit IRC16:04
bauzascfriesen: the reason is "historical"16:11
bauzasin other words, two efforts made by two teams16:11
*** trinaths has left #openstack-nova16:13
cfriesenbauzas: why am I not surprised. :)16:15
bauzasthe problem is that changing a filter behaviour is something tricky16:16
bauzasI'd prefer providing a new filter that would do both image and flavor checks with the same behaviour, and in the meantime deprecate the old two filters16:17
bauzasso we would be super clear that the behaviour is changing16:17
bauzaskeep those filters for a couple of releases, and then remove them from the tree16:17
bauzasif people want to keep them out-of-tree, I'm fine16:17
bauzascfriesen: fancy proposing that ? :)16:18
bauzasafter all, it's just filters16:18
*** hemna_ has quit IRC16:18
cfriesenbauzas: Will propose it internally.  What do you think of my proposal for  to have the "strictness" of the isolation be scoped to individual keys?16:19
mriedemi thought that's what the new psec was16:19
mriedemyeah that one16:19
cfriesenmriedem: they're just proposing adding a new boolean flag on either the image or flavor to say the matching is strict....but that doesn't factor in that the behaviour of the two filters is different16:20
dansmithedleafe: are you revising your alt hosts thing?16:20
cfriesenmriedem: and I think it'd make more sense to allow strict matching on a per-key basis, specified in the aggregate metadata16:20
edleafedansmith: which alt hosts thing?16:20
edleafedansmith: yeah, I'm working through the whole series to incorporate the changes to the Selection object based on the spec changes16:21
dansmithedleafe: okay cool16:21
bauzascfriesen: well, I think I said "meh" in my comment16:23
bauzascfriesen: so, basically, I'm not opiniated16:23
bauzascfriesen: I tend to avoid having filter behaviours driven by keys16:23
bauzasbut looks like we need to be pragmatic16:23
bauzasanother option could be a config option, but that would be worst I think for interop16:24
bauzasbecause two clouds would behave differently16:24
*** gbarros has joined #openstack-nova16:24
bauzascfriesen: so, honestly, maybe a key is okay16:25
bauzasI dunno, I need more time to think about that16:25
cfriesenbauzas: I was thinking that we might want to be strictly isolationist for some keys but not for others (as opposed to the all-or-nothing that the current spec proposes)16:25
bauzaswhen you say "isolationist for a specific *key*", you mean either an aggregate extraspec key for matching the flavor, or an image property?16:27
bauzascfriesen: ^16:27
*** penick has joined #openstack-nova16:28
bauzascfriesen: I just wonder how you would express that in the aggregate metadata16:28
bauzasbecause of the k=v pair16:28
cfriesenbauzas: I was thinking that the aggregate key could be something like '{"strict:os": "windows"}', in which case only instances with image property or flavor extra-spec of "os:windows" would match16:29
cfriesenbasically just a "strict:" namespace on the aggregate key16:29
bauzasnamespacing ? I don't like that16:29
bauzaswe already namespace keys AFAIKK16:29
cfriesenwe namespace them on the flavor/image, but not on the aggregate currently I think16:30
bauzasnevermind, we namespace the image properties or the flavor specs16:30
bauzasyeah that16:30
bauzascfriesen: the problem is that if you do that, you change the API16:30
cfriesenbauzas: the nice thing about that is that it works with existing flavors/images16:30
cfriesenbauzas: we're talking about a new filter anyway16:31
bauzascfriesen: say I already have aggregates that are tagged and filters16:31
bauzascfriesen: how can that work in an upgrade way?16:31
bauzasyou would default no namespace to the current behaviour ?16:31
cfriesenbauzas: yes16:31
bauzascfriesen: honesty, I don't like that16:31
cfriesenin the existing filters16:31
bauzasyou can create as many aggregates as you want16:32
cfriesenin the new filter we'd want consistent behaviour for both image/flavor16:32
bauzasand a host can be part of 1:N aggs16:32
bauzasso, if you need strict isolation for only a couple of keys, why just not define two aggregates, one containing keys with no strictness, and the other with keys needing to be strict ?16:32
*** larainema has quit IRC16:32
cfriesenbauzas: with strict matching you'd need the flavor/image to match all the strict keys from all the aggregates the host is in16:32
cfriesenbauzas: you're thinking a "strict-match" boolean flag on the aggregate?  yeah, that could work.16:33
bauzasI'm just talking of the current proposal16:34
bauzashe proposes to add new keys that are global per-aggregate16:34
bauzascfriesen: in that case, if you need some keys with strict isolation, and some with not, just define two aggregates and only apply the new metadata tag image_strict_isolation=True to the aggregate containing the keys you want to be strict16:35
mriedemdansmith: looking at - it occurs to me that if prep_resize fails, i don't think we ever set the migration status to 'failed'16:36
mriedemdansmith: which i think is just a latent bug16:36
mriedemif resize_instance fails it will, but we might not get that far16:37
cfriesenbauzas: I guess.   Although AggregateInstanceExtraSpecsFilter doesn't do isolation currently, and AggregateImagePropertiesIsolation doesn't ensure that what you specify in the image is present in the aggregate.16:38
mriedemdansmith: oh i know why - because we never had the migration before that point, because the RT always created it16:38
cfriesenbauzas: hence the rationale for a new filter with common behaviour16:39
*** jgriffith is now known as Guest7311116:39
*** jgriffit1 is now known as jgriffith16:39
dansmithmriedem: right, because _prep_resize's  with resize_claim would do that right?16:39
*** jgriffith_ has joined #openstack-nova16:39
mriedembut now you're passing down a migration record and not handling errors16:39
openstackgerritRodolfo Alonso Hernandez proposed openstack/os-vif master: Add support for Windows network commands
dansmiththe bottom one if this series was just kicked out of the gate anyway, so I'll rebase and freshen since we're like 8 hours from merge anyway16:42
*** felipemonteiro__ has quit IRC16:44
bauzascfriesen: I don't disagree16:46
*** trinaths1 has joined #openstack-nova16:46
mriedemdansmith: still going through this if you want to hold up16:46
mriedeminto the conductor stuff now16:46
cfriesenbauzas: will comment on the review16:46
dansmithmriedem: sure16:46
*** gbarros has quit IRC16:47
*** dtantsur is now known as dtantsur|afk16:47
* bauzas drops ofrf16:48
cfriesenbauzas: it occurs to me that combining flavor extra-specs and image properties is tricky...some of that logic is way down in the virt code.16:48
*** tonygunk has quit IRC16:50
mriedemdansmith: ok done :)16:53
dansmithmriedem: I can't wait to see what gifts you have left for me16:54
mriedemthey are bountiful16:54
mriedemi'm going to go pat myself on the back with lunch16:54
*** sambetts is now known as sambetts|afk16:55
mriedemjaypipes: cdent: fyi
cdentmriedem: mlph16:57
cdentthis shit is too confusing16:57
*** Swami has joined #openstack-nova16:58
*** jpena is now known as jpena|off16:58
*** yamahata has quit IRC17:08
mriedemcdent: yeah, there are like 10 things that happen outside the scenes of everything...17:10
*** lpetrut has quit IRC17:10
*** hemna_ has joined #openstack-nova17:13
*** manasm has joined #openstack-nova17:18
*** trinaths has joined #openstack-nova17:18
efriedjaypipes ( cdent ) I finished reviewing the series starting at -- is there anything else to look at for NRP or related at the moment?17:18
openstackgerritRodolfo Alonso Hernandez proposed openstack/nova-specs master: Network bandwitdh resource provider
cdentefried: have you seen jay’s orm removal stack?17:19
cdentit’s tangentially related17:19
efriedWhat's an orm/17:19
efried(But I guess not)17:19
efriedcdent rgr17:21
*** trinaths1 has quit IRC17:21
cdentefried: you’re already aware of alex’s traits work I think?17:22
efriedcdent Ah, sort of, but need to plug into it.  Got a starter patch?17:22
cdentit’s currently merge conflict, but:
efriedAmago eat, will dig in after.17:24
*** andrewbogott has left #openstack-nova17:25
*** efried is now known as efried_nomnom17:25
* cdent now wants sushi17:26
cfriesenquestion for someone with more sqlalchemy-foo than I....will the soft_delete() call at affect both the "aggregate" table and the "aggregate_hosts" table?  Or just the "aggregate" table?17:28
*** ralonsoh has quit IRC17:30
*** markvoelker has quit IRC17:30
*** markvoelker has joined #openstack-nova17:30
*** trinaths1 has joined #openstack-nova17:33
*** lbragstad has quit IRC17:34
*** trinaths has quit IRC17:34
*** felipemonteiro_ has joined #openstack-nova17:35
openstackgerritpriyaduggirala proposed openstack/nova master: Rename parameters in call() of nova/image/
*** felipemonteiro__ has joined #openstack-nova17:36
*** smatzek has quit IRC17:37
*** smatzek has joined #openstack-nova17:38
openstackgerritMerged openstack/nova master: Read from console ptys using privsep.
*** felipemonteiro_ has quit IRC17:40
openstackgerritMerged openstack/nova stable/ocata: Fix 500 if list servers called with empty regex pattern
*** smatzek_ has joined #openstack-nova17:40
*** esberglu has quit IRC17:41
*** manasm has quit IRC17:42
openstackgerritDan Smith proposed openstack/nova master: Make allocation cleanup honor new by-migration rules
openstackgerritDan Smith proposed openstack/nova master: Pre-create migration object
openstackgerritDan Smith proposed openstack/nova master: Revert allocations by migration uuid
openstackgerritDan Smith proposed openstack/nova master: Refactor resource tracker to account for migration allocations
openstackgerritDan Smith proposed openstack/nova master: Make migration uuid hold allocations for migrating instances
openstackgerritDan Smith proposed openstack/nova master: Make live migration hold resources with a migration allocation
*** smatzek has quit IRC17:42
*** esberglu has joined #openstack-nova17:42
*** esberglu has quit IRC17:42
*** smatzek_ has quit IRC17:42
*** esberglu has joined #openstack-nova17:42
*** trinaths1 has left #openstack-nova17:43
*** smatzek has joined #openstack-nova17:44
*** gjayavelu has joined #openstack-nova17:44
*** trinaths has joined #openstack-nova17:44
*** yamahata has joined #openstack-nova17:44
*** jgriffith has quit IRC17:45
mriedemcfriesen: the latter17:45
mriedemthere is no such thing as cascading soft deletes17:45
*** jgriffith_ is now known as jgriffith17:45
mriedemcfriesen: you could write a simple db api unit test to recreate that17:46
*** smatzek_ has joined #openstack-nova17:46
*** smatzek__ has joined #openstack-nova17:48
*** smatzek has quit IRC17:48
*** itlinux has joined #openstack-nova17:49
*** acormier has joined #openstack-nova17:50
*** smatzek_ has quit IRC17:50
*** smatzek__ has quit IRC17:51
*** trinaths has left #openstack-nova17:51
cfriesenmriedem: I think that our online_data_migration will end up leaving a bunch of entries in the "aggregate_hosts" table after a migration....though properly written code shouldn't care.17:52
cfriesen /s/migration/upgrade17:52
*** smatzek has joined #openstack-nova17:53
mriedemwrite a test to show that17:53
*** itlinux has quit IRC17:53
*** smatzek has quit IRC17:54
*** smatzek has joined #openstack-nova17:55
*** trinaths has joined #openstack-nova17:57
dansmithmriedem: are you sure that dest_host is the ip in the case of cold migration?
*** lbragstad has joined #openstack-nova17:57
dansmithI was surprised to find this was true for live migration a couple weeks ago when debugging a real thing,17:58
dansmithbut I'm not sure it's the same for cold migration17:58
openstackgerritMatt Riedemann proposed openstack/nova master: Only query BDMs once in API during rebuild
mriedemdansmith: yeah, that's what the RT sets it to17:58
mriedemdoes live migration even set the migration.dest_host anywhere?17:59
melwittmriedem: it looks like the ceph package from the pike uca pulls in python-pastedeploy which gets install in parallel with a pip installed PasteDeploy package and keystone can't start because it has trouble referencing when there are multiple? (see last comment on
dansmithactually we don't use moveclaim for live migration right?17:59
mriedemdansmith: correct18:00
dansmithmaybe I had that reversed18:00
mriedemthat's why live migration creates the migratoin in conductor18:00
dansmithI think it's wrong for us to be setting the ip in that host field18:00
dansmithin a "it's wrong but I'm not going to fix it any time soon" sort of way18:00
mriedemdansmith: heh agree that it's surely confusing as hell18:00
*** MasterOfBugs has joined #openstack-nova18:00
mriedemhence my putting a patch to comment on those damn fields18:01
mriedemb/c i have to re-look it up every time18:01
clarkbmelwitt: on ubuntu (where uca is a thing) the deb packages should install to /usr and pip installed things to /usr/local. So if you get the order sorted out I expect it would just work18:02
mriedemmelwitt: weird, is there an ubuntu bug report?18:03
mriedemwe could ask jamespage about a fix18:03
mriedemmelwitt: alternatively we could workaround that maybe in the devstack-plugin-ceph repo by monkeying with the packages18:03
*** itlinux has joined #openstack-nova18:04
*** ijw has joined #openstack-nova18:04
mriedemheh, wtf, the compute rpcapi method for reboot_instance has a 'block_device_info' parameter that is always None from the API18:08
*** lpetrut has joined #openstack-nova18:09
*** Sukhdev_ has quit IRC18:11
*** itlinux has quit IRC18:12
openstackgerritJay Pipes proposed openstack/nova master: rp: move RP._set_traits() to module scope
openstackgerritJay Pipes proposed openstack/nova master: rp: remove CRUD operations on Inventory class
openstackgerritJay Pipes proposed openstack/nova master: rp: streamline InventoryList.get_all_by_rp_uuid()
openstackgerritJay Pipes proposed openstack/nova master: rp: remove dead code in Allocation._create_in_db()
openstackgerritJay Pipes proposed openstack/nova master: rp: remove ability to delete 1 allocation record
openstackgerritJay Pipes proposed openstack/nova master: rp: fix up AllocList.get_by_resource_provider_uuid
openstackgerritJay Pipes proposed openstack/nova master: rp: rework AllocList.get_all_by_consumer_id()
openstackgerritJay Pipes proposed openstack/nova master: rp: remove _HasAResourceProvider mixin
*** jdwidari has joined #openstack-nova18:15
*** jdwidari has quit IRC18:15
*** tonygunk has joined #openstack-nova18:16
*** jdwidari has joined #openstack-nova18:18
*** jdwidari has quit IRC18:18
cdentjaypipes: is that orm-removal stuff basically ready for review, or are you still working through the kinks? i’m out of juice to review it now, but tomorrow morning might be good18:19
*** edmondsw has quit IRC18:20
jaypipescdent: ready for review, thanks.18:20
jaypipescdent: even doing three or four of the bottom ones would be useful.18:21
jaypipesthey're pretty smallish patches.18:21
*** jdwidari has joined #openstack-nova18:21
*** ociuhandu has quit IRC18:21
jaypipescdent: fwiw, the HEAD of that series gets the file to the point where I can do the root_provider_id/parent_provider_id fields (instead of the root_provider_uuid/parent_provider_uuid fields) that dansmith has asked for on the nested series.18:22
jaypipesthat dansmith has requested and that I agree with, btw18:22
mriedemthat reminds me,18:23
mriedemandreykurilin: is there a devstack plugin for rally to make it easy to get an env setup?18:23
*** edmondsw has joined #openstack-nova18:23
mriedemoooo there is
mriedemdoes it work?18:24
andreykurilinmriedem: sure18:24
mriedemwhat is the default backend when doing that? like do i need to also have ceilometer enabled?18:24
andreykurilinmriedem: yes. it is used in all our gates18:24
andreykurilinhm... ceilometer? rally has nothing related to ceilometer except test plugins for it. the default database backend is sqllite18:25
mriedemok yeah i was wondering about a db, i don't know much about rally, and i thought there was something to do with configuring it to send metrics collection to ceilometer18:25
mriedemi'm looking for something simple to setup so i can compare test runs before and after some nova patch is applied18:26
mriedemor, comparing scheduling times between ocata and pike18:26
mriedemfor example18:26
*** edmondsw has quit IRC18:28
mriedemi guess i should shut up and just read
andreykurilinmriedem: so rally is quite simple and doesn't require any additional services to be installed. All the results are stored in the sql-dabase. One rally installation can check different clusters. The latest clients work perfectly with older releases of openstack. For example we have regular job which launch the scenarios at mitaka release18:29
mriedemok, i'm mostly interested in scheduling performance between ocata and pike,18:29
*** eharney has quit IRC18:29
mriedemi only have a single node devstack, so i'd have to get creative with mocking out computes18:30
*** eharney has joined #openstack-nova18:30
*** sean-k-mooney has joined #openstack-nova18:30
mriedemi know devstack has some flags for mocking out fake compute services when using the fake virt driver, but not sure if that works anymore with the tiered conductor stuff in devstack18:31
mriedemi could run it in singleconductor mode to workaround that18:31
*** efried_nomnom is now known as efried18:32
*** vvargaszte has joined #openstack-nova18:32
openstackgerritMerged openstack/nova master: Log consumer uuid when retrying claims in the scheduler
openstackgerritMerged openstack/nova master: [placement] gabbi tests for shared custom resource class
andreykurilinmriedem: btw, I know some guys from huawei who are working on inner performance testing of openstack(as far as I know). I'll contact them about the hardware for such testing as you need18:38
mriedemandreykurilin: are they in toronto?18:39
mriedemhongbin's team?18:39
andreykurilinmriedem: they are from China18:39
mriedemwell, if you can get answers on internal openstack perf testing at huawei faster than i can, kudos to you sir18:40
andreykurilinmriedem: heh. At least I'll try, but it will happen tomorrow, since they are slipping now18:41
mriedemit's a chinese national holiday this week too18:41
openstackgerritpriyaduggirala proposed openstack/nova master: Rename parameters in call() of nova/image/
melwittclarkb: sorry, was on a call. what did you mean by get the order sorted out between the uca package being installed and the pip package being installed?18:42
melwittmriedem: yeah, there's no ubuntu bug I don't think, because I think this is a unique problem to having devstack install PasteDeploy via pip and then the installation of ceph packages in the devstack plugin installing the uca packages18:43
melwittI thought the same, that I could probably do something like try uninstalling the uca python-pastedeploy package in the ceph devstack plugin and see if that would work18:44
clarkbmelwitt: basically make sure that apt-get happens first, then pip install, then keystone start18:44
melwittclarkb: ah, okay thanks. I'll see if that's possible in devstack plugin land, if there's a hook point for plugins that occurs before the pip installs that devstack does18:45
andreykurilinmriedem: oh... yes, I forgot about it18:46
melwittbecause the devstack ceph plugin is doing the apt-get installs18:46
*** eharney has quit IRC18:47
mriedemmelwitt: should be in this phase
mriedemcheck the logs and see if that happens before or after the pip install happens18:48
*** edmondsw has joined #openstack-nova18:48
melwitthm, yeah ... so it should be doing the right thing already. looking18:49
melwittsays "Requirement already satisfied: PasteDeploy===1.5.2 in /usr/lib/python2.7/dist-packages" which means it skipped the pip install18:54
melwitt(I think)18:54
melwittso how is keystone not finding it18:54
openstackgerritMerged openstack/nova master: fix nova accepting invalid availability zone name with ':'
melwittI had thought that because PasteDeploy was in the pip freeze output, that it had been pip installed too
*** cdent has quit IRC18:55
melwittbut I guess that's not necessarily true18:55
mriedemnot necessarily18:55
mriedemi remember yum installed packages show up in pip freeze output too18:55
mriedemyou just generally aren't using pip with rpms because things get wonky18:55
melwittyeah, huh.18:56
mriedemright so it's setup via deb first
mriedemand then pip ignores it
clarkbhrm pastedeploy 1.5.2 is latest18:57
clarkbso is the ubuntu package just broken and not including all the bits?18:57
mriedemoops first link is
melwittdo you think it has anything to do with this? wsgi-file = /usr/local/bin/keystone-wsgi-public
mriedemwe are using pike uca now18:58
mriedemas of last week18:58
*** gouthamr has quit IRC18:58
*** vvargaszte has quit IRC18:58
melwittis the /usr/local/ there causing keystone to only look under /usr/local for paste deploy?18:58
*** dillaman has joined #openstack-nova18:58
*** gouthamr_ has joined #openstack-nova18:59
*** MasterOfBugs has quit IRC18:59
melwittclarkb: hm, maybe. I hadn't considered the package could be broken18:59
cfriesenis there a way to specify running a single functional test class?19:00
clarkbcfriesen: tox -e functional -- ClassNameHere ?19:00
melwittmriedem: yeah, the ceph-mgr started pulling in python-pastedeploy starting in the pike uca19:00
clarkbthat actually takes a regex (or should)19:00
cfriesenclarkb: will try it...I had done that with the full dotted path to the class and it didn't like it.19:00
*** artom_ is now known as artom19:01
artomcfriesen, clarkb's right, it should be a regex19:01
artomSo just ClassName should do it19:01
artomOr even assName, because it matches as wlel19:01
mriedemsorry make that monday
cfriesenartom: clarkb: nope.  I ran "tox -e functional -- AggregateMigrationTestCase" and it started doing tests from nova.tests.functional.api.openstack.placement.test_placement_api19:02
*** baoli has quit IRC19:02
melwittcfriesen: do you have an old checkout? there was some issues around the regex functionality that got fixed while we were at the PTG19:02
artomcfriesen, hrmm - try without the --?19:02
mriedemcfriesen: rebuild19:02
cfriesenmelwitt: entirely possible19:02
mriedemtox -r -e19:02
*** baoli has joined #openstack-nova19:03
*** eharney has joined #openstack-nova19:03
melwittcfriesen: I'd make sure you have the latest version of tox.ini19:03
*** felipemonteiro__ has quit IRC19:04
*** felipemonteiro__ has joined #openstack-nova19:04
*** oomichi is now known as oomichi_afk19:05
*** edmondsw has quit IRC19:06
*** edmondsw has joined #openstack-nova19:06
*** crushil has quit IRC19:07
*** crushil has joined #openstack-nova19:07
cfriesencombination of a fresh pull and tox -r seems to have done the trick19:08
openstackgerritMatt Riedemann proposed openstack/nova master: use already loaded BDM in instance.<action>
openstackgerritMatt Riedemann proposed openstack/nova master: use already loaded BDM in instance.<action> (2)
openstackgerritMatt Riedemann proposed openstack/nova master: use already loaded BDM in instance.create
*** slaweq_ has joined #openstack-nova19:14
*** krypto has joined #openstack-nova19:17
kryptohello all i have a strange problem with Mitaka release ,on one compute if i enable debug to true nova-compute is restarting continously with error "UnicodeDecodeError: 'ascii' codec can't decode byte 0xe2 in position 335: ordinal not in range(128)" with out debug everything is working as expected ,any idea?19:18
mriedemkrypto: there is likely some debug-only log message hitting something that's not encoded properly19:20
mriedemi probably know what / where it is too19:20
*** tonygunk has quit IRC19:20
mriedem is similar19:20
openstackLaunchpad bug 1624277 in OpenStack Compute (nova) mitaka "nova-scheduler: UnicodeDecodeError in host aggregates handling" [Medium,Confirmed] - Assigned to Nazeema Begum (nazeema123)19:20
mriedemkrypto: look at this fix in oslo.log which was never released in the mitaka series19:21
cfriesenmriedem: I tweaked the code slightly to test the host aggregate migration, and confirmed that we don't soft-delete the entries in table 'aggregate_hosts'.   Diff is here, with functional test output at bottom:     This shouldn't affect normal code, but it messed up one of our audits that verified DB consistency.19:21
mriedemlooks like oslo.log 3.9.0 has that fix19:21
mriedemkrypto: i've also seen one in nova-compute when logging the vif19:22
mriedemi can't find the bug for it though19:23
kryptoi see thank you mriedem ,thanks alot i didnt know that turning debug=true could bring down nova-compute :) not sure whats wrong with my compute19:23
*** tonygunk has joined #openstack-nova19:23
openstackgerritEd Leafe proposed openstack/nova master: Add alternate hosts
openstackgerritEd Leafe proposed openstack/nova master: Add Selection objects
openstackgerritEd Leafe proposed openstack/nova master: Return Selection objects from the scheduler driver
*** bnemec is now known as bnemec-backlater19:24
mriedemkrypto: you would need to capture the stacktrace from the n-cpu logs and put those in a paste to know where the actual problem needs to be fixed, or if it's already fixed and you can cherry pick that fix to your mitaka env, since mitaka is eol upstream19:25
mriedemit might not be in nova, it could be in glanceclient, i see fixed bugs for the same thing in glanceclient from several years ago19:26
kryptookk let me check how to do that19:26
openstackgerritMatt Riedemann proposed openstack/nova master: use already loaded BDM in instance.<action> (2)
openstackgerritMatt Riedemann proposed openstack/nova master: use already loaded BDM in instance.create
cfriesenis there a way to boot an instance from a volume-exposed-as-cdrom, install to another volume-exposed-as-disk, then unmount the CDROM volume?  Or to change boot order of volumes in the instance after it's created?19:28
*** trinaths has left #openstack-nova19:28
*** hemna_ has quit IRC19:29
*** READ10 has quit IRC19:34
*** lbragstad has quit IRC19:35
rybridgesHello. Did we remove the ability to list servers by project/tenant id in the novaclient?19:36
rybridgesnova list --tenant-id <tenant_id> is not working. same if i try with openstack client19:36
rybridgeshave tried all combinations of tenant_id, tenant-id, project_id, project-id in both openstack client and novaclient19:37
rybridgesbut I see it in the documentation here
mriedemrybridges: which version of novaclient?19:37
*** vladikr has quit IRC19:38
mriedemnova help list19:38
mriedemshows me:19:38
mriedem --tenant [<tenant>]           Display information from single tenant (Admin19:38
mriedem                                only).19:38
melwittyou might have to specific --all-tenants too. and make sure tenant is a UUID19:39
mriedemnova list --tenant f7832bfb6daf4a00b6c5687b2ab756fd totally works for me19:39
mriedemas the admin and ^ is the demo tenant in devstack19:39
melwittoh, cool19:39
rybridgesI also tried using the raw python binding like so: nova_client.servers.list(detailed=True, search_opts={'tenant_id': '9dfb1c98d6224351b55ca6dd8e4246c6'}) but it does not seem to be filtering. it only returns the servers in the tenant which i instantiated the novaclient with. that tenant btw is admin so it should be able to query machines from other tenants.19:39
*** Sukhdev has joined #openstack-nova19:39
*** baoli_ has joined #openstack-nova19:39
*** sridharg has quit IRC19:40
rybridgesnovaclient version is 7.1.319:40
mriedemi'm on 9.1.019:40
rybridgesahh ok nova list --tenant is working19:40
mriedemyeah, use the help to see what's supported19:40
mriedemnova help list19:40
rybridgesused to be --tenant_id19:41
rybridgeswas a bit confusing b/c docs make it sound like tenant_id will work19:41
*** baoli has quit IRC19:41
mriedemthe api reference docs != the nova cli19:41
rybridgeswell, nova list --tenant <id> does seem to work19:42
rybridgesnow how can i do the same thing from the python binding19:42
mriedemfor review, gibi's bdm perf fixup series has my +2, - it's pretty mechanical plumbing stuff19:42
mriedemeasy review19:42
rybridgesthe docs just pointed me to the page i linked above19:42
rybridgesfor a full list of search opts19:42
rybridgesbut tenant, tenant_id, tenant-id, project, project_id, project-id all do not filter properly when using the python binding directly19:43
mriedemrybridges: project_id and tenant_id are valid search opts to the GET /servers API19:44
mriedemhowever, as noted in the docs, they only work if you're an admin making the request19:44
mriedemso, are you doing this with an admin token19:44
mriedemis the first question19:44
rybridgeswhen i instantiated the novaclient object in python19:45
rybridgesi passed it admin user/password19:45
mriedemi guess the docs don't say this, but the nova cli help does19:45
rybridgesand i can list servers in the admin tenant19:45
mriedemcan you do: nova service-list?19:45
rybridgeswhen i pass a tenant id as a search opt, it still just returns the same servers that are in the admin tenant, not those in the project i specified19:45
mriedemok i think i see the issue19:46
*** ijw has quit IRC19:46
rybridgesyes, does work19:46
rybridgesfrom the python binding19:46
rybridgeswith the instance of nova client that i created19:46
mriedemyup so we suck at documenting things in the api reference...19:46
mriedemrybridges: so try searching with project_id19:47
rybridgesso i must specify all_tenants19:47
mriedemrather than tenant_id19:47
mriedemwell try ^ first19:47
rybridgesi have tried with project_id and i get the same behavior19:47
mriedemok probably because of search_opts['project_id'] = context.project_id19:48
mriedemit overwrites the requested project_id filter with the one in the token19:48
mriedemwhich is your admin project19:48
rybridgesokay i see now19:48
mriedemso yeah you have to specify all_tenants19:48
rybridgesthis is working: nova.servers.list(detailed=True,  search_opts={'project_id': '9dfb1c98d6224351b55ca6dd8e4246c6', 'all_tenants': True})19:48
rybridgesoof that sucks =(19:48
mriedemyeah, we need to note this in the docs19:48
mriedemi can push a patch in a bit19:48
mriedemfor the docs, not the actual api behavior19:49
rybridgesthat would be great19:49
melwittI think it used to be that all_tenants had to be passed to the CLI too but someone added automagic to do it if --tenant was passed since it's required for --tenant to work19:49
*** Sukhdev has quit IRC19:51
rybridgesokay one more thing19:51
rybridgessince in the python binding i have to pass all_tenants19:51
rybridgeswhenever i want to filter by project id19:51
rybridgesdoes that mean that novaclient is actually pulling all of the instances for all tenants back, and then filtering based on project id on the front end?19:51
rybridgeswe dont want that, because we have tens of thousands of instances19:52
rybridgesand filtering all of that on the front end would be very slow19:52
mriedemrybridges: it is filtering server-side19:52
*** nicolasbock has joined #openstack-nova19:52
rybridgesthank you ^.^19:52
mriedemwhich release are you testing this on?19:52
mriedemis this penick's cluster?19:53
rybridgesheh yes, penick is our architect19:53
mriedemhe is our lord and savior19:53
mriedemignore that19:54
rybridgesyes he is my lord and savior as well19:54
jaypipeswell, with penick, most people are always looking UP at him. :)19:54
rybridgesi am one of the few who does not have to look up19:54
jaypipesrybridges: wow, lucky you :)19:55
rybridgesi have 6'7" so he is only a couple inches taller than me19:55
rybridgesi am**19:55
jaypipesrybridges: basically, all of us nova cores have capitulated to all of penick's demands because we're all afraid he will crush us with his pinky finger if we give him any lip.19:56
*** Sukhdev_ has joined #openstack-nova19:56
jaypipesrybridges: well, that and, you know, melwitt :)19:56
jaypipesrybridges: though melwitt could easily stomp on any of us with her boots :)19:57
mriedemjaypipes: he doesn't have demands, because he know we can't fulfill anything for juno19:57
jaypipesmriedem: lol, dig!19:57
rybridgesyea... freakin juno man19:57
dansmithrybridges: that's what we call him19:57
dansmithrybridges: "juno man"19:57
dansmithlike a superhero but...sadder19:57
jaypipesedleafe: we ready on the alternate hosts/selection objects patches? just a rebase just now?19:58
jaypipesedleafe: or blueprint changed...19:58
*** gszasz has quit IRC20:00
edleafejaypipes: I changed the structure of the Selection object to match the spec20:02
jaypipesgot it.20:02
edleafejaypipes: should be ready for review20:02
edleafejaypipes: working on the code that changes the return from select_destinations now20:02
edleafeit's... involved :)20:02
*** tbachman has quit IRC20:03
*** vvargaszte has joined #openstack-nova20:04
*** ianw is now known as ianw|pto20:04
*** vvargaszte has quit IRC20:04
*** felipemonteiro_ has joined #openstack-nova20:08
*** pchavva has quit IRC20:09
*** felipemonteiro__ has quit IRC20:11
*** ijw has joined #openstack-nova20:17
openstackgerritMerged openstack/nova master: Fix inconsistency of 'NOTE:' description
*** baoli_ has quit IRC20:20
melwittclarkb: I think python-pastedeploy isn't being installed from the uca repo based on this and trying installing it locally I'm able to do "from paste import deploy". so AFAICT the package isn't broken20:22
*** burt has joined #openstack-nova20:23
dansmithjaypipes: edleafe: slap me if I missed something on any of these comments:
*** flanders_ has joined #openstack-nova20:26
openstackgerritMerged openstack/nova master: [placement] Update the placement deployment instructions
openstackgerritMerged openstack/nova master: api-ref: remove redundant preserve_ephemeral mention from rebuild docs
*** lbragstad has joined #openstack-nova20:27
openstackgerritMerged openstack/nova master: api-ref: add note about rebuild not replacing volume-backed root disk
*** smatzek has quit IRC20:28
*** smatzek has joined #openstack-nova20:28
*** Nel1x has quit IRC20:29
mriedemaha, found a bug in a test20:29
penickTake that, stupid test20:32
*** smatzek has quit IRC20:33
melwittdoes anyone know anything about wsgi_scripts in setup.cfg? I found that it's a directive for pbr to install scripts and they get installed to what setuptools considers to be the "install directory" and they end up in /usr/local/bin20:34
melwittis there any reason installing them that way would make it so they can't import modules that had been installed via apt-get?20:34
mriedemsounds like a sdague question20:37
sdaguemelwitt: import path will be determined by the python interpretter20:38
sdaguewhat's the execute line on it?20:38
*** zzzeek has quit IRC20:39
openstackgerritMatt Riedemann proposed openstack/nova master: Avoid redundant BDM lookup in check_can_live_migrate_source
edleafedansmith: tempting, but I think those are good points20:39
dansmithedleafe: \o/20:40
melwittsdague: on the script? I'm not sure, it's generated by pbr from this
sdagueon my devstack it's #!/usr/bin/python20:40
*** ijw has quit IRC20:40
sdaguethe only thing I can imagine is if it got a weird value during build20:41
melwittI'm trying to debug a ceph job failure so I don't have info like that for that env20:41
sdaguemelwitt: url?20:41
melwittthe only difference I see between it and normal gate jobs is paste deploy is being installed from apt-get instead of pip20:41
sdaguethat shouldn't really matter, as long as we aren't talking about a venv20:42
melwittsdague: here's how keystone fails to start
jaypipesdansmith: no, I think those are accurate issues you've identified.20:43
melwittsdague: I don't understand how it's not able to import paste deploy if paste deploy is installed (with apt-get)20:43
melwittand I don't think it's double installed with apt-get AND pip. I think the pip installs are skipped because it was already installed with apt-get20:43
*** yassine has quit IRC20:44
melwittlike this is it being skipped later
sdaguemelwitt: so... the more concerning thing is paste gets deployed again20:44
sdagueI think this is the python namespace issue20:44
sdaguepaste & paste-deploy (which is really paste.deploy using a python namespace)20:45
sdagueend up being installed by apt20:45
sdaguethen pip install new paste20:45
*** ijw has joined #openstack-nova20:45
sdaguein /usr/local/lib20:45
sdaguewhich now means the search path for paste.* is in /usr/local/lib, the new one20:45
*** cleong has quit IRC20:45
sdaguepaste-deploy is not installed again, because the version looks the same to pip20:45
sdaguebut it's in a different namespaced part of the tree20:46
sdagueso when you start importing it's not visible20:46
melwittoh, I see. damn20:46
sdaguethis is why we got oslo to stop doing namespaces20:46
*** gouthamr_ has quit IRC20:46
melwittI was focused only on paste.deploy20:46
*** ijw has quit IRC20:46
sdaguewhen I saw the upgrade of paste, that reminded me of this funky issue20:46
sdaguefor this reason we try really hard to install very little python from system packages20:47
sdaguebecause you end up in exactly this situation20:47
melwittokay, so paste install wasn't skipped by pip for already existing because the version probably didn't satisfy20:47
sdagueI think the answer is to remove paste & paste-deploy from the dpkg list20:47
sdaguebecause of upper-constraints20:47
*** tbachman has joined #openstack-nova20:48
clarkbisn't paste-deploy coming in transitively in the dpkg list though? that could get tricky to remove20:48
melwittit's coming in through ceph package20:48
sdaguemelwitt: yeh, well that's going to be non compatible with the way we do things20:48
sdagueyou could hack in a: pip install -U --force paste.deploy20:49
sdagueat the right part of the process20:49
sdaguethat's probably the best option if the ceph package is dragging that stuff in20:50
melwittsdague: okay, yeah thanks. I'll see where I could put something like that in the devstack ceph plugin20:51
melwittthanks for your help cracking this20:51
*** tssurya_ has joined #openstack-nova20:53
*** lpetrut has quit IRC20:54
*** ijw has joined #openstack-nova20:55
mriedemomfg cells meeting in 5 minutes20:55
openstackmelwitt: Error: "!!" is not a valid command.20:56
melwittmy alarm alarmed openstack too20:56
*** ijw has quit IRC20:56
*** ijw has joined #openstack-nova20:56
*** zzzeek has joined #openstack-nova20:57
*** smatzek has joined #openstack-nova20:57
*** smatzek has quit IRC21:00
openstackgerritEric Berglund proposed openstack/nova master: WIP(5): PowerVM driver: ovs vif
*** vladikr has joined #openstack-nova21:08
*** tssurya_ has quit IRC21:14
*** felipemonteiro_ has quit IRC21:19
*** felipemonteiro_ has joined #openstack-nova21:19
*** zzzeek has quit IRC21:19
*** krypto has quit IRC21:20
*** ociuhandu has joined #openstack-nova21:20
*** zzzeek has joined #openstack-nova21:20
*** tbachman has quit IRC21:21
edleafedansmith: I may get to slap you yet21:22
edleafedansmith: The test removed from test_filter_scheduler was testing pre-pike conductors that didn't pass instance_uuids. Now that we're in Queens, that test is no longer needed21:23
*** zzzeek has quit IRC21:23
*** cali_boxer has joined #openstack-nova21:23
cali_boxerhello all21:24
*** zzzeek has joined #openstack-nova21:25
*** crushil has quit IRC21:26
mriedemedleafe: if the test is no longer needed, did you also remove the code that's checking for empty instance_uuids?21:29
mriedemedleafe: and if so, that should be split into a separate change21:30
*** sahid has quit IRC21:30
*** edmondsw has quit IRC21:32
dansmithedleafe: the rpc api still allows for it to be empty right?21:32
*** edmondsw has joined #openstack-nova21:32
melwitthm, devstack post-config happens after keystone is started. would not have expected that21:32
dansmithedleafe: you don't just get to drop rpc compat when it's been long enough, you have to do the version work21:33
clarkbmelwitt: we were just talking about why that is in the keystone channel21:33
clarkbmelwitt: long story short there is no default default domain21:33
dansmithedleafe: otherwise people that have version mismatches get uncool error messages instead of "this is too old for that"21:33
clarkbmelwitt: so you have to get keystone fully up and running before you can configure anything else21:33
dansmithedleafe: I just dealt with one of those yesterday, where they thought they had upgraded conductor but hadn't21:33
melwittclarkb: hah, good coincidence. thanks for sharing21:34
*** slaweq_ has quit IRC21:34
openstackgerritMatt Riedemann proposed openstack/nova master: api-ref: note that project_id filter only works with all_tenants
mriedemrybridges: ^21:36
*** edmondsw has quit IRC21:37
*** gszasz has joined #openstack-nova21:38
*** tonygunk has quit IRC21:39
*** catintheroof has quit IRC21:41
openstackgerritMerged openstack/nova master: Remove unused get_all_instance_*metadata methods
openstackgerritMerged openstack/nova master: Remove old compat code from servers ViewBuilder._get_metadata
edleafedansmith: so if filter_scheduler gets a request without instance_uiuds, it has to return the old-style host_states without alternates?21:44
*** sree has joined #openstack-nova21:49
*** takashin has joined #openstack-nova21:52
*** ijw has quit IRC21:53
*** sree has quit IRC21:53
*** ijw has joined #openstack-nova21:54
*** lyan has quit IRC21:54
melwittsigh, I need to be able to do something after devstack does pip installs but before keystone starts and I'm not seeing a way21:57
melwitt(in a devstack plugin)21:57
*** ijw has quit IRC21:58
mriedem ?22:00
mriedemmelwitt: look at the run phases in stack.sh22:00
mriedemthose are the hook points i think22:00
melwittyeah. but I noticed keystone starts before post-config, which it's not supposed to (or originally wasn't). clarkb explained about that a little bit ago22:01
mriedemit looks like everything is started after post-config, EXCEPT keystone22:01
* melwitt dies22:02
mriedemah yup here
*** kylek3h has quit IRC22:02
mriedemi'm guessing b/c the accounts are needed to configure the other services?22:03
mriedemlike configure nova to talk to cinder/glance/neutron22:03
mriedemyou have to have those things created in keystone first, which means starting keystone22:03
melwitthe said it's because there's no default default domain22:03
mriedemso besides a pre-start phase being added,22:03
mriedemyou could probably hack something up in to at least test the theory,22:03
mriedemby checking to see if ceph is configured for the backend22:04
*** slaweq_ has joined #openstack-nova22:04
mriedemlike, if [[ $ENABLE_CEPH_CINDER == "True" ]]; then22:05
*** Sukhdev_ has quit IRC22:05
melwittah, I see. yeah, thanks22:06
mriedemso like check that right here
mriedemand if ceph is enabled, do the dirty package thing22:06
melwittyeah, thanks. I was thinking, "how can I tell that if it's the ceph devstack plugin" but you were already on it22:06
openstackgerritMatt Riedemann proposed openstack/nova stable/pike: fix nova accepting invalid availability zone name with ':'
mriedemthere is probably also a way to find out if the ceph plugin is enabled, but this is an easier hack most likely22:07
mriedemoh heh22:08
melwittoh hey22:08
mriedemif [[ is_plugin_enabled "ceph" ]]; then22:08
mriedem   do that dirty package thing22:08
mriedemmelwitt: devstack also has the ceph job in it's experimental queue so you can test it right in the devstack change itself22:09
*** brault has joined #openstack-nova22:10
mriedemeglynn can send checks to my house directly22:10
*** awaugama has quit IRC22:10
*** acormier has quit IRC22:11
*** eharney has quit IRC22:11
openstackgerritMatt Riedemann proposed openstack/nova stable/ocata: fix nova accepting invalid availability zone name with ':'
efriedjaypipes Is <SELECT, FROM Traits JOIN ResourceProviderTraits ON == ResourceProviderTraits.trait_id and ResourceProviderTraits.resource_provider_id == $rp_id> more efficient than (or equivalent to) <SELECT id, name from Traits where id in (SELECT trait_id from ResourceProviderTraits WHERE resource_provider_id == $rp_id)> ?)>22:11
*** tbachman has joined #openstack-nova22:12
*** acormier has joined #openstack-nova22:12
*** gszasz has quit IRC22:14
*** brault has quit IRC22:14
*** felipemonteiro_ has quit IRC22:15
*** acormier has quit IRC22:17
jaypipesefried: yes.22:19
efriedjaypipes Which, more efficient, or equivalent?22:19
jaypipesefried: former is more efficient.22:20
efriedBecause no sub-query22:20
efriedDig, thanks.22:20
jaypipesefried: though many modern optimizers will rewrite the subquery in the WHERE clause to be a JOIN22:20
efriedYeah, I figured that was probably the case.22:20
efriedjaypipes The stuff we're using here, that looks like python code doing the joins and stuff - is that actually doing the magic in the database layer or in python?22:21
jaypipesefried: DB22:21
efriedMust be some pretty heavy logic in there, like when some of those "methods" have conditionals embedded in 'em.22:22
jaypipesefried: when using ORM queries that use things like joined_load eager and all that jazz, the SQL produced is less efficient.22:22
efriedjaypipes Like, how does something like ``sa.and_( == rpt.c.trait_id, rpt.c.resource_provider_id == rp_id)`` *not* interpret those `==`s in python?22:23
dansmithedleafe: yeah, you need to adjust the return value of that rpc method as well22:26
jaypipesefried: oh, that's actually the beauty of SQLAlchemy's core expression API... it *does* interpret those things actually :) it's just what rpt.c.resource_provider_id == rp_id turns into is an expression object in SQLAlchemy that is processed with the __eq__ magic function...22:26
efriedahhhh, __eq__, of course.  Very cool.22:27
*** gouthamr has joined #openstack-nova22:27
efriedjaypipes So I guess that means the order of the arguments in there is crucial.  (rpt.c.resource_provider_id == rp_id) is cool, but (rp_id == rpt.c.resource_provider_id) is nonsense.22:28
jaypipesefried: no...22:28
dansmithedleafe: that's the next patch though right?22:28
efriedWouldn't it try to use rp_id's __eq__ in that case?22:28
*** gjayavelu has quit IRC22:29
jaypipesefried: there's some magic reflection happening in the sa.and_() function.22:29
jaypipesefried: where it's looking at the structure of the parameters supplied to it.22:29
efriedbefore the interpreter gets hold of it?  That *is* magic.22:29
efriedjaypipes Because
dansmithefried: if rp_id is an integer, then yes22:30
dansmiththe order would matter22:30
efrieddansmith Phew, that makes me feel more sane.22:31
*** tbachman has quit IRC22:32
jaypipesefried: there's lots o magic happening in
jaypipesefried: have fun reading :)22:32
efriedThis is my first day looking at this sqlalchemy business.  Appreciate the pointers.22:32
*** ociuhandu has quit IRC22:32
jaypipesno prob. BTW, zzzeek is the author of SQLAlchemy. I am sure he would be the best person to ask about magicalities in the core expression API. :)22:32
*** archit has quit IRC22:33
*** gjayavelu has joined #openstack-nova22:34
*** penick has quit IRC22:35
efriedI'm pleased to report that nothing in that file blew my freaking mind, or made me expect (int == magic_sqlalchemy_object) not to explode.22:37
*** lbragstad has quit IRC22:38
mriedemso, dansmith,22:38
mriedemhave you and edleafe been up to something you're not telling the rest of us?22:38
*** tbachman has joined #openstack-nova22:38
*** slaweq_ has quit IRC22:39
dansmithmriedem: um, what?22:40
dansmithmriedem: the banter was about the alternate hosts patch22:40
efrieddansmith Oh.  On the other hand, it looks like __eq__ may internally handle the left-hand side not understanding the right:
efriedSo it probably *would* work.22:41
dansmithefried: I dunno how that works,22:41
*** edmondsw has joined #openstack-nova22:41
dansmithunless it only works against primitives or something22:41
melwittmriedem: lol22:41
efriedI imagine it's like 'except TypeError: try_the_other_guy's___eq___method'22:42
mriedemi enjoy getting to do a pull request once per year22:42
dansmithefried: see the second answer:
edleafemriedem: dansmith and I were trying to keep that a secret until the Forum!22:43
*** dillaman has quit IRC22:44
* dansmith scowls at edleafe 22:44
efrieddansmith Beaut, thanks.22:44
*** slaweq_ has joined #openstack-nova22:44
dansmithefried: I did not think that worked that way22:44
efrieddansmith TIL, for sure.22:45
dansmithand I might be remembering really old behavior, pre-new-style classes and all22:45
efriedMakes sense.  Very prescient of those who wrote python itself.22:45
edleafedansmith: well, both patches change what is returned: this adds alternates, and the next changes them all to Selection objects22:45
edleafeOld conductors won't know what to do with alternates22:46
*** edmondsw has quit IRC22:46
dansmithedleafe: yeah, then you need to pass a flag that says "give me the new stuff", or as mikal would say "do it to me big boy"22:46
dansmithedleafe: I was thinking with all the "make this look like the existing return, change in next patch" stuff that we were remaining compatible here22:47
*** tbachman has quit IRC22:49
*** gbarros has joined #openstack-nova22:49
edleafedansmith: what's returned from the scheduler manager is the same. These changes are all internal to the scheduler22:50
edleafethe next patch in the series is where all hell breaks loose22:50
dansmithedleafe: okay but not all internal to the scheduler if the rpc api changes22:50
edleafeI hope to have that in a decent state by tomorrow22:50
edleafethe rpc api isn't changing in these22:50
*** gbarros has quit IRC22:54
dansmithI'm confused22:55
*** ijw has joined #openstack-nova22:55
openstackgerritOpenStack Proposal Bot proposed openstack/os-vif stable/pike: Updated from global requirements
dansmithedleafe: you said that new conductors will always be passing uuids now.. that is an rpc interface between conductor and scheduler.. we're agreed that you need to keep the test and compat behavior there, yes?22:55
dansmithand the other question is if you're changing the return type of the rpc call that conductor is making..22:56
edleafedansmith: the instance_uuid change was in Pike22:56
edleafeyeah, I'm keeping the test22:57
dansmithokay, so .. the return value of the rpc call is changed here or no?22:57
edleafenot yet22:57
edleafeI am working on the next patch in the series where is changes22:57
edleafeNeedless to say, that breaks a lot of tests22:57
dansmithedleafe: this one right?
dansmithor that one is just internal still and then another one after will change what the rpc consumer gets?22:59
edleafeno, that just changes the scheduler driver's return value to the manager22:59
edleafethe next patch will change what the manager returns, which is the rpc boundary22:59
dansmithright, okay22:59
dansmithso in _that_ one you'll need a flag to say "give me the new stuff"22:59
*** sdague has quit IRC23:00
dansmithand in these before, as long as you keep that test for the empty uuids I think you're good23:00
*** esberglu has quit IRC23:00
openstackgerritEd Leafe proposed openstack/nova master: Add alternate hosts
openstackgerritEd Leafe proposed openstack/nova master: Add Selection objects
openstackgerritEd Leafe proposed openstack/nova master: Return Selection objects from the scheduler driver
*** ijw has quit IRC23:01
edleafedansmith: ^^ fixed per your comments23:01
*** esberglu has joined #openstack-nova23:01
edleafeonly the first patch changed; the rest are rebases23:01
*** armax has quit IRC23:01
edleafeand with that it's time to make dinner23:03
dansmithedleafe: cool, thanks23:05
dansmithjaypipes: ^23:05
*** esberglu has quit IRC23:05
*** brault has joined #openstack-nova23:08
openstackgerritOpenStack Proposal Bot proposed openstack/os-vif stable/pike: Updated from global requirements
*** brault has quit IRC23:13
*** edmondsw has joined #openstack-nova23:13
*** slaweq__ has joined #openstack-nova23:16
*** mingyu has joined #openstack-nova23:16
*** slaweq_ has quit IRC23:17
*** crushil has joined #openstack-nova23:17
*** hongbin has quit IRC23:18
*** felipemonteiro_ has joined #openstack-nova23:20
openstackgerritOpenStack Proposal Bot proposed openstack/os-vif stable/newton: Updated from global requirements
*** slaweq__ has quit IRC23:21
*** felipemonteiro__ has joined #openstack-nova23:21
*** acormier has joined #openstack-nova23:24
*** felipemonteiro_ has quit IRC23:25
*** acormier has quit IRC23:28
*** gmann_sleep is now known as gmann23:32
*** dillaman has joined #openstack-nova23:33
*** edmondsw has quit IRC23:35
openstackgerritMerged openstack/nova stable/pike: Add nova-manage db command for ironic flavor migrations
*** chyka has quit IRC23:40
openstackgerritLIU Yulong proposed openstack/nova-specs master: Reset the instance keypair while rebuilding (spec)
*** chyka has joined #openstack-nova23:41
*** chyka has quit IRC23:45
*** slaweq_ has joined #openstack-nova23:46
openstackgerritMerged openstack/nova stable/ocata: Fix --max-count handling for nova-manage cell_v2 map_instances
gmannmriedem: yes, that is kind of dead code. those are not going to be in filter_mapping as API 400 before it23:51
*** namnh_ has quit IRC23:52
*** ijw has joined #openstack-nova23:53
gmannmriedem: we left this while doing schema for index. you have patch up for that cleanup otherwise i can push one.23:54
*** jdg has joined #openstack-nova23:56
*** baoli has joined #openstack-nova23:58
*** armax has joined #openstack-nova23:58
*** ijw has quit IRC23:59
*** baoli has quit IRC23:59

Generated by 2.15.3 by Marius Gedminas - find it at!