| openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Add docs and releasenotes for BP policy-defaults-refresh https://review.opendev.org/720129 | 00:01 |
|---|---|---|
| *** tetsuro has joined #openstack-nova | 00:03 | |
| *** brinzhang has joined #openstack-nova | 00:13 | |
| *** hongbin has quit IRC | 00:22 | |
| *** avolkov has quit IRC | 00:26 | |
| *** hongbin has joined #openstack-nova | 00:29 | |
| *** ttsiouts has joined #openstack-nova | 00:30 | |
| *** prometheanfire has quit IRC | 01:00 | |
| *** ttsiouts has quit IRC | 01:04 | |
| *** prometheanfire has joined #openstack-nova | 01:07 | |
| *** Liang__ has joined #openstack-nova | 01:09 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Add docs and releasenotes for BP policy-defaults-refresh https://review.opendev.org/720129 | 01:22 |
| *** tetsuro has quit IRC | 01:47 | |
| *** tetsuro has joined #openstack-nova | 01:54 | |
| *** mkrai has joined #openstack-nova | 01:59 | |
| *** brinzhang_ has joined #openstack-nova | 02:04 | |
| *** brinzhang has quit IRC | 02:08 | |
| *** brinzhang has joined #openstack-nova | 02:08 | |
| *** brinzhang_ has quit IRC | 02:09 | |
| *** yaawang has quit IRC | 02:10 | |
| *** yaawang has joined #openstack-nova | 02:11 | |
| *** ttsiouts has joined #openstack-nova | 02:15 | |
| *** sapd1 has joined #openstack-nova | 02:28 | |
| *** artom has quit IRC | 02:30 | |
| *** lifeless_ is now known as liffeless | 02:40 | |
| *** liffeless is now known as lifeless | 02:40 | |
| *** JamesBenson has quit IRC | 02:47 | |
| *** ttsiouts has quit IRC | 02:49 | |
| *** hongbin has quit IRC | 02:50 | |
| *** ociuhandu has joined #openstack-nova | 03:14 | |
| *** prometheanfire has left #openstack-nova | 03:15 | |
| *** ociuhandu has quit IRC | 03:19 | |
| *** maohongbo1 has joined #openstack-nova | 03:22 | |
| *** ttsiouts has joined #openstack-nova | 03:23 | |
| *** maohongbo has quit IRC | 03:23 | |
| *** maohongbo1 is now known as maohongbo | 03:23 | |
| *** psachin has joined #openstack-nova | 03:26 | |
| *** ttsiouts has quit IRC | 03:28 | |
| *** threestrands has joined #openstack-nova | 03:29 | |
| *** tetsuro has quit IRC | 03:30 | |
| *** maohongbo1 has joined #openstack-nova | 03:35 | |
| *** maohongbo has quit IRC | 03:35 | |
| *** maohongbo1 is now known as maohongbo | 03:35 | |
| *** gyee has quit IRC | 04:02 | |
| *** ttsiouts has joined #openstack-nova | 04:03 | |
| *** ttsiouts has quit IRC | 04:37 | |
| *** evrardjp has quit IRC | 04:37 | |
| *** evrardjp has joined #openstack-nova | 04:37 | |
| *** udesale has joined #openstack-nova | 04:43 | |
| *** tetsuro has joined #openstack-nova | 04:48 | |
| *** ratailor has joined #openstack-nova | 05:00 | |
| *** ratailor has quit IRC | 05:01 | |
| *** vishalmanchanda has joined #openstack-nova | 05:01 | |
| *** ratailor has joined #openstack-nova | 05:06 | |
| *** maohongbo has quit IRC | 05:10 | |
| *** maohongbo has joined #openstack-nova | 05:10 | |
| *** ttsiouts has joined #openstack-nova | 05:41 | |
| *** ttsiouts has quit IRC | 05:46 | |
| *** maohongbo1 has joined #openstack-nova | 05:49 | |
| *** maohongbo has quit IRC | 05:50 | |
| *** maohongbo1 is now known as maohongbo | 05:50 | |
| *** ralonsoh has joined #openstack-nova | 05:56 | |
| *** ociuhandu has joined #openstack-nova | 06:11 | |
| *** ttsiouts has joined #openstack-nova | 06:17 | |
| *** ttsiouts has quit IRC | 06:22 | |
| *** dklyle has quit IRC | 06:36 | |
| *** damien_r has joined #openstack-nova | 06:37 | |
| *** ociuhandu has quit IRC | 06:38 | |
| *** ociuhandu has joined #openstack-nova | 06:39 | |
| gibi | good morning | 06:40 |
| *** damien_r has quit IRC | 06:42 | |
| *** ociuhandu has quit IRC | 06:44 | |
| *** nightmare_unreal has joined #openstack-nova | 06:47 | |
| *** dpawlik has joined #openstack-nova | 06:49 | |
| *** maohongbo has quit IRC | 06:50 | |
| *** maohongbo has joined #openstack-nova | 06:51 | |
| *** ttsiouts has joined #openstack-nova | 06:56 | |
| *** ccamacho has joined #openstack-nova | 07:01 | |
| *** tesseract has joined #openstack-nova | 07:06 | |
| *** ttsiouts has quit IRC | 07:09 | |
| *** ttsiouts has joined #openstack-nova | 07:09 | |
| *** slaweq has joined #openstack-nova | 07:12 | |
| *** maciejjozefczyk has joined #openstack-nova | 07:13 | |
| *** ociuhandu has joined #openstack-nova | 07:30 | |
| openstackgerrit | Merged openstack/nova master: Pass the actual target in flavor access policy https://review.opendev.org/719607 | 07:34 |
| *** rpittau|afk is now known as rpittau | 07:34 | |
| *** tosky has joined #openstack-nova | 07:38 | |
| *** maohongbo has quit IRC | 07:39 | |
| *** maohongbo has joined #openstack-nova | 07:39 | |
| *** ttsiouts has quit IRC | 07:47 | |
| *** maohongbo1 has joined #openstack-nova | 07:50 | |
| *** maohongbo has quit IRC | 07:51 | |
| *** maohongbo1 is now known as maohongbo | 07:51 | |
| *** ttsiouts has joined #openstack-nova | 07:55 | |
| *** links has joined #openstack-nova | 08:02 | |
| bauzas | good morning Nova | 08:03 |
| bauzas | gibi: permission to lock the semaphore for triaging bugs ? | 08:10 |
| gibi | bauzas: granted, I'm on a call | 08:12 |
| bauzas | gibi: -EINTR | 08:13 |
| *** happyhemant has joined #openstack-nova | 08:14 | |
| bauzas | sem_wait(*bugs) | 08:16 |
| *** maohongbo has quit IRC | 08:16 | |
| *** maohongbo has joined #openstack-nova | 08:17 | |
| openstackgerrit | Merged openstack/nova master: Fix server actions to be system and project scoped https://review.opendev.org/720123 | 08:19 |
| *** abhishek has joined #openstack-nova | 08:20 | |
| *** ociuhandu has quit IRC | 08:25 | |
| *** huaqiang has quit IRC | 08:26 | |
| *** dtantsur|afk is now known as dtantsur | 08:28 | |
| gibi | bauzas: before I lost the lock I was looking at https://bugs.launchpad.net/nova/+bug/1861493 | 08:28 |
| openstack | Launchpad bug 1861493 in OpenStack Compute (nova) "Nova sends an "X-Service-Token" header when "send_service_user_token" is disabled" [Undecided,New] | 08:28 |
| * bauzas thinks https://bugs.launchpad.net/nova/+bug/1863844 seems legit but wonders whether the boot call is right | 08:29 | |
| openstack | Launchpad bug 1863844 in OpenStack Compute (nova) " The virtual machine with the config_driver device cannot properly mount the disk using the virtio-scsi driver" [Undecided,New] | 08:29 |
| bauzas | sorry, I'm really not a volume expert :( | 08:29 |
| bauzas | gibi: ack ok | 08:29 |
| gibi | me neither | 08:30 |
| bauzas | lyarwood: not sure you're up yet but could you tell me whether https://bugs.launchpad.net/nova/+bug/1863844 is legit ? | 08:30 |
| openstack | Launchpad bug 1863844 in OpenStack Compute (nova) " The virtual machine with the config_driver device cannot properly mount the disk using the virtio-scsi driver" [Undecided,New] | 08:30 |
| openstackgerrit | Ivaylo Mitev proposed openstack/nova master: Faults from cell DB missing in GET /servers/detail https://review.opendev.org/699176 | 08:33 |
| *** ociuhandu has joined #openstack-nova | 08:36 | |
| lyarwood | bauzas: morning, yeah it's valid and I think I know why | 08:37 |
| * lyarwood takes a look | 08:37 | |
| bauzas | lyarwood: k, no regression hopefully ? | 08:37 |
| lyarwood | bauzas: actually they set --property hw_scsi_model='virtio-scsi' that I assumed they had missed | 08:38 |
| lyarwood | bauzas: I've reproduced it FWIW with q35 | 08:38 |
| lyarwood | bauzas: -> coffee and I'll look closer | 08:38 |
| bauzas | ack, gtk | 08:39 |
| bauzas | and thanks | 08:40 |
| *** avolkov has joined #openstack-nova | 08:41 | |
| *** threestrands has quit IRC | 08:45 | |
| *** tkajinam has quit IRC | 08:50 | |
| *** martinkennelly has joined #openstack-nova | 08:59 | |
| openstackgerrit | Ivaylo Mitev proposed openstack/nova master: Faults from cell DB missing in GET /servers/detail https://review.opendev.org/699176 | 09:01 |
| *** ociuhandu has quit IRC | 09:05 | |
| *** ociuhandu has joined #openstack-nova | 09:08 | |
| *** mkrai has quit IRC | 09:09 | |
| *** mkrai has joined #openstack-nova | 09:23 | |
| bauzas | argh, I'm surprised to see the virtual PTG planned for only one single week... | 09:38 |
| bauzas | I thought we had an agreement on having more than just one week | 09:38 |
| bauzas | :/ | 09:38 |
| hrw | lyarwood: I am waiting for x86 world to migrate to q35/uefi... | 09:38 |
| lyarwood | hrw: trying to for V | 09:40 |
| lyarwood | hrw: it's on my plate at least | 09:40 |
| openstackgerrit | Dincer Celik proposed openstack/nova stable/train: Abort live-migration during instance_init https://review.opendev.org/720414 | 09:41 |
| *** Liang__ has quit IRC | 09:42 | |
| hrw | :) | 09:46 |
| *** smcginnis has quit IRC | 09:52 | |
| *** tetsuro has quit IRC | 09:52 | |
| *** tetsuro has joined #openstack-nova | 09:54 | |
| *** lpetrut has joined #openstack-nova | 09:56 | |
| nightmare_unreal | can someone follow up on this : https://review.opendev.org/#/c/715395/ | 09:57 |
| nightmare_unreal | Thanks :) | 09:57 |
| *** liuyulong has quit IRC | 09:59 | |
| *** ociuhandu has quit IRC | 10:02 | |
| *** ociuhandu has joined #openstack-nova | 10:03 | |
| bauzas | lyarwood: fwiw, i'm about to classify https://bugs.launchpad.net/nova/+bug/1862375 as invalid/wishlist since it seems an improvement | 10:08 |
| openstack | Launchpad bug 1862375 in OpenStack Compute (nova) "Subsequent nova-api volume attach request waiting for previous one to complete" [Undecided,Invalid] | 10:08 |
| bauzas | (for the timeout, i'll explain the option) | 10:08 |
| *** elod_ has joined #openstack-nova | 10:09 | |
| *** elod_ has quit IRC | 10:09 | |
| gibi | bauzas: I agree and raced with you on that. sorry. I forget to take the lock. I'm stopping now | 10:10 |
| bauzas | gibi: all cool | 10:11 |
| lyarwood | bauzas / gibi ; meh, the whole device naming thing is pointless and in an ideal world should be removed | 10:13 |
| gibi | lyarwood: removing that would need an API change as nova returns the device name in the POST rsp | 10:13 |
| lyarwood | bauzas / gibi ; the computes actually serialise these requests anyway so we could remove this in the future | 10:13 |
| lyarwood | gibi: right, we could remove it for virt drivers that can't guarantee it with a microversion right? | 10:15 |
| lyarwood | gibi: such as libvirt | 10:15 |
| lyarwood | actually can any virt drivers guarantee it | 10:15 |
| gibi | I'm fine removing that in a microversion but that needs spec | 10:17 |
| gibi | and I cannot really comment on the virt driver side so I trust you | 10:18 |
| *** rpittau is now known as rpittau|bbl | 10:19 | |
| bauzas | huh, unless I misunderstand something, a snapshot is not intended to be written again by the instance that was snapshoted ? https://bugs.launchpad.net/nova/+bug/1860913 | 10:20 |
| openstack | Launchpad bug 1860913 in OpenStack Compute (nova) "Instance uses base image file when it is rebooted after snapshot creation if cinder nfs backend is used" [Undecided,New] | 10:20 |
| bauzas | so, Invalid, right ? ^ | 10:20 |
| bauzas | a snapshot is by definition immutable | 10:20 |
| lyarwood | the bug suggests we are writing to the snapshot and that would be a bug | 10:23 |
| lyarwood | it might be that they are just looking at the QEMU command line that isn't updated | 10:23 |
| bauzas | I'll ask them more then => Incomplete | 10:27 |
| *** tetsuro has quit IRC | 10:28 | |
| *** ttsiouts has quit IRC | 10:28 | |
| *** mkrai has quit IRC | 10:29 | |
| *** mkrai has joined #openstack-nova | 10:36 | |
| *** ttsiouts has joined #openstack-nova | 10:38 | |
| *** avolkov has quit IRC | 10:51 | |
| *** derekh has joined #openstack-nova | 10:58 | |
| hrw | is there a way to set cpu_mode/cpu_model in CI jobs? | 10:59 |
| *** mkrai has quit IRC | 11:16 | |
| *** mkrai has joined #openstack-nova | 11:16 | |
| sean-k-mooney | hrw: you can just set it in the devstack local.conf | 11:16 |
| hrw | sean-k-mooney: thanks | 11:17 |
| sean-k-mooney | hrw: like this https://opendev.org/x/networking-ovs-dpdk/src/branch/master/.zuul.yaml#L72-L86 | 11:17 |
| hrw | great! | 11:18 |
| sean-k-mooney | set cpu_mode: custome then cpu_model: whatever | 11:18 |
| hrw | yay | 11:18 |
| sean-k-mooney | also spelling custom correctly helps :) | 11:19 |
| hrw | nova-next has it too ;d | 11:19 |
| sean-k-mooney | im guessing its using an old model | 11:20 |
| sean-k-mooney | oh you just ment setting config values https://github.com/openstack/nova/blob/master/.zuul.yaml#L184-L195 | 11:21 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/nova master: [WIP] CI: add tempest-integrated-compute-aarch64 job https://review.opendev.org/714439 | 11:22 |
| hrw | let's check ;d | 11:22 |
| hrw | now it should be able to run VM instances | 11:22 |
| hrw | ~curse aarch64 for lack of nested virt | 11:23 |
| openstackgerrit | Wenping Song proposed openstack/nova master: Accurately clean up ARQs resources during build_instances() in conductor https://review.opendev.org/720439 | 11:23 |
| sean-k-mooney | oh nova next has bandwith aware schulding configured resource_provider_bandwidths: br-ex:1000000:1000000 | 11:23 |
| sean-k-mooney | i did not know we were running tempest test for that in nova next | 11:24 |
| gibi | sean-k-mooney: we have qos tempet tests | 11:24 |
| gibi | tempest | 11:24 |
| sean-k-mooney | gibi: i just did not know we were runing them in the gate. is there a reason to only run those in nova-next instead of all nova tempest jobs | 11:24 |
| sean-k-mooney | i gues it does not matter it should noe be affaced by say using ceph image backend or qcow | 11:26 |
| sean-k-mooney | so nova next is fine | 11:26 |
| gibi | sean-k-mooney: I think the original reason was to see if they are stable enoug | 11:26 |
| sean-k-mooney | well nova next is voting so if they would fail it would fail the ci run anyway | 11:26 |
| gibi | good point | 11:27 |
| sean-k-mooney | we dont really have a nova base job anymore that is based on zull v3 so without that it would be annoying to configure this in all the job anyway so its fine | 11:27 |
| gibi | yeah, it needs some zuul config to enable the tests | 11:28 |
| gibi | I can play with it after Ussuri is done | 11:28 |
| sean-k-mooney | this move operation support for this land this cycle or is that still pending | 11:28 |
| sean-k-mooney | i kind of lost track of that | 11:28 |
| *** artom has joined #openstack-nova | 11:29 | |
| sean-k-mooney | did you also need the allocation like we do for vgpus https://review.opendev.org/#/q/topic:bug/1778563+(status:open+OR+status:merged) | 11:30 |
| bauzas | gibi: I'm done for today with bug triage, we're down to 69 | 11:38 |
| bauzas | nothing really urgent afaics | 11:38 |
| bauzas | any other reviews needed ? should I say | 11:39 |
| bauzas | sean-k-mooney: hopefully, only a very few server actions miss allocations I think | 11:39 |
| bauzas | and that being said... | 11:40 |
| gibi | sean-k-mooney: support for move operations landed in Ussuri (tempest test is open) | 11:40 |
| bauzas | gibi: sean-k-mooney: stephenfin: I'd indeed appreciate reviews of https://review.opendev.org/#/q/topic:bug/1778563+(status:open+OR+status:merged) | 11:40 |
| gibi | sean-k-mooney: I don't think we need any extra support from the virt dirver side for qos | 11:40 |
| gibi | bauzas: thanks for the triage. | 11:40 |
| gibi | bauzas: and yes, I still have your patch series open | 11:41 |
| bauzas | gibi: I'll continue tomorrow morning | 11:41 |
| gibi | bauzas: thanks | 11:41 |
| bauzas | but I think I'll open my review dashboard for bugs and see what to chime in | 11:41 |
| bauzas | folks, a nova-core is looking for bugfixes to review, please hassle him <= | 11:41 |
| bauzas | err, | 11:42 |
| bauzas | <= please hassle him | 11:42 |
| gibi | :) | 11:43 |
| lyarwood | bauzas: do you also take trivial cleanups and fups? | 11:50 |
| *** ratailor has quit IRC | 11:50 | |
| lyarwood | bauzas: https://review.opendev.org/#/c/702021/ & https://review.opendev.org/#/c/711679/ for example | 11:50 |
| *** dpawlik has quit IRC | 11:50 | |
| *** dpawlik has joined #openstack-nova | 11:50 | |
| openstackgerrit | Lee Yarwood proposed openstack/nova stable/train: Avoid spurious error logging in _get_compute_nodes_in_db https://review.opendev.org/702902 | 11:52 |
| openstackgerrit | Lee Yarwood proposed openstack/nova stable/train: Reject boot request for unsupported images https://review.opendev.org/708577 | 11:53 |
| *** mkrai has quit IRC | 11:54 | |
| *** JamesBenson has joined #openstack-nova | 11:57 | |
| *** spsurya_ has quit IRC | 11:58 | |
| *** smcginnis has joined #openstack-nova | 12:01 | |
| *** ociuhandu has quit IRC | 12:04 | |
| *** ociuhandu has joined #openstack-nova | 12:05 | |
| *** nweinber has joined #openstack-nova | 12:06 | |
| *** rpittau|bbl is now known as rpittau | 12:07 | |
| *** ociuhandu has quit IRC | 12:09 | |
| *** dpawlik has quit IRC | 12:09 | |
| *** dpawlik has joined #openstack-nova | 12:10 | |
| *** psachin has quit IRC | 12:16 | |
| *** ociuhandu has joined #openstack-nova | 12:28 | |
| *** udesale_ has joined #openstack-nova | 12:31 | |
| *** mkrai has joined #openstack-nova | 12:33 | |
| *** udesale has quit IRC | 12:34 | |
| *** tbachman has quit IRC | 12:44 | |
| *** tbachman has joined #openstack-nova | 12:45 | |
| *** links has quit IRC | 12:45 | |
| *** ociuhandu has quit IRC | 12:46 | |
| bauzas | lyarwood: sorry, was at lunch, but indeed, everything can be accepted, I'm not the kind of person who tell you at the door 'sorry, but we can't accept you because of your shoes' | 12:46 |
| *** ociuhandu has joined #openstack-nova | 12:46 | |
| stephenfin | lyarwood: those are dodgy shoes though | 12:47 |
| *** vishalmanchanda has quit IRC | 12:50 | |
| *** ociuhandu has quit IRC | 12:51 | |
| *** mkrai has quit IRC | 12:55 | |
| *** links has joined #openstack-nova | 12:56 | |
| *** mriedem has joined #openstack-nova | 13:04 | |
| *** mkrai has joined #openstack-nova | 13:04 | |
| lyarwood | stephenfin: white trainers ftw | 13:09 |
| bauzas | lyarwood: get off my lawn ! | 13:11 |
| bauzas | :p | 13:11 |
| * bauzas actually has a gate. And a dog, which barfs. | 13:11 | |
| bauzas | heh, s/barf/bark of course | 13:12 |
| bauzas | (pardon my French (c) ) | 13:12 |
| *** ociuhandu has joined #openstack-nova | 13:16 | |
| kashyap | lyarwood: Zuul (grenade-py3) failure on this: https://review.opendev.org/#/c/702021/ | 13:21 |
| kashyap | If you've already seen it, disregard me. | 13:21 |
| lyarwood | kashyap: yeah unrelated | 13:24 |
| lyarwood | thanks for looking | 13:25 |
| *** eharney has joined #openstack-nova | 13:26 | |
| kashyap | bauzas: FWIW, this is straightforward to just merge this: https://review.opendev.org/#/c/702021/ (libvirt: Remove VIR_DOMAIN [...]) | 13:26 |
| kashyap | (And the 'qemu-img' one, too.) | 13:28 |
| *** mkrai has quit IRC | 13:36 | |
| *** kevinz has quit IRC | 13:51 | |
| *** yaawang_ has joined #openstack-nova | 13:59 | |
| *** yaawang has quit IRC | 14:01 | |
| *** ociuhandu has quit IRC | 14:17 | |
| *** jdillaman has joined #openstack-nova | 14:22 | |
| *** dklyle has joined #openstack-nova | 14:24 | |
| *** amodi has quit IRC | 14:25 | |
| *** amodi has joined #openstack-nova | 14:26 | |
| *** lpetrut has quit IRC | 14:34 | |
| *** martinkennelly has quit IRC | 14:36 | |
| *** martinkennelly has joined #openstack-nova | 14:37 | |
| *** mlavalle has joined #openstack-nova | 14:38 | |
| *** gyee has joined #openstack-nova | 14:42 | |
| *** ttsiouts has quit IRC | 14:43 | |
| *** mkrai has joined #openstack-nova | 14:43 | |
| *** ttsiouts has joined #openstack-nova | 14:46 | |
| gibi | lyarwood, bauzas, stephenfin: Am I missing something here https://review.opendev.org/#/c/711679/5/nova/virt/images.py@42 ? | 14:47 |
| stephenfin | Oh, it looks like it | 14:49 |
| lyarwood | gibi: nope it's unused | 14:50 |
| sean-k-mooney | well its used here https://review.opendev.org/#/c/711679/5/nova/virt/images.py@48 | 14:50 |
| sean-k-mooney | but i dont know if its everset to anything other then json | 14:50 |
| bauzas | gibi: lyarwood: wait, it can be | 14:50 |
| bauzas | sec, finding the github link | 14:51 |
| *** ttsiouts has quit IRC | 14:51 | |
| bauzas | gibi: lyarwood: stephenfin: https://github.com/openstack/nova/blob/e1359567e4985e9a671359d4c0d53404a8ba64ab/nova/virt/libvirt/utils.py#L224 | 14:51 |
| stephenfin | bauzas: that's format, not output_format | 14:52 |
| * bauzas facepalms | 14:52 | |
| sean-k-mooney | looking at the reference on github the only place output_formate is ever set today | 14:53 |
| sean-k-mooney | is https://github.com/openstack/nova/blob/e1359567e4985e9a671359d4c0d53404a8ba64ab/nova/tests/unit/virt/libvirt/test_utils.py#L167 | 14:53 |
| sean-k-mooney | where it is currently set to json | 14:53 |
| gibi | sean-k-mooney: at https://review.opendev.org/#/c/711679/5/nova/virt/images.py@48 we can hardcode the output_format to 'json' | 14:54 |
| sean-k-mooney | gibi: yes | 14:54 |
| sean-k-mooney | and remove the parmanter form our method | 14:54 |
| stephenfin | burn it with fire | 14:54 |
| lyarwood | there's actually a place in the driver as well | 14:54 |
| lyarwood | but we can drop it from there | 14:54 |
| lyarwood | 1 sec | 14:54 |
| * stephenfin actually <3s functional tests | 14:55 | |
| *** ttsiouts has joined #openstack-nova | 14:55 | |
| sean-k-mooney | so useing codesearch http://codesearch.openstack.org/?q=qemu_img_info&i=nope&files=&repos=openstack/nova | 14:55 |
| sean-k-mooney | its either not set or set to json our side of the funtion definitions | 14:57 |
| sean-k-mooney | since you change the defualt to json in the patch any fucntion that did not set it already works with json | 14:58 |
| sean-k-mooney | so ya i think we can just hardcode it and drop it | 14:58 |
| sean-k-mooney | or at least push a patch to do that and see what breaks | 14:58 |
| *** ttsiouts has quit IRC | 14:59 | |
| stephenfin | bauzas: comments left on https://review.opendev.org/#/c/712741/ | 14:59 |
| stephenfin | Looks like they're mostly the same as sean-k-mooney's (de-duping stuff) | 14:59 |
| bauzas | stephenfin: ack, looking | 14:59 |
| *** ttsiouts has joined #openstack-nova | 14:59 | |
| stephenfin | dansmith, gibi, melwitt: Thoughts on merging https://review.opendev.org/#/c/589085/ ? | 15:00 |
| stephenfin | given that it touches the driver API. Is it still okay to land after feature freeze? | 15:01 |
| sean-k-mooney | ya i had that question too but the notifcation was also already sent to the mailing list of the change about a month ago | 15:06 |
| gibi | we need to figure out how risky is this change. what can we break? | 15:07 |
| dansmith | what's the point of merging that, if not just to enable a feature? | 15:08 |
| dansmith | IMHO, unless it gets us something by landing it in U to set up for fewer migration problems in V, it's really not worth doing at this point | 15:08 |
| dansmith | meaning, if this got us a cycle earlier on data migration or something, but I don't see that unless I'm missing something | 15:09 |
| bauzas | gibi: dansmith: sean-k-mooney: stephenfin: I wrote a ML thread 1 month ago about this and honestly it's an internal API | 15:13 |
| sean-k-mooney | this is required for resize its not related to the multi gpu types so it has no impact on upgrde or data migrations | 15:13 |
| bauzas | and there is no object impact or RPC call involved in any matter | 15:13 |
| dansmith | sean-k-mooney: right | 15:14 |
| *** happyhemant has quit IRC | 15:14 | |
| dansmith | sean-k-mooney: that's my point.. if we're not going to merge the feature it enables, and it doesn't set us up for something early, then merging it now vs. as soon as V opens doesn't matter, except for risk and change yeah? | 15:14 |
| bauzas | there is no feature behind it | 15:15 |
| bauzas | we already support instances resizes | 15:15 |
| dansmith | bauzas: oh is it a bug fix? | 15:15 |
| bauzas | yup | 15:15 |
| bauzas | I wouldn't beg for a FFE | 15:16 |
| bauzas | if so | 15:16 |
| bauzas | https://bugs.launchpad.net/nova/+bug/1778563 | 15:16 |
| openstack | Launchpad bug 1778563 in OpenStack Compute (nova) "Resize/Cold-migrate doesn't recreate vGPUs" [High,In progress] - Assigned to Sylvain Bauza (sylvain-bauza) | 15:16 |
| bauzas | we just blindly resize the instance with forgetting the vgpus | 15:16 |
| dansmith | ah, sorry I figured this was for new stuff | 15:16 |
| bauzas | so, 'whoops, sorry' basically | 15:16 |
| bauzas | adding a new call to _allocate_mdevs() will just return if no VGPUs involvedf | 15:17 |
| bauzas | that's basically a noop | 15:17 |
| bauzas | (for instances with no VGPUs in there) | 15:17 |
| dansmith | is this new in U or latent since T? | 15:17 |
| bauzas | latent since Queens :) | 15:17 |
| bauzas | https://review.opendev.org/#/c/720250/1/doc/source/admin/virtual-gpu.rst | 15:18 |
| dansmith | ah, then probably not worth rushing into U then | 15:18 |
| dansmith | I haven't reviewed the stack so I'm not sure how much risk, granted | 15:18 |
| dansmith | but if no object, rpc, or db changes then it could be backported even right? | 15:18 |
| stephenfin | I don't think so. It has virt driver API changes, which means a backport would break out-of-tree drivers | 15:19 |
| bauzas | that's the issue i see ^ | 15:19 |
| stephenfin | I'd have approved this already were it not for that | 15:19 |
| bauzas | I warned the out-of-tree driver maintainers that we were changing the resize signature | 15:20 |
| stephenfin | this hunk, basically https://review.opendev.org/#/c/589085/8/nova/virt/driver.py | 15:20 |
| dansmith | I'm not aware of ever blocking a backport on those grounds | 15:20 |
| dansmith | not sure why that would matter at all, when it doesn't for master | 15:20 |
| stephenfin | o rly? | 15:20 |
| bauzas | I dunno, honestly | 15:20 |
| bauzas | either way, if folks prefer to punt it to V, fair enough | 15:21 |
| stephenfin | I figured people had notice on master and would expect to have to do some work with a new major version | 15:21 |
| stephenfin | not so a new minor one | 15:21 |
| bauzas | I just thought it wouldn't harm (c) :p | 15:21 |
| artom | dansmith, not sure we ever blocked a backport, but there was a case where we made a point of *not* changing the driver interface to enable the backport | 15:22 |
| artom | And then in a FUP re-did the thing the "correct" way | 15:22 |
| artom | https://review.opendev.org/#/c/670645/2/nova/virt/libvirt/driver.py@9045 | 15:22 |
| dansmith | artom: because it was doable without I guess? either way, I bet I wasn't the one worried about it :) | 15:23 |
| bauzas | dansmith: stephenfin: gibi: shouldn't we just discuss this as a 'Stuck review' during the nova meeting in 48 mins ? | 15:23 |
| bauzas | and find a consensus there so we move on | 15:23 |
| artom | dansmith, it may have been mriedem, don't remember | 15:23 |
| dansmith | sounds like him :) | 15:23 |
| stephenfin | bauzas: um, sure, why not | 15:23 |
| bauzas | again, that doesn't really change my business | 15:24 |
| bauzas | if we go accepting this change, that's cool | 15:24 |
| artom | (And here's the "correct" fup: https://review.opendev.org/#/c/668631/5 ) | 15:24 |
| bauzas | if we prefer be conservative, then meh | 15:24 |
| * artom avoids the conservative MANA joke | 15:24 | |
| bauzas | but I just feel we somehow need to decide on what to do with internal virt API changes when we're post FF | 15:24 |
| * artom not really | 15:24 | |
| artom | ... and I totally messed it up anyways | 15:25 |
| *** dpawlik has quit IRC | 15:26 | |
| dansmith | the actual fix is the third patch right? | 15:26 |
| *** dpawlik has joined #openstack-nova | 15:26 | |
| sean-k-mooney | yes the rest is just passing the allocations | 15:28 |
| * gibi is reading back | 15:28 | |
| bauzas | dansmith: yup | 15:28 |
| sean-k-mooney | you could actully just use the reportclient to get the allcoation in the driver to avoid the virt api change too but that is less clean. | 15:30 |
| dansmith | I'll have to go learn what _allocate_mdevs() does in order to review that patch I guess | 15:30 |
| dansmith | sean-k-mooney: yeah, don't do that, IMHO | 15:30 |
| bauzas | sean-k-mooney: NO | 15:30 |
| sean-k-mooney | bauzas: im not saying we should | 15:31 |
| bauzas | the virt driver is an island :) | 15:31 |
| *** ttsiouts_ has joined #openstack-nova | 15:31 | |
| bauzas | dansmith: _allocate_mdevs() is the process that checks the configuration to see whether it should assign some kernel mediated devices (in other words, vgpus) to the guest XML | 15:32 |
| dansmith | bauzas: right, but if we're reverting, it had an mdev before right? | 15:33 |
| dansmith | I just need to look at how all that works is all | 15:33 |
| mriedem | what did i do? | 15:33 |
| gibi | bauzas: if we not merge the virt driver interface change in U would you like to backport the V change to stable/U later? | 15:33 |
| mriedem | oh the TypeError check on a backport? | 15:34 |
| bauzas | gibi: yeah, if that doesn't break the stable policy, which I'm unsure | 15:34 |
| bauzas | stable rules* policy | 15:34 |
| *** ttsiouts has quit IRC | 15:34 | |
| alex_xu | gibi: sean-k-mooney, we are just working on ensuring the requirement on https://review.opendev.org/#/q/topic:bp/provider-config-file+(status:open+OR+status:merged | 15:35 |
| sean-k-mooney | bauzas: altering the virt driver internal api i think should be a stable policy violation even if its not a public api | 15:35 |
| dansmith | totes disagree :) | 15:35 |
| dansmith | as you all know | 15:35 |
| gibi | it is basically a decision between changing the virt driver interface in a major versior (now) or in a minor version (later in a backport) | 15:36 |
| sean-k-mooney | yes although if we really dont want to provide any support for out of tree driver i would suggest we also stop sending the email when we change things | 15:36 |
| sean-k-mooney | which i think you are also in faovr of | 15:36 |
| dansmith | indeed :) | 15:36 |
| bauzas | that's the exact reason why I want to raise it as a stuck review | 15:37 |
| bauzas | go fight it ! | 15:37 |
| bauzas | and honestly, I don't really care of shipping it in U if you wanna get MHO | 15:37 |
| bauzas | because Red Hat... | 15:37 |
| sean-k-mooney | bauzas: we are not planning to back port this downstream beyond ussuri right? | 15:37 |
| bauzas | sean-k-mooney: don't say gross words | 15:38 |
| bauzas | we *could* but do we want, honestly ? :) | 15:38 |
| *** ttx has quit IRC | 15:38 | |
| *** tbarron_ has joined #openstack-nova | 15:38 | |
| *** ttx has joined #openstack-nova | 15:38 | |
| bauzas | dansmith: for the revert case, that's an interesting question | 15:38 |
| sean-k-mooney | just triple checing that we will deliver this to customers in 17 which will be based on victoria | 15:39 |
| gibi | alex_xu: thanks. If it turns out that there are requirements on Intel side then will there be some dev from intel to push the patches in V? | 15:39 |
| bauzas | oh man, don't get me spilling my business here :) | 15:39 |
| bauzas | but yeah, 17 so U or V, I doesn't matter | 15:39 |
| alex_xu | gibi: yes, I think so | 15:39 |
| gibi | alex_xu: cool. thanks. let me know the result of your investigation | 15:40 |
| alex_xu | gibi: yea, I will let you know whatever the result coming out | 15:40 |
| openstackgerrit | Lee Yarwood proposed openstack/nova master: images: Make JSON the default output format of calls to qemu-img info https://review.opendev.org/711679 | 15:40 |
| lyarwood | ^ now with added unit tests, thanks again all. | 15:40 |
| *** martinkennelly has quit IRC | 15:43 | |
| *** martinkennelly has joined #openstack-nova | 15:44 | |
| sean-k-mooney | bauzas: for the revert case we have not undeifed the domain on the source node when we trigger revert | 15:45 |
| *** ociuhandu has joined #openstack-nova | 15:45 | |
| sean-k-mooney | so im not sure the allocation are truly needed as we should not need to regenerate the xml | 15:46 |
| sean-k-mooney | that said we might do anyway depending on how we restart the vm | 15:46 |
| bauzas | sean-k-mooney: we regenerate the XML https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L10259 | 15:46 |
| sean-k-mooney | if we just call power on it call hard reboot which would | 15:47 |
| *** ttsiouts_ has quit IRC | 15:47 | |
| sean-k-mooney | ya ok so that is why we need it in this case. | 15:48 |
| bauzas | yuuuuuup | 15:48 |
| bauzas | that does sound crazy, but we blindy recreate the guest when reverting | 15:49 |
| bauzas | sean-k-mooney: see my functtest, you'll see evidence of it | 15:49 |
| sean-k-mooney | im not sure if we really need to regenerate the xml in that case i cant think of why but lets not get distracted by that for now | 15:49 |
| dansmith | sean-k-mooney: right that was my exact concern | 15:49 |
| *** martinkennelly has quit IRC | 15:50 | |
| sean-k-mooney | oh i know why i think | 15:50 |
| dansmith | where does allocate_mdevs persist? does it look at all the other mdevs currently being used? | 15:50 |
| *** ociuhandu has quit IRC | 15:50 | |
| dansmith | meaning, when we undefine the original and re-define with a new one, will the old one be available again, or might we leak something? | 15:50 |
| sean-k-mooney | we regenerate the xml becasue snapshots | 15:50 |
| sean-k-mooney | this code https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L10249-L10254 | 15:50 |
| bauzas | dansmith: mdevs aren't persisted | 15:51 |
| bauzas | mdevs can be even pre-provisioned | 15:51 |
| dansmith | sean-k-mooney: also maybe because you can attach a volume to the new one before revert | 15:51 |
| dansmith | bauzas: how do we select one that isn't in use? that's what I'm getting at | 15:51 |
| bauzas | "allocating a mdev" just means modifying the guest XML domain to use a specific sysfs object | 15:51 |
| bauzas | dansmith: oh this, sec | 15:52 |
| dansmith | right, I know, but we need to make sure we hand it one not in use | 15:52 |
| bauzas | dansmith: https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L7206 | 15:52 |
| dansmith | bauzas: so we create a new one in sysfs? | 15:52 |
| sean-k-mooney | ya i guess that is ture although im realy not sure why we allow volume or interface attach in resize_verify or whatever that state is called | 15:53 |
| dansmith | bauzas: how does the old one get reaped when we redefine? | 15:53 |
| bauzas | dansmith: which is made by comparing all the existing mdevs and the ones that are in use by each domain | 15:53 |
| bauzas | https://github.com/openstack/nova/blob/e1359567e4985e9a671359d4c0d53404a8ba64ab/nova/virt/libvirt/driver.py#L7088-L7102 | 15:53 |
| sean-k-mooney | but given that we do we cant assume the current xml is correct | 15:53 |
| dansmith | sean-k-mooney: we do, but it's crazy | 15:53 |
| dansmith | bauzas: right so when we revert, we'll see that our original one is in use (by us) and allocate a new one right? | 15:54 |
| bauzas | dansmith: during the resize, before the confirm, we have a doubled allocation of mdevs | 15:54 |
| bauzas | (and a double placement allocation) | 15:55 |
| bauzas | dansmith: when we revert, the target mdev allocation disappears because we undefine the domain | 15:55 |
| dansmith | bauzas: but that doubling is on two different hosts | 15:55 |
| sean-k-mooney | https://github.com/openstack/nova/blob/e1359567e4985e9a671359d4c0d53404a8ba64ab/nova/virt/libvirt/driver.py#L6349 | 15:55 |
| sean-k-mooney | it look like the domain creation fucntion will overwrite existing domains | 15:56 |
| dansmith | bauzas: I'm talking about when we revert, don't we hit the "no free mdevs, allocate a new one" in _allocate_mdevs() since our original mdev is still in use by the libvirt domain we're about to redefine? | 15:56 |
| bauzas | dansmith: yeah https://review.opendev.org/#/c/712741/2/nova/tests/functional/libvirt/test_vgpu.py@224 | 15:56 |
| bauzas | dansmith: I see your point and I honestly have to double check | 15:56 |
| sean-k-mooney | oh you were asking about the allcotaion fo the mdevs | 15:57 |
| dansmith | sean-k-mooney: yes | 15:57 |
| bauzas | dansmith: found why | 15:57 |
| sean-k-mooney | ya it should get the exising ones and reuse them | 15:57 |
| bauzas | dansmith: sec, giving you the link | 15:57 |
| gibi | nova meeting starts in 2 minutes on #openstack-meeting-3 | 15:59 |
| *** ociuhandu has joined #openstack-nova | 16:02 | |
| *** ttsiouts has joined #openstack-nova | 16:04 | |
| melwitt | lbragstad: hey, would you be able to take a gander at the proposed doc patch for policy scopes https://review.opendev.org/#/c/720129/4/doc/source/configuration/policy-new-defaults.rst we want to have it clearly explain what are the action items for operators to be able to use scope types. i.e. will it involve creating new keystone roles and stuff like that | 16:15 |
| lbragstad | melwitt sure - i can take a look | 16:15 |
| melwitt | thanks | 16:15 |
| *** rpittau is now known as rpittau|afk | 16:18 | |
| gmann | melwitt: quota class policy is ready, fixed the johnthetubaguy comment of removing the non-working 'quota_class' from check_str - https://review.opendev.org/#/c/720042/1 | 16:21 |
| gmann | and its base pacth | 16:21 |
| *** tesseract has quit IRC | 16:21 | |
| melwitt | gmann: ok, I wanted johnthetubaguy to look at it first and make sure he's ok with it since he found issue earlier | 16:21 |
| gmann | melwitt: ok | 16:22 |
| *** abhishek has quit IRC | 16:25 | |
| *** mgariepy has quit IRC | 16:29 | |
| * bauzas loves how the libvirt driver is cryptic | 16:30 | |
| *** yaawang has joined #openstack-nova | 16:32 | |
| gmann | melwitt: thanks for nice feedback on doc patch, I will wait for other comments and update it together. | 16:32 |
| melwitt | gmann: cool sounds good | 16:33 |
| *** yaawang_ has quit IRC | 16:33 | |
| *** udesale_ has quit IRC | 16:35 | |
| *** dtantsur is now known as dtantsur|afk | 16:36 | |
| *** evrardjp has quit IRC | 16:37 | |
| *** evrardjp has joined #openstack-nova | 16:37 | |
| *** ttsiouts has quit IRC | 16:42 | |
| *** ttsiouts has joined #openstack-nova | 16:42 | |
| bauzas | dansmith: hah, found why it wasn't leaving a VGPU allocated when you revert https://github.com/openstack/nova/blob/e1359567e4985e9a671359d4c0d53404a8ba64ab/nova/virt/libvirt/driver.py#L6359 | 16:54 |
| bauzas | dansmith: we redefine the existing domain here ^ | 16:54 |
| dansmith | I don't think that answers my question | 16:56 |
| dansmith | when we delete a domain, do we unallocate the mdev, or do we just leave it so the next spawn that needs one will find it? | 16:57 |
| bauzas | dansmith: when we undefine the domain, the mdev is consider being freed | 17:00 |
| *** tframbo has joined #openstack-nova | 17:00 | |
| dansmith | bauzas: "considered free" because it's not assigned to any other domain, but still allocated in sysfs yeah? | 17:00 |
| sean-k-mooney | tframbo: so first the namespace should be hw not os | 17:00 |
| bauzas | dansmith: yup, you're correct | 17:00 |
| sean-k-mooney | that extra specs as you know alter the cpu feature flags reported to the guest and hide the fact its a vm form guest os | 17:01 |
| tframbo | so it is add compat for "hw:hide_hypervisor_id” | 17:01 |
| tframbo | ? | 17:01 |
| sean-k-mooney | yes so we woudl deprecate hide_hypervisor_id and add "hw:hide_hypervisor_id” | 17:01 |
| sean-k-mooney | make the diver accept both and then we could special case hide_hypervior_id so it is ignored | 17:02 |
| bauzas | sean-k-mooney: wait, like I said, the bug is about a scheduling decision | 17:02 |
| bauzas | sean-k-mooney: not a boot problem | 17:02 |
| sean-k-mooney | you would use aggregate_instance_extra_specs:hide_hypervior_id in the flavor if you wanted to match on it in that filter | 17:02 |
| bauzas | OVH and a couple of other folks already use this knob | 17:03 |
| bauzas | which was merged a while ago | 17:03 |
| sean-k-mooney | bauzas: yes i know but the current filter check all unamesapced extra specs and assert they are in the metadta | 17:03 |
| bauzas | but here, what's wanted is to trick the scheduler to automatically accept hosts on such things | 17:03 |
| bauzas | sean-k-mooney: I explained this can be achieved without amending with this filter | 17:04 |
| sean-k-mooney | how | 17:04 |
| bauzas | tbc : create a flavor with a random keypair | 17:04 |
| bauzas | create aggregates with hosts having the same keypair in their metadata | 17:05 |
| bauzas | and you're done | 17:05 |
| bauzas | you don't need to use this specific property | 17:05 |
| sean-k-mooney | your missing the point | 17:05 |
| bauzas | which is ? | 17:06 |
| sean-k-mooney | we cant remove hide_hypervisor_id form the current flavor without changing the behavior | 17:06 |
| sean-k-mooney | and if it si set in the falvor then the AggregateInstanceExtraSpecsFilter will check it against the aggreate metadata | 17:06 |
| dansmith | bauzas: so my point is, when we revert, we will allocate a second mdev, and then abandon the original one when we redefine.. but I guess like delete, that one then will get used the next time we go to spawn something that needs one | 17:07 |
| sean-k-mooney | we will return false here https://github.com/openstack/nova/blob/46a3bcd80b41e99ec4923c7cf3d0f8dd8505e97c/nova/scheduler/filters/aggregate_instance_extra_specs.py#L62-L68 | 17:07 |
| sean-k-mooney | tframbo: the other workaround is to add the metadata key to the aggreate | 17:07 |
| bauzas | dansmith: you're probably correct, we could have a window in time where we would allocate a new mdev and then releasing the other | 17:08 |
| bauzas | which could lead to a capacity issue if you're out of available mdevs | 17:08 |
| sean-k-mooney | tframbo: basically create two two new agggreats that conatin all hosts, then add hide_hypervisor_id=true to one and hide_hypervisor_id=false to the other | 17:08 |
| bauzas | dansmith: s/probably/absolutely after re-thinking of it | 17:09 |
| dansmith | bauzas: right | 17:09 |
| sean-k-mooney | tframbo: if you do that we will get to here https://github.com/openstack/nova/blob/46a3bcd80b41e99ec4923c7cf3d0f8dd8505e97c/nova/scheduler/filters/aggregate_instance_extra_specs.py#L69-L71 and break then return ture and pass the host | 17:09 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Add new default roles in remaining servers policies https://review.opendev.org/720116 | 17:10 |
| openstackgerrit | Merged openstack/nova master: Fix servers policy for admin_or_owner https://review.opendev.org/718501 | 17:10 |
| bauzas | dansmith: this being said, I don't exactly know how to fix this, since this is the whole revert workflow | 17:10 |
| openstackgerrit | Merged openstack/nova master: Add test coverage of existing server policies https://review.opendev.org/718348 | 17:10 |
| bauzas | and I guess the problem remains for all other resources | 17:10 |
| openstackgerrit | Merged openstack/nova master: Introduce scope_types in servers Policies https://review.opendev.org/662968 | 17:10 |
| openstackgerrit | Merged openstack/nova master: Add new default roles in servers policies https://review.opendev.org/662971 | 17:10 |
| openstackgerrit | Merged openstack/nova master: Add test coverage of existing server attributes policies https://review.opendev.org/719726 | 17:10 |
| openstackgerrit | Merged openstack/nova stable/train: Reset the cell cache for database access in Service https://review.opendev.org/718934 | 17:11 |
| openstackgerrit | Merged openstack/nova stable/stein: Use placement stable version for functional job https://review.opendev.org/719121 | 17:11 |
| bauzas | dansmith: we could consider to special case the revert and find the existing mdev so we would pass it again instead of calling _allocate_mdevs() tho | 17:11 |
| dansmith | bauzas: seems like that would be better | 17:12 |
| dansmith | also avoids needing to grab the allocations right? | 17:12 |
| *** mkrai has quit IRC | 17:12 | |
| bauzas | indeed | 17:12 |
| bauzas | dansmith: <3 with love | 17:12 |
| bauzas | that's what I call a top-notch review | 17:12 |
| *** ociuhandu has quit IRC | 17:13 | |
| sean-k-mooney | tframbo: does my responce on the bug or above make sense too you | 17:15 |
| tframbo | sean-k-mooney: yeah, we could also add the aggregate metadata "hide_hypervisor_id=true" to solve the bug. | 17:15 |
| sean-k-mooney | yes | 17:16 |
| tframbo | thank you sean-k-mooney and bauzas | 17:16 |
| sean-k-mooney | do you also use the compute capability filter by the way | 17:16 |
| sean-k-mooney | if not be aware it has the same behavior | 17:16 |
| tframbo | I think the hide_hypervisor_id is bool before , sorry | 17:16 |
| sean-k-mooney | tframbo: its is a bool that default to false if unset | 17:17 |
| sean-k-mooney | so there isnt really a reason to ever set it to false explcitly | 17:18 |
| bauzas | tframbo: sorry I dropped focus | 17:18 |
| sean-k-mooney | unless you want to prevent images form settin git too true | 17:18 |
| bauzas | sean-k-mooney: tframbo: so, again, the intent is to have the aggregate metadata key to be set | 17:20 |
| tframbo | sean-k-mooney: yes ,we alse use the compute capability filter by the way | 17:21 |
| tframbo | bauzas: yes | 17:21 |
| bauzas | I honestly don't get the problem | 17:22 |
| sean-k-mooney | tframbo: ok so the point i wanted to make in that case was that if you use both then you are required to not use unnamespaced keys or they can conflict | 17:22 |
| *** mgariepy has joined #openstack-nova | 17:22 | |
| bauzas | wait | 17:23 |
| bauzas | the namespace is only for specifying that the key is for this filter | 17:23 |
| sean-k-mooney | bauzas: the problem is the AggregateInstanceExtraSpecsFilter require that an aggreate assocaited with a host must have all extra specs in the aggregate_instance_extra_specs as well as all non namespaced extra specs in the metadta | 17:23 |
| bauzas | I just feel it's a configuration problem | 17:23 |
| bauzas | or prefixed by aggregate_instance_extra_specs https://github.com/openstack/nova/blob/46a3bcd80b41e99ec4923c7cf3d0f8dd8505e97c/nova/scheduler/filters/aggregate_instance_extra_specs.py#L54 | 17:24 |
| bauzas | and this is not a bug | 17:24 |
| sean-k-mooney | bauzas: the filter also looks at non namespced extra specs and we only have 1 standard extra specs that falls into that catagory | 17:24 |
| bauzas | this *works* with any random key | 17:25 |
| bauzas | I don't get the point again | 17:25 |
| bauzas | if you're using this flavor key for passing QEMU, fine | 17:25 |
| bauzas | that's how it's intended to be used | 17:25 |
| sean-k-mooney | hide_hypervisor_id is not a random custom key | 17:25 |
| bauzas | but then, indeed, you need to have aggregates matching it | 17:25 |
| bauzas | (if you use this filter) | 17:26 |
| sean-k-mooney | its a stanard one and its the only standard extra spec without a namesapce | 17:26 |
| tframbo | https://docs.openstack.org/nova/latest/user/flavors.html | 17:26 |
| sean-k-mooney | so its also the only stanard extrapec the filter uncondtionally checks | 17:26 |
| bauzas | this filter doesn't care a single bit about standard extra specs | 17:26 |
| bauzas | you're confusing with the other filters that do care of those prefixes | 17:27 |
| sean-k-mooney | bauzas: no im not | 17:27 |
| bauzas | honestly, I'm done for the day | 17:27 |
| sean-k-mooney | to be clear the bug was intoduceing hide_hypervisor_id without a namespace | 17:27 |
| tframbo | sean-k-mooney: thank you, I will sleep , good night | 17:27 |
| sean-k-mooney | tframbo: night o/ | 17:28 |
| *** tframbo has quit IRC | 17:28 | |
| melwitt | sorry I didn't understand all the previous details but why does the filter require a namespace in order to work? should it be able to work with namespaced and non-namespaced extra specs? | 17:28 |
| melwitt | *shouldn't it | 17:28 |
| sean-k-mooney | melwitt: the filter iterages over all extra specs in the flavor and then asserts they match the metadta if and only if the start with the filters namespace or they have no namespce | 17:29 |
| bauzas | melwitt: that's my thinkings | 17:29 |
| sean-k-mooney | so for all other standard extra specs they are ingore because they have a namespace which is not the filters one | 17:30 |
| bauzas | but apparently we need to namespace now... | 17:30 |
| sean-k-mooney | since this extra spec has no namespace its check by defualt which no other standar extraspec is | 17:30 |
| sean-k-mooney | so there is a behavioral difference | 17:30 |
| bauzas | sean-k-mooney: so | 17:30 |
| bauzas | sean-k-mooney: https://github.com/openstack/nova/blob/46a3bcd80b41e99ec4923c7cf3d0f8dd8505e97c/nova/scheduler/filters/aggregate_instance_extra_specs.py#L55-L58 | 17:30 |
| melwitt | sean-k-mooney: wait but you say "if they have no namespace", doesn't this have no namespace and therefore should be considered? | 17:30 |
| melwitt | gah this is so confusing | 17:31 |
| bauzas | sean-k-mooney: this conditional is here to *PREVENT* other standard keys are ARE prefixed to be read | 17:31 |
| sean-k-mooney | melwitt yes this has no namespace an by the filter logic should be check | 17:31 |
| sean-k-mooney | melwitt: however if you add any other standard extra spec you do not have to update the metadta to boot a vm | 17:31 |
| melwitt | so ... what's the problem? that makes it sound like there's no bug | 17:31 |
| sean-k-mooney | for this extra spec you do | 17:31 |
| bauzas | melwitt: there is NO bug in my mind | 17:32 |
| bauzas | IMHO the bug should be consider Invalid if not Expired | 17:32 |
| sean-k-mooney | bauzas: i strongly dissagre. as i said the bug is not in the filter | 17:32 |
| sean-k-mooney | the bug is that we added a stanard extra spec without a namespace | 17:32 |
| bauzas | and what's the impact then ? | 17:33 |
| sean-k-mooney | by intoducing a flavor extra spec without a namespace, to use the feature enable by that extra spec it addtionally required the operator to update the aggreate metatad and host capabilty if they use the ComputeCapabilitiesFilter or AggregateInstanceExtraSpecsFilter | 17:35 |
| bauzas | oh, the fact that you need to create aggregates in order to use it, let me bet ? | 17:35 |
| sean-k-mooney | yes | 17:35 |
| sean-k-mooney | that is a behavioral differen form every other extra spec that is defined by nova | 17:35 |
| bauzas | holy f..., gotcha | 17:35 |
| sean-k-mooney | melwitt: do you follow too ^ | 17:36 |
| bauzas | okay, so indeed the fix is not about the filter | 17:36 |
| bauzas | it's about where we use this extraspec for booting | 17:36 |
| melwitt | I see, ok. just read through the bug and comments again. yeah, I think so. this is an extra spec meant to be used without having to add a matching metadata key on an aggregate | 17:37 |
| sean-k-mooney | ya so the fix is add a namespaced version of the extra spec and compat code | 17:37 |
| *** rcernin has quit IRC | 17:37 | |
| sean-k-mooney | then deprecate teh non namespaced one | 17:37 |
| melwitt | yeah, I think that makes sense. agree that's the way to fix | 17:37 |
| bauzas | sean-k-mooney: easy fix then https://review.opendev.org/#/c/555861/10/nova/virt/libvirt/driver.py | 17:37 |
| bauzas | sean-k-mooney: just add another key there with a prefix and just provide a deprecation warning for the existing one | 17:38 |
| sean-k-mooney | yep | 17:38 |
| bauzas | that's all flavors and aggregates, we don't need to care about the interop | 17:38 |
| sean-k-mooney | im wondering if we should also add a config option for the filter to disable checking unnamesapced extraspecs | 17:39 |
| bauzas | tlrambo dropped but I'll leave a comment in the bug | 17:39 |
| bauzas | sean-k-mooney: oh please don't | 17:39 |
| *** maciejjozefczyk has quit IRC | 17:39 | |
| bauzas | sean-k-mooney: it was a review problme | 17:39 |
| bauzas | not a code problem | 17:39 |
| sean-k-mooney | well the reason for doing it is i would prefer to drop the non namespced approch entirely eventually or maybe depreate the filters | 17:40 |
| bauzas | NO again in capitals :) | 17:40 |
| sean-k-mooney | given custom traits could used for this. you also suggested this last week by the way | 17:40 |
| sean-k-mooney | this is why im tinking about it | 17:40 |
| sean-k-mooney | anyway lets jus tdo the minima dirver fix for now | 17:41 |
| bauzas | sean-k-mooney: the only difference is that we don't have traits on placement aggregates, right? | 17:41 |
| sean-k-mooney | bauzas: correct they live on RPs | 17:42 |
| sean-k-mooney | so i thikn the compute capablity filter can defiently go. | 17:42 |
| bauzas | from what I understood from the very-long-standing battle of allocation ratios is that some operators do care about having a grouping system for managing their fleet of computes | 17:42 |
| sean-k-mooney | this one woudl requirte us to creat a sharing resouce provider per host aggreate | 17:42 |
| bauzas | (even if that can be done programmatically by something else) | 17:42 |
| bauzas | my old grandma' was sayin' : "if that works, don't touch it" | 17:43 |
| bauzas | and loooots of ops do manage aggregates thru this filter | 17:43 |
| sean-k-mooney | we have had custoemr bitten by this in the past as an fyi. specifcly the conflict betwwen the capablity filter and aggreate one | 17:43 |
| bauzas | so unless we come up with a solid upgrade plan for replacing it with very simple abstractions, don't touch it | 17:44 |
| sean-k-mooney | bauzas: yep agree | 17:44 |
| bauzas | sean-k-mooney: we resolved it with namespaces, right? | 17:44 |
| sean-k-mooney | yes | 17:44 |
| sean-k-mooney | basicaly they were adding pinned=true | 17:44 |
| *** links has quit IRC | 17:44 | |
| bauzas | problem solved. | 17:44 |
| sean-k-mooney | they just namespaced it | 17:44 |
| * bauzas is thirsty | 17:45 | |
| bauzas | ++ | 17:45 |
| dansmith | man, so busy this morning I missed out on 50% of my usual coffee consumption.. it must be TEOTWAWKI | 17:51 |
| sean-k-mooney | i try to some degree contol my caffein intake including normaly not drinking coffee at the weekends but i can totally feel teh difference when i dont have any | 17:53 |
| sean-k-mooney | given i only drink 1-2 cups a day i dont know if it woudl be more noticable if i drank more or less | 17:53 |
| sean-k-mooney | if i drank more i think it would have less of an effect when i drank it but likely more of an effect when i didnt which is why i reduced my cafee intake in the first place | 17:55 |
| *** ociuhandu has joined #openstack-nova | 18:09 | |
| *** ralonsoh has quit IRC | 18:16 | |
| *** nightmare_unreal has quit IRC | 18:23 | |
| *** icarusfactor has joined #openstack-nova | 18:24 | |
| *** factor has quit IRC | 18:27 | |
| *** ociuhandu has quit IRC | 18:31 | |
| *** ociuhandu has joined #openstack-nova | 19:16 | |
| *** ociuhandu has quit IRC | 19:22 | |
| *** nweinber has quit IRC | 19:37 | |
| *** mgariepy has quit IRC | 19:44 | |
| openstackgerrit | melanie witt proposed openstack/nova stable/stein: Reset the cell cache for database access in Service https://review.opendev.org/720587 | 19:55 |
| *** derekh has quit IRC | 20:26 | |
| *** derekh has joined #openstack-nova | 20:35 | |
| *** ccamacho has quit IRC | 20:38 | |
| openstackgerrit | melanie witt proposed openstack/nova stable/rocky: Reset the cell cache for database access in Service https://review.opendev.org/720592 | 20:39 |
| *** eharney has quit IRC | 20:40 | |
| *** ociuhandu has joined #openstack-nova | 20:45 | |
| *** xek has quit IRC | 20:47 | |
| *** ociuhandu has quit IRC | 20:49 | |
| openstackgerrit | melanie witt proposed openstack/nova stable/queens: Reset the cell cache for database access in Service https://review.opendev.org/720596 | 20:56 |
| openstackgerrit | melanie witt proposed openstack/nova stable/rocky: Reset the cell cache for database access in Service https://review.opendev.org/720592 | 20:56 |
| *** slaweq has quit IRC | 21:14 | |
| *** ttsiouts has quit IRC | 21:24 | |
| *** ttsiouts has joined #openstack-nova | 21:27 | |
| *** ttsiouts has quit IRC | 21:28 | |
| *** ttsiouts has joined #openstack-nova | 21:28 | |
| *** ociuhandu has joined #openstack-nova | 21:38 | |
| openstackgerrit | Merged openstack/nova master: libvirt: Remove VIR_DOMAIN_BLOCK_REBASE_RELATIVE flag check https://review.opendev.org/702021 | 21:40 |
| *** ociuhandu has quit IRC | 21:44 | |
| *** ttsiouts has quit IRC | 21:44 | |
| *** ttsiouts has joined #openstack-nova | 21:45 | |
| *** mriedem has left #openstack-nova | 22:26 | |
| openstackgerrit | Merged openstack/nova master: images: Make JSON the default output format of calls to qemu-img info https://review.opendev.org/711679 | 22:37 |
| *** rcernin has joined #openstack-nova | 22:39 | |
| *** tkajinam has joined #openstack-nova | 22:39 | |
| *** tosky has quit IRC | 23:09 | |
| *** threestrands has joined #openstack-nova | 23:14 | |
| *** threestrands has quit IRC | 23:15 | |
| *** threestrands has joined #openstack-nova | 23:15 | |
| *** sapd1 has quit IRC | 23:39 | |
| *** mlavalle has quit IRC | 23:50 | |
| *** sapd1 has joined #openstack-nova | 23:51 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!