Thursday, 2022-02-24

clarkbsean-k-mooney[m]: it actually can be made to work with tox00:22
clarkbbut ya a lot of projects don't bother with that00:22
clarkb(we have what we call tox siblings support in the tox jobs which enables this)00:22
*** efried1 is now known as efried00:55
opendevreviewmelanie witt proposed openstack/nova master: libvirt: Register defaults for undefined hw image properties  https://review.opendev.org/c/openstack/nova/+/80070801:02
opendevreviewmelanie witt proposed openstack/nova master: manage: Add image_property commands  https://review.opendev.org/c/openstack/nova/+/82439201:02
opendevreviewGhanshyam proposed openstack/nova master: Add DB and scheduler filter method for getting nodes,RP  https://review.opendev.org/c/openstack/nova/+/83070401:07
*** lajoskatona_ is now known as lajoskatona02:25
melwittgibi: I respun https://review.opendev.org/c/openstack/nova/+/824392 and https://review.opendev.org/c/openstack/nova/+/800708 to address stephenfin's comments04:32
melwittgibi: also the top two unified limits patches have +2 from bauzas if you might be able to take a look04:32
opendevreviewGhanshyam proposed openstack/nova master: API change to allow project admin to boot server on specific host  https://review.opendev.org/c/openstack/nova/+/83054304:46
*** amoralej|off is now known as amoralej07:09
gibimelwitt: o/ sure I will take  look at both07:55
yuvalGuys good morning09:04
gibiyuval: o/ good morning09:04
yuvalthis have +1 from zuul passed lightbits ci and +2 from gibi, can another core check it out: https://review.opendev.org/c/openstack/nova/+/82160609:04
yuvaltoday is the deadline right?09:05
gibiyuval: yes, today is the deadline09:05
yuvalok09:05
gibibut I think you are in a good position 09:05
yuvalthank you09:06
bauzasyuval: gibi: looking at it10:20
bauzassean-k-mooney: ha11:00
bauzasyou already +Wd the lightbits driver11:01
bauzasI had a comment 11:01
bauzaswe need a reno file11:01
sean-k-mooney[m]ah you are right11:02
sean-k-mooney[m]i can pull +w or we can add a followup patch11:02
bauzasyuval: https://review.opendev.org/c/openstack/nova/+/821606/16/nova/tests/unit/virt/libvirt/volume/test_lightos.py#2811:02
bauzassean-k-mooney: well, given it's already going, and given the deadline is today, that's OK11:02
bauzassean-k-mooney: but I want yuval to create it 11:03
bauzasand also I had a comment https://review.opendev.org/c/openstack/nova/+/821606/16/nova/tests/unit/virt/libvirt/volume/test_lightos.py#2811:03
bauzasyuval: around ?11:03
sean-k-mooney[m]ok yuval if you add a follow up patch with a release note we can review that quickly11:03
sean-k-mooney[m]ya just saw the question re flags11:04
sean-k-mooney[m]i thihk that was a leftover form a previous patch11:04
sean-k-mooney[m]oh its for the retires11:05
sean-k-mooney[m]that is what they ment to set in the test11:05
sean-k-mooney[m]although they asserted its default value in the test so not entirly needed11:07
yuvalhey11:17
yuvalI am here11:17
yuvalrelease note ok11:17
bauzasthanks 11:18
bauzasyuval: do you know how to do it ?11:18
yuvallast time I add it in cinder I copied an existing release note and edited it.11:18
yuvalneed to keep the correct format etc11:19
sean-k-mooney[m]we have a tool11:19
sean-k-mooney[m]you can do tox -e venv reno new lightos11:19
sean-k-mooney[m]that will create a release note you can edit11:20
sean-k-mooney[m]then you can test that its valied with tox -e releasenotes11:20
bauzasyuval: no, let me give you the doc11:20
yuvalok11:21
sean-k-mooney[m]in this case you can delete all sections except feature where you should provide a short overview of the feature and any version requirements11:21
sean-k-mooney[m]e.g. this need lightos version X11:21
bauzasyuval: https://docs.openstack.org/nova/latest/contributor/releasenotes.html11:22
bauzasyuval: just use the 'features' section for the YAML file11:22
bauzasyuval: also, I had a nit for a test11:23
bauzasyuval: could you just reply on gerrit ?11:23
yuvalyes, I saw I didnt understand it actually11:23
bauzashttps://review.opendev.org/c/openstack/nova/+/821606/16/nova/tests/unit/virt/libvirt/volume/test_lightos.py#2811:23
sean-k-mooney[m]ok going to grab coffee. brb11:24
gibibauzas: I'm +2 on the unified limits series. Is it OK with you if I go back and plug the missing +As to the series?11:31
bauzasgibi: sure11:31
gibiack, on it11:31
bauzasdo this if you want11:31
sean-k-mooney:) i care about that landign but also know i do not have the time before FF to load the context required to review it properly so i have been staying away from the series11:53
gibimelwitt: I have a small question in https://review.opendev.org/c/openstack/nova/+/789963 11:56
*** bhagyashris_ is now known as bhagyashris11:57
*** amoralej is now known as amoralej|lunch12:19
opendevreviewJonathan Race proposed openstack/nova master: driver/secheduler/docs for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82205312:23
opendevreviewJonathan Race proposed openstack/nova master: zuul-job for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82837212:23
chateaulavgibi: fixed the pep8 error, i promise this is the last time the driver patch will be updated, focus is on the zuul patch today12:24
gibichateaulav: no worries :)12:26
sean-k-mooneygiven gibi has +2'd the first two patches i take it this series is generally in a good postion for review12:32
gibisean-k-mooney: yes12:33
sean-k-mooneycool i will try an take a look at them shortly so12:33
gibithe last patch adds the gate testing, that still has some failing tempest test but most of the tempest already green12:33
gibiwhich is a good sign that the emulation actually working12:33
sean-k-mooneyack do we want to hold +w until that last ci patch is green im not going to +w until we are happy with all 3 fo the feature patches as i think this should all merge at once anyway but is the zuul job part of the DOD12:35
sean-k-mooneyoh the feature is now only 2 patches12:35
sean-k-mooneyoh right the third is the ci patch12:35
chateaulavsean-k-mooney: correct12:35
chateaulavi figured once that one is green ill remove it from check, that way you guys can implement it the weekly and periodic\12:36
sean-k-mooneyhonestly im ok with havign one emulated env in check12:37
chateaulavok12:37
sean-k-mooneybut if we want to test others i would put them in the periodic weekly pipeline12:37
gibisean-k-mooney: I'm OK to land the feature today even without the CI path, I think the current state of the CI result on the CI patch already looks good enough 12:38
sean-k-mooneyits very unlikely that a change woudl break mips emulation but not arm12:38
sean-k-mooneygibi: ack ok ill focus on the first too so then take a look at the ci patch12:38
sean-k-mooneychateaulav: the m1.tiny flavor size might break some senario tests due to sapce if ti does i think m1.micro is big enough for arm cirros but small then m1.tiny12:39
sean-k-mooneyfor now lets leave it as you have it and see if it breaks anything12:40
chateaulavok12:40
yuval  File "/home/yuval/workspace/opendev/nova/nova/virt/driver.py", line 129, in <module>12:42
yuval    "supports_remote_managed_ports": os_traits.COMPUTE_REMOTE_MANAGED_PORTS,12:42
yuvalAttributeError: module 'os_traits' has no attribute 'COMPUTE_REMOTE_MANAGED_PORTS'12:42
yuvalI get this while running unittests12:42
yuvalwhat I am doing wrong?12:42
opendevreviewMerged openstack/nova master: Lightbits LightOS driver  https://review.opendev.org/c/openstack/nova/+/82160612:43
gibiyuval: you need to rebuild you tox env12:43
gibiyuval: it probably has an old os-traits lib 12:43
yuvaltox -e ?12:43
gibiyuval: try tox -r -e <env>12:43
yuvalor remove it12:43
gibithe -r is the rebuild12:43
yuvalok12:43
gibior you can remove .tox dir 12:43
yuvalthanks12:43
yuvalyes I will try it if it wont work12:43
sean-k-mooneyyuval: ya this is likely just an old os-traits version12:44
sean-k-mooneythe other way to fix it is to manually install it12:44
sean-k-mooney.tox/<env>/bin/python3 -m pip install -U os-traits12:44
sean-k-mooneythat is also how you can locally test with os-triats changes that are not released12:45
gibibauzas, gmann: I'v started re-reading the policy series and I have concerns about assisted volume snapshot https://review.opendev.org/c/openstack/nova/+/828994/7/nova/api/openstack/compute/assisted_volume_snapshots.py#4212:45
sean-k-mooneyjust replace os-triats with the path to the git repo on disk12:45
*** carloss is now known as carloss|afk12:56
opendevreviewMerged openstack/nova master: libvirt: Register defaults for undefined hw image properties  https://review.opendev.org/c/openstack/nova/+/80070813:00
opendevreviewMerged openstack/nova master: manage: Add image_property commands  https://review.opendev.org/c/openstack/nova/+/82439213:00
yuvalhey added release note13:02
yuvalI got a little mixed up with git issues13:05
yuvalI edited the changes without pulling latest version13:06
yuvalthen had to rebase13:06
yuvalahhh13:06
yuvalits merged before I added the release note13:06
yuvalok, I am uploading a followup13:07
gibiyuval: release not as a follow up is OK13:10
sean-k-mooneyyuval: since you have not pushed yet can you reference the gerrit change id for the lightos feature patch in the commit13:11
yuvalits a special syntax or just followup for: <changeid>13:11
sean-k-mooneyno special syntax so followup for: ... is fine13:12
sean-k-mooneyjust makes it simpler to corralate them in the future13:12
sean-k-mooneyyou will still need the normal change id13:12
sean-k-mooneyin the normal way but if you reference the other one in the commit message body we can copy paste and find the relevent review if we ever need too in the future13:13
opendevreviewyuval proposed openstack/nova master: Lightos driver release note  https://review.opendev.org/c/openstack/nova/+/83081713:14
sean-k-mooneycool commit message looks fine ill review the rest later13:15
opendevreviewyuval proposed openstack/nova master: Lightos driver release note  https://review.opendev.org/c/openstack/nova/+/83081713:16
yuvalthank you13:16
*** amoralej|lunch is now known as amoralej13:19
sean-k-mooneygibi: chateaulav im also +2 on the emulation patches. i think the testign coudl be imporved in the futrue but just wanted to point out that https://review.opendev.org/c/openstack/nova/+/822053/63 is in merge conflict13:33
sean-k-mooneyso likely we will need to rebase the series before it can proceed13:34
chateaulavok13:34
sean-k-mooneychateaulav: if you rebase it im ok with fast approving or im sure gibi woudl be happy to reapove too13:35
chateaulavok, and just to make sure i dont screw this up i should follow: https://docs.opendev.org/opendev/infra-manual/latest/developers.html#rebasing-a-commit ?13:41
sean-k-mooneychateaulav: just do the following13:42
sean-k-mooneygit fetch --all13:42
sean-k-mooneygit rebase origin/master13:42
sean-k-mooneygit review13:42
sean-k-mooneyfrom the top patch in the series13:42
chateaulavok, 13:43
sean-k-mooneyalthough you have a syntax error in the zuul patch where you defien the regex13:43
chateaulavsounds good13:43
sean-k-mooneyso you might want to fix that before you do the git review13:43
sean-k-mooneyhttps://zuul.opendev.org/t/openstack/build/fdf39c2132e844799465d04c6200429c/log/job-output.txt#3009113:43
sean-k-mooneychateaulav: actully i think that issue is coming form the parent job13:45
sean-k-mooneychateaulav: can you add tox_envlist: 'all' https://review.opendev.org/c/openstack/nova/+/828372/28/.zuul.yaml13:51
sean-k-mooneychateaulav: that should fix the job issue13:51
chateaulavok13:51
gibichateaulav, sean-k-mooney: I can reapprove sure13:52
sean-k-mooneychateaulav: i think the job is defaulting to the smoke env https://github.com/openstack/tempest/blob/master/tox.ini#L227-L23513:53
sean-k-mooneyall https://github.com/openstack/tempest/blob/master/tox.ini#L58-L68 will allow you to fully contol what tests run via the zuul job13:53
opendevreviewJonathan Race proposed openstack/nova master: object/notification for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82836914:14
opendevreviewJonathan Race proposed openstack/nova master: driver/secheduler/docs for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82205314:14
opendevreviewJonathan Race proposed openstack/nova master: zuul-job for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82837214:14
chateaulavmost stressfull rebase ever... everything is updated, waiting on checks to run14:17
sean-k-mooney:) im sure it will be fine14:18
sean-k-mooneyah it was the image property regesistration patch it conflicted with14:20
chateaulavyeah14:20
sean-k-mooneyok well im +2 on the rebase it looks correct to me14:21
chateaulavthanks sean-k-mooney and gibi14:21
*** carloss|afk is now known as carloss14:22
sean-k-mooneyill check back on the ci results later for the zuul patch 14:22
opendevreviewsean mooney proposed openstack/nova master: update default overcommit  https://review.opendev.org/c/openstack/nova/+/83082914:46
dmitriissean-k-mooney: RE https://review.opendev.org/c/openstack/nova/+/829974 I have a fix for the migration & unshelve scenarios here (binding:profile updates) along with func tests. I could work on making vf_num and pf_mac_address to be stored in extra_info but need to clarify the ordering. It's close to freeze time so I just wanted to know whether to14:53
dmitriisdo the move to extra_info first or after this change is in.14:53
gmanngibi: thanks for catching that, I missed that it is called from cinder.14:53
gmanngibi: I will update14:53
sean-k-mooneydmitriis: im not sure of the status of gibis change to store the mac in extra info14:58
gibisean-k-mooney: it is in the works14:59
sean-k-mooneyi woudl proably defer moving it to extra_info until after this change14:59
gibisean-k-mooney: I can publish the code today the funct testing will take more time14:59
dmitriissean-k-mooney, gibi: ack15:00
gibiI'm OK moving forward with dmitriis fix that is up and refactor it later15:00
sean-k-mooneyack that is what i was going to ask15:00
sean-k-mooneyhttps://review.opendev.org/c/openstack/nova/+/829974 is addin more call to the pci utils but that can be refactored later15:01
sean-k-mooneyand i think having move operation work is proably more imporant then purity in this case15:01
sean-k-mooneygiven we should be able to backport the refactor after FF i think15:01
sean-k-mooneydmitriis: ill try and review that fully later today15:02
dmitriissean-k-mooney: ack, ty15:03
gibiack I will review too15:15
sean-k-mooneydmitriis: +1 on https://review.opendev.org/c/openstack/nova/+/829974 comments inline15:55
gibidmitriis: I also left comments inline https://review.opendev.org/c/openstack/nova/+/829974 just now15:56
melwittgibi: answered. I also see the series is in merge conflict so I have to rebase it 😩15:58
sean-k-mooneychateaulav: looking at the realtime status of https://review.opendev.org/c/openstack/nova/+/822053/64 in zuul the tox jobs are going to fail15:59
chateaulavtracking, taking a look now16:00
dmitriissean-k-mooney, gibi: looking16:00
sean-k-mooneymelwitt: do we need to start stacking series and or wait for unified limists to merge before mergeing others16:00
gibimelwitt: thanks for the answer I missed the fact that it ius differnt coverage. Then I think it make sense to keep both coverage in place16:00
bauzasmelwitt: don't be afraid, we can merge your series tomorrow if needed16:01
bauzasmelwitt: I mean, if gibi says +2 of course16:01
gibimelwitt: ping me after the rebase and I can put back the +A on the series16:02
bauzasme too16:02
gibibauzas: I'm +2 on the series16:02
melwittsean-k-mooney: dunno actually, I hadn't checked conflict with other things that need to go now16:02
bauzasgibi: cool then16:02
* bauzas tries to look at https://review.opendev.org/c/openstack/nova/+/76429216:02
bauzasgmann: wdyt if we merge a few changes from this tenant-id series ?16:03
melwittbauzas, gibi: that sounds great, thank you!16:03
gibibauzas: is the tenant-id series an all-or-nothing situation as the microversion is in the first patch?16:04
gmannbauzas: humm that will be like need two microversion for single type of change.16:04
gmanngibi: indeed16:04
gmannwe cannot split the single microversion changes in two series16:05
gmanntwo cycle16:05
gmannand it change the many APIs16:05
chateaulavany insight for resolution on a couple of the issues16:15
chateaulavhttps://www.irccloud.com/pastebin/sdXpnsVs/16:15
gibichateaulav: where do you see that error? on the gate?16:15
chateaulavi know yuval was doing the lightos stuff, just not sure how that gets triggered on my rebase16:16
chateaulavthe driver patch check after rebase16:16
yuvalwhats the issue?16:16
gibibut locally16:16
gibiI assume16:16
gibiso I think you have to rebuild your tox env16:16
gibitox -r -e <env>16:16
gibiyou have an old os-brick in your env16:16
chateaulavyeah, running that now to make sure, takes a min16:16
chateaulavok yeah, i see. ill focus on the 3 i see in zuul and then we should be good.16:17
sean-k-mooneyim just on a dowstream call ill try and take a look after16:17
chateaulavfor a sec I was like WTF im supposed to do with that16:18
gibiyeah, you tox is not always smart enought to detect that deps are changed 16:18
gibis/you//16:18
chateaulavfact16:18
gibiand now we are in a busy period integrating stuff on master16:18
bauzasgmann: gibi: sorry was in meeting (and still are :) )16:18
gibibauzas: ack16:19
gibino worries16:19
bauzasgmann: gibi: I was asking for it as I knew it would mean two microversions for the same project_id change16:19
yuvalgibi thanks for the fast response, chateaulav sorry about that16:19
chateaulavyour good. just a sec of panic on my side16:19
bauzasgmann: gibi: but I was wondering about merging a few things for Yoga as we don't have any new microversions and just telling within the microversion what APIs were touched16:19
bauzasgmann: gibi: but if you guys prefer to merge it at once, fine by me16:20
gibibauzas: we can technically split to two microversions it is a bit of a meh UX but yeah16:20
bauzasthis is just unrealistic to merge the whole branch before today EOB 16:20
gibiack16:20
bauzasmy only concern is that brinzhang isn't around as we speak16:21
gibiit is pretty late for him16:21
bauzasso we can't really ask him to change the microversion change for saying which APIs are modified16:21
bauzasthis would mean some kind of exception16:21
bauzasso maybe not a priority16:21
bauzasas you can see, I'm not opinionated16:21
gmannyeah, I also prefer to merge it together 16:21
bauzasOK, if there is no huge interesting in merging by pieces, let's then punt 16:22
gmannas it is cleanup, I think we can move it to next cycle and make it priority in Zed since starting.?16:22
gibimy only fear that we never merge it ever and brinzhang gets mad and drops it16:22
gmanngibi: yeah, it has been going in many cycle. 16:23
gibiyeah we somehow need tof force ourselve to look at that series early in the cycle16:23
gibibut we failed on it this time16:23
gibior at least I failed16:23
gmannI too failed, but I will commit on this for Zed for sure.16:24
gmannthere are few last patches not ready at least server tenant usage url one16:24
gmannwhich brin asked me to look and I did not get chance 16:24
melwittapologies for the upcoming spam16:26
opendevreviewmelanie witt proposed openstack/nova master: Move keypair quota error message into exception  https://review.opendev.org/c/openstack/nova/+/82818616:26
opendevreviewmelanie witt proposed openstack/nova master: Add stub unified limits driver  https://review.opendev.org/c/openstack/nova/+/71213716:26
opendevreviewmelanie witt proposed openstack/nova master: Assert quota related API behavior when noop  https://review.opendev.org/c/openstack/nova/+/71214016:26
opendevreviewmelanie witt proposed openstack/nova master: Make unified limits APIs return reserved of 0  https://review.opendev.org/c/openstack/nova/+/71214116:26
opendevreviewmelanie witt proposed openstack/nova master: Add logic to enforce local api and db limits  https://review.opendev.org/c/openstack/nova/+/71213916:26
opendevreviewmelanie witt proposed openstack/nova master: Enforce api and db limits  https://review.opendev.org/c/openstack/nova/+/71214216:26
opendevreviewmelanie witt proposed openstack/nova master: Update quota_class APIs for db and api limits  https://review.opendev.org/c/openstack/nova/+/71214316:26
opendevreviewmelanie witt proposed openstack/nova master: Update limit APIs  https://review.opendev.org/c/openstack/nova/+/71270716:26
opendevreviewmelanie witt proposed openstack/nova master: Update quota sets APIs  https://review.opendev.org/c/openstack/nova/+/71274916:26
opendevreviewmelanie witt proposed openstack/nova master: Tell oslo.limit how to count nova resources  https://review.opendev.org/c/openstack/nova/+/71330116:26
opendevreviewmelanie witt proposed openstack/nova master: Enforce resource limits using oslo.limit  https://review.opendev.org/c/openstack/nova/+/61518016:26
opendevreviewmelanie witt proposed openstack/nova master: Add legacy limits and usage to placement unified limits  https://review.opendev.org/c/openstack/nova/+/71349816:26
opendevreviewmelanie witt proposed openstack/nova master: Update quota apis with keystone limits and usage  https://review.opendev.org/c/openstack/nova/+/71349916:26
opendevreviewmelanie witt proposed openstack/nova master: Add reno for unified limits  https://review.opendev.org/c/openstack/nova/+/71527116:26
opendevreviewmelanie witt proposed openstack/nova master: Enable unified limits in the nova-next job  https://review.opendev.org/c/openstack/nova/+/78996316:26
melwittgibi, bauzas: merge conflict was super easy, was due to requirements bump so only requirements.txt and test-requirements.txt16:28
bauzas++16:28
gibiOK, On it16:29
opendevreviewGhanshyam proposed openstack/nova master: Modify remaining APIs as per RBAC new guidelines  https://review.opendev.org/c/openstack/nova/+/82899416:32
dansmithbauzas: did you see my reply comment here about splitting? https://review.opendev.org/c/openstack/nova/+/82036816:32
gmanngibi: bauzas  fixed the assisted volume snapshot one  https://review.opendev.org/c/openstack/nova/+/82899416:33
dansmithbauzas: I think that's what you intended and I was picking the wrong seam, but just want clarification before I go trying to help split that as described in my most recent one16:33
opendevreviewGhanshyam proposed openstack/nova master: Separate flavor extra specs policy for server APIs  https://review.opendev.org/c/openstack/nova/+/82962616:33
opendevreviewGhanshyam proposed openstack/nova master: Complete phase-1 of RBAC community-wide goal  https://review.opendev.org/c/openstack/nova/+/82986616:33
gibimelwitt: rebase looks good, I did the paperwork to send the series to the gate16:35
melwittthank you gibi!16:36
gibigmann: you are next :)16:36
gmanngibi: thanks :)16:36
gibiI have to stop in about an hour16:36
gibigmann: thanks for the fix16:54
gibibauzas: this morning I re-read the policy series and I find no problems (except the assisted snapshot that is fixed now) so I upgraded my vote now to +2 on the policy series16:55
gibibauzas: we still need you to check back to https://review.opendev.org/c/openstack/nova/+/828994/8 as that was updated recently 16:55
bauzasgibi: ok16:55
gibibut otherwise the policy series is also going to the gate16:55
bauzasdone16:56
gibinow zuul has things to do :D16:56
bauzasdansmith: sorry in an internal meeting but looking16:58
bauzasdansmith: so,17:01
bauzaswe have two different things 17:01
bauzasone is the REST API modification adding a new microversion17:01
bauzasthe other is the new instance external event type we add in the related ovo17:02
dansmithwell, the microversion for the event is specifically for the event api, not the ovo directly, but yeah17:02
bauzasagreed, it's in the event itself17:03
bauzasI'm speaking of https://review.opendev.org/c/openstack/nova/+/820368/9/nova/objects/external_event.py17:03
bauzasvs. https://review.opendev.org/c/openstack/nova/+/820368/9/nova/api/openstack/compute/rest_api_version_history.rst17:04
dansmithyah17:04
bauzaswe were about discussing 5 mins ago about merging one microversion for tenant_id17:05
bauzaswhich could be 2.9117:05
bauzasfortunately, this won't happen17:05
bauzasso, technically, the 2.91 slot is free to anyone17:05
gmannbauzas: gibi thanks for review. 17:05
bauzasdansmith: just take note of https://review.opendev.org/c/openstack/nova/+/764292/34/nova/api/openstack/compute/rest_api_version_history.rst17:06
bauzasbut we agreed 5 mins ago with gmann to not merge it17:06
bauzasthat said, I'll have to leave in 10 mins17:07
dansmithbauzas: yeah, I really wish we shouldn't do those sorts of things17:07
dansmithbut regardles17:07
dansmithall I'm trying to figure out is whether you actually wanted *two* microversions, one for the event and one for the rebuild change17:07
bauzastwo REST microversions ?17:08
dansmithor if you meant to split the plumbing underneath (i.e. the actual event, the rebuild code) from the api changes17:08
bauzasno, I didn't ask for it17:08
bauzasI asked for splitting the plumbing as you speak17:08
dansmithI forgot that we started microversions for the events (which I'm not sure why we started doing that), but..17:08
dansmithright okay17:08
bauzaslet's call it event versioning, if you prefer :)17:08
dansmithcool cool, my mistake for forgetting that initially so I was on a slightly wrong track17:09
bauzasbut gibi knows more than me on it17:09
bauzasdansmith: again, that said, I won't be able to review it honestly today17:09
dansmithbauzas: that's fine17:09
bauzasas I have an hard stop in 5 mins17:09
dansmithwhoami-rajat: are you cool with me trying to split your patch accordingly or do you want to do it?17:09
gibibauzas: I think you are conflating versioning notifications emitted by nova and microversion bump for the external_events REST API. 17:11
whoami-rajatdansmith, either way is fine with me, maybe better if you do it because I still don't have full idea of the ask here or if guided i can17:11
dansmithwhoami-rajat: ack, partially my fault anyway :D17:12
whoami-rajatdansmith, i think it's because the codebase is fairly new to me :)17:13
dansmithwhoami-rajat: you should seize the opportunity to blame me ;P17:13
bauzasgibi: I'm not conflating, but I wrote a comment which was unclear :)17:13
gibiOK. I don't have the full context sorry :)17:13
bauzashttps://review.opendev.org/c/openstack/nova/+/820368/9#message-f4b92b3310dee2af9598f108dc761f7fabafd34a17:14
whoami-rajat:D17:14
bauzas"event one" was unclear17:14
bauzasI was referring to the patch17:14
gibiOK. 17:14
gibiI see now17:14
bauzasnot to any other microversion17:14
* bauzas has to disappear for the night17:17
bauzasother cores, I trust you17:17
bauzasif you wanna continue to review, fine by me17:17
* bauzas will look at all the blueprints by tomorrow and decide to use the -2 hammer only if none of the cores said +217:18
ralonsohgibi, https://review.opendev.org/c/openstack/neutron/+/829247 just a heads-up, if you didn't see my reply17:19
gibiralonsoh: thank, I haven't seen it yet. I will check tomorrow or next week, as we are in feature freeze crazy :)17:22
ralonsohsure, no rush17:22
ralonsohping me whenever you can17:22
opendevreviewJonathan Race proposed openstack/nova master: object/notification for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82836917:28
opendevreviewJonathan Race proposed openstack/nova master: driver/secheduler/docs for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82205317:28
opendevreviewJonathan Race proposed openstack/nova master: zuul-job for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82837217:28
gibiralonsoh: thanks17:29
chateaulavsean-k-mooney: that should do, sorry it took a litle bit the rebase got wonked so started fresh.17:29
chateaulavgibi: thanks as well17:29
gibichateaulav: rebaswe looks clear17:30
gibiso I'm +217:30
gibibut now I have to stop for today17:30
gibisee you all tomorrow o/17:30
chateaulavthanks17:31
opendevreviewMerged openstack/nova master: Allow per-context rule in error messages  https://review.opendev.org/c/openstack/nova/+/81686517:41
*** amoralej is now known as amoralej|off17:42
dansmithwhoami-rajat: so, functional tests fail without the api change in place, which I think means that we're not properly honoring the old microversion behavior17:45
dansmithwhich is a good reason for this split17:45
whoami-rajatdansmith, ack17:53
opendevreviewDan Smith proposed openstack/nova master: Add support for volume backed server rebuild  https://review.opendev.org/c/openstack/nova/+/82036817:59
opendevreviewDan Smith proposed openstack/nova master: Add volume-rebuild support to cinder module  https://review.opendev.org/c/openstack/nova/+/83088217:59
opendevreviewDan Smith proposed openstack/nova master: Add API support for rebuilding BFV instances  https://review.opendev.org/c/openstack/nova/+/83088317:59
dansmithwhoami-rajat: ^17:59
dansmithwhoami-rajat: assuming that first mini patch is good, we should be able to get that merged ASAP and cut down what remains18:00
dansmithhmm, did I lose the reno?18:00
opendevreviewDan Smith proposed openstack/nova master: Add API support for rebuilding BFV instances  https://review.opendev.org/c/openstack/nova/+/83088318:01
dansmiththere we go18:01
whoami-rajatack, thanks for splitting it up18:01
dansmithwhoami-rajat: so I think we need to get the rebuild flag down to the compute manager so it knows whether or not to trigger the new behavior18:02
whoami-rajatdansmith, i agree with the usage of an old version of API but if they don't pass the flag, it will never reach the compute manager and should fail at API level18:05
whoami-rajatbut we need to still consider the old API new compute case18:05
dansmithwhoami-rajat: see the test failure in that middle patch :)18:05
whoami-rajatwill look into that18:05
dansmithwhoami-rajat: if the image ref does not change, we should rebuild the instance but not the volume,18:05
dansmithbut your code makes it still call rebuild, which will erase the root volume18:05
whoami-rajatAh ok, we support rebuilding volume backed instances having same image as the provided one18:06
whoami-rajatso reimage_boot_volume should be checked in the manager, got it18:07
dansmithright, the image can't change, but we do that because we don't destroy the root disk, so we require the image to be the same18:07
dansmithyep18:07
whoami-rajatack, will work on that18:07
dansmithalso,18:07
dansmithyour api will only check that field if the image-ref changes, so even with the new version of the API, you can trigger the accidental root volume erase18:08
dansmithso something needs to change there to make sure we never ever erase the root disk unless they specifically ask for it18:08
dansmithlemme comment on that to record that18:08
whoami-rajatif we don't trigger the rebuild_volume_backed_instance code in compute, something like, if reimage_boot_volume and is_volume_backed: , then we're safe from the same image case right?18:10
whoami-rajatbut if you want the operation to get blocked at the API layer only then we can change that18:11
dansmithwe have to support the old and new behavior in manager18:11
dansmithwe need to make sure the api properly communicates what should happen to the manager18:11
dansmiththe api may enforce certain behaviors of the api client, but we have to make sure manager knows what the right thing to do is18:11
whoami-rajatok18:12
dansmithwhoami-rajat:  be sure to put your co-authored-by on those new patches when you revise.. I meant to do that, sorry18:14
dansmithI don't want to look like I took credit for your work18:15
whoami-rajatdansmith, np, will do it and doesn't matter much :)18:15
whoami-rajat:D18:15
whoami-rajatdoesn't look like that at all18:15
dansmithit matters A LOT :)18:15
whoami-rajat:)18:16
opendevreviewMerged openstack/nova master: Revert project-specific APIs for servers  https://review.opendev.org/c/openstack/nova/+/81620618:17
opendevreviewMerged openstack/nova master: Test PROJECT_ADMIN APIs with no legacy rule case  https://review.opendev.org/c/openstack/nova/+/82484518:17
opendevreviewMerged openstack/nova master: Move rule_if_system() method to base test class  https://review.opendev.org/c/openstack/nova/+/82447518:18
opendevreviewRajat Dhasmana proposed openstack/nova master: Add volume-rebuild support to cinder module  https://review.opendev.org/c/openstack/nova/+/83088218:19
whoami-rajatdansmith, so does the first patch looks good or anything to change there? (just added the co-authored-by to commit msg ^)18:21
dansmithwhoami-rajat: I didn't actually look, but it's probably straightforward18:22
whoami-rajatyeah just sending the request to cinderclient18:22
dansmithyep, +2d18:23
whoami-rajatgreat thanks18:23
opendevreviewMerged openstack/nova master: Convert SYSTEM_ADMIN|READER to Admin and system scope  https://review.opendev.org/c/openstack/nova/+/81939018:26
opendevreviewRajat Dhasmana proposed openstack/nova master: Add support for volume backed server rebuild  https://review.opendev.org/c/openstack/nova/+/82036818:51
whoami-rajatdansmith, i think test_api makes more sense in the 3rd patch where we introduce the API changes (since those are required to pass the reimage_boot_volume flag down to manager)?19:13
opendevreviewJonathan Race proposed openstack/nova master: zuul-job for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82837219:14
chateaulavsean-k-mooney: Driver patch is good (+1 verified), and then ci should be good this go around watched zuul live and added the last test considerations19:16
chateaulavill keep watching the ci patchset though19:17
opendevreviewRajat Dhasmana proposed openstack/nova master: Add support for volume backed server rebuild  https://review.opendev.org/c/openstack/nova/+/83089419:44
opendevreviewRajat Dhasmana proposed openstack/nova master: Add API support for rebuilding BFV instances  https://review.opendev.org/c/openstack/nova/+/83088319:50
opendevreviewmelanie witt proposed openstack/nova master: Follow up for nova-manage image property commands  https://review.opendev.org/c/openstack/nova/+/83089519:55
sean-k-mooneychateaulav: one test failed but thats pretty good https://zuul.opendev.org/t/openstack/build/3cba136d39924d73b775a2b3a8d626ba20:03
chateaulavok, ill get that20:07
sean-k-mooney nova.exception.InterfaceAttachFailed: Failed to attach network adapter device to 61a0c10a-930f-4b46-8651-c8583d1d9ff720:08
sean-k-mooneycaused by r libvirt.libvirtError: internal error: No more available PCI slots20:09
sean-k-mooneychateaulav: so that is a simple fix20:09
sean-k-mooneyi suspec that just as we have to preacllocatre pcie root ports for x86 with q35 machine type20:10
sean-k-mooneywe likely need to do that too for aarch64 and virt machien type20:10
sean-k-mooneychateaulav: for q35 we use https://docs.openstack.org/nova/latest/configuration/config.html#libvirt.num_pcie_ports to contol this20:11
sean-k-mooneyso you proably need to set that to 2820:11
sean-k-mooneyor similar20:11
sean-k-mooneyin the ci job and that should be all that is requried20:11
chateaulavalright20:11
sean-k-mooneyits possibel we have to alter the driver slightly but it shoudl be trivial that is what i woudl try first to adress it20:12
sean-k-mooneychateaulav: i approved the previous patch and rechecke the first patch20:12
sean-k-mooneyso they shoudl proceed 20:12
chateaulavgreat!20:12
chateaulavsean-k-mooney: appreciate the help20:12
sean-k-mooneyno worries20:13
sean-k-mooneyim proably going to call it a day now20:13
sean-k-mooneythe failure is here by the way if you were wonderign https://zuul.opendev.org/t/openstack/build/3cba136d39924d73b775a2b3a8d626ba/log/controller/logs/screen-n-cpu.txt#921920:13
sean-k-mooneylooking at the xml https://zuul.opendev.org/t/openstack/build/3cba136d39924d73b775a2b3a8d626ba/log/controller/logs/screen-n-cpu.txt#903520:14
sean-k-mooneyi dont see any addtional pcie root ports in the xml we sent to libvirt20:14
sean-k-mooneynormally libvirt will allcoate 1 addtional pcie endpoint but that might only be for q3520:15
sean-k-mooneyanyway o/20:15
chateaulav\o20:16
opendevreviewJonathan Race proposed openstack/nova master: zuul-job for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82837220:18
opendevreviewRajat Dhasmana proposed openstack/nova master: Add volume-rebuild support to cinder module  https://review.opendev.org/c/openstack/nova/+/83088220:46
opendevreviewRajat Dhasmana proposed openstack/nova master: Add support for volume backed server rebuild  https://review.opendev.org/c/openstack/nova/+/83089420:46
opendevreviewRajat Dhasmana proposed openstack/nova master: Add API support for rebuilding BFV instances  https://review.opendev.org/c/openstack/nova/+/83088320:46
melwittgmann: do you know if there is any known issue with gate failure "[ERROR] /opt/stack/devstack/lib/neutron_plugins/ovn_agent:178 Socket /var/run/openvswitch/ovnnb_db.sock not found"? I've hit it twice today21:09
gmannmelwitt: not i am aware of. seems new to me21:09
melwittack21:10
gmannlatest i saw my rbac patches passed gate ~2hr ago21:11
melwittthanks21:14
opendevreviewMerged openstack/nova master: object/notification for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82836923:35
opendevreviewMerged openstack/nova master: driver/secheduler/docs for Adds Pick guest CPU architecture based on host arch in libvirt driver support  https://review.opendev.org/c/openstack/nova/+/82205323:35
sean-k-mooneymelwitt: locally or in the gate23:50
melwittsean-k-mooney: gate. but looks like I was just unlucky23:50
sean-k-mooneylocally devstack is not always proeperly cleaning up when you unstack23:51
sean-k-mooneyack23:51
sean-k-mooneyim not sure what would cause that on a clean install23:51

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!