Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2022-04-06
08:55:30 kashyap gibi[m]: Hey, when you get 25 minutes, I've addrressed various things from your past feedback here: https://review.opendev.org/c/openstack/nova/+/762330/
08:56:39 kashyap gibi[m]: It requires to reload some context, so do let me know if you see any further glaring holes there. Also see my last comment on PS24 for a short changelog
08:57:13 kashyap I'm badgering you because you actually gave it a close look in the past. :)
09:03:57 gibi kashyap: ack, I will look at it
09:09:11 opendevreview kiran pawar proposed openstack/nova master: VMware: Early fail spawn if memory is not multiple of 4. https://review.opendev.org/c/openstack/nova/+/835739
09:53:55 opendevreview Balazs Gibizer proposed openstack/nova master: DNM: trigger nova-emulation on the parent patch https://review.opendev.org/c/openstack/nova/+/836764
10:04:45 opendevreview Stephen Finucane proposed openstack/python-novaclient master: Deprecate nova CLI https://review.opendev.org/c/openstack/python-novaclient/+/836682
10:04:46 opendevreview Stephen Finucane proposed openstack/python-novaclient master: docs: Update docs to reflect deprecation status https://review.opendev.org/c/openstack/python-novaclient/+/836683
10:04:46 opendevreview Stephen Finucane proposed openstack/python-novaclient master: Update pre-commit hook, hacking versions https://review.opendev.org/c/openstack/python-novaclient/+/836768
10:06:58 gibi kashyap: replied
10:07:50 kashyap gibi: Thank you! Will look shortly
10:07:53 gibi kashyap: and I have to update your estimator, it took ~1 hour to load the context and review the patch again :) Maybe I need more coffee
10:08:32 kashyap gibi: Haha, really sorry; I was being selfish there. I almost wrote 40 minutes, then I thought: "No, that might dissuade Gibi" :D
10:08:45 gibi no worries :)
10:11:41 kashyap gibi: Just read; good questions. I need a couple more hours to study it carefully to answer your questions well
10:11:54 kashyap This whole CPU area is a minefield... :D
10:12:49 gibi indeed
10:13:15 gibi honestly I'm not even feeling confident to review it alone so I'm happy stephenfin also looked at it
10:14:52 sean-k-mooney i will also try and review it again
10:16:56 gibi what is the magic word for gerrit to trigger an experimental job on a review?
10:19:53 gibi "check experimental" is the answer :)
10:20:14 sean-k-mooney yes
10:20:19 opendevreview Balazs Gibizer proposed openstack/nova master: Add nova-emulation to the experimental queue too https://review.opendev.org/c/openstack/nova/+/836792
10:20:34 sean-k-mooney defined in the pipline defintionin project-config
10:42:42 kashyap gibi: Also thanks for running the nova-emulation job on it. On confidence - I definitely wanna run a couple of real integration/migration tests before merging it.
10:47:40 gibi yeah, I saw the comment thread about emulated archs in the review and wondered if we have a result from the recently added emulation job
10:48:05 kashyap gibi: Right, I saw your inline comment there.
10:48:30 kashyap For now, the main goal here is to get this working for x86 on x86, and AArch64 on AArch64
10:49:11 sean-k-mooney right but we cant break emulation either
10:49:57 sean-k-mooney i guess it would mainly be an issue if we were using traits to schdule based on emulated cpu capablities
11:03:20 sean-k-mooney gibi: bauzas so openstack-tox-lower-constraints on stable/xena seams to be broken now
11:03:41 sean-k-mooney its depending on python3.6 which appear to nolonger exist on ubuntu 20.04
11:03:47 sean-k-mooney it alway default to 3.8
11:04:17 sean-k-mooney so either the job disto has recently updated or the package wa aviable and has now been remvoed since 3.6 is eol
11:04:40 sean-k-mooney should i add this to the ptg adgenda
11:05:08 sean-k-mooney we basicly have 3 options. pin to bionic, use 3.8 or delete the job
11:05:41 gibi hm
11:05:55 gibi based on https://governance.openstack.org/tc/reference/runtimes/xena.html 3.6 could work on centos8 if we want
11:06:13 gibi but I'm OK to simply drop the job
11:06:14 sean-k-mooney ya it could i would just have to change the nodeset
11:06:38 sean-k-mooney elodilles:^ this is sable so you might have an opion too
11:06:49 sean-k-mooney https://review.opendev.org/c/openstack/nova/+/828407/
11:06:55 sean-k-mooney is what failed on it
11:07:53 sean-k-mooney gibi: funally enough the centos 8 py 36 non voting job also failed
11:08:16 sean-k-mooney oh...
11:08:18 sean-k-mooney The conflict is caused by:
11:08:20 sean-k-mooney oslo-vmware 3.9.2 depends on suds-jurko>=0.6
11:08:22 sean-k-mooney The user requested (constraint) suds-jurko===0.6
11:08:43 sean-k-mooney i tought we had fixed that already
11:09:23 sean-k-mooney that is clearly unrelated and the job is non-voting anyway so we can ignore that for this speicic patch
11:31:19 opendevreview kiran pawar proposed openstack/nova master: VMware: Split out VMwareAPISession https://review.opendev.org/c/openstack/nova/+/832156
11:31:20 opendevreview kiran pawar proposed openstack/nova master: VMware: StableMoRefProxy for moref recovery https://review.opendev.org/c/openstack/nova/+/832164
11:33:21 lajoskatona bauzas: Hi, do you see any Nova-Neutron cross-project topics?
11:48:53 sean-k-mooney lajoskatona: in general no
11:49:15 sean-k-mooney but ill be attening the neutron room until the nova stuff start an hour later
11:49:47 lajoskatona sean-k-mooney: ok, thanks, do you think wee need to sit around a table to discuss anything? Earlier with bauzas we agreed to keep a slot for Thursday
11:49:52 sean-k-mooney so i suggested if we have nova/neutron topics we could jsut do them in the first hour of the neutron session any day that works for ye
11:50:41 sean-k-mooney am the only topic i have are ones we have discussed many time before
11:50:54 sean-k-mooney e.g. how can we star requiring extnetion in neutron
11:51:18 sean-k-mooney spcificaly how can we require ml2 plugins and core plugings to implemet multiple port bidnings
11:51:27 lajoskatona sean-k-mooney: as it is today Friday is free for us, we have no topics for that day
11:52:17 elodilles sean-k-mooney gibi : yepp, py36 was dropped from ubuntu focal image. we only see that at stable/xena's l-c job, as we explicitly pin there py36 in tox.ini
11:52:20 sean-k-mooney are there any pain points that neutron want to raise with nova? i joined for the os-vif topic yesterday
11:52:25 sean-k-mooney but i dont know if there are others
11:52:54 sean-k-mooney elodilles: ack do you have a preference on how to move forward
11:53:14 sean-k-mooney elodilles: we can unpin, drop the job or swap to centos8 stream nodeset
11:54:10 lajoskatona sean-k-mooney: thanks for it, I don't know any
11:54:20 elodilles sean-k-mooney gibi : melwitt discussed this with infra during weekend and the suggestion was to drop the l-c from xena :) with which I'm OK, but wanted to check whether i can add somehow py36 to the l-c job somehow, but i haven't got there yet o:)
11:55:46 elodilles sean-k-mooney: i'm OK with the other options, too. either we could use py38, or centos8 stream nodeset, both could work too
11:55:56 sean-k-mooney lajoskatona: well the way to do it is to make a policy desicsion and delare any driver that does not comply unsupproted
11:56:05 sean-k-mooney with a deprecation period
11:57:07 sean-k-mooney that woudl allow nova assume its always available there are other extneions that shoudl be treated as required eventually too
11:57:37 sean-k-mooney im thinking of the exteions for port resouce requests
11:58:17 sean-k-mooney lajoskatona: if we want to discuss this topic i am happy to on friday if you think we will actullly make prgress on it
11:58:36 lajoskatona sean-k-mooney: yeah that was / is tricky with resource_request and resource_request_groups
11:59:33 sean-k-mooney i think fundemtally neutron need a way to graduate extention into the core api
12:13:35 bauzas lajoskatona: sean-k-mooney: as you want, folks
12:15:53 lajoskatona sean-k-mooney, bauzas: Is Friday ok? We have sessions 14:00-1600 UTC
12:16:08 bauzas lajoskatona: OK, we can reserve one timeslot
12:17:40 lajoskatona bauzas: ok, cool, Friday 1400 UTC than
12:17:48 lajoskatona bauzas: ? --^
12:17:54 bauzas wfm
12:25:19 opendevreview Alexey Stupnikov proposed openstack/nova stable/xena: Fix openstack-tox-lower-constraints in stable/xena https://review.opendev.org/c/openstack/nova/+/836650
12:57:26 bauzas lajoskatona: do you have any agenda for our x-p session ?
12:57:42 bauzas lajoskatona: just updated our etherpad : https://etherpad.opendev.org/p/nova-zed-ptg
12:58:20 lajoskatona bauzas: only from sean-k-mooney: https://etherpad.opendev.org/p/neutron-zed-ptg#L259
12:58:45 bauzas lajoskatona: ok, would we meet in the Nova room or your own ?
12:59:12 lajoskatona bauzas: we can go to Nova room, we have no other topic for Friday
12:59:59 bauzas lajoskatona: OK
13:10:04 pdeore #openstack-nova, may I know where to join today's PTG session?
13:13:00 stephenfin I will miss the first ~30 mins of the PTG session today. If there's anything for me please hold it until I can join :)
13:13:16 gmann pdeore: this link https://zoom.us/j/96855299345?pwd=bG5KZHloRzA3bkhCQVJBNmZCQUltZz09#success
13:13:51 gmann pdeore: you can click on the projects/track and it will take you to the right video link of that track https://ptg.opendev.org/ptg.html
13:13:59 stephenfin gibi: I suspect that patch will need to be broken down into a few smaller patches. It's rather bloated at this point (not surprising, given the number of iterations it's been through). I will try working with kashyap on it, perhaps this Friday
13:14:09 pdeore gmann, yeah I'm trying that but not able to get in?
13:14:27 gmann pdeore: oh we will start at 14 UTC
13:14:30 kashyap stephenfin: I agree. I have been neglecting to break it down into pieces...
13:15:09 opendevreview Stephen Finucane proposed openstack/python-novaclient master: Deprecate nova CLI https://review.opendev.org/c/openstack/python-novaclient/+/836682
13:15:09 opendevreview Stephen Finucane proposed openstack/python-novaclient master: docs: Update docs to reflect deprecation status https://review.opendev.org/c/openstack/python-novaclient/+/836683
13:15:10 opendevreview Stephen Finucane proposed openstack/python-novaclient master: Update pre-commit hook, hacking versions https://review.opendev.org/c/openstack/python-novaclient/+/836768

Earlier   Later