Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2020-07-02
17:02:58 elod melwitt: oh, i see. then I will try to see if that is possible to raise it to 1.4.0 in Stein (if i'm not mistaken it's only a problem with py35)
17:03:18 sean-k-mooney we are pretty familar with it upstram too considering how much pain it has caused around Feature Freaze for thae last few cycles
17:03:50 stephenfin sean-k-mooney: melwitt: You're welcome 🌟 ;)
17:04:01 sean-k-mooney elod: nope it was a proablem on py27 and several versions of py3 if i recall
17:04:02 melwitt elod: yeah I was thinking we could propose it to openstack/requirements upper-constraints. we can't bump lower bounds on stable but we can bump upper bounds
17:04:21 melwitt stephenfin: heh, gold star worthy indeed
17:04:49 elod melwitt: and crossing fingers that it won't break other constraints ;)
17:05:01 sean-k-mooney elod: basically the subunit parser bug can happen anytime the output of the test exceed the size of a single frame in the subunit protocol
17:05:50 sean-k-mooney elod: stephenfin found a tricky race in how it processed the io streams
17:05:52 melwitt elod: yeah... I'm not sure whether it will work out but I don't have any other ideas
17:06:11 stephenfin elod: https://github.com/testing-cabal/subunit/pull/40
17:07:22 elod sean-k-mooney melwitt stephenfin : nice! thanks for the details, now the picture is somewhat complete :]
17:08:13 elod i've added then the 'subunit raise to 1.4.0' to my TODOs :]
18:40:37 openstackgerrit Merged openstack/nova master: Fix cherry-pick check for merge patch https://review.opendev.org/738271
18:40:47 openstackgerrit Merged openstack/nova master: Fix user creation with GRANT in MySQL 8.0(Ubuntu Focal) https://review.opendev.org/738723
18:50:14 openstackgerrit sean mooney proposed openstack/nova master: lookup nic feature by pci address https://review.opendev.org/739131
18:53:08 sean-k-mooney melwitt: if you have time can you take a look at ^
18:53:28 melwitt ack
18:54:28 sean-k-mooney thanks im going to go have dinner. ill be deploying that and testing it on my sriov host tomorrow so ill adress any feedback then
#openstack-nova - 2020-07-03
01:18:19 openstackgerrit Sam Morrison proposed openstack/nova master: WIP: Lock down evacuate request spec to the same cell https://review.opendev.org/739157
02:32:58 openstackgerrit Kevin Zhao proposed openstack/nova master: [WIP] CI: add tempest-integrated-compute-aarch64 job https://review.opendev.org/714439
03:17:28 tony_su /quit
03:49:46 openstackgerrit Sam Morrison proposed openstack/nova master: Lock down evacuate request spec to the same cell https://review.opendev.org/739157
08:03:15 openstackgerrit Elod Illes proposed openstack/nova stable/ussuri: Fix cherry-pick check for merge patch https://review.opendev.org/739176
08:23:00 brinzhang_ gibi: good morning
08:24:51 brinzhang_ gibi: In addition to this simple upgrade document, where is there a more detailed record? For example, how to modify the configuration file, yum source configuration, etc.
08:25:01 brinzhang_ https://docs.openstack.org/nova/stein/user/upgrade.html
08:25:03 openstackgerrit Elod Illes proposed openstack/nova stable/train: Fix cherry-pick check for merge patch https://review.opendev.org/739177
08:44:50 gibi brinzhang_: I don't think we have such document in nova. I think the deployment engines (ansible, kolla, tripleo) does that for us
08:49:50 brinzhang_ gibi: yeah, I pinged kolla, and get some useful info, thanks ^
09:02:37 bauzas brinzhang_: we have installation docs that are related to the OS
09:02:51 bauzas eg. https://docs.openstack.org/nova/stein/install/controller-install.html
09:03:15 brinzhang_ bauzas: ack, thanks
09:03:23 bauzas but those docs are old
09:03:31 bauzas and I'm not sure they are all good
09:04:29 brinzhang_ bauzas: got it, and just as reference, if there are some place need to be updated, I think I can submit it
09:14:27 stephenfin gibi: I think I'm going to split up https://review.opendev.org/#/c/631363/ since it's rather hefty. Is that okay with you?
09:47:19 gibi stephenfin: sure
09:47:21 gibi thanks
10:03:08 priteau Hi Nova devs, could you confirm that nova-conductor doesn't need access to a policy file? I don't see any call to context.can() in its source code
10:03:30 openstackgerrit Balazs Gibizer proposed openstack/nova stable/train: Guard against missing image cache directory https://review.opendev.org/738455
10:05:39 gibi priteau: I think only the nova-api service needs the policy file but gmann is the best person to confirm it
10:08:15 priteau Based a comment in kolla-ansible, it looks like there is *one* occurence in code that makes it required by nova-compute: https://opendev.org/openstack/nova/src/commit/b72980960e62e05ad4dcd9af196105e3af0ac43f/nova/network/neutron.py#L605
10:08:37 sean-k-mooney stephenfin: im currently adressing the comments on the libvirt patch but over all are you ok with the way i adressed it. other then the normal style nits/typos i didnt really see any significant changes requested
10:09:56 gibi priteau: ohh, that sounds bad but most probably necessary
10:09:57 stephenfin sean-k-mooney: I agree with all the comments but yeah, the approach makes sense to me
10:10:39 sean-k-mooney cool ill be deploying it on my sriov systems shortly and then after i test it there ill push the update
10:12:23 sean-k-mooney hehe did you notice that the vm im currently using does not have ispell currently installed :P
10:30:08 stephenfin sean-k-mooney: I may have, yes :)
10:31:01 sean-k-mooney is the mypy check runable via tox? we have an env right
10:31:34 sean-k-mooney stephenfin: i was thinking that it would be nice to add it to the precommit config too the flake8 check in that did not run it
10:31:51 sean-k-mooney in a seperate patch obviously
10:32:12 stephenfin yeah, you can use 'tox -e mypy'
10:32:24 sean-k-mooney yep just found it
10:33:15 stephenfin adding it to pre-commit makes sense too, though I'm not sure if it'd work: pre-commit usually only checks the files you modified but we'd want to test all files listed in mypy-files.txt because of the interdependencies
10:33:56 sean-k-mooney i think you can disable that in the precomit config
10:34:11 sean-k-mooney but ya im not sure
10:34:16 stephenfin so long as we can do that for a single check
10:34:29 sean-k-mooney ya i think we could do it in the check definiton
10:34:30 stephenfin we don't want to run flake8 on every file everytime you commit. It's slooowwww
10:34:47 sean-k-mooney ya
10:38:27 sean-k-mooney i cant help thinking we should break up the libvirt driver.py file
10:38:46 stephenfin We _really_ need to
10:38:56 stephenfin it's far too big for its own good
10:38:57 sean-k-mooney partly because test file is 26K lines long
10:41:14 sean-k-mooney emacs promts me to ask if im sure i want to use python mode with it incase i have bad performance
10:41:39 sean-k-mooney it works fine more or less but still
10:44:36 openstackgerrit Sylvain Bauza proposed openstack/nova-specs master: Routed network scheduling spec https://review.opendev.org/733703
11:12:43 openstackgerrit sean mooney proposed openstack/nova master: lookup nic feature by pci address https://review.opendev.org/739131
11:13:16 stephenfin lyarwood: care to look at https://review.opendev.org/#/c/683043/ today?
11:16:43 openstackgerrit Stephen Finucane proposed openstack/nova master: Add support for resize and cold migration of emulated TPM files https://review.opendev.org/639934
11:16:43 openstackgerrit Stephen Finucane proposed openstack/nova master: Add emulated TPM support to Nova https://review.opendev.org/631363
11:16:44 openstackgerrit Stephen Finucane proposed openstack/nova master: libvirt: Split '_create_domain' function https://review.opendev.org/739208
11:16:44 openstackgerrit Stephen Finucane proposed openstack/nova master: crypto: Add type hints https://review.opendev.org/739207
11:16:45 openstackgerrit Stephen Finucane proposed openstack/nova master: scheduler: Request vTPM trait based on flavor or image https://review.opendev.org/739210
11:16:45 openstackgerrit Stephen Finucane proposed openstack/nova master: libvirt: Add vTPM config support https://review.opendev.org/739209
11:16:46 openstackgerrit Stephen Finucane proposed openstack/nova master: manager: Prevent compute startup on invalid vTPM config https://review.opendev.org/739212
11:16:46 openstackgerrit Stephen Finucane proposed openstack/nova master: crypto: Add support for creating, destroying vTPM secrets https://review.opendev.org/739211
11:16:47 openstackgerrit Stephen Finucane proposed openstack/nova master: docs: Add docs for vTPM support https://review.opendev.org/739213
11:30:56 lyarwood stephenfin: not today sorry, downstream bug backlog fun.
11:33:06 stephenfin ack
11:42:41 openstackgerrit Jiri Suchomel proposed openstack/nova master: Move image verification and writing out of download method https://review.opendev.org/738738
11:45:31 sean-k-mooney stephenfin: i was going to rework the previous patch but i dont see the point given im rewriting it in https://review.opendev.org/#/c/739131/ and i cant test the excption raising eaisly
11:46:08 sean-k-mooney do you have time to take a look before you finish for the day
11:48:29 openstackgerrit Wenping Song proposed openstack/nova master: Admin auth token to call Cyborg API by Nova https://review.opendev.org/739011
12:30:32 sean-k-mooney bauzas: im debating if i should abandon https://review.opendev.org/#/c/739017/1 and just go with https://review.opendev.org/#/c/739131/2
12:31:57 sean-k-mooney bauzas: what do you think? i cant easily test raising the excption as the a fakelibivt.libvirtError will not be caught correctly and i cant mock raising libvirt.libvirtError without adding libvirt-python as a test dep
12:33:13 sean-k-mooney hum altough this works for some reason https://review.opendev.org/#/c/739131/2/nova/tests/unit/virt/libvirt/test_host.py@1200
12:44:52 sean-k-mooney damb i have a bug..
12:57:29 bauzas sean-k-mooney: sorry, was not looking at the chan
12:58:26 sean-k-mooney i need to figure out what the function is renamed too but basically im wondering if there si any point doing the incomplete fix when i more or less have the correct fix written
12:58:29 bauzas sean-k-mooney: well, it's a bit larger
12:58:58 sean-k-mooney right but we will be backporting both in either case
13:01:24 sean-k-mooney virConnectListAllNodeDevices becomes listAllDevices
13:02:38 sean-k-mooney i guess it makes sense but still kind of wish the python bindings had docs that were actully useful
13:27:30 openstackgerrit sean mooney proposed openstack/nova master: lookup nic feature by pci address https://review.opendev.org/739131
14:23:53 openstackgerrit Lucio Seki proposed openstack/nova master: Implement extend_volume for libvirt NFS volume driver https://review.opendev.org/739077
14:34:13 gmann priteau gibi yeah, servers_policies.NETWORK_ATTACH_EXTERNAL, is the only policy which is enforced outside API service. May be we can move that to API layer and pass the info via parameter.
14:34:52 gmann but until you are not overriding it, you do not need the policy file and just reply on default polocy in code.
14:34:56 gmann policy

Earlier   Later