| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2021-02-17 | |||
| 18:27:17 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: Add `socket` PCI NUMA affinity policy request prefilter https://review.opendev.org/c/openstack/nova/+/772749 | |
| 18:27:22 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: Track host NUMA topology in PCI manager https://review.opendev.org/c/openstack/nova/+/774149 | |
| 18:27:27 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: WIP: pci: implement the `socket` NUMA affinity policy https://review.opendev.org/c/openstack/nova/+/772779 | |
| 18:28:08 | stephenfin | sean-k-mooney: If you do look at the patch, ignore the fact that it's using neutronclient for a bit and tell me it doesn't make e.g. the NeutronFixture easier to grok | |
| 18:28:28 | stephenfin | that's where I'm coming from | |
| 18:28:58 | stephenfin | I'll do the sdk work eventually, once the OSC gaps are closed, but I think this is a win for now | |
| 18:29:32 | sean-k-mooney | well it not going to change the public api of that module so im not sure it will | |
| 18:29:42 | sean-k-mooney | im looking at it now | |
| 18:30:32 | sean-k-mooney | but the frecatorign you are ding im not sure require the code changes | |
| 18:31:58 | sean-k-mooney | none of those private api shoudl be needed in the generic neutron fixture https://review.opendev.org/c/openstack/nova/+/706295/10/nova/tests/fixtures.py#b1769 | |
| 18:32:11 | sean-k-mooney | they should not be called outside the neutron moduule | |
| 18:32:22 | sean-k-mooney | only the public funtion will be | |
| 18:33:05 | stephenfin | They're called inside the neutron module | |
| 18:33:21 | sean-k-mooney | yep which can stub them in its own unit tests | |
| 18:33:35 | stephenfin | NeutronFixture is weird. It mocks APIs from both neutronclient and nova.network.neutron | |
| 18:33:48 | sean-k-mooney | ya | |
| 18:34:07 | sean-k-mooney | i then to use the libvirt version of it too at least when wriging func test | |
| 18:34:18 | sean-k-mooney | there is value in this yes | |
| 18:34:18 | stephenfin | no, it we don't stub them out then we need to provide an entire mock of e.g. 'bind_ports_to_host' | |
| 18:34:48 | sean-k-mooney | well we proably shoudl do htat since we need it for live migration right | |
| 18:35:01 | stephenfin | well we do | |
| 18:35:09 | stephenfin | by mocking out the private APIs it calls | |
| 18:35:39 | stephenfin | that API is the glue. We don't want to mock that out | |
| 18:35:50 | stephenfin | but we do want to mock out the neutron API calls | |
| 18:36:09 | stephenfin | I said above that the fixture is weird because it mocks APIs from both neutronclient and nova.network.neutron | |
| 18:36:10 | sean-k-mooney | yes which is what it was doing before | |
| 18:36:38 | stephenfin | but it really shouldn't be doing that. The reason it has to is because we're providing our own neutronclient implementation of sorts using KSA | |
| 18:36:45 | stephenfin | so it has to mock those | |
| 18:36:53 | stephenfin | things like _create_port_binding | |
| 18:37:22 | stephenfin | which, I'll note, is also what neutronclient uses now that I look at it | |
| 18:37:24 | sean-k-mooney | im surpised that has actully been added to neutronclint | |
| 18:37:38 | sean-k-mooney | it was ment to back in rocky but never happened | |
| 18:38:05 | sean-k-mooney | i cant argue that its not shorter | |
| 18:38:27 | sean-k-mooney | and it appear to more or less be functional | |
| 18:39:05 | stephenfin | yeah, I added it because the ksa stuff confused me | |
| 18:39:21 | stephenfin | What's the easiest thing to do if something confuses you? Delete it! | |
| 18:39:23 | stephenfin | :) | |
| 18:39:45 | sean-k-mooney | well the reason i brougt this up is i am concerned by what it might conflict with | |
| 18:39:54 | sean-k-mooney | in terms of backports | |
| 18:40:46 | sean-k-mooney | stephenfin: it conclitc with https://review.opendev.org/c/openstack/nova/+/653506 | |
| 18:41:00 | sean-k-mooney | whcih we partly fixed a different way but we proably shoudl stil merge this | |
| 18:41:19 | stephenfin | It's easy to rework these things and we bring them back | |
| 18:41:27 | stephenfin | no different to using f-strings in code, surely? :) | |
| 18:41:43 | stephenfin | see what I did there? | |
| 18:42:33 | sean-k-mooney | readding ksa is alot harder | |
| 18:43:05 | sean-k-mooney | i guess on hte backport it will already be there | |
| 18:43:37 | sean-k-mooney | and if the neutron client fucntio you are using is not supprot by the older version of neutron clinet on the stable branch you would have to use ksa anyway | |
| 18:43:56 | stephenfin | yeah, unfortunately so. We can't bump deps either | |
| 18:44:01 | stephenfin | Anyhow, /me really needs to get to the gym | |
| 18:44:08 | stephenfin | (home gym, sadly) | |
| 18:44:09 | stephenfin | o/ | |
| 18:44:13 | sean-k-mooney | i guess we can proceed with thsi but i feel somewhat blind sided by it | |
| 18:44:24 | sean-k-mooney | i just wasnt expecting you to be reworking that | |
| 18:44:36 | sean-k-mooney | enjoy o/ | |
| 21:45:26 | openstackgerrit | melanie witt proposed openstack/nova master: Add regression test for bug 1914777 https://review.opendev.org/c/openstack/nova/+/775307 | |
| 21:45:27 | openstackgerrit | melanie witt proposed openstack/nova master: Handle instance = None in _local_delete_cleanup https://review.opendev.org/c/openstack/nova/+/775308 | |
| 21:45:27 | openstack | bug 1914777 in OpenStack Compute (nova) "Possible race condition between n-cpu and n-api when deleting a building instance" [High,In progress] https://launchpad.net/bugs/1914777 - Assigned to melanie witt (melwitt) | |
| 22:35:17 | openstackgerrit | Belmiro Moreira proposed openstack/nova master: Change API unexpected exception message https://review.opendev.org/c/openstack/nova/+/628061 | |
| #openstack-nova - 2021-02-18 | |||
| 01:12:27 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: WIP: pci: implement the `socket` NUMA affinity policy https://review.opendev.org/c/openstack/nova/+/772779 | |
| 07:07:40 | openstackgerrit | MIKI Nobuhiro proposed openstack/nova master: libvirt: add IP address to libvirt guest metadata https://review.opendev.org/c/openstack/nova/+/750552 | |
| 09:42:15 | swp20 | gibi: morning, please check the NOVNC patch https://review.opendev.org/c/openstack/nova/+/622336 if you have spare time. | |
| 10:00:34 | openstackgerrit | Lucian Petrut proposed openstack/nova master: hyper-v rbd volume support https://review.opendev.org/c/openstack/nova/+/763550 | |
| 10:01:35 | stephenfin | gibi: bauzas: I'm +2 on https://review.opendev.org/c/openstack/nova/+/750552 but have held off on +W because you both reviewed it previously | |
| 10:01:54 | stephenfin | Ah, wait, lyarwood dropped his +2 while I was reviewing so not as urgent after all | |
| 10:02:16 | stephenfin | lyarwood: Left comments inline but fwiw I'm fine with camelCase. We can discuss in the team meeting today if we want to though | |
| 10:02:40 | bauzas | stephenfin: ack, just fixing your concerns about negative tests missing on routed networks and then hopefully I should be on the review side | |
| 10:03:14 | gibi | bauzas, stephenfin: the general direction is OK to me in https://review.opendev.org/c/openstack/nova/+/706295 (neutron client bump) I just had some question inline | |
| 10:03:50 | gibi | swp20: I will try | |
| 10:04:25 | gibi | stephenfin: ack, I will check the libvirt metadata too | |
| 10:04:59 | stephenfin | gibi: cool, I'll take a look at the noVNC patch too | |
| 10:05:22 | lyarwood | stephenfin: yup sorry didn't want to jump the gun on that | |
| 10:15:39 | openstackgerrit | Vishal Manchanda proposed openstack/nova master: Docs: Correct ``Password injection using the dashboard`` Explanation https://review.opendev.org/c/openstack/nova/+/775084 | |
| 10:21:27 | gibi | stephenfin, lyarwood: I'm +2 on the libvirt ip metadata (including ipVersion). lyarwood do you want to get back to that patch or should I send it through? | |
| 10:22:24 | lyarwood | gibi: I just wanted to ensure Sean was okay with it tbh | |
| 10:22:47 | lyarwood | gibi: they left a comment on the spec about talking it over in the review | |
| 10:23:19 | gibi | lyarwood: OK let's wait for sean-k-mooney's vote on the review then | |
| 10:23:39 | lyarwood | gibi: https://review.opendev.org/c/openstack/nova/+/763550 - FWIW the HyperV RBD change looks ready now | |
| 10:23:45 | gibi | lyarwood: on it | |
| 10:26:02 | lpetrut | thanks for the reviews | |
| 10:27:56 | gibi | lpetrut: is there any other patch needed in nova for the rbd support? or can I close the bp once https://review.opendev.org/c/openstack/nova/+/763550 merges? | |
| 10:28:16 | lpetrut | nope, that's it | |
| 10:28:26 | gibi | lyarwood: thanks for the review on the noVNC patch | |
| 10:28:47 | gibi | lpetrut: ack, I will close the bp accordingly | |
| 10:45:34 | bauzas | oh, man, why _get_provider_aggregates() is so cryptic ? | |
| 10:46:00 | bauzas | it can return either None, a tuple with empty list, or an exception | |
| 10:46:42 | bauzas | stephenfin: thanks for having letting me rework on negative tests, I need to check all of the above | |
| 10:46:46 | gibi | because nobody cared in the past to have a proper return type. but now with mypy... :D | |
| 10:47:37 | bauzas | gibi: technically, it's a private method, so that's my call | |
| 10:47:52 | gibi | also we don't have a clear guideline (or pattern) when to return None for error case and when to raise for it | |
| 10:48:10 | gibi | s/None/falsy value/ | |
| 10:59:08 | openstackgerrit | Lucas Alvares Gomes proposed openstack/nova master: [OVN] Adapt the live-migration job scripts to work with OVN https://review.opendev.org/c/openstack/nova/+/776419 | |
| 11:16:07 | stephenfin | lyarwood: gibi: I have some pretty serious concerns about the VNC patch. Why would one ever recommend an 8 ASCII character, trivially brute forceable password over TLS encryption? | |
| 11:17:26 | stephenfin | Lots of comments left. I don't know if it's possible yet (starting to investigate), but this feels like something the proxy layer should be providing for us (i.e. HTTPS and a password to get to the proxy, and then TLS from the proxy to the guest) | |
| 11:19:39 | lyarwood | stephenfin: hmm how do we lose the ability to encrypt traffic between the proxy and guest? | |
| 11:19:46 | stephenfin | per my reading of things, yes | |
| 11:20:26 | stephenfin | the presence of an entirely separate auth scheme suggests as much https://review.opendev.org/c/openstack/nova/+/622336/30/nova/console/rfb/auths.py#30 | |
| 11:20:39 | stephenfin | as does the QEMU doc on available security methods https://qemu.readthedocs.io/en/latest/system/vnc-security.html | |
| 11:24:38 | lyarwood | stephenfin: ah crap sorry I didn't look at RFBAuthSchemeVeNCrypt | |
| 11:32:38 | openstackgerrit | Lee Yarwood proposed openstack/nova stable/ussuri: only wait for plugtime events in pre-live-migration https://review.opendev.org/c/openstack/nova/+/770745 | |
| 11:33:59 | stephenfin | Man, my security know how is not what it should be, but it sounds like there's a TLSNone subtype we can use with VeNCrypt https://github.com/rfbproto/rfbproto/blob/master/rfbproto.rst#7243subtypes-with-vnc-suffix | |