| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-08-20 | |||
| 00:53:19 | sean-k-mooney | or fix it before the relese | |
| 00:53:57 | sean-k-mooney | we cant assume multiple prot bindigns is supported by neutron | |
| 00:54:47 | sean-k-mooney | apparently neutron is not filtering it properly for ml2 drivers on older releases | |
| 00:55:17 | sean-k-mooney | also core plugins e.g. where you dont use ml2 may or may not support it | |
| 00:55:24 | sean-k-mooney | contrail does not currently | |
| 01:05:46 | sean-k-mooney | artom: so ya https://review.opendev.org/#/c/729763/ | |
| 01:06:12 | sean-k-mooney | we cant assume multiple port bindings is used | |
| 01:06:20 | sean-k-mooney | so that fix is not valid | |
| 01:08:09 | melwitt | sean-k-mooney: pls -1 the backport so reviewers know | |
| 01:08:26 | sean-k-mooney | just did | |
| 01:09:33 | sean-k-mooney | melwitt: i have been diging into https://review.opendev.org/#/c/742180/4/nova/objects/migrate_data.py@97 | |
| 01:10:12 | sean-k-mooney | for the last cople of days which seam to be cause by port-bindings being reported as supported when its not | |
| 01:10:37 | sean-k-mooney | the contrial core plugin does not support multipel port bidnigns so we have at least one concreate case | |
| 01:11:17 | sean-k-mooney | there are some ml2 driver that also dont in cluding the alternitive way to integrate vrouter/contrial via an ml2 driver | |
| 01:12:03 | sean-k-mooney | ovn also did not support it until recently. not sure if they added support or just removed it form the list. | |
| 01:12:31 | sean-k-mooney | but it broke 16 and they had to fix it quickly before the 16.0 release | |
| 01:12:57 | sean-k-mooney | so i dont know if they still have the quick fix or actully impemented support | |
| 04:58:49 | openstackgerrit | Merged openstack/nova master: trivial: Remove unused attributes https://review.opendev.org/743284 | |
| 04:59:00 | openstackgerrit | Merged openstack/nova master: Remove warning filters for legacy Python version https://review.opendev.org/743285 | |
| 04:59:12 | openstackgerrit | Merged openstack/nova master: libvirt: Remove MIN_LIBVIRT_ETHERNET_SCRIPT_PATH_NONE https://review.opendev.org/730865 | |
| 07:58:51 | stephenfin | gmann++ Thanks or the pointer | |
| 07:59:05 | stephenfin | *for | |
| 07:59:07 | stephenfin | and the reviews | |
| 08:02:55 | openstackgerrit | Stephen Finucane proposed openstack/os-vif master: windows: Add missing return https://review.opendev.org/746870 | |
| 08:02:56 | openstackgerrit | Stephen Finucane proposed openstack/os-vif master: mypy: Add type annotations https://review.opendev.org/745162 | |
| 08:02:57 | openstackgerrit | Stephen Finucane proposed openstack/os-vif master: mypy: Extend type annotations to 'vif_plug_noop' https://review.opendev.org/745163 | |
| 08:11:41 | lyarwood | kashyap: https://review.opendev.org/#/c/694821/ - Morning, do you recall why you picked NEXT_MIN_LIBVIRT_VERSION v5.0.0 and NEXT_MIN_QEMU_VERSION 4.0.0 in U? | |
| 08:12:26 | lyarwood | kashyap: https://review.opendev.org/#/c/746981/ - I started bumping things yesterday and noticed that as we don't use UCA anymore it's impossible for us to move to these until we move to focal | |
| 08:12:34 | kashyap | lyarwood: Hi, will answer in a bit. | |
| 08:14:36 | kashyap | lyarwood: On the first 'why', I always add the rationale - it was the common version available across the distros we care | |
| 08:15:54 | kashyap | lyarwood: Reasons documented here: http://lists.openstack.org/pipermail/openstack-discuss/2019-November/010907.html | |
| 08:16:14 | kashyap | lyarwood: On the "not using to UCA anymore" -- oops | |
| 08:16:25 | lyarwood | kashyap: yeah the wiki page is wrong | |
| 08:16:33 | lyarwood | kashyap: well, not wrong it does list UCA | |
| 08:16:36 | kashyap | Wrong how? | |
| 08:16:37 | lyarwood | kashyap: but we don't use it | |
| 08:16:38 | kashyap | Yeah, exactly | |
| 08:16:49 | kashyap | Right; I lost track when we stopped using UCA - do you know why? | |
| 08:17:12 | lyarwood | kashyap: we stopped using it a while ago, let me find the changes | |
| 08:18:48 | lyarwood | kashyap: https://review.opendev.org/#/c/695114/4/tools/fixup_stuff.sh - we removed it when we moved to bionic | |
| 08:20:41 | kashyap | lyarwood: Nod; I'll come back here in a bit. I need to do some time-sensitive adminstrative work :-( | |
| 08:21:17 | openstackgerrit | Merged openstack/nova master: Add checks for volume status when rebuilding https://review.opendev.org/734734 | |
| 08:23:35 | lyarwood | kashyap: np :) | |
| 08:25:09 | openstackgerrit | Hao Sun proposed openstack/nova stable/ussuri: Add checks for volume status when rebuilding https://review.opendev.org/747122 | |
| 08:26:17 | frickler | lyarwood: uca was removed in that patch because there is no uca for queens, bionic contains queens pkgs natively. if you want to re-add that code with some newer UCA now in order to get updated qemu/libvirt, I'd guess that'd make perfect sense | |
| 08:33:24 | lyarwood | frickler: yeah understood, I've posted a WIP to see if this breaks anything | |
| 08:42:28 | lyarwood | stephenfin, sean-k-mooney, kashyap : https://review.opendev.org/#/q/topic:bump-libvirt-qemu-victoria - The first two in this series are simple test fixes if you have time this morning btw | |
| 08:44:20 | openstackgerrit | Lee Yarwood proposed openstack/nova master: libvirt: Log exception when unable to import rbd https://review.opendev.org/746873 | |
| 08:50:56 | openstackgerrit | Stephen Finucane proposed openstack/nova master: functional: Clean up PCI tests https://review.opendev.org/746945 | |
| 08:50:56 | openstackgerrit | Stephen Finucane proposed openstack/nova master: functional: Add resize tests for PCI servers https://review.opendev.org/746946 | |
| 08:50:57 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conductor: Don't use setattr https://review.opendev.org/746947 | |
| 08:50:57 | openstackgerrit | Stephen Finucane proposed openstack/nova master: functional: Add test for SR-IOV neutron ports https://review.opendev.org/746948 | |
| 08:50:58 | openstackgerrit | Stephen Finucane proposed openstack/nova master: fixtures: Handle binding of first port https://review.opendev.org/746949 | |
| 08:50:58 | openstackgerrit | Stephen Finucane proposed openstack/nova master: functional: Add live migration tests for PCI, SR-IOV servers https://review.opendev.org/746950 | |
| 09:32:38 | stephenfin | alex_xu: Thanks for the review. Replied on https://review.opendev.org/#/c/744958/ | |
| 09:38:32 | openstackgerrit | Merged openstack/nova master: Add new default roles in baremetal nodes policies https://review.opendev.org/741716 | |
| 09:38:40 | openstackgerrit | Merged openstack/nova master: tests: Add reproducer for bug #1879878 https://review.opendev.org/744950 | |
| 09:38:40 | openstack | bug 1879878 in OpenStack Compute (nova) "VM become Error after confirming resize with Error info CPUUnpinningInvalid on source node " [Medium,In progress] https://launchpad.net/bugs/1879878 - Assigned to Stephen Finucane (stephenfinucane) | |
| 10:17:57 | artom | sean-k-mooney, yeah, you'll have to explain the multiple port bindings thing when I'm more awake | |
| 10:53:57 | stephenfin | luyao: Reviewed that series. Let me know if anything doesn't make sense | |
| 11:46:49 | admin0 | lyarwood, i have narrowed down my ceph issue, when i try to manually mount it, its asking me to disable object-map fast-diff deep-flatten | |
| 11:47:04 | admin0 | trying to do rbd map --pool vms --id cinder --conf /etc/ceph/ceph.conf $uuid_disk from the hypervsior directly | |
| 11:47:36 | lyarwood | admin0: you shouldn't need to mount it to export volumes | |
| 11:48:01 | lyarwood | admin0: rbd can export directly from the backend to a local file | |
| 11:48:02 | admin0 | my issue is when using raw, the instances do not boot up .. they are stuck in boot .. | |
| 11:48:13 | admin0 | even after a reformat, the behaviour is like that | |
| 11:48:37 | lyarwood | admin0: right, as we spoke about the other day the rbd import (not export sorry) command was stalling right? | |
| 11:55:12 | frickler | admin0: that sounds like you are trying to use rbd with an old kernel module, but I don't understand where this is needed, usually qemu should handle the rbd connection without kernel support | |
| 11:55:33 | frickler | but maybe you have a mismatch in ceph versions between your cluster and the hypervisor? | |
| 11:58:24 | admin0 | ceph is on 15.x on centos, clinet is 14.x on ubuntu | |
| 12:01:42 | lyarwood | admin0: did you stop using config drives? | |
| 12:01:48 | admin0 | yep | |
| 12:01:57 | lyarwood | admin0: and where is it failing now? | |
| 12:02:04 | admin0 | by stopping config drives, they can at least go active and show the console | |
| 12:03:25 | lyarwood | admin0: and by show in the console are you saying the instance boots as expected? | |
| 12:04:35 | admin0 | lyarwood, i have to run to a meeting now .. i will get back on this later | |
| 12:04:43 | admin0 | i think its the hypervisor -> osd connectivity that i think is the issue | |
| 12:27:52 | sean-k-mooney | admin0: did you figure out your nova-mange issue you are using openstack-helm they are configuring cells incorrectly | |
| 12:40:43 | admin0 | it was a network routing issue between hypervisor <-> osd | |
| 12:45:15 | sean-k-mooney | ok | |
| 12:46:07 | sean-k-mooney | there was someone else asking the same question later the same day that was cuased by helml not actully generig the correct config and not handeling the case wehre you go form a non tls deploymet to a tls deployment | |
| 12:46:12 | sean-k-mooney | as a result | |
| 12:46:28 | sean-k-mooney | admin0: i was jsut wondering if you were hitting the same issue. | |
| 12:48:46 | openstackgerrit | Merged openstack/nova master: zuul: use the new barbican simple-crypto job https://review.opendev.org/746583 | |
| 13:29:53 | gmann | stephenfin: np!, thank you too for re-reviewing policy patches. | |
| 13:30:00 | stephenfin | yw :) | |
| 13:38:54 | openstackgerrit | Vlad Gusev proposed openstack/nova stable/stein: libvirt: Use 'writeback' QEMU cache mode when 'none' is not viable https://review.opendev.org/747163 | |
| 13:44:36 | openstackgerrit | Mohammed Naser proposed openstack/nova master: db: fix database migrations when name includes dash https://review.opendev.org/747164 | |
| 13:45:29 | openstackgerrit | Mohammed Naser proposed openstack/nova master: db: fix database migrations when name includes dash https://review.opendev.org/747164 | |
| 13:45:30 | mnaser | ^ very trivial, honestly, i don't know how we'd test it | |
| 14:00:19 | stephenfin | alex_xu: I think there are a whole class of bugs here to do with races between setting the migration status to foo and dropping a move claim | |
| 14:00:26 | stephenfin | Thanks for the rabbit hole \o/ :P | |
| 14:06:02 | kashyap | lyarwood: Or anyone who recalls stable rules (my mind is ill right now, can't recall), does the above 'writeback' backport (https://review.opendev.org/#/c/747163/) look okay? | |
| 14:13:10 | lyarwood | kashyap: we would retain the existing cachemodes during a LM right? | |
| 14:13:34 | kashyap | lyarwood: Correct, until the instance is cold-rebooted on the destination. | |
| 14:13:40 | lyarwood | kashyap: if it doesn't impact running instances during LM then I'd be okay with this assuming it's now doing the correct thing | |
| 14:14:28 | lyarwood | kashyap: I'll review it in detail later | |
| 14:14:33 | kashyap | Yeah, it does the right thing now. I've spent several days extensively documenting it | |
| 14:14:48 | kashyap | lyarwood: No rush at all for it; just noticed only when Vlad (above) posted it | |
| 14:32:41 | artom | sean-k-mooney, whenever you're around, wanna talk about the multiple vif bindings thing? | |