| opendevreview | VO LE HUY proposed openstack/nova master: Shouldn't use the abs(elapsed) function when elapsed is a negative number. Wrong operating principle but nova service still shows up status. https://review.opendev.org/c/openstack/nova/+/908536 | 00:35 |
|---|---|---|
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Detect vtpm support by libvirt https://review.opendev.org/c/openstack/nova/+/908508 | 02:03 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Detect vtpm support by libvirt https://review.opendev.org/c/openstack/nova/+/908508 | 02:46 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Ensure both swtpm and swtpm_setup exist for vTPM support https://review.opendev.org/c/openstack/nova/+/908541 | 03:01 |
| opendevreview | VO LE HUY proposed openstack/nova master: pep8: E501 line too long (80 > 79 characters) https://review.opendev.org/c/openstack/nova/+/908542 | 03:02 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Ensure both swtpm and swtpm_setup exist for vTPM support https://review.opendev.org/c/openstack/nova/+/908541 | 03:06 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Ensure swtpm_ioctl is available for vTPM support https://review.opendev.org/c/openstack/nova/+/908546 | 03:18 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Ensure swtpm_ioctl is available for vTPM support https://review.opendev.org/c/openstack/nova/+/908546 | 03:19 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Detect vtpm support by libvirt https://review.opendev.org/c/openstack/nova/+/908508 | 03:44 |
| opendevreview | VO LE HUY proposed openstack/nova master: Add unit tests https://review.opendev.org/c/openstack/nova/+/908548 | 04:05 |
| opendevreview | VO LE HUY proposed openstack/nova master: It may hide time sync problems causing backword clocks. https://review.opendev.org/c/openstack/nova/+/908536 | 04:18 |
| opendevreview | VO LE HUY proposed openstack/nova master: Add unit test and fix pep8: E501 line too long https://review.opendev.org/c/openstack/nova/+/908553 | 04:18 |
| opendevreview | VO LE HUY proposed openstack/nova master: Add unit test and fix pep8: E501 line too long https://review.opendev.org/c/openstack/nova/+/908553 | 04:18 |
| opendevreview | VO LE HUY proposed openstack/nova master: Treat future service timestap as error https://review.opendev.org/c/openstack/nova/+/908536 | 04:27 |
| opendevreview | VO LE HUY proposed openstack/nova master: Treat future service timestap as error https://review.opendev.org/c/openstack/nova/+/908554 | 04:34 |
| opendevreview | melanie witt proposed openstack/nova master: WIP Support encrypted backing files for qcow2 https://review.opendev.org/c/openstack/nova/+/907961 | 04:38 |
| opendevreview | melanie witt proposed openstack/nova master: libvirt: Introduce support for raw with LUKS https://review.opendev.org/c/openstack/nova/+/884313 | 04:38 |
| opendevreview | melanie witt proposed openstack/nova master: libvirt: Introduce support for rbd with LUKS https://review.opendev.org/c/openstack/nova/+/889912 | 04:38 |
| opendevreview | VO LE HUY proposed openstack/nova master: Treat future service timestap as error https://review.opendev.org/c/openstack/nova/+/908554 | 06:07 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Detect vtpm support by libvirt https://review.opendev.org/c/openstack/nova/+/908508 | 07:13 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Detect vtpm support by libvirt https://review.opendev.org/c/openstack/nova/+/908508 | 08:14 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Detect vtpm support by libvirt https://review.opendev.org/c/openstack/nova/+/908508 | 08:21 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Ensure both swtpm and swtpm_setup exist for vTPM support https://review.opendev.org/c/openstack/nova/+/908541 | 08:23 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Ensure swtpm_ioctl is available for vTPM support https://review.opendev.org/c/openstack/nova/+/908546 | 08:23 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: libvirt: Detect vtpm support by libvirt https://review.opendev.org/c/openstack/nova/+/908508 | 08:23 |
| opendevreview | VO LE HUY proposed openstack/nova master: Treat future service timestap as error https://review.opendev.org/c/openstack/nova/+/908554 | 09:05 |
| opendevreview | VO LE HUY proposed openstack/nova master: Treat future service timestap as error https://review.opendev.org/c/openstack/nova/+/908554 | 09:07 |
| opendevreview | VO LE HUY proposed openstack/nova master: Treat future service timestamp as error https://review.opendev.org/c/openstack/nova/+/908554 | 09:07 |
| opendevreview | VO LE HUY proposed openstack/nova master: Treat future service timestamp as error https://review.opendev.org/c/openstack/nova/+/908554 | 09:10 |
| opendevreview | Takashi Kajinami proposed openstack/nova master: Remove old warning log about UEFI support https://review.opendev.org/c/openstack/nova/+/908563 | 09:20 |
| tkajinam | I wonder if UEFI support is still experimental really or we can just remove that warning ^^^ | 09:21 |
| noonedeadpunk | hey folks, do you know smth about centos9stream quite recent failures? | 09:38 |
| noonedeadpunk | in terms that we've started getting smth like that for nova-compute recently https://zuul.opendev.org/t/openstack/build/4cacd5600c0e43b0be4e443ad33a9169/log/logs/host/nova-compute.service.journal-22-34-38.log.txt#6333 | 09:39 |
| noonedeadpunk | this does not affect rocky linux nor ubuntu or debian | 09:39 |
| tkajinam | noonedeadpunk, maybe https://issues.redhat.com/browse/RHEL-20609 ? | 09:59 |
| tkajinam | the page is still loading for me, though | 09:59 |
| noonedeadpunk | tkajinam: ok, yes, that's likely it:) | 10:07 |
| tkajinam | yeah I see repeated unicode error in nova-compute logs | 10:08 |
| tkajinam | we excluded that known bad version in puppet CI to workaround the issue | 10:08 |
| noonedeadpunk | We mostly switched jobs to rocky linux, as it tends not to go on this bleeding edge as centos does. | 10:24 |
| noonedeadpunk | So rocky is unaffected... | 10:24 |
| noonedeadpunk | but trying to keep an eye on centos as well, so that issue won't propogate to rocky | 10:24 |
| opendevreview | Amit Uniyal proposed openstack/nova master: Allow swap resize from non-zero to zero https://review.opendev.org/c/openstack/nova/+/857339 | 10:28 |
| tkajinam | noonedeadpunk, Is that in OSA right ? I wonder if you are using RDO with Rocky or a different install source ? | 10:38 |
| gibi | tkajinam: noonedeadpunk: yes libvirt 9.10 has a regression | 10:39 |
| * tkajinam is hoping 10.0.0 is released very soon in mirrors | 10:39 | |
| gibi | we needed to pin to a lower version https://github.com/openstack-k8s-operators/edpm-ansible/pull/560 | 10:39 |
| noonedeadpunk | doesn't libvirt 10.0.0 giving only new modular version? | 10:47 |
| noonedeadpunk | ie https://fedoraproject.org/wiki/Changes/LibvirtModularDaemons | 10:48 |
| tkajinam | noonedeadpunk, afaik it still supports modular one | 10:49 |
| noonedeadpunk | or well, integrated one?:) | 10:49 |
| noonedeadpunk | old one... | 10:49 |
| tkajinam | single libvirtd instead of virt(qemu|node|...)d | 10:49 |
| noonedeadpunk | yeah | 10:49 |
| noonedeadpunk | ok, good then | 10:49 |
| tkajinam | https://review.opendev.org/c/openstack/puppet-openstack-integration/+/908291?tab=change-view-tab-header-zuul-results-summary | 10:49 |
| noonedeadpunk | As I really would hate to do such split in a middle of OS lifecycle | 10:50 |
| tkajinam | we tried the test build of 10.0.0 and it worked. (the failure in scenario004 job is irrelevant. 001-003 use single libvirtd) | 10:50 |
| tkajinam | I can't agree more | 10:50 |
| noonedeadpunk | ++ | 10:50 |
| tkajinam | noonedeadpunk, you can still use single libvirtd in c9s but at some point they required us to install an additional package to use it | 10:51 |
| noonedeadpunk | yes, true | 10:51 |
| tkajinam | the libvirt-daemon package, it seems | 10:51 |
| noonedeadpunk | yep, exactly | 10:52 |
| noonedeadpunk | since 9.3 | 10:52 |
| noonedeadpunk | it was removed from dependencies | 10:52 |
| tkajinam | yeah | 10:53 |
| tkajinam | I'm also not much fun of their switch to .socket units because it killed multiple assumptions in idempotency and service refresh but they believe that's better :-( | 10:53 |
| noonedeadpunk | ah, yes, that you need to stop socket/service/other sockets in order not to get service started by socket... | 10:54 |
| noonedeadpunk | ugh | 10:54 |
| noonedeadpunk | so much fun :) | 10:54 |
| opendevreview | VO LE HUY proposed openstack/nova master: Treat future service timestamp as error https://review.opendev.org/c/openstack/nova/+/908554 | 11:22 |
| sean-k-mooney | noonedeadpunk: i recrently decontainerised libvirt for our next gen installer and yes socket activation was... fun | 12:55 |
| artom | sean-k-mooney, https://review.opendev.org/c/openstack/nova/+/904183 | 15:18 |
| artom | gibi, 5 line diff needs a second +2 :) | 15:19 |
| gibi | artom: done. | 15:59 |
| artom | \o/ | 16:00 |
| bauzas | dansmith: good news, it looks we already have a guest XML on the target | 16:24 |
| bauzas | when we migrate | 16:24 |
| dansmith | ... / | 16:25 |
| bauzas | so I think we even don't need to use a dict | 16:25 |
| dansmith | oh, cool | 16:25 |
| bauzas | I'll just doublecheck that the method we already have for knowing which mdevs are assigned is getting *all* the guests, even if paused | 16:25 |
| bauzas | if so, this was already calculating that the target mdev was used | 16:26 |
| dansmith | yeah that'd be much nicer, IMHO | 16:26 |
| bauzas | if so, I'll just abandon the related change | 16:26 |
| bauzas | ... but fwiw the migration takes a lot of time because of the mdevs :) | 16:27 |
| dansmith | I imagine they dirty memory/state very quickly | 16:28 |
| bauzas | yeah the A100 GPU doesn't support a dirty memory support | 16:29 |
| bauzas | dirty memory tracking* | 16:30 |
| bauzas | L4s are other cards do, not that one | 16:30 |
| sean-k-mooney | melwitt: quick update. i have tested live/cold migate (confirm and revert) all are looking good | 18:26 |
| sean-k-mooney | i forgot to start an etherpad before doign that so ill likely start again form scratch and build that up | 18:27 |
| melwitt | sean-k-mooney: ok, cool. that will be good too bc I fixed some things with the latest respin | 19:23 |
| sean-k-mooney | ack i started lookign at the base of the serise again but realised i was tired do im calling it there for today | 19:28 |
| melwitt | get some rest 🙂 | 19:31 |
| opendevreview | Jay Faulkner proposed openstack/nova-specs master: Sharding: More detailed test plan https://review.opendev.org/c/openstack/nova-specs/+/908673 | 20:25 |
| JayF | sean-k-mooney, others ^ updated a detailed test plan in there | 20:25 |
| JayF | Figured you all might appreciate us being explicit about how we're testing, and finding a way to test it without doing an additional server boot. | 20:26 |
| JayF | (I'll note that Ironic has also added full API contract tests for /v1/node endpoint around anything relating to shards and querying for nodes in shards) | 20:29 |
| opendevreview | Jay Faulkner proposed openstack/nova-specs master: Sharding: More detailed test plan https://review.opendev.org/c/openstack/nova-specs/+/908673 | 21:12 |
| opendevreview | melanie witt proposed openstack/nova master: libvirt: Introduce support for rbd with LUKS https://review.opendev.org/c/openstack/nova/+/889912 | 21:21 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!