Tuesday, 2026-08-18

*** erlon9 is now known as erlon00:04
opendevreviewGoutham Pacha Ravi proposed openstack/nova master: Support cold migration with virtiofs shares  https://review.opendev.org/c/openstack/nova/+/98963301:02
opendevreviewMerged openstack/nova master: Stabilize functional test for slow instance.save  https://review.opendev.org/c/openstack/nova/+/99912101:59
opendevreviewGoutham Pacha Ravi proposed openstack/nova master: Support cold migration with virtiofs shares  https://review.opendev.org/c/openstack/nova/+/98963306:15
*** erlon1 is now known as erlon07:34
gibigmaan: I did some digging into the logs of https://bugs.launchpad.net/nova/+bug/2163552 it is strange as everything seems to correct, except that nova-api does not forwards the vif-plugged event to the compute even though the compute already waiting for it. 09:09
gibigmaan: funnily that sounds very similar to the downstream bug report you triaged recently https://redhat.atlassian.net/browse/OSPRH-3135409:14
gibiand that is reported to Antelope09:14
gibiand we see it on master09:14
gibicrazy09:14
opendevreviewBalazs Gibizer proposed openstack/nova master: [DNM]: Troubleshoot race condition in unshelve vif plug  https://review.opendev.org/c/openstack/nova/+/100129409:44
gibigmaan: ^^ I added extra logs 09:44
opendevreviewBalazs Gibizer proposed openstack/nova master: [DNM]: Troubleshoot race condition in unshelve vif plug  https://review.opendev.org/c/openstack/nova/+/100129411:33
opendevreviewMerged openstack/nova master: Make functional-py313-threading voting  https://review.opendev.org/c/openstack/nova/+/99721612:45
dansmithgibi: what is the purpose of the StaticallyDelayingCancellableTaskExecutorWrapper ?13:35
dansmithoh for filtering out stops when a start is right on the heels13:36
opendevreviewMerged openstack/nova master: libvirt: Check only x86_64 arch for SEV capability detection  https://review.opendev.org/c/openstack/nova/+/99828513:47
gibidansmith: yeah that is an old behavior of the libvirt virt driver13:51
gibijust eventlet had a native submit_after facility to handle that13:51
dansmithI remember discussing it now13:51
dansmithit's a very (impressively) long class name13:51
gibiI'm open to suggestions :) (we can still change futurist before the deadline);5;13~13:52
dansmithare you building that into futurist?13:53
gibiyes13:58
gibiit is in now13:58
gibiwaiting for a lib release to use it from there13:58
gibihttps://github.com/openstack/futurist/commit/350e39d788a04fdbd38529564a0f44f3e395f80f13:58
gibihttps://review.opendev.org/c/openstack/nova/+/997410 this is the patch that will usit if from futurist13:59
dansmithokay those seem like more reasonable names already :)13:59
gibi*/use it from/13:59
gibithanks to Kamil :)13:59
gibiand thanks to the fact that the delay is not static now14:00
gibiwe basically implemented submit_after for native thread pools14:00
dansmithcool14:07
*** chandank` is now known as chandankumar14:08
opendevreviewAshish Gupta proposed openstack/nova master: Fix init_host crash migration test under native threading  https://review.opendev.org/c/openstack/nova/+/100136914:58
gmaangibi: ok, i see, it seems similar to that downstream bug. I have some idea to add retry in server external event when it drop the network-vif-plugged  but not sure if that will solve the actual root cause or hide. because form downstream bug, it was ~23 sec when resource claim happened (which set the host on instance) and nova-api still did not see the host set and dropped the event15:27
gmaanthanks for the adding more logging, will do multiple recheck if we can hit it15:28
gibi23 seconds very long15:28
gmaanyeah15:28
gibithe current run of the DNM patch hit the issue but I haven't had the time to check the logs15:29
*** M00SE1 is now known as M00SE15:30
gmaannice, will check in noon15:51
opendevreviewDaniel Marques Caires proposed openstack/nova master: Add functional regression test for bug #2156927  https://review.opendev.org/c/openstack/nova/+/100054816:17
opendevreviewDaniel Marques Caires proposed openstack/nova master: nova: cast set_host_enabled on disable  https://review.opendev.org/c/openstack/nova/+/99846416:17
opendevreviewDaniel Marques Caires proposed openstack/nova master: Add functional regression test for bug #2156927  https://review.opendev.org/c/openstack/nova/+/100054816:20
opendevreviewDaniel Marques Caires proposed openstack/nova master: nova: cast set_host_enabled on disable  https://review.opendev.org/c/openstack/nova/+/99846416:47
opendevreviewClif Houck proposed openstack/nova master: perf(ironic): eliminate O(N²) ProviderTree deepcopy at startup  https://review.opendev.org/c/openstack/nova/+/98067617:21
opendevreviewClif Houck proposed openstack/nova master: Parallelize per-node resource updates  https://review.opendev.org/c/openstack/nova/+/98067917:21
opendevreviewMerged openstack/nova stable/2026.1: Restore 'fork' start method for daemon mode on Python 3.14  https://review.opendev.org/c/openstack/nova/+/99966717:43
gmaandansmith: I replied to the unify thread pool comments, I see the value of initializing executors during service start so I am ok to change it in this effort only instead of doing it again. we still have 2 weeks for FF, i hope that is enough time to review.   but let me know if you are busy for other things to review before FF then I will add the missing test coverage and refactor the initialization part later.19:14
gmaanmelwitt: this might fix the failure you saw for graceful shutdown job fail to stop service https://review.opendev.org/c/openstack/nova/+/100101719:23
melwittah cool, will look19:24
gmaanmelwitt: and on top of it I improved that job tasks name so that it is easy to know which operation tests it is running https://review.opendev.org/c/openstack/nova/+/100121219:24
gmaancurrently ansible top level block name are not appended in console logs so adding operation name in front of those tasks19:25
gmaanthanks19:25
dansmithgmaan: don't change it now.19:45
gmaandansmith: ok19:48
dansmithI mean, you can change it now, but don't change it in this commit.. let's not jeopardize landing the functionality :)19:49
gmaandansmith: got it, let me add the missing test coverage in this change and propose new one independent of graceful shutdown series for refactoring 19:50
dansmithyup thanks19:51
opendevreviewGhanshyam Maan proposed openstack/nova master: Unify the thread pool executors into a central factory  https://review.opendev.org/c/openstack/nova/+/99857121:34
gmaandansmith: ^^ a few jobs are still running in CI but it is ready22:38

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!