Monday, 2025-07-21

*** mhen_ is now known as mhen01:51
*** elodilles_pto is now known as elodilles05:53
dtantsurAnother futurist question folks. Was it intended that stopping one worker thread always stop all threads?14:03
dtantsurI'm trying to understand if I should fix that or work around it.14:04
dtantsurhberaud[m]: if you have an opinion ^^14:04
hberaud[m]no sorry, no idea14:05
dtantsurmagical futurist :D14:05
dtantsurThreadWorker is not a part of the public API, I suspect I can change it14:06
TheJuliaI *suspect* it was always expected if you base it on a direct mapping14:11
hberaud[m]probably14:13
opendevreviewDmitry Tantsur proposed openstack/futurist master: Fix ThreadWorker.stop to stop only this worker  https://review.opendev.org/c/openstack/futurist/+/95547814:15
TheJuliaAt the same time, when I was looking at it last week as well, I was wondering the same thing as Dmitry14:16
opendevreviewDmitry Tantsur proposed openstack/futurist master: WIP: Add DynamicThreadPoolExecutor that resizes itself  https://review.opendev.org/c/openstack/futurist/+/95521714:17
dtantsurokay, added the stop() fix to the chain. Now a few more unit tests, and we're good to go.14:17
TheJuliavery cool14:20
opendevreviewMerged openstack/futurist master: Trivial: avoid noqa in __init__.py  https://review.opendev.org/c/openstack/futurist/+/95537415:33
opendevreviewDmitry Tantsur proposed openstack/futurist master: Add DynamicThreadPoolExecutor that resizes itself  https://review.opendev.org/c/openstack/futurist/+/95521715:34
dtantsurready for review ^^ (cc TheJulia)15:34
TheJuliaThanks, restarted my local ironic with that. So far so good16:52
TheJuliaaveraging ~7.5 minutes per pass17:17
TheJuliaI'll let it run for a while, if you want logs I can dump them out, but in the grand scheme of the universe I'm pretty good with it as-is17:17

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!