| *** mhen_ is now known as mhen | 01:15 | |
| *** ralonsoh_ is now known as ralonsoh | 08:43 | |
| elodilles | hi Oslo team, i see that the upper constraint version bump of oslo.serialization failed and the 2 job that is breaking seem to be related to the change: https://review.opendev.org/c/openstack/requirements/+/958742 | 12:20 |
|---|---|---|
| elodilles | it seems that (if i understand correctly) heat and openstacksdk need some code changes | 12:22 |
| yusufgungor_ | hi everyone, after 2025.1 upgrade cinder-scheduler_fanout, scheduler_fanout AMQP queues keeps increasing. Is there anyone else experiencing this situation? | 13:28 |
| yusufgungor_ | https://usercontent.irccloud-cdn.com/file/cdrUcacK/image.png | 13:29 |
| yusufgungor_ | we have talked on #openstack-kolla and @mnasiadka helped us to understand that situation is normal when using stream type of queues. you can ignore this messages | 13:42 |
| tkajinam | elodilles, the problem with heat is a "known" one caused by immediate renaming of public modules... I insistent that we should have avoided that but the other cores preferred it to avoid redundant deprecation caused by intermediate changes during this cycle... | 14:15 |
| tkajinam | elodilles, I've proposed the transition fix for heat which you can find as a dependency of that req bump. it seems the job is unblocked so I'll take care of merging that heat update | 14:15 |
| tkajinam | elodilles, I didn't see anything related to openstacksdk (I guess it's the one actually related to oslo.serialization ?) If you share its pointer, that is helpful | 14:16 |
| tkajinam | (the req bump you posted is related to heat and oslo.service, and looks irrelevant to openstacksdk+oslo.serialization | 14:16 |
| elodilles | tkajinam: thanks for working on that! | 14:56 |
| elodilles | tkajinam: this is the openstacksdk job failure on the patch i linked: https://zuul.opendev.org/t/openstack/build/e9aaad0166f0401b8d5c46500f03f7ce | 14:57 |
| elodilles | tkajinam: but you are right, the error is actually coming from heat-engine | 14:58 |
| elodilles | tkajinam: heat-engine[93189]: ERROR heat-engine ModuleNotFoundError: No module named 'oslo_service.backend.threading' | 14:58 |
| elodilles | i hope this renaming won't block other things (that are not tested by requirments gate). the best is to merge the upper constraint as soon as possible so that we will have time to find other possible errors and fix them before the coordinated 2025.2 Flamingo release :S | 15:03 |
| elodilles | (and of course, i wanted to write 'oslo.service' not serialization... sorry for that... :S) | 15:07 |
| tkajinam | elodilles, heat is a kind of special which requires extended ThreadGroup... the other projects may not touch that class directly but only via the remaining public interface | 15:08 |
| elodilles | tkajinam: ahh, good to hear that \o/ | 15:09 |
| elodilles | thanks again for spending time with fixing this! | 15:10 |
| tkajinam | elodilles, there is a potentially better approach proposed by ramishra ( https://review.opendev.org/c/openstack/heat/+/958773 ) so I'll check if it can pass CI, and then approve that one instead | 15:11 |
| tkajinam | anyway I'll take care of that req bump.... I might leave it for tomorrow, though | 15:11 |
| elodilles | tkajinam: sure, it's perfectly OK for tomorrow as well | 15:12 |
| opendevreview | Merged openstack/futurist master: Fix `shutdown(wait=True)` to wait for queued tasks https://review.opendev.org/c/openstack/futurist/+/958689 | 15:49 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
