Thursday, 2024-04-04

opendevreviewMerged openstack/oslo.log master: Imported Translations from Zanata  https://review.opendev.org/c/openstack/oslo.log/+/91270501:03
opendevreviewAdam McArthur proposed openstack/oslo.service master: Removing oslo_wsgi eventlet dependency  https://review.opendev.org/c/openstack/oslo.service/+/91492701:24
opendevreviewXiang Wang proposed openstack/oslo.messaging master: Add HTTP driver to oslo.messaging.drivers  https://review.opendev.org/c/openstack/oslo.messaging/+/91484004:12
opendevreviewXiang Wang proposed openstack/oslo.messaging master: Add HTTP driver to oslo.messaging.drivers  https://review.opendev.org/c/openstack/oslo.messaging/+/91484005:59
opendevreviewMerged openstack/oslo.messaging stable/2024.1: Fix incorrect desc of rabbit_stream_fanout option  https://review.opendev.org/c/openstack/oslo.messaging/+/91458308:46
opendevreviewXiang Wang proposed openstack/oslo.messaging master: Implement HTTPListener and HTTPClient and add OpenAPI specs  https://review.opendev.org/c/openstack/oslo.messaging/+/91505611:13
opendevreviewMerged openstack/oslo.messaging stable/2023.1: kafka: Fix invalid hostaddr format for IPv6 address  https://review.opendev.org/c/openstack/oslo.messaging/+/91445011:13
*** sfinucan is now known as stephenfin12:25
hberaudelodilles: o/ would be good to have your thinking here https://review.opendev.org/c/openstack/oslo.log/+/91488013:17
elodilleshberaud: looking13:22
elodillesnice :/ in the first place, deprecations should happen on master branch. furthermore, there were some changes around deprecation (of features) due to SLURP releases (something like: things should be deprecated in a SLURP release, and can only be removed on another SLURP release, or something like that)... given though that this is needed for py312 compatibility, it's quite hard situation13:26
hberaudyep13:33
hberaudthat's why I requete13:33
hberaudrequested your expertize$13:33
hberaudesp13:33
hberaudexpertise13:34
* hberaud should boughts new fingers13:34
elodilles:)13:36
elodillesi've commented on the patch, but further ideas are welcome13:37
hberaudthanks13:37
elodilles(i don't think deprecation warning should not be backported, but if the feature needs to be removed as that is 1) causing errors 2) not even functional, then just simply remove it maybe?)13:40
elodilles*i don't think deprecation warning should be backported13:40
elodillesso let's NOT backport it o:)13:40
tkajinamMy plan was following what was done in nova for hyper-v removal. they did backport deprecation to stable/2023.1 and removed the feature in stable/2023.2 after creating a new 2023.1 release afair13:54
tkajinamMaking it noop is another option and probably a bit simpler approach13:59
opendevreviewTakashi Kajinami proposed openstack/oslo.log master: Disable implementation for watch_log_file  https://review.opendev.org/c/openstack/oslo.log/+/91506814:10
tkajinamelodilles hberaud, I'll send an email to the list (replying to https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/AQ7HH5KNLMLY66NY5TXQST6EXSUYB6VM/ )  I thought to do so but was pulled by other stuffs during my day14:14
hberaudack thanks14:16
elodillestkajinam: thanks!14:35
tkajinamI wish I could have found it early... but I didn't notice the situation until I looked into py312 job which was added recently...14:40
elodilleswe just started to test py312 as non-voting job, so i'm sure we'll see more similar issues in the coming months :/14:51
tkajinamyeah14:52
tkajinamI spent some time during 2024.1 cycle to review items in upper constraints and global requirements. hope I could caught some of such problems early by that work.14:52
elodillestkajinam: thanks for mentioning nova's case about deprecations. i see that there was a similar discussion in the comments ( https://review.opendev.org/c/openstack/nova/+/899159 ), so yes, there we agreed to backport the deprecation warning :S14:53
elodillesthough it was to mark a *driver* to be deprecated and due to the new rule about SLURP releases14:54
tkajinamyeah14:54
elodilles(so that it was marked as deprecated in Antelope and could be removed in Caracal)14:55
tkajinamI think we can take a simpler approach for this case because we are talking about a single optional feature which has small impact14:55
elodilleshere if i understand correctly we can't wait with the removal for E release anyway... :/14:56
elodillestkajinam: yepp, i agree ++14:56
tkajinamI'll see if anyone else brings any better idea, but if not then I'll update the mocking patch I proposed so that we can move it forward15:02
opendevreviewMerged openstack/oslo.messaging stable/2023.2: kafka: Fix invalid hostaddr format for IPv6 address  https://review.opendev.org/c/openstack/oslo.messaging/+/91444816:00
tkajinamhberaud, fyi. I found a new interesting eventlet problem https://bugs.launchpad.net/oslo.cache/+bug/206021516:17
tkajinamhberaud, this may probably need to be reported in eventlet but created one in oslo.cache first to quickly capture my findings...16:17
tkajinamhmm. ignore it. it seems the issue should be fixed in newer eventlet by https://github.com/eventlet/eventlet/commit/caf9f9983a43c537efff5c4c9ff1d543af6e122016:20
tkajinamok it turned out eventlet in rdo is a bit old16:31
amorinhey team, is there any room/talk on PTG next week for olso topics? Like oslo.messaging?21:23
opendevreviewXiang Wang proposed openstack/oslo.messaging master: Implement HTTPListener and HTTPClient and add OpenAPI specs  https://review.opendev.org/c/openstack/oslo.messaging/+/91505623:08
opendevreviewXiang Wang proposed openstack/oslo.messaging master: Implement HTTPListener and HTTPClient and add OpenAPI specs  https://review.opendev.org/c/openstack/oslo.messaging/+/91505623:42

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!