| opendevreview | Merged openstack/oslo.messaging master: Utilize the new RequestContext redacted_copy method https://review.opendev.org/c/openstack/oslo.messaging/+/894574 | 14:47 |
|---|---|---|
| opendevreview | Hervé Beraud proposed openstack/oslo.privsep master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/oslo.privsep/+/905930 | 15:21 |
| opendevreview | Hervé Beraud proposed openstack/oslo.service master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/oslo.service/+/905931 | 15:22 |
| opendevreview | Hervé Beraud proposed openstack/oslo.vmware master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/oslo.vmware/+/905933 | 15:24 |
| opendevreview | Hervé Beraud proposed openstack/futurist master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/futurist/+/905957 | 15:41 |
| jamespage | o/ | 15:42 |
| jamespage | is there an existing spot where I could refactor some copy/pasted code from clients into somewhere shared | 15:43 |
| jamespage | https://codesearch.openstack.org/?q=HTTPSConnection.__init__&i=nope&literal=nope&files=&excludeFiles=&repos= | 15:43 |
| jamespage | needs a minor refactor to support 3.12 and its in a few places | 15:43 |
| opendevreview | Hervé Beraud proposed openstack/oslo.concurrency master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/oslo.concurrency/+/905961 | 15:46 |
| tkajinam | jamespage, probably in openstacksdk ? seeing these are mostly client repos | 15:46 |
| tkajinam | ideally they should replace the current implementation by one dependent on sdk code but I'm not sure if that happens very soon | 15:46 |
| opendevreview | Hervé Beraud proposed openstack/oslo.db master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/oslo.db/+/905962 | 15:46 |
| tkajinam | hberaud, wondering if we have any actual problems without that eventlet bump. I expect no, as long as we control the version used in CI using upper-constraints | 15:47 |
| opendevreview | Hervé Beraud proposed openstack/oslo.log master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/oslo.log/+/905963 | 15:48 |
| opendevreview | Hervé Beraud proposed openstack/oslo.messaging master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/oslo.messaging/+/905964 | 15:49 |
| tkajinam | we haven't done such bump for the other libraries afair and usually distros are responsible for selecting the right version according to the runtime version they have | 15:49 |
| opendevreview | Hervé Beraud proposed openstack/oslo.reports master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/oslo.reports/+/905965 | 15:50 |
| JayF | tkajinam: In this case, there is value in bumping as eventlet-upstream has improved greatly. | 15:51 |
| opendevreview | Hervé Beraud proposed openstack/oslo.rootwrap master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/oslo.rootwrap/+/905966 | 15:51 |
| JayF | tkajinam: it esssentially went from no maintenance for 1y+, no working CI in years, to maintained with working CI in the last handful of months | 15:52 |
| JayF | the bump hberaud is pushing out gets us on the eventlet version that has these enhancements (in addition to a boatload of bug fixes, specifically impacting python 3.11/python 3.12) | 15:52 |
| tkajinam | ok | 15:52 |
| opendevreview | Hervé Beraud proposed openstack/oslo.utils master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/oslo.utils/+/905967 | 15:53 |
| tkajinam | if the new eventlet brings benefit for python < 3.12 and we strongly recommend using it then it may make sense to bump it. AFAIK we don't actively bump minimum versions unless we are aware of the interface change (eg. new features) required but this case may be special | 15:55 |
| opendevreview | Hervé Beraud proposed openstack/taskflow master: bump eventlet to latest version that support python 3.12 https://review.opendev.org/c/openstack/taskflow/+/905970 | 15:55 |
| JayF | Yeah, the improvements to upstream eventlet are significant and well beyond a simple python 3.12 bump, although that's the easiest thing to headline the PRs with :D | 15:58 |
| JayF | Herve, Itamar, and several other folks have been reinforcing the foundation of eventlet in the last while and Caracal will be better for it :) | 15:59 |
| tkajinam | ok | 16:00 |
| tkajinam | we can try merging these. I was thinking of trade off between benefit and effort of these patches but now I tend to agree it's good | 16:00 |
| tkajinam | we may face some problems with swift which still declares python 2 support but we can skip it if there is no functional problem without that bump | 16:00 |
| tkajinam | probably it's worth announcing in the ml so that we get attention (and agreement) about that global bump | 16:02 |
| JayF | There's been a lot of noise on the ML about the Eventlet lack of maintenance and fixing of that problem; this to me seems like a logical continuation... but more communication is never a bad thing :) | 16:06 |
| JayF | tkajinam: hberaud: on further reflection, and based on similar feedback from dtantsur; an email might be valuable. I think folks are missing the idea that this is an opportunistic improvement | 16:10 |
| dtantsur | As we just discussed on the Ironic IRC, https://review.opendev.org/c/openstack/ironic/+/905887 is probably the right thing to do because of how broken eventlet was | 16:10 |
| JayF | Literally we repeated, almost verbatim, the conversation tkajinam and I just had in here ;) | 16:11 |
| dtantsur | but the explanation "to support python 3.12" is utterly wrong and caused me to nearly -2 them | 16:11 |
| dtantsur | yep, sure, I was not here | 16:11 |
| JayF | Oh, I'm just letting you know context is basically the same without it being repeated | 16:11 |
| JayF | It's strangely comforting for everyone to be aligned in their confusion :D | 16:12 |
| dtantsur | :D | 16:13 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!