| cid | JayF: I will include that once I'm fully on seat. | 07:14 |
|---|---|---|
| cid | Good morning ironic o/ | 07:15 |
| rpittau | good morning ironic! o/ | 07:27 |
| adam-metal3 | good morning Ironic! | 07:34 |
| rpittau | hey adam-metal3 :) | 07:49 |
| rpittau | adam-metal3: the spec is a very interesting reading, I didn't leave any review for the time being, but I added myself to the reviewers list, hopefully I'll be able to tackle that a bit deeper before EOW | 07:49 |
| adam-metal3 | rpittau: thanks | 07:53 |
| rpittau | mmm I just realized I forgot a topic for yesterday meeting | 08:08 |
| rpittau | let's ask now! | 08:08 |
| rpittau | dtantsur, JayF, TheJulia, iurygregory: I'd like to stop cutting bugfix branches for ironic-inspector at this point, we don't use it anymore on metal3 side which was the main (probably only) consumer, wdyt? | 08:11 |
| opendevreview | Verification of a change to openstack/ironic master failed: Require hashed passwords for rescue by default https://review.opendev.org/c/openstack/ironic/+/923452 | 08:23 |
| opendevreview | cid proposed openstack/ironic master: Self-Service via Runbooks https://review.opendev.org/c/openstack/ironic/+/925170 | 08:27 |
| opendevreview | cid proposed openstack/python-ironicclient master: Add CLI support for runbooks https://review.opendev.org/c/openstack/python-ironicclient/+/925172 | 08:29 |
| opendevreview | cid proposed openstack/python-ironicclient master: Add CLI support for runbooks https://review.opendev.org/c/openstack/python-ironicclient/+/924101 | 08:31 |
| opendevreview | cid proposed openstack/ironic master: Self-Service via Runbooks https://review.opendev.org/c/openstack/ironic/+/922142 | 08:33 |
| * cid Finallllly! I was like, gerrit, thank you but I already have a changeid. | 08:33 | |
| opendevreview | Mohammed Boukhalfa proposed openstack/sushy-tools master: Extend the fake systems driver with external systems status notifier https://review.opendev.org/c/openstack/sushy-tools/+/923111 | 08:40 |
| opendevreview | Jacob Anders proposed openstack/sushy-tools master: Translate VM name to UUID to avoid redirect issue https://review.opendev.org/c/openstack/sushy-tools/+/924297 | 09:14 |
| opendevreview | Merged openstack/ironic master: Test devstack OSCaas against all Ironic jobs https://review.opendev.org/c/openstack/ironic/+/919620 | 09:17 |
| dtantsur | rpittau: I guess with it officially deprecated, it makes sense. Let's just make sure we announce it somehow (ML? release notes?) | 09:53 |
| jssfr | ironic-inspector is deprecated? | 09:56 |
| iurygregory | rpittau, ++ makes sense to me | 09:56 |
| dtantsur | jssfr: https://docs.openstack.org/releasenotes/ironic-inspector/unreleased.html#deprecation-notes | 09:57 |
| dtantsur | also keep in mind that the actual end of life will not happen any soon | 09:57 |
| dtantsur | so the messaging is "migrate but not in panic" | 09:57 |
| * jssfr goes to read, thanks | 10:02 | |
| opendevreview | Merged openstack/ironic master: Require hashed passwords for rescue by default https://review.opendev.org/c/openstack/ironic/+/923452 | 10:38 |
| iurygregory | rpittau, I think elod comment makes sense in small patch to sushy-tools https://review.opendev.org/c/openstack/sushy-tools/+/925069 | 11:24 |
| iurygregory | lol i pasted the wrong message :D | 12:06 |
| iurygregory | ctrl+c you lied to me :D | 12:07 |
| iurygregory | https://review.opendev.org/c/openstack/releases/+/925171 this was the patch | 12:07 |
| rpittau | dtantsur: I will announce it in the ML to see if we get a reply from some other consumer | 12:32 |
| rpittau | but considering that it's deprecated I doubt it | 12:32 |
| dtantsur | rpittau: we could even consider cycle-with-rc for inspector | 12:59 |
| rpittau | dtantsur: mmm yeah just wondering if it's worth changing it now, how far are we from closing the project? | 13:00 |
| dtantsur | rpittau: well... not soon for sure. Several release cycles. | 13:00 |
| rpittau | oh ok, I thought it would be sooner | 13:01 |
| rpittau | then we can take that into consideration | 13:01 |
| opendevreview | Riccardo Pittau proposed openstack/ironic master: Remove ironic-inspector from bugfix branches https://review.opendev.org/c/openstack/ironic/+/925200 | 13:05 |
| opendevreview | Jacob Anders proposed openstack/sushy-tools master: Translate VM name to UUID to avoid redirect issue https://review.opendev.org/c/openstack/sushy-tools/+/924297 | 13:09 |
| mohammed | Thanks for feedback on the patch https://review.opendev.org/c/openstack/sushy-tools/+/923111 and on the yesterday weekly I addressed the comment on the patch and again as I explained on the weekly this tiny minimal interface making fake system extensibale and push notifications to a configured endpoint. We’ve improved the documentation, please let me know if you think further refinements are needed. Otherwise, this | 13:12 |
| mohammed | interface in the fake system is ready to make the most of it in maximizing testing ironic at scale and minimizing | 13:12 |
| mohammed | cardoe: crowd strike concerns in production. | 13:12 |
| TheJulia | Good morning | 13:18 |
| jamesdenton_alt | good morning | 13:21 |
| opendevreview | OpenStack Release Bot proposed openstack/ironic-python-agent bugfix/9.13: Update .gitreview for bugfix/9.13 https://review.opendev.org/c/openstack/ironic-python-agent/+/925210 | 13:53 |
| opendevreview | Doug Goldstein proposed openstack/sushy master: avoid MAC addresses that are the empty string https://review.opendev.org/c/openstack/sushy/+/925130 | 14:05 |
| opendevreview | Merged openstack/ironic-python-agent bugfix/9.13: Update .gitreview for bugfix/9.13 https://review.opendev.org/c/openstack/ironic-python-agent/+/925210 | 14:22 |
| mohammed | Jayf: Good morning! The patch got +2 but it can't be merged without your approval ':D | 15:41 |
| JayF | I don't have time to give it a quality review, but if you link it I can make sure that my negative vote is removed | 15:44 |
| mohammed | I don’t see any negative reviews on the patch. Please double check here: https://review.opendev.org/c/openstack/sushy-tools/+/923111 | 15:48 |
| dtantsur | any core can add the 2nd +2 | 15:49 |
| rpittau | mohammed: approved :) | 16:06 |
| rpittau | good night! o/ | 16:06 |
| mohammed | rpittau: thanks switching to fake-ipa now | 16:07 |
| mohammed | o/ | 16:07 |
| JayF | okay cool, I didn't think I had put a - on that but if I had was gonna pull it | 16:28 |
| JayF | glad you got it reviewed | 16:28 |
| opendevreview | Merged openstack/sushy-tools master: Extend the fake systems driver with external systems status notifier https://review.opendev.org/c/openstack/sushy-tools/+/923111 | 16:36 |
| mohammed | Thanks Jay now I can head off for my vacation in peace xD | 16:53 |
| cardoe | How do people feel about "python_requires" constraints? | 17:57 |
| cardoe | I see for example sushy just has >=3.8 | 17:57 |
| cardoe | But sushy doesn't work on Python 3.12 because of it's implicit use of pkg_resources which comes from setuptools which won't be installed by default in 3.12 | 17:58 |
| cardoe | If you cap the upper limit then it would require a new release to support the newer version of Python e.g. python_requires >=3.8,<3.12 | 17:59 |
| clarkb | it still works if you install setuptools though | 17:59 |
| cardoe | For now | 17:59 |
| clarkb | I would not put that restriction in place | 17:59 |
| clarkb | you would make it more difficult for people to use the software when it may just work | 18:00 |
| cardoe | Yep. Agreed. | 18:00 |
| clarkb | you can also write a pyproject.toml that makes the setuptools dependency explicit | 18:00 |
| clarkb | and if the use of pkg_resources also happens at runtime you can list setuptools as a dependency | 18:01 |
| cardoe | Well I'm submitting a fix to sushy | 18:02 |
| cardoe | Python 3.9 and newer provides importlib.resources with the functionality. While Python 3.8 would need importlib_resources package. | 18:02 |
| cardoe | sushy's usage of pkg_resources is deprecated and slated for removal from setuptools. | 18:03 |
| clarkb | ah yup if the things pkg_resources are used for have been replaced by importlib then you can switch over to that too | 18:03 |
| clarkb | when it comes to python_requires I would limit that to actual incomaptibilities with the language and not just things like library shuffling | 18:04 |
| clarkb | for example "we use more modern python type specifications and therefore aren't compatible with python older than 3.6" | 18:04 |
| JayF | cardoe: if you find a fix that is happy with sushy, if you could pattern that fix across Ironic projects that'd be incredibly helpful | 18:29 |
| cid | o/ | 20:59 |
| opendevreview | Doug Goldstein proposed openstack/sushy master: drop usage of pkg_resource for newer python compat https://review.opendev.org/c/openstack/sushy/+/925304 | 21:44 |
| cardoe | How about ^, JayF? | 21:45 |
| JayF | cardoe: I am the messenger of doom, sorry | 21:49 |
| cardoe | So I doubt they'll accept my change | 21:53 |
| TheJulia | JayF: shouldn't that be DOOOOOooooommmm? | 21:53 |
| cardoe | There's already a change they're working on to drop Python 3.8 support since it goes EOL 2024-10-01, which is after the 2024.2 release but apparently a bunch of projects are dropping Python 3.8 support for the 2024.2 release so they don't have the weird conflict of supporting it at release and then not supporting it post release. | 21:54 |
| cardoe | Hence my question yesterday of can we bump up some things to require Python 3.9 and newer instead of 3.8? | 21:54 |
| cardoe | Cause if I can nudge sushy up to 3.9 and newer then I don't need an extra depend and this all works with stdlib. | 21:55 |
| JayF | I have not had the capacity to fully contextualize this situation (and likely won't today, either) | 21:55 |
| JayF | and mainly was trying to give you a quick review, which usually is a positive one | 21:55 |
| JayF | but :( | 21:55 |
| JayF | so I can't speak to the larger idea r/n | 21:55 |
| TheJulia | I, suspect, change wise it is just the need to take a shot at updating https://github.com/openstack/requirements/blob/master/global-requirements.txt | 21:56 |
| clarkb | couple of things. I think they should accept adding it to global requirements (importlib-metadata is already there). The dropping of 3.8 support needs to happen in a specific order with libraries going last. Sushy is a libary it should be at the tail end of the support drop. Also just because the runtimes list for 2024.2 requires 3.9 doesn't mean 3.8 is disallowed | 21:57 |
| clarkb | we're likely to see a long tail from projects like swift since they have historically been in that boat | 21:57 |
| TheJulia | ++ | 21:58 |
| TheJulia | clarkb: well put | 21:58 |
| clarkb | I see the change you are referring to and I'll make a note there too | 21:59 |
| cardoe | okay I'll submit it to them. | 21:59 |
| clarkb | the last time large scale uncoordinated python version drop was attempted it crashed and burned. THis whole process needs to be more of a coordinated dance rather than a hodge podge of cleanups | 22:00 |
| * TheJulia glares at dns | 22:07 | |
| TheJulia | https://www.reddit.com/r/networkingmemes/comments/hx9rnc/dns_haiku/ | 22:08 |
| opendevreview | Doug Goldstein proposed openstack/sushy master: drop usage of pkg_resource for newer python compat https://review.opendev.org/c/openstack/sushy/+/925304 | 23:33 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!