| *** ociuhandu has joined #openstack-infra | 00:10 | |
| *** ociuhandu has quit IRC | 00:12 | |
| *** ociuhandu has joined #openstack-infra | 00:14 | |
| *** ociuhandu has quit IRC | 00:19 | |
| *** tetsuro has joined #openstack-infra | 00:31 | |
| *** dave-mccowan has joined #openstack-infra | 00:37 | |
| *** yamamoto has joined #openstack-infra | 01:15 | |
| *** auristor has quit IRC | 01:40 | |
| *** Lucas_Gray has quit IRC | 01:40 | |
| *** dave-mccowan has quit IRC | 01:45 | |
| kevinz | ianw: new arm64 cluster is ready, I'd like to configure projects for openstack-infra | 01:50 |
|---|---|---|
| *** stephenfin has quit IRC | 01:50 | |
| *** dosaboy has quit IRC | 01:50 | |
| *** stephenfin has joined #openstack-infra | 01:50 | |
| kevinz | ianw: so you need os-control and os-jobs right? we just have less than 20 IPV4 public IP, others are all IPV6 ips | 01:50 |
| *** smcginnis|PTO has quit IRC | 01:51 | |
| *** exsdev0 has joined #openstack-infra | 01:51 | |
| *** exsdev has quit IRC | 01:51 | |
| *** exsdev0 is now known as exsdev | 01:51 | |
| *** smcginnis|PTO has joined #openstack-infra | 01:52 | |
| *** dosaboy has joined #openstack-infra | 01:59 | |
| *** yamamoto has quit IRC | 02:18 | |
| *** prometheanfire has quit IRC | 02:39 | |
| *** diablo_rojo has joined #openstack-infra | 02:43 | |
| *** prometheanfire has joined #openstack-infra | 03:03 | |
| *** yamamoto has joined #openstack-infra | 03:15 | |
| *** yamamoto has quit IRC | 03:59 | |
| *** ramishra has joined #openstack-infra | 03:59 | |
| *** yamamoto has joined #openstack-infra | 04:01 | |
| *** udesale has joined #openstack-infra | 04:18 | |
| *** raukadah is now known as chandankumar | 04:51 | |
| *** rcernin_ has joined #openstack-infra | 04:56 | |
| *** lxkong has joined #openstack-infra | 04:58 | |
| *** udesale_ has joined #openstack-infra | 04:58 | |
| *** ociuhandu has joined #openstack-infra | 04:59 | |
| *** rcernin has quit IRC | 04:59 | |
| *** udesale has quit IRC | 05:02 | |
| *** ociuhandu has quit IRC | 05:04 | |
| *** surpatil has joined #openstack-infra | 05:07 | |
| *** ramishra has quit IRC | 05:14 | |
| *** ramishra has joined #openstack-infra | 05:21 | |
| *** ramishra has quit IRC | 05:32 | |
| *** evrardjp has quit IRC | 05:34 | |
| *** evrardjp has joined #openstack-infra | 05:34 | |
| *** tkajinam has quit IRC | 05:52 | |
| *** tkajinam has joined #openstack-infra | 05:53 | |
| openstackgerrit | Simon Westphahl proposed zuul/nodepool master: Cleanup exception logging in static provider https://review.opendev.org/702828 | 06:14 |
| *** ramishra has joined #openstack-infra | 06:30 | |
| *** ykarel|away is now known as ykarel | 06:32 | |
| openstackgerrit | Rico Lin proposed openstack/project-config master: Add ircbot and repo for Multi-Arch SIG https://review.opendev.org/703323 | 06:32 |
| openstackgerrit | Rico Lin proposed opendev/system-config master: Add irc bots for Multi-Arch SIG https://review.opendev.org/703325 | 06:36 |
| *** ramishra has quit IRC | 06:49 | |
| *** lmiccini has joined #openstack-infra | 06:51 | |
| *** ramishra has joined #openstack-infra | 07:05 | |
| *** ramishra has quit IRC | 07:09 | |
| *** rcernin_ has quit IRC | 07:14 | |
| *** ramishra has joined #openstack-infra | 07:17 | |
| *** pgaxatte has joined #openstack-infra | 07:27 | |
| *** piotrowskim has joined #openstack-infra | 07:36 | |
| *** ramishra has quit IRC | 07:44 | |
| *** ramishra has joined #openstack-infra | 07:47 | |
| *** rpittau|afk is now known as rpittau | 07:51 | |
| *** ykarel is now known as ykarel|lunch | 07:51 | |
| *** diablo_rojo has quit IRC | 07:55 | |
| *** slaweq has joined #openstack-infra | 08:00 | |
| *** ociuhandu has joined #openstack-infra | 08:02 | |
| *** ociuhandu has quit IRC | 08:02 | |
| *** ociuhandu has joined #openstack-infra | 08:03 | |
| *** ociuhandu has quit IRC | 08:05 | |
| *** slaweq has quit IRC | 08:05 | |
| *** ociuhandu has joined #openstack-infra | 08:05 | |
| *** jtomasek has joined #openstack-infra | 08:07 | |
| *** dchen has quit IRC | 08:08 | |
| *** ociuhandu has quit IRC | 08:11 | |
| *** ociuhandu has joined #openstack-infra | 08:12 | |
| *** tesseract has joined #openstack-infra | 08:16 | |
| *** tosky has joined #openstack-infra | 08:17 | |
| *** ociuhandu has quit IRC | 08:18 | |
| *** ociuhandu has joined #openstack-infra | 08:19 | |
| *** ociuhandu has quit IRC | 08:22 | |
| *** ociuhandu has joined #openstack-infra | 08:23 | |
| *** slaweq has joined #openstack-infra | 08:24 | |
| *** ralonsoh has joined #openstack-infra | 08:26 | |
| *** ociuhandu has quit IRC | 08:27 | |
| *** tkajinam has quit IRC | 08:28 | |
| *** ociuhandu has joined #openstack-infra | 08:29 | |
| *** pkopec has joined #openstack-infra | 08:29 | |
| *** ociuhandu has quit IRC | 08:30 | |
| *** ociuhandu has joined #openstack-infra | 08:30 | |
| *** gfidente has joined #openstack-infra | 08:31 | |
| *** xek_ has joined #openstack-infra | 08:33 | |
| *** xek has quit IRC | 08:36 | |
| *** dpawlik has joined #openstack-infra | 08:37 | |
| *** iurygregory has quit IRC | 08:49 | |
| *** jpena|off is now known as jpena | 08:50 | |
| *** jaosorior has joined #openstack-infra | 08:51 | |
| *** ociuhandu has quit IRC | 08:51 | |
| *** ociuhandu has joined #openstack-infra | 08:52 | |
| *** ociuhandu has quit IRC | 08:53 | |
| *** ociuhandu has joined #openstack-infra | 08:54 | |
| *** lucasagomes has joined #openstack-infra | 08:54 | |
| *** ociuhandu has quit IRC | 08:55 | |
| *** ociuhandu has joined #openstack-infra | 08:58 | |
| *** ykarel|lunch is now known as ykarel | 09:00 | |
| *** apetrich has joined #openstack-infra | 09:01 | |
| openstackgerrit | Carlos Goncalves proposed openstack/diskimage-builder master: Add CentOS 8 support https://review.opendev.org/684308 | 09:01 |
| *** ociuhandu has quit IRC | 09:03 | |
| *** ricolin has quit IRC | 09:04 | |
| *** yamamoto has quit IRC | 09:06 | |
| *** yamamoto has joined #openstack-infra | 09:08 | |
| *** iurygregory has joined #openstack-infra | 09:09 | |
| *** ociuhandu has joined #openstack-infra | 09:10 | |
| *** ociuhandu has quit IRC | 09:14 | |
| *** ricolin has joined #openstack-infra | 09:15 | |
| *** arxcruz|off is now known as arxcruz | 09:22 | |
| *** jhesketh has quit IRC | 09:32 | |
| *** jhesketh has joined #openstack-infra | 09:34 | |
| *** david-lyle has joined #openstack-infra | 09:38 | |
| *** dklyle has quit IRC | 09:41 | |
| *** derekh has joined #openstack-infra | 09:47 | |
| *** yamamoto has quit IRC | 09:59 | |
| *** surpatil has quit IRC | 09:59 | |
| openstackgerrit | Carlos Goncalves proposed openstack/diskimage-builder master: Add CentOS 8 support https://review.opendev.org/684308 | 10:01 |
| *** xek_ has quit IRC | 10:01 | |
| *** yamamoto has joined #openstack-infra | 10:05 | |
| *** yamamoto has quit IRC | 10:05 | |
| *** AJaeger has quit IRC | 10:05 | |
| *** jaosorior has quit IRC | 10:17 | |
| openstackgerrit | Miklos Vajna proposed opendev/git-review master: Make it possible to specify who is notified https://review.opendev.org/702861 | 10:36 |
| *** gfidente has quit IRC | 10:38 | |
| openstackgerrit | Miklos Vajna proposed opendev/git-review master: Make it possible to specify who is notified https://review.opendev.org/702861 | 10:38 |
| openstackgerrit | Antoine Musso proposed zuul/zuul master: doc: add links to components documentation https://review.opendev.org/703105 | 10:40 |
| *** gfidente has joined #openstack-infra | 10:41 | |
| openstackgerrit | Antoine Musso proposed openstack/diskimage-builder master: dib-lint: test elements have README.rst file https://review.opendev.org/177832 | 10:48 |
| *** ociuhandu has joined #openstack-infra | 10:51 | |
| *** ricolin has quit IRC | 10:55 | |
| *** udesale_ has quit IRC | 11:02 | |
| openstackgerrit | Miklos Vajna proposed opendev/git-review master: Make it possible to specify who is notified https://review.opendev.org/702861 | 11:03 |
| *** ociuhandu has quit IRC | 11:08 | |
| *** ociuhandu has joined #openstack-infra | 11:09 | |
| *** dtantsur|afk is now known as dtantsur | 11:13 | |
| *** ociuhandu has quit IRC | 11:14 | |
| *** ociuhandu has joined #openstack-infra | 11:17 | |
| *** rpittau is now known as rpittau|bbl | 11:20 | |
| *** dpawlik has quit IRC | 11:22 | |
| openstackgerrit | Miklos Vajna proposed opendev/git-review master: Make it possible to specify who is notified https://review.opendev.org/702861 | 11:24 |
| *** ociuhandu has quit IRC | 11:24 | |
| *** dpawlik has joined #openstack-infra | 11:34 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul master: WIP: docs: improve job.role documentation https://review.opendev.org/703372 | 11:39 |
| *** ykarel is now known as ykarel|afk | 11:40 | |
| *** AJaeger has joined #openstack-infra | 11:56 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul master: docs: improve job.role documentation https://review.opendev.org/703372 | 12:00 |
| *** ociuhandu has joined #openstack-infra | 12:02 | |
| *** Lucas_Gray has joined #openstack-infra | 12:04 | |
| *** rosmaita has joined #openstack-infra | 12:05 | |
| *** roman_g has joined #openstack-infra | 12:06 | |
| *** ociuhandu has quit IRC | 12:08 | |
| AJaeger | config-core, please review https://review.opendev.org/702818 and https://review.opendev.org/703189 to retire Ubuntu Trusty testing. Once those are in, we need an infra-root to delete the AFS volume mirror.wheel.trustyx64 . | 12:17 |
| *** ykarel|afk is now known as ykarel | 12:24 | |
| *** mnaser has quit IRC | 12:27 | |
| *** adam_g has quit IRC | 12:27 | |
| *** mnaser has joined #openstack-infra | 12:27 | |
| *** adam_g has joined #openstack-infra | 12:28 | |
| *** rfolco has joined #openstack-infra | 12:29 | |
| *** jpena is now known as jpena|lunch | 12:35 | |
| *** yamamoto has joined #openstack-infra | 12:37 | |
| *** yamamoto has quit IRC | 12:37 | |
| *** lxkong has quit IRC | 12:38 | |
| *** xek has joined #openstack-infra | 12:41 | |
| openstackgerrit | Miklos Vajna proposed opendev/git-review master: Make it possible to specify who is notified https://review.opendev.org/702861 | 12:42 |
| *** yamamoto has joined #openstack-infra | 12:42 | |
| openstackgerrit | Merged opendev/base-jobs master: Remove ubuntu-trusty nodeset https://review.opendev.org/702818 | 12:46 |
| *** slaweq has quit IRC | 12:48 | |
| *** slaweq has joined #openstack-infra | 12:51 | |
| *** rfolco is now known as rfolco|bbl | 12:56 | |
| *** rh-jelabarre has joined #openstack-infra | 12:58 | |
| *** rlandy has joined #openstack-infra | 13:00 | |
| *** ykarel is now known as ykarel|afk | 13:01 | |
| *** electrofelix has joined #openstack-infra | 13:01 | |
| *** electrofelix has quit IRC | 13:01 | |
| *** Lucas_Gray has quit IRC | 13:02 | |
| *** jaosorior has joined #openstack-infra | 13:04 | |
| *** yamamoto has quit IRC | 13:04 | |
| *** Lucas_Gray has joined #openstack-infra | 13:06 | |
| *** smcginnis|PTO is now known as smcginnis | 13:09 | |
| *** yamamoto has joined #openstack-infra | 13:09 | |
| openstackgerrit | Rodolfo Alonso Hernandez proposed openstack/openstack-zuul-jobs master: Increase openstack-tox-py3{6,7} timeout to 3600 seconds https://review.opendev.org/703386 | 13:11 |
| *** Lucas_Gray has quit IRC | 13:11 | |
| *** rpittau|bbl is now known as rpittau | 13:11 | |
| *** yamamoto has quit IRC | 13:14 | |
| *** Lucas_Gray has joined #openstack-infra | 13:15 | |
| *** jaosorior has quit IRC | 13:26 | |
| *** jpena|lunch is now known as jpena | 13:30 | |
| *** dpawlik has quit IRC | 13:34 | |
| *** dpawlik has joined #openstack-infra | 13:35 | |
| *** kjackal has joined #openstack-infra | 13:36 | |
| *** ociuhandu has joined #openstack-infra | 13:39 | |
| *** yamamoto has joined #openstack-infra | 13:41 | |
| *** yamamoto has quit IRC | 13:42 | |
| *** yamamoto has joined #openstack-infra | 13:42 | |
| *** yamamoto has quit IRC | 13:43 | |
| *** yamamoto has joined #openstack-infra | 13:43 | |
| *** ociuhandu has quit IRC | 13:44 | |
| openstackgerrit | Merged openstack/project-config master: Bye, Bye, Trusty https://review.opendev.org/703189 | 13:50 |
| AJaeger | infra-root, please delete the AFS volume mirror.wheel.trustyx64 . We just removed ubuntu-trusty images from OpenDev ^ | 13:52 |
| AJaeger | #success Ubuntu Trusty images have been removed | 13:52 |
| openstackstatus | AJaeger: Added success to Success page (https://wiki.openstack.org/wiki/Successes) | 13:52 |
| AJaeger | #status log Ubuntu Trusty images have been removed | 13:53 |
| openstackstatus | AJaeger: finished logging | 13:53 |
| *** kjackal has quit IRC | 13:55 | |
| *** ykarel|afk is now known as ykarel | 13:56 | |
| *** ricolin has joined #openstack-infra | 13:57 | |
| *** jaosorior has joined #openstack-infra | 13:58 | |
| *** dciabrin_ is now known as dciabrin | 13:58 | |
| *** ricolin has quit IRC | 14:07 | |
| *** Lucas_Gray has quit IRC | 14:09 | |
| *** ricolin has joined #openstack-infra | 14:11 | |
| *** ociuhandu has joined #openstack-infra | 14:11 | |
| *** rfolco|bbl is now known as rfolco | 14:16 | |
| *** ociuhandu has quit IRC | 14:18 | |
| *** ricolin_ has joined #openstack-infra | 14:24 | |
| *** ricolin_ has quit IRC | 14:26 | |
| *** ricolin has quit IRC | 14:26 | |
| *** ricolin has joined #openstack-infra | 14:26 | |
| *** ab-a has joined #openstack-infra | 14:27 | |
| openstackgerrit | Merged openstack/reviewstats master: Update ironic subprojects https://review.opendev.org/700890 | 14:29 |
| openstackgerrit | Rico Lin proposed openstack/project-config master: Add ircbot and repo for Multi-Arch SIG https://review.opendev.org/703323 | 14:29 |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul master: bindep: fixed wrong dep names on rpm platform https://review.opendev.org/703403 | 14:30 |
| ricolin | AJaeger, https://review.opendev.org/703323 updated | 14:31 |
| ricolin | do you have more background information about governance/reference/programs.yaml | 14:32 |
| ricolin | I can't find it in current governance repo now | 14:33 |
| *** zxiiro has joined #openstack-infra | 14:36 | |
| *** udesale has joined #openstack-infra | 14:42 | |
| openstackgerrit | Simon Westphahl proposed zuul/nodepool master: Handle event id in node requests https://review.opendev.org/703406 | 14:46 |
| openstackgerrit | Simon Westphahl proposed zuul/nodepool master: Centralize logging adapters https://review.opendev.org/703407 | 14:46 |
| AJaeger | ricolin: let me update that part... | 14:53 |
| ricolin | AJaeger, thx | 14:54 |
| openstackgerrit | Simon Westphahl proposed zuul/nodepool master: Make flake8 config compatible with latest version https://review.opendev.org/703410 | 14:58 |
| *** jaosorior has quit IRC | 15:05 | |
| *** ykarel is now known as ykarel|away | 15:05 | |
| frickler | dansmith: mugsie: (cc infra-root) do you still have interest in using #openstack-merges ? the project list seems pretty outdated, also it looks broken since openstackgerrit doesn't have a voice there | 15:06 |
| *** yamamoto has quit IRC | 15:07 | |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul master: Unlock urllib pinning https://review.opendev.org/703414 | 15:07 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Update list for #openstack-merges https://review.opendev.org/703416 | 15:08 |
| AJaeger | frickler: this updates the list ^ | 15:08 |
| frickler | AJaeger: I'm thinking whether we can instead discard that channel. with at most two subscribers, that setup doesn't seem too effective to me | 15:10 |
| frickler | if we decide to continue it, updates should be automated IMO | 15:10 |
| openstackgerrit | Sorin Sbarnea proposed zuul/zuul master: Unpin urllib https://review.opendev.org/703414 | 15:11 |
| *** artom has joined #openstack-infra | 15:12 | |
| openstackgerrit | Sorin Sbarnea proposed opendev/gear master: tox: enable extra python versions https://review.opendev.org/703418 | 15:14 |
| *** chandankumar is now known as raukadah | 15:15 | |
| AJaeger | frickler: good point, happy to abandon ;) | 15:17 |
| AJaeger | frickler: let me propose an alternative as well | 15:18 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Disable #openstack-merges https://review.opendev.org/703421 | 15:20 |
| AJaeger | frickler, mugsie, dansmith ^ | 15:20 |
| *** eharney has joined #openstack-infra | 15:21 | |
| openstackgerrit | Sorin Sbarnea proposed opendev/gear master: packaging: updated project urls https://review.opendev.org/703422 | 15:21 |
| frickler | AJaeger: Dan and Graham were the only subscribers before, waiting for their feedback. the channel setup does actually seem to work, though, seems I was misinterpreting the flags | 15:21 |
| AJaeger | frickler: I added both to the review, let's wait for them... | 15:22 |
| *** kjackal has joined #openstack-infra | 15:22 | |
| ricolin | AJaeger, fungi it will be nice if we can get this patch (https://review.opendev.org/#/c/703323) and our storyboard created for SIG soon, so I can start add tasks in storyboard :) | 15:23 |
| AJaeger | ricolin: please add a "needed-By"to that one | 15:23 |
| AJaeger | I really like to see the link to the government change in the review as stated in the infra-manual | 15:23 |
| openstackgerrit | Rico Lin proposed openstack/project-config master: Add ircbot and repo for Multi-Arch SIG https://review.opendev.org/703323 | 15:24 |
| ricolin | AJaeger, thx for the reminding and~ done!^^^ | 15:25 |
| *** ociuhandu has joined #openstack-infra | 15:27 | |
| openstackgerrit | Sorin Sbarnea proposed opendev/gear master: WIP: test extra jobs https://review.opendev.org/703425 | 15:29 |
| *** ociuhandu has quit IRC | 15:32 | |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Update list for #openstack-merges https://review.opendev.org/703416 | 15:35 |
| *** bauzas has quit IRC | 15:43 | |
| *** jaosorior has joined #openstack-infra | 15:43 | |
| *** yamamoto has joined #openstack-infra | 15:45 | |
| clarkb | frickler: if no one is using it then turning it off is probably best | 15:51 |
| AJaeger | clarkb: only mugsie and dansmith - and mugsie gave a +1 already on https://review.opendev.org/703421 | 15:52 |
| AJaeger | infra-root, please delete the AFS volume mirror.wheel.trustyx64 . We removed ubuntu-trusty images from OpenDev | 15:53 |
| *** yamamoto has quit IRC | 15:55 | |
| AJaeger | config-core, two small reviews for openstack-zuul-jobs to update testing, please: https://review.opendev.org/703297 and https://review.opendev.org/703298 | 15:58 |
| openstackgerrit | Andreas Jaeger proposed opendev/system-config master: Remove #openstack-merges https://review.opendev.org/703434 | 15:58 |
| *** diablo_rojo has joined #openstack-infra | 15:58 | |
| openstackgerrit | Merged openstack/project-config master: Add ircbot and repo for Multi-Arch SIG https://review.opendev.org/703323 | 16:01 |
| *** ociuhandu has joined #openstack-infra | 16:03 | |
| openstackgerrit | Merged openstack/openstack-zuul-jobs master: Update docs building https://review.opendev.org/703297 | 16:08 |
| *** udesale has quit IRC | 16:09 | |
| openstackgerrit | Merged openstack/openstack-zuul-jobs master: Increase hacking version https://review.opendev.org/703298 | 16:11 |
| *** openstackgerrit has quit IRC | 16:13 | |
| *** lpetrut has quit IRC | 16:20 | |
| *** KeithMnemonic has joined #openstack-infra | 16:21 | |
| *** ociuhandu has quit IRC | 16:21 | |
| *** ociuhandu has joined #openstack-infra | 16:22 | |
| *** piotrowskim has quit IRC | 16:22 | |
| *** pgaxatte has quit IRC | 16:22 | |
| *** tosky has quit IRC | 16:23 | |
| *** slaweq has quit IRC | 16:24 | |
| *** iurygregory has quit IRC | 16:29 | |
| jrosser | what was the outcome of the required python version metadata stuff for the infra wheel mirrors? | 16:31 |
| fungi | jrosser: step 1 was we blacklisted setuptools, pip and virtualenv from our wheel cache | 16:32 |
| fungi | (and deleted setuptools 45.0.0 from it) | 16:32 |
| clarkb | step 2 was discover all the way it is difficult to write the metadata on those packages ourselves | 16:32 |
| fungi | yeah. i've worked out how to mostly make wheel builds reproducible, and also how to filter out wheels we didn't build ourselves by parsing pip's log | 16:33 |
| jrosser | ah ok, becasue i was just looking into OSA tempest fails on stable/stein because it pulls in a py3 neutron-lib out of the wheel mirror | 16:33 |
| jrosser | but an idential job on debian stretch works becasue it hits upstream pypi and i assume the correct metadata is present there | 16:34 |
| clarkb | we are not writing that metadata | 16:34 |
| fungi | i think with those we have the ability to perform index generation locally on the build node, and only write a relatively small amount of data into afs (which is a bottleneck) | 16:34 |
| clarkb | jrosser: openstack produced wheels should specify if they are python2, python3 or both though | 16:34 |
| clarkb | if neutron-lib doesn't do that properly we likely have a bug in package config or possibly pbr | 16:34 |
| *** lmiccini has quit IRC | 16:35 | |
| jrosser | oh hmm http://mirror.ord.rax.opendev.org/wheel/ubuntu-18.04-x86_64/neutron-lib/ | 16:36 |
| jrosser | neutron_lib-2.0.0-py2.py3-none-any.whl <- suspicious | 16:36 |
| clarkb | yes that is a bug if it can't run under python2 | 16:36 |
| jrosser | when i look here its >=3.6 https://pypi.org/project/neutron-lib/2.0.0/ | 16:37 |
| clarkb | metadata is probably getting set properly then. Possible this is a pbr wheel generation bug | 16:38 |
| jrosser | neutron-lib appears to do something sensible https://github.com/openstack/neutron-lib/blob/master/setup.cfg#L9 | 16:41 |
| jrosser | where do i look next, not super familiar with how this works | 16:41 |
| clarkb | jrosser: in openstack/pbr I think | 16:43 |
| clarkb | since that is what works with setuptools to generate the wheel | 16:43 |
| *** rpittau is now known as rpittau|afk | 16:44 | |
| *** bdodd_ has quit IRC | 16:45 | |
| *** bauzas has joined #openstack-infra | 16:50 | |
| *** bdodd has joined #openstack-infra | 16:52 | |
| *** jaosorior has quit IRC | 16:55 | |
| *** ricolin has quit IRC | 16:55 | |
| jrosser | i am wondering if this is a similar situation to setuptools - becasue pip(2) takes neutron-lib 2.0.0 as a valid installation candidate from the wheel mirror due to lack of metadata but does not deem it a candidate when using upstream pypi | 16:56 |
| clarkb | jrosser: it is, except since we control our own wheel generation we should be able to properly create a py3 only wheel | 16:57 |
| clarkb | which would avoid this problem | 16:57 |
| jrosser | the METADATA inside the .py2.py3 file looks OK, "Requires-Python: >=3.6" and is likely what causes the install to fail ultimately | 16:57 |
| clarkb | also we tightly control all our deps using requirements and constraints | 16:57 |
| clarkb | we shoudl exclude versions that are not valid there | 16:57 |
| jrosser | well, "tempest" :( | 16:57 |
| fungi | yeah, gmann is working on this problem | 16:57 |
| clarkb | not sure why tempest matters. We should have requirements lines with env marker | 16:58 |
| prometheanfire | it seems like the requirements bot is failing | 16:58 |
| clarkb | neutron-lib<2.0.0 ; python_version<3.6 | 16:58 |
| fungi | basically tempest is branchless but we install it in a virtualenv on xenial where python3 is 3.5, but it has dependencies which claim to only work on 3.6 and later | 16:58 |
| clarkb | neutron-lib>=2.0.0 ; python_version>=3.6 | 16:58 |
| clarkb | those two requirements lines should fix it | 16:59 |
| fungi | so, yeah, adding something like that in requirements could solve it | 16:59 |
| jrosser | and OSA rocky and stein are unchangeably py2 only | 16:59 |
| gmann | yeah, we can cap in requirement also but that will be needed for all such dependency. | 16:59 |
| fungi | jrosser: that part is addressed by installing tempest under python3 | 16:59 |
| clarkb | gmann: we already do it for all dependencies | 16:59 |
| gmann | i am trying to install py3.6 interpreter for tempest venv | 16:59 |
| clarkb | we've just not had problems with most of them yet | 16:59 |
| fungi | (it has its own virtualenv separate from the system context) | 17:00 |
| *** lucasagomes has quit IRC | 17:00 | |
| *** tesseract has quit IRC | 17:00 | |
| *** openstackgerrit has joined #openstack-infra | 17:03 | |
| openstackgerrit | James E. Blair proposed zuul/zuul-website master: Update redirects https://review.opendev.org/703454 | 17:03 |
| clarkb | jrosser: fungi: pip wheel seems to respect requires python by default https://pip.pypa.io/en/stable/reference/pip_wheel/#cmdoption-ignore-requires-python we use python setup.py bdist_wheel though | 17:04 |
| clarkb | setuptools docs imply that it respects requires-python too | 17:05 |
| clarkb | https://packaging.python.org/tutorials/packaging-projects/ note that project has requiers python set to >=3.6 and the bdist_wheel result is python3 only | 17:06 |
| fungi | we do use `pip wheel` to build platform-specific wheels for our wheel cache, but yeah pbr uses the old way | 17:07 |
| clarkb | if I build the wheel locally I get neutron_lib-2.0.0-py3-none-any.whl | 17:07 |
| clarkb | using python setup.py bdist_wheel | 17:07 |
| clarkb | that implies to me that pbr is working as is setuptools so the issue is at a higher level | 17:08 |
| clarkb | (possibly in the wheel mirror generation script) | 17:08 |
| fungi | https://pypi.org/project/neutron-lib/#files | 17:08 |
| fungi | the job building the wheel for release is making it universal | 17:09 |
| clarkb | ah so that may be a problem there too | 17:09 |
| fungi | probably worth digging up the build log from the release pipeline | 17:09 |
| openstackgerrit | James E. Blair proposed zuul/zuul-website master: Update redirects https://review.opendev.org/703456 | 17:09 |
| fungi | pypi does pep 530 metadata in its simple index, which is why it's only a problem for pip fetching from our wheel cache | 17:09 |
| clarkb | d9a0ce5d7273ed9414d3d6a2127ac4b6a983523e removed the unviersal flag from setup.cfg and that change is included in 2.0.0 | 17:11 |
| openstackgerrit | James E. Blair proposed zuul/zuul-website master: Update redirects https://review.opendev.org/703456 | 17:13 |
| openstackgerrit | James E. Blair proposed zuul/zuul-website master: Remove some redirects https://review.opendev.org/703457 | 17:13 |
| *** tobiash has quit IRC | 17:14 | |
| *** kjackal has quit IRC | 17:14 | |
| clarkb | http://zuul.openstack.org/build/9d6ee88b51514ca4a96d8278ed434327/console#2/0/0/ubuntu-bionic our release jobs set --universal | 17:16 |
| clarkb | https://opendev.org/openstack/project-config/src/branch/master/zuul.d/jobs.yaml#L97-L114 is the issue | 17:16 |
| clarkb | for our uploads to pypi at least | 17:16 |
| gmann | clarkb: prometheanfire not sure if u-c of ussuri should be removing the py3.4, py3.5 versions cap also along with py2.7 as they are not supported version for ussuri releases. only cap the compatibility of py3.6, 3.7 and any other supported py version. | 17:17 |
| clarkb | the wheel builder just grabs from pypi if the wheel exists there I think | 17:17 |
| clarkb | gmann: I think you have to wait for everything to have fully transitioned to make that change since global requirements are global | 17:18 |
| clarkb | gmann: so maybe at the end of the cycle do the older python removal | 17:18 |
| clarkb | but until then people may be transitioning? | 17:18 |
| gmann | yeah, at the end we should. that is why i was concern on doing neutron-lib cap for < or > = py3.6 | 17:19 |
| smcginnis | We need to keep swift's py2 needs too. | 17:20 |
| gmann | and instead fix the Tempest installation way to either create py3.6 venv or pin Tempest py2 compatible version and so does u-c too. | 17:20 |
| clarkb | gmann: env markers allow us to have different caps for different python versions | 17:20 |
| fungi | clarkb: if memory serves, the reason we added --universal (years ago) was that projects were not getting around to adding it consistently to their setup.cfg files. now that's coming back to bite us i guess | 17:21 |
| *** kjackal has joined #openstack-infra | 17:21 | |
| clarkb | gmann: neutron-lib<2.0.0 ; python_version<3.6 and neutron-lib>2.0.0 ; python_version>=3.6 | 17:21 |
| clarkb | fungi: ya its going to make this a giant pain to work around I expect | 17:21 |
| gmann | clarkb: yeah but my question is do we really need to do that when ussuri release do not support <py3.6 | 17:22 |
| fungi | well, we need to drop the --universal from the job asap | 17:22 |
| gmann | clarkb: on ussuri u-c. | 17:22 |
| clarkb | gmann: yes, because usurri is still in a state of transition | 17:22 |
| clarkb | fungi: ya I've taken the wheel conversation to the releases channel | 17:22 |
| gmann | clarkb: ohk, for transition it make sense. | 17:23 |
| *** tobiash has joined #openstack-infra | 17:23 | |
| AJaeger | speaking about wheels: infra-root, please delete the AFS volume mirror.wheel.trustyx64 . We removed ubuntu-trusty images from OpenDev | 17:24 |
| *** ociuhandu_ has joined #openstack-infra | 17:26 | |
| corvus | AJaeger, infra-root: i'll delete that volume | 17:26 |
| fungi | thanks corvus! | 17:27 |
| clarkb | tyty | 17:27 |
| AJaeger | thanks! | 17:27 |
| jrosser | gmann: from an openstack-ansible perspective it is not possible to deploy tempest in py3 for either rocky or stein | 17:28 |
| fungi | why? | 17:29 |
| fungi | does openstack-ansible not deploy tempest in a virtualenv like devstack does? | 17:29 |
| *** ociuhandu has quit IRC | 17:29 | |
| fungi | tempest will inevitably have dependency version conflicts with the installed openstack services so they need separate environments regardless | 17:30 |
| *** ociuhandu_ has quit IRC | 17:30 | |
| gmann | yeah, otherwise you would not be able use tempest master with stable u-c | 17:30 |
| AJaeger | config-core, here's an update for gerritbot to use two new flags, please review: https://review.opendev.org/698091 | 17:33 |
| *** dtantsur is now known as dtantsur|afk | 17:33 | |
| jrosser | fungi: yes it does, but the entire setup is py2 only until train which is our py2/p3 transition release | 17:33 |
| *** evrardjp has quit IRC | 17:34 | |
| jrosser | there was massive work needed to bring py3 in and is inconcievable to backport | 17:34 |
| openstackgerrit | Sean McGinnis proposed openstack/project-config master: Stop forcing universal wheel releases https://review.opendev.org/703464 | 17:34 |
| *** evrardjp has joined #openstack-infra | 17:34 | |
| clarkb | jrosser: I think you'd need to install stable tempest then? | 17:35 |
| clarkb | on those older stable branches | 17:35 |
| jrosser | yes, i'm just looking at that now | 17:35 |
| fungi | there is no stable tempest | 17:36 |
| fungi | you'd have to install tempest from an old tag | 17:36 |
| gmann | jrosser: you can use the Tempest tag in that case like we did for ocata and pike- https://review.opendev.org/#/c/681950/ | 17:36 |
| clarkb | fungi: right that | 17:37 |
| gmann | we have Tempest tag for each release wise compatibility if master cannot be used for any rason | 17:37 |
| fungi | jrosser: by "work needed to bring py3 in" you mean installing the interpreter? because that should in theory be all tempest needs for a py3 virtualenv | 17:37 |
| gmann | and for all Tempest plugins also | 17:37 |
| jrosser | we have an ansible role that build python venv <- patches | 17:38 |
| jrosser | and our own wheel building <- patches | 17:38 |
| jrosser | and centos7 <- difficult | 17:38 |
| clarkb | centos7 actually has python3 in the main distro no | 17:38 |
| clarkb | *now | 17:39 |
| jrosser | and ubuntu/debian/suse/centos to work universally | 17:39 |
| fungi | ahh, yeah, doing it on centos7 was tough until the latest centos7 release | 17:39 |
| jrosser | whilst deploy fro either source code or distro packages | 17:39 |
| jrosser | "switch to py3" is a big deal for deployment tooling | 17:39 |
| jrosser | and upgrades must still work and so on | 17:39 |
| clarkb | ya I think using old tempest is probably simplest | 17:39 |
| fungi | agreed | 17:40 |
| clarkb | jrosser: https://review.opendev.org/703464 is the fix for the wheel problem fwiw | 17:40 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: docs: remove generated toc from the main index https://review.opendev.org/703468 | 17:40 |
| clarkb | we'll stop generating python2 compat wheels for python3 only projects | 17:40 |
| clarkb | (we'll only make python3 wheels actually, but thats ok python2 can simply use the sdists) | 17:41 |
| jrosser | clarkb: is there a behaviour where something uses the py2.py3 from the filename rather than the metadata in the wheel? | 17:41 |
| clarkb | jrosser: yes pip checks both | 17:41 |
| jrosser | ah right | 17:41 |
| clarkb | I believe the filename was the original method for addressing this problem | 17:41 |
| jrosser | it's interesting that my debian job which doesnt use the infra wheel works fine | 17:42 |
| clarkb | then they realized python3.4 and 3.7 aren't necessarily compat | 17:42 |
| clarkb | so added the more expressive thing? | 17:42 |
| fungi | pip won't install a py3 wheel on python2.x nor a py2 wheel on python3.x, but will install a py2.py3 wheel on both | 17:42 |
| clarkb | jrosser: your debian job is using the metadata from pypi | 17:42 |
| jrosser | yes, that was my conclusion | 17:42 |
| fungi | jrosser: that's because it's hitting pypi's simple index, which additionally has pep 530 metadata about supported python versions | 17:42 |
| corvus | #status log deleted afs volume mirror.wheel.trustyx64 | 17:42 |
| openstackstatus | corvus: finished logging | 17:42 |
| clarkb | fungi: any idea if setuptools itself fixed their wheel name? | 17:43 |
| clarkb | or if we filed a bug on that? | 17:43 |
| fungi | they said "open a new issue if you want universal=1 removed" | 17:43 |
| openstackgerrit | James E. Blair proposed opendev/system-config master: Add docs for deleting an AFS volume https://review.opendev.org/703470 | 17:43 |
| fungi | clarkb: and i don't believe anyone has | 17:43 |
| fungi | https://github.com/pypa/setuptools/issues/1964 | 17:44 |
| clarkb | https://github.com/pypa/setuptools/pull/1978 | 17:44 |
| clarkb | cool I guess setuptools 46 will also be fixed | 17:44 |
| fungi | ahh, nice | 17:44 |
| fungi | anyway, i think if we started stripping wheels we've downloaded from pypi our of our wheel cache, this would also be a non-issue (though we would have taken a lot longer to notice we're incorrectly setting --universal all the time) | 17:45 |
| *** Goneri has quit IRC | 17:47 | |
| *** gouthamr_ has joined #openstack-infra | 17:47 | |
| *** jpena is now known as jpena|off | 17:48 | |
| openstackgerrit | James E. Blair proposed zuul/zuul master: Docs: change "config" title https://review.opendev.org/703471 | 17:51 |
| *** dpawlik has quit IRC | 17:52 | |
| *** bnemec has joined #openstack-infra | 17:52 | |
| openstackgerrit | Merged openstack/project-config master: Stop forcing universal wheel releases https://review.opendev.org/703464 | 17:57 |
| openstackgerrit | Tristan Cacqueray proposed zuul/zuul master: docs: remove generated toc from the main index https://review.opendev.org/703468 | 17:58 |
| *** derekh has quit IRC | 18:01 | |
| *** jamesmcarthur has joined #openstack-infra | 18:03 | |
| *** gfidente is now known as gfidente|afk | 18:08 | |
| *** bnemec has quit IRC | 18:08 | |
| *** gouthamr_ has quit IRC | 18:10 | |
| *** sdoran has quit IRC | 18:11 | |
| *** nickv1985 has quit IRC | 18:11 | |
| *** Shrews has quit IRC | 18:11 | |
| *** sdoran has joined #openstack-infra | 18:11 | |
| *** Shrews has joined #openstack-infra | 18:11 | |
| *** nickv1985 has joined #openstack-infra | 18:12 | |
| clarkb | seems like the openstack TC would like a formalized opendev governance document. Does opendev/system-config/doc seem like a good place for that? | 18:22 |
| clarkb | also linaro has their new arm64 cloud up and running. If someone else would like to help bootstrap that let me know. I'm going to try and get the connection details up on bridge in the near future | 18:23 |
| * clarkb is juggling a bunch of administrivia today as well as family going "isn't it a holiday?" :) | 18:23 | |
| fungi | yeah, i'm still trying to catch up on e-mail. today has been crazier than anticipated | 18:34 |
| *** jamesmcarthur has quit IRC | 18:36 | |
| corvus | clarkb: re opendev governance, where was that feedback? | 18:40 |
| corvus | (i didn't see a reply to your friday email) | 18:40 |
| fungi | corvus: it was in review comments on on https://review.opendev.org/703134 | 18:40 |
| corvus | ah thx | 18:41 |
| corvus | clarkb: yeah, that seems like a good place. https://docs.openstack.org/infra/system-config/project.html is essentially our current governance. | 18:43 |
| fungi | makes sense to me too | 18:43 |
| clarkb | alright I'll get the distilled thread version pushed up there | 18:44 |
| clarkb | then maybe we don't approve it until TC governance is no longer a conflict | 18:44 |
| *** tosky has joined #openstack-infra | 18:50 | |
| *** kjackal has quit IRC | 18:54 | |
| *** hamzy__ has joined #openstack-infra | 18:55 | |
| *** david-lyle has quit IRC | 18:56 | |
| *** lpetrut has joined #openstack-infra | 18:56 | |
| *** eharney has quit IRC | 18:56 | |
| *** hamzy_ has quit IRC | 18:57 | |
| *** ralonsoh has quit IRC | 19:00 | |
| *** eharney has joined #openstack-infra | 19:09 | |
| *** addyess has quit IRC | 19:09 | |
| *** addyess has joined #openstack-infra | 19:11 | |
| fungi | clarkb: slight improvement on the wheel filtering plan... i can parse the filenames out of the per-branch logs we're already generating... looks like for publish-wheel-mirror-ubuntu-bionic as an example we're stashing 1669 unique wheels which were downloaded directly from pypi | 19:20 |
| fungi | no need for a separate appending log since we retain them all individually | 19:21 |
| *** igordc has joined #openstack-infra | 19:23 | |
| *** stevebaker has joined #openstack-infra | 19:26 | |
| *** adam_g has quit IRC | 19:28 | |
| openstackgerrit | Jeremy Stanley proposed openstack/project-config master: Don't cache downloaded wheels in our wheels mirror https://review.opendev.org/703487 | 19:33 |
| fungi | clarkb: jrosser: smcginnis: ^ | 19:33 |
| *** goldyfruit has joined #openstack-infra | 19:35 | |
| fungi | you can also play the home game, by downloading a build-logs.tar.gz from a recent run like http://zuul.opendev.org/t/openstack/build/f2e33b555c704628ad9a434dc216d66e/logs | 19:36 |
| openstackgerrit | Clark Boylan proposed opendev/system-config master: Update project doc to reflect OpenDev changes https://review.opendev.org/703488 | 19:38 |
| clarkb | fungi: corvus ^ and all other infra-root config-core and infra folks I think that is the next step ish in the oepndev split proposal | 19:39 |
| clarkb | I don't think we want to merge that until the split happens so I will go ahead and WIP it now | 19:39 |
| *** goldyfruit has quit IRC | 19:39 | |
| clarkb | fungi: looking | 19:39 |
| *** goldyfruit has joined #openstack-infra | 19:40 | |
| clarkb | fungi: one note on the change. Also I realized if we stop building universal wheels those jobs will end up building python2 wheels | 19:47 |
| clarkb | so we won't delete everything as expected. But I think that is fine as we'll get wheels for python2 of versions that work with python2 aiui | 19:47 |
| openstackgerrit | Merged zuul/zuul-helm master: Added support for configuring disk_limit_per_job https://review.opendev.org/702963 | 19:48 |
| openstackgerrit | Merged zuul/zuul-helm master: Add extra files for secret https://review.opendev.org/702965 | 19:48 |
| fungi | well, we won't build wheels if they can be downloaded from pypi, and we'll still avoid saving those into the cache at the end | 19:49 |
| clarkb | fungi: right but there won't by python2 wheels on pypi anymore but we'll build python2 wheels in our jobs | 19:49 |
| clarkb | in the subset of cases where requires-python isn't set | 19:49 |
| clarkb | I don't expect it to be a problem | 19:50 |
| fungi | but yeah, if we stop publishing wheels which support python2 and something claims to need those versions, we'll build and cache python2-supporing versions of them... but that should hopefully be a dwindling set | 19:50 |
| fungi | and projects whose releases properly support python2 (it will mostly be stable branch point releases) can set universal=1 in setup.cfg and things happen as expected then | 19:51 |
| clarkb | yup setting universal=1 is definitely the best fix here | 19:51 |
| fungi | but i don't think 703487 changes any of that anyway | 19:51 |
| fungi | it would happen regardless | 19:51 |
| clarkb | ya mostly just pointing out there maybe straggler wheels we don't expect to stick around as a result of this | 19:52 |
| clarkb | more than lxml, libvirt-python, numpy, cryptography, etc | 19:52 |
| AJaeger | fungi, want to update your change (or followup) with the sanity check that clarkb proposes? Or is that not needed? | 19:54 |
| clarkb | fwiw my concern there is we'll stop deleting things if the pip log format changes and we won't notice | 19:54 |
| clarkb | I figure higher chance of noticing if the job starts failing | 19:54 |
| fungi | oh, sure | 19:56 |
| *** hashar has joined #openstack-infra | 19:58 | |
| AJaeger | clarkb: you still left the OpenStack Infra project team in governance - do we need a governing document for them? Your change updates the governing document for these... | 20:01 |
| clarkb | AJaeger: ya, its a bit awkward because we put that content in opendev/system-config so I think the content there should be opendev specific. But it is likely that openstack infra team may want a separate doc (maybe in openstack/project-config?) to document anything different there | 20:02 |
| clarkb | AJaeger: fwiw I don't think a lot of the infra governance/teams stuff applies much to what will remain in openstack. refstack, health, dib don't really necessarily need a cohesive service perspective | 20:03 |
| clarkb | thats a long way to say I think the openstack infra team that remains can get away with lightweight default openstack governance? | 20:03 |
| clarkb | more similar to other openstack projects that way | 20:03 |
| clarkb | AJaeger: does that make sense? | 20:04 |
| AJaeger | clarkb: yes, makes sense - and I agree, something lightweight might be enough... That's why I formulated it as question ;) | 20:04 |
| clarkb | I think my preference would be to start with default openstack as the assumption and if any of that changes due to need we can add a new doc, likely in project-config | 20:05 |
| AJaeger | works for me | 20:05 |
| AJaeger | clarkb: you might want to add that to the commit message | 20:06 |
| AJaeger | (something along the lines above) | 20:06 |
| clarkb | can do | 20:06 |
| clarkb | I'll incorporate that with edits from yall's feedback | 20:07 |
| *** apetrich has quit IRC | 20:07 | |
| openstackgerrit | Jeremy Stanley proposed openstack/project-config master: Don't cache downloaded wheels in our wheels mirror https://review.opendev.org/703487 | 20:07 |
| fungi | clarkb: AJaeger: done ^ | 20:07 |
| fungi | i had originally not done that so that the job won't fail due to unexpected parsing errors which didn't show up locally for me (since we can't easily test this without merging) but i suppose the worst that happens is our wheel mirrors stop updating until we fix it | 20:08 |
| clarkb | fungi: ya and I think that will help us notice vs the other situation where things mostly work until 9 months later when a random package python version interaction causes it to fail | 20:10 |
| *** exsdev has quit IRC | 20:10 | |
| *** icey has quit IRC | 20:10 | |
| fungi | i'll also go ahead and push up my reproducibility fixes too before i forget what they were | 20:11 |
| *** Lucas_Gray has joined #openstack-infra | 20:11 | |
| *** slaweq has joined #openstack-infra | 20:12 | |
| *** dosaboy has quit IRC | 20:12 | |
| *** rfarr has joined #openstack-infra | 20:14 | |
| *** Lucas_Gray has quit IRC | 20:14 | |
| AJaeger | fungi: +2A | 20:14 |
| *** Lucas_Gray has joined #openstack-infra | 20:16 | |
| *** lxkong has joined #openstack-infra | 20:21 | |
| openstackgerrit | Jeremy Stanley proposed openstack/project-config master: Build reproducible wheels for our wheel cache https://review.opendev.org/703495 | 20:21 |
| fungi | there's the hacks to make many of our wheels consistent between builds | 20:21 |
| *** kjackal has joined #openstack-infra | 20:22 | |
| *** adam_g has joined #openstack-infra | 20:23 | |
| fungi | the other thing we can consider doing is removing wheels from the cache which aren't in the list of wheels we built, so that these caches don't grow unbounded | 20:25 |
| AJaeger | fungi, CFLAGS=-g0 means also no optimization, what about saying "CFLAGS=-g0 -O2"? | 20:25 |
| openstackgerrit | Sorin Sbarnea proposed opendev/gear master: Add extra python jobs https://review.opendev.org/703425 | 20:26 |
| fungi | AJaeger: yeah, i'm seeing some examples of pairing it with -O3 as well | 20:28 |
| openstackgerrit | Merged openstack/project-config master: Don't cache downloaded wheels in our wheels mirror https://review.opendev.org/703487 | 20:28 |
| AJaeger | AFAIR -O2 has only safe optimizatoins, so would use that one for everything | 20:30 |
| *** kjackal has quit IRC | 20:30 | |
| openstackgerrit | Sorin Sbarnea proposed opendev/gear master: Add extra python jobs https://review.opendev.org/703425 | 20:34 |
| fungi | AJaeger: where are you seeing -g0 negate optimizations? the gcc docs don't seem to mention that anywhere i can find | 20:36 |
| openstackgerrit | Clark Boylan proposed opendev/system-config master: Update project doc to reflect OpenDev changes https://review.opendev.org/703488 | 20:36 |
| clarkb | AJaeger: fungi ^ thank you for the review.s I think I got it all there | 20:36 |
| fungi | AJaeger: neither https://gcc.gnu.org/onlinedocs/gcc/Debugging-Options.html#Debugging-Options nor https://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html#Optimize-Options seem to mention any interaction between -g and -O | 20:38 |
| fungi | (it's not that i don't believe you, i just want to know precisely what the interaction is between them) | 20:39 |
| *** goldyfruit has quit IRC | 20:40 | |
| *** goldyfruit has joined #openstack-infra | 20:40 | |
| AJaeger | fungi: what I mean is: You have default CFLAGS for a project like "-O2 -g". If you set CFLAGS=-g0, then you *remove* the -O2 and the code is not optimized. | 20:40 |
| fungi | oh! | 20:40 |
| AJaeger | There's no interaction between those two - the default is -O0 (last time I looked). | 20:41 |
| fungi | i get you, i should instead export CFLAGS="-g0 $CFLAGS" | 20:41 |
| AJaeger | I guess that won't work, since CFLAGS are overwritten. | 20:41 |
| AJaeger | But that needs experimentation... | 20:42 |
| AJaeger | And I would use "$CFLAGS -g0", the latter overwrite... | 20:42 |
| fungi | do we assume this script might run somewhere that CFLAGS is already set in the calling environment? | 20:42 |
| openstackgerrit | Sorin Sbarnea proposed opendev/gear master: Add extra python jobs https://review.opendev.org/703425 | 20:43 |
| AJaeger | fungi: My understanding is that the script *calls* pip wheel building, and thus you pass in from outside the CFLAGS. | 20:44 |
| AJaeger | so, no way to override - you replace. But please check it on an example build | 20:45 |
| fungi | yeah, when pip wheel ends up running gcc to compile extensions, the CFLAGS exported by the script get used (unless the package overwrites CFLAGS itself, but generally that's where they should be prepending or appending the existing CFLAGS var instead of overwriting) | 20:45 |
| AJaeger | do you have some packages that compile? | 20:46 |
| fungi | basically if a package doesn't obey CFLAGS from the calling environment, that would be a bug worth filing against the package's source repositories | 20:46 |
| AJaeger | yep | 20:46 |
| fungi | i was using pyyaml as a test case since it was mentioned in https://github.com/pypa/pip/issues/6505 | 20:48 |
| *** eharney has quit IRC | 20:48 | |
| AJaeger | interesting, default compile vs. CFLAGS=-g0: http://paste.openstack.org/show/788616/ | 20:55 |
| AJaeger | that seems to duplicate CFLAGS if you don't pass them in... | 20:55 |
| *** exsdev has joined #openstack-infra | 20:56 | |
| AJaeger | Sorry, no idea how wheel building handles flags - in this case it looks ok with external CFLAGS, but less so without | 20:56 |
| *** rfarr has quit IRC | 20:58 | |
| fungi | yeah, i suspect there's some error in their makefile where cflags is being set conditionally | 20:59 |
| *** dosaboy has joined #openstack-infra | 21:01 | |
| fungi | i'm not finding it though | 21:02 |
| fungi | might be something in how setuptools works | 21:03 |
| fungi | it does do some handling of cflags, looks like: https://github.com/pypa/setuptools/blob/master/setuptools/command/build_clib.py#L19-L20 | 21:04 |
| *** Wryhder has joined #openstack-infra | 21:11 | |
| *** Lucas_Gray has quit IRC | 21:11 | |
| *** Wryhder is now known as Lucas_Gray | 21:12 | |
| *** lpetrut has quit IRC | 21:12 | |
| *** slaweq_ has joined #openstack-infra | 21:14 | |
| *** gfidente|afk has quit IRC | 21:15 | |
| *** slaweq has quit IRC | 21:15 | |
| *** slaweq_ is now known as slaweq | 21:15 | |
| ianw | hello | 21:33 |
| *** Wryhder has joined #openstack-infra | 21:38 | |
| *** Lucas_Gray has quit IRC | 21:39 | |
| *** Wryhder is now known as Lucas_Gray | 21:39 | |
| ianw | sorry, catching up. | 21:41 |
| ianw | clarkb: i can look at new arm64 cloud if you like | 21:41 |
| ianw | ahh, i see i got the details via mail, will look at | 21:42 |
| ianw | fungi: looks like you've made good progress on modernising our wheel infra a bit ... still think we want a small spec? i have that on my todo | 21:43 |
| openstackgerrit | Antoine Musso proposed opendev/gear master: inline helps for interactive commands https://review.opendev.org/66591 | 21:45 |
| fungi | ianw: mainly hitting incremental improvements which are independent of the indexing problem but can also make indexing easier | 21:45 |
| ianw | AJaeger: thanks for pushing on the trusty removal! i think we can purge all that from dib now too | 21:45 |
| fungi | spec could probably be limited to the indexing implementation? | 21:45 |
| ianw | fungi: ok, yep i think still useful so we can get feedback from everyone in a more formal manner | 21:46 |
| *** rcernin has joined #openstack-infra | 21:49 | |
| ianw | p.s. just like to note that as i prediced @ http://eavesdrop.openstack.org/irclogs/%23openstack-infra/%23openstack-infra.2020-01-15.log.html#t2020-01-15T23:48:47 the cable provider was unable to figure out how to provide a cable | 21:50 |
| *** eharney has joined #openstack-infra | 21:50 | |
| fungi | well, providing cables is far from what i would expect from a cable provider | 21:52 |
| *** xek has quit IRC | 22:03 | |
| hashar | fungi: ""we're out of zuul stickers"" in french that is: "nous n'avons plus d'autocollants Zuul" ;] | 22:06 |
| *** jamesmcarthur has joined #openstack-infra | 22:16 | |
| *** Lucas_Gray has quit IRC | 22:20 | |
| *** Lucas_Gray has joined #openstack-infra | 22:22 | |
| fungi | hashar: perfect! i'm all set for fosdem now | 22:25 |
| hashar | fungi: ah fosdem, that is a good even though overwheelming. If you get 10k stickers, you would be set | 22:30 |
| hashar | is there a Zuul booth there? | 22:31 |
| fungi | openstack foundation will have a community table for all osf projects, including zuul | 22:31 |
| fungi | it's on the second level, so may have less traffic, but it's both days of the conference | 22:31 |
| fungi | if anybody's planning to be there and wants to help, see https://etherpad.openstack.org/p/fosdem-2020 where we're starting to coordinate | 22:32 |
| *** pkopec has quit IRC | 22:33 | |
| hashar | I was supposedly in the us at that time but change of plan .. | 22:34 |
| hashar | it is a bit too late to find a hotel / travel there though ;D | 22:35 |
| hashar | it is definitely a great conference anyway | 22:35 |
| hashar | have fun there :] | 22:35 |
| fungi | there's always next year! | 22:35 |
| fungi | thanks, i hope to! | 22:35 |
| hashar | yeah 2021 surely | 22:35 |
| *** slaweq has quit IRC | 22:36 | |
| *** adam_g has quit IRC | 22:40 | |
| *** adam_g has joined #openstack-infra | 22:43 | |
| *** apetrich has joined #openstack-infra | 22:43 | |
| openstackgerrit | Sorin Sbarnea proposed opendev/gear master: packaging: updated project urls https://review.opendev.org/703422 | 22:44 |
| openstackgerrit | Sorin Sbarnea proposed opendev/gear master: Add extra python jobs https://review.opendev.org/703425 | 22:46 |
| openstackgerrit | Sorin Sbarnea proposed opendev/gear master: packaging: updated project urls https://review.opendev.org/703422 | 22:47 |
| *** tkajinam has joined #openstack-infra | 22:51 | |
| *** jtomasek has quit IRC | 22:53 | |
| *** hashar has quit IRC | 23:14 | |
| clarkb | ianw: cool thanks | 23:15 |
| * clarkb is back from a long bike ride and realizes that he probably won't catch up on all the backlog items | 23:40 | |
| clarkb | feel free to ping me if I can help with anything urgent | 23:40 |
| * clarkb gets infra meeting agenda out | 23:40 | |
| *** dchen has joined #openstack-infra | 23:48 | |
| *** tosky has quit IRC | 23:51 | |
| openstackgerrit | Clark Boylan proposed opendev/system-config master: Update project doc to reflect OpenDev changes https://review.opendev.org/703488 | 23:58 |
| clarkb | ianw: ^ thank you for the feedback. Does that read better now? | 23:58 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!