Monday, 2020-07-27

« Sunday, 2020-07-26 Index Tuesday, 2020-07-28 »
*** tetsuro has joined #openstack-release00:02
*** e0ne has joined #openstack-release01:53
*** e0ne has quit IRC01:58
*** e0ne has joined #openstack-release02:04
*** e0ne has quit IRC02:08
*** e0ne has joined #openstack-release02:23
*** e0ne has quit IRC02:28
*** e0ne has joined #openstack-release02:33
*** armax has joined #openstack-release02:35
*** armax has quit IRC02:37
*** e0ne has quit IRC02:38
*** e0ne has joined #openstack-release02:43
*** e0ne has quit IRC02:48
*** e0ne has joined #openstack-release02:53
*** e0ne has quit IRC02:58
*** ricolin has joined #openstack-release03:41
*** udesale has joined #openstack-release04:07
*** ykarel has joined #openstack-release04:26
*** dave-mccowan has quit IRC04:37
*** e0ne has joined #openstack-release05:13
*** e0ne has quit IRC05:17
*** e0ne has joined #openstack-release05:23
*** e0ne has quit IRC05:27
*** e0ne has joined #openstack-release05:43
*** e0ne has quit IRC05:47
*** e0ne has joined #openstack-release05:53
openstackgerritMerged openstack/releases master: tripleo victoria-1 release  https://review.opendev.org/74249205:55
openstackgerritMerged openstack/releases master: Re-release recent xstatic packages  https://review.opendev.org/74284905:55
*** e0ne has quit IRC05:58
*** udesale has quit IRC06:08
*** udesale has joined #openstack-release06:10
openstackgerritHervé Beraud proposed openstack/releases master: Release oslo for victoria - 2020-07-27  https://review.opendev.org/74315106:11
*** e0ne has joined #openstack-release06:41
*** tosky has joined #openstack-release06:42
*** amoralej|off is now known as amoralej07:08
*** jbadiapa has joined #openstack-release07:37
*** ricolin has quit IRC07:43
*** tosky has quit IRC08:03
*** dtantsur|afk is now known as dtantsur08:05
*** vishalmanchanda has joined #openstack-release08:29
*** e0ne has quit IRC08:34
*** rpittau has joined #openstack-release08:34
*** tetsuro has quit IRC09:08
*** tetsuro has joined #openstack-release09:09
*** e0ne has joined #openstack-release09:10
*** tetsuro has quit IRC09:16
openstackgerritHervé Beraud proposed openstack/releases master: Add some whitespaces to helps to fix rendering.  https://review.opendev.org/74317209:21
openstackgerritHervé Beraud proposed openstack/releases master: Add some whitespaces on helps to fix rendering.  https://review.opendev.org/74317209:23
*** tosky has joined #openstack-release09:33
*** e0ne has quit IRC10:09
*** tkajinam has quit IRC10:10
*** e0ne has joined #openstack-release10:10
openstackgerritHervé Beraud proposed openstack/releases master: Fix a bug on list_unreleased_changes during tag retrieving  https://review.opendev.org/74318510:24
*** e0ne has quit IRC10:26
*** iurygregory has quit IRC10:59
*** iurygregory has joined #openstack-release11:01
*** e0ne has joined #openstack-release11:48
*** amoralej is now known as amoralej|lunch11:57
*** udesale_ has joined #openstack-release12:00
*** udesale has quit IRC12:03
*** evrardjp has quit IRC12:32
*** evrardjp has joined #openstack-release12:32
*** evrardjp has quit IRC12:32
*** evrardjp has joined #openstack-release12:33
*** KeithMnemonic has joined #openstack-release12:46
smcginnishberaud: I saw some failures. What's the situation with those?12:56
hberaudsmcginnis: you speak about https://review.opendev.org/#/c/743185/ ?12:57
*** amoralej|lunch is now known as amoralej12:59
smcginnishberaud: I don't know. I just saw several failure emails.13:01
smcginnishberaud: Not that patch. That one isn't merged yet. I'm referring to the release failures.13:03
hberaudsmcginnis: ah, no ideas, do you've an example?13:04
smcginnisExample?13:05
hberaudfailure emails13:05
hberaudof13:05
hberaudI just seen this one => Release of openstack/oslo.messaging for ref refs/tags/12.2.2 failed13:05
smcginnisYeah, there were several of those this morning.13:06
hberaudlemme check13:06
smcginnisI'll be back a little later. I can't dig in right now.13:06
hberaudI need to dash early today, but I'll try to take a look13:07
*** dtantsur is now known as dtantsur|brb13:22
*** mnasiadka has joined #openstack-release13:50
smcginnishberaud: I have a little time now. Did you find out anything?14:22
*** ricolin has joined #openstack-release14:31
hberaudsmcginnis: nope, error looks like on the project-config side and related to AFS14:36
smcginnishberaud: Yeah, looks like a couple of them were AFS related.14:37
smcginnisStarted collecting failures in the tracking etherpad.14:37
hberaudsmcginnis: but AFAIK I don't seen recent changes on that14:37
hberaud(related to AFS)14:37
smcginnisfungi: We've had a few AFS failures that will likely need to be reenqueed. Going through them in the etherpad.14:37
smcginnisA couple xstatic package failures due to the README not being formatted properly.14:38
smcginnisWe must be missing something in our twine check to catch those.14:38
fungiyeah, twine check should normally spot that14:38
fungii'll take a look at the afs failures14:39
smcginnisThose ones will need a fix in the repo and a new release to correct.14:39
smcginnisSeveral were release note updates. Not sure if we need to reenqueue those, but probably wouldn't hurt.14:39
smcginnisBut they will just get updated on the next release, so maybe we can wait. I was thinking they were stable branches that might not have new releases any time soon, but actually looks like it's the tripleo release that probably will.14:40
hberaudIIRC some of our issues were with tarball and I seen this twine bug related to tarball => https://github.com/pypa/twine/issues/67714:41
smcginnishberaud: What was the issue with the tarballs?14:41
smcginnisThe twine ones seem related to the README being README.txt instead of README.rst.14:42
smcginnisAnd maybe some minor formatting issues.14:42
hberaudoslo.messaging's tarball not published => http://lists.openstack.org/pipermail/openstack-discuss/2020-July/016074.html14:42
smcginnisYeah, that bug looks applicable.14:43
smcginnisTwine really should catch that, IMO.14:43
hberaudbut seems related to the AFS stuff14:43
smcginnisI don't see how it's related to the AFS stuff. Can you explain that?14:44
smcginnisfungi: I've placed (reenqueue) next to the failures that I think we need to have rerun.14:45
smcginnisfungi: Or probably more likely, that we will need to have the whole download from pypi, check valid package, upload to tarballs.o.o dance.14:46
hberaudSorry, ignore my previous comment I mixed topics14:46
smcginnisAh, OK. :)14:46
hberaudSo I can see that we use twine 3.2.0 like in the previous twine bug ^^^14:47
hberaudmaybe a track to follow14:47
fungiall four of those similar afs-related errors look like they happened on ze11, so it does seem to be an executor-specific problem14:48
smcginnisIt seems twine can't catch things when setup.py does "long_description = open('README.txt').read() "14:48
smcginnisExecute the executor? :)14:48
fungismcginnis: twine can check the metadata in the generated package14:48
smcginnisOnly other issue I see with those README.txt files is the title characters are not long enough.14:49
smcginnisI didn't think that would be a significant enough error for pypi to complain.14:49
fungifor my personal projects i do this in a tox env: `rm -fr dist;python setup.py bdist_wheel sdist;twine check dist/*` (with usedevelop = False and whitelist_externals = rm and deps = pbr twine wheel)14:50
fungihttp://paste.openstack.org/show/796332/14:51
smcginnisThe only difference I see between the failed ones and the xstatic packages that were fine is the short title marker.14:51
smcginnisfungi: Thanks, good to check that.14:51
smcginnisI *think* that's basically what our validation jobs do, but I guess we'll have to double check that.14:52
fungiyep, just suggesting as a possible way to locally sanity-check14:52
smcginnisHmm, some of these maybe have other issues: envlist = py27,py3414:52
smcginnisfungi: I ran that tox -e dist on xstatic-moment-timezone, and twine indeed passed just fine.14:53
hberaudwe also published oslo.messaging ~3 weeks ago, maybe we could compare related jobs to seen difference with the one in error => https://github.com/openstack/releases/commit/e4ed69da018b3f1d9c707694953ba32984ad856214:54
smcginnisThey set long_description_content_type="text/markdown" even though README.txt is in RST.14:55
fungioh, ew14:56
fungiyeah, i'd suggest "long_description = file: README.rst" and "long_description_content_type = text/x-rst; charset=UTF-8"14:57
* hberaud eject, I need to go early today14:59
smcginniso/14:59
*** beekneemech is now known as bnemec15:00
smcginnisThe content_type = markdown is what was throwing it off. So PyPi must do a different kind of check of the long description than what twine does. Twine thought it was fine (because *technically* I guess it is just fine markdown) but PyPi still treats it as RST.15:01
*** dtantsur|brb is now known as dtantsur15:06
fungiinteresting, and yeah the checks warehouse (pypi) performs aren't libified such that tools like twine could repeat them reliably, last i checked15:10
fungianother easy thing it misses is that warehouse will reject packages with unknown trove classifiers in their metadata15:11
fungifor that you need to extract the list of valid trove classifiers from the warehouse source code or query its api to get the current list15:11
smcginnisSounds like we are probably doing the best we can then.15:18
smcginnisAt least the checks we do have did catch a lot of actual issues when we first added them. So they are still valuable.15:18
*** armax has joined #openstack-release15:26
*** ykarel is now known as ykarel|away15:33
*** e0ne has quit IRC15:43
*** ykarel|away has quit IRC16:15
*** ricolin has quit IRC16:16
*** udesale_ has quit IRC16:21
*** amoralej is now known as amoralej|off16:25
*** dtantsur is now known as dtantsur|afk16:59
*** nicolasbock has joined #openstack-release16:59
*** jbadiapa has quit IRC17:12
*** ykarel has joined #openstack-release17:53
*** ykarel has quit IRC18:01
fungismcginnis: good news! we tracked down what was causing the afs failures. boot-time race between docker-compose bind-mounting things into the executor container and afsd creating /afs18:09
fungithe executors where those errors were occurring were ones where their containers won the race against afsd the last time they rebooted18:10
fungithey've been manually repaired while we work out a permanent boot ordering solution for it18:11
*** vishalmanchanda has quit IRC18:18
*** evrardjp has quit IRC18:21
*** evrardjp has joined #openstack-release18:24
*** tosky has quit IRC18:33
smcginnisfungi: That's great. Those are tricky.19:01
smcginnisI know container start up ordering torpedoed at least one keynote. ;)19:02
fungiheh19:03
fungianyway, now that's sorted, i'll get to work fixing the failed tarball uploads (hopefully for the last time)19:03
smcginnisfungi: Thanks. Just let me know if there isn't enough info captured in the etherpad.19:07
fungiall the info i need is in the build results anyway19:08
fungibut thanks19:08
smcginnis++19:08
fungi#status log copied tripleo-image-elements 12.1.0 and os-apply-config 11.3.0 wheels/sdists from pypi to the tarballs site after verifying with release key signatures (also uploaded) from build failure logs19:20
openstackstatusfungi: finished logging19:20
fungismcginnis: ^19:20
fungii'll go ahead and reenqueue the release notes jobs next19:21
fungisudo zuul enqueue-ref --tenant=openstack --trigger=gerrit --pipeline=tag --project=openstack/tripleo-image-elements --ref=refs/tags/12.1.0 --newrev=2348cec68d63be837d0c2217febbf8d80b5f59e819:22
smcginnisAwesome, thanks. That looks correct to me.19:23
fungisudo zuul enqueue-ref --tenant=openstack --trigger=gerrit --pipeline=tag --project=openstack/python-tripleoclient --ref=refs/tags/13.5.0 --newrev=18fa8f6e9643159c815aac5e812e698a60427f8819:24
smcginnis+119:24
fungi#status log reenqueued refs for tripleo-image-elements 12.1.0 and python-tripleoclient 13.5.0 into the tag pipeline to rerun failed release notes jobs19:25
openstackstatusfungi: finished logging19:25
fungidid that catch everything that needed fixing up? (assuming those succeed this time)19:25
smcginnisfungi: There's os-net-config too. Waiting to make sure the other two are good first?19:25
smcginnishttps://zuul.opendev.org/t/openstack/build/9751cb52205c4d2dbd8e2bb0359b110d19:25
fungioh, i must have missed that in the etherpad, thanks19:26
smcginnisThe other two left are the xstatic ones, and those will require changes merged to the repos and a new release done before that's all cleared up.19:26
fungiahh, i think it was still being filled in when i grabbed urls earlier19:26
fungior i may have just skimmed past it19:27
smcginnisEasy to do.19:27
fungisudo zuul enqueue-ref --tenant=openstack --trigger=gerrit --pipeline=tag --project=openstack/os-net-config --ref=refs/tags/12.4.0 --newrev=8b54c2f47525c9d58995dbcf3408cc14cc6ef3ee19:28
smcginnis+119:28
fungi#status log reenqueued ref for os-net-config 12.4.0 into the tag pipeline to rerun failed release notes job19:28
openstackstatusfungi: finished logging19:28
smcginnisfungi: Looks like we have one left from last week too. Though they've released a newer version since then it appears.19:29
smcginnisnova19:29
smcginnisNot nova. That's a failing clipboard.19:30
smcginnishttps://zuul.opendev.org/t/openstack/build/9751cb52205c4d2dbd8e2bb0359b110d19:30
smcginnisThat one.19:30
smcginnispython-tripleoclient 13.4.019:30
fungiyour clipboard really wants to paste that build url. it's the one you pasted for os-net-config too, but it's really xstatic-moment-timezone19:35
smcginnisI don't know what the deal is with my clipboard on this machine. Pretty much stock Fedora 32.19:36
smcginnishttps://zuul.opendev.org/t/openstack/build/74ecc0b3a4d3492ea4d20fb6b826039019:36
smcginnisThat's the one.19:36
smcginnisLatest stable/ussuri release of python-tripleoclient.19:37
fungiahh, cool, i thought i had gotten all the ones last week but i'll take care of that in a sec\19:37
smcginnisI did my best to hide it in the list too. ;)19:37
fungiheh19:38
fungi#status log copied python-tripleoclient 13.4.0 wheel/sdist from pypi to the tarballs site after verifying with release key signatures (also uploaded) from build failure log19:43
openstackstatusfungi: finished logging19:43
fungismcginnis: hopefully that's the last of it19:43
* smcginnis knocks on wood veneer19:44
openstackgerritMerged openstack/releases master: Release networking-ovn for stable/stein  https://review.opendev.org/74175619:51
openstackgerritMerged openstack/releases master: Add some whitespaces on helps to fix rendering.  https://review.opendev.org/74317219:57
openstackgerritMerged openstack/releases master: Release python-troveclient 5.0.0 for Victoria  https://review.opendev.org/74309719:57
openstackgerritMerged openstack/releases master: Release oslo for victoria - 2020-07-27  https://review.opendev.org/74315119:57
openstackgerritHervé Beraud proposed openstack/releases master: Fix a bug on list_unreleased_changes during tag retrieving  https://review.opendev.org/74318520:04
openstackgerritMerged openstack/releases master: Release tripleo-operator-ansible  https://review.opendev.org/74289120:07
*** e0ne has joined #openstack-release20:31
*** slaweq has quit IRC20:36
openstackgerritSean McGinnis proposed openstack/releases master: Victoria-2 Release for automaton  https://review.opendev.org/74340320:46
openstackgerritSean McGinnis proposed openstack/releases master: Victoria-2 Release for cliff  https://review.opendev.org/74340420:46
openstackgerritSean McGinnis proposed openstack/releases master: Victoria-2 Release for debtcollector  https://review.opendev.org/74340520:47
openstackgerritSean McGinnis proposed openstack/releases master: Victoria-2 Release for keystonemiddleware  https://review.opendev.org/74340620:47
*** slaweq has joined #openstack-release20:47
openstackgerritSean McGinnis proposed openstack/releases master: Victoria-2 Release for metalsmith  https://review.opendev.org/74340720:48
openstackgerritSean McGinnis proposed openstack/releases master: Victoria-2 Release for os-brick  https://review.opendev.org/74340820:49
openstackgerritMerged openstack/releases master: Fix a bug on list_unreleased_changes during tag retrieving  https://review.opendev.org/74318520:51
*** slaweq has quit IRC20:52
openstackgerritSean McGinnis proposed openstack/releases master: Victoria-2 Release for python-designateclient  https://review.opendev.org/74340920:54
openstackgerritSean McGinnis proposed openstack/releases master: Victoria-2 Release for python-karborclient  https://review.opendev.org/74341020:55
openstackgerritSean McGinnis proposed openstack/releases master: Victoria-2 Release for python-magnumclient  https://review.opendev.org/74341120:55
*** tosky has joined #openstack-release21:25
openstackgerritSean McGinnis proposed openstack/releases master: Switch to PyYaml safe_load  https://review.opendev.org/74342021:33
openstackgerritSean McGinnis proposed openstack/releases master: Update to latest hacking for pep8 checks  https://review.opendev.org/74342121:33
*** e0ne has quit IRC21:50
openstackgerritGoutham Pacha Ravi proposed openstack/releases master: [manila] release 10.0.1 from stable/ussuri  https://review.opendev.org/74342822:13
*** tkajinam has joined #openstack-release22:52
*** tosky has quit IRC23:25
« Sunday, 2020-07-26 Index Tuesday, 2020-07-28 »

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!