Friday, 2020-02-21

*** tosky has quit IRC00:09
*** mlavalle has quit IRC00:26
*** armstrong has quit IRC00:42
*** tetsuro has joined #openstack-release01:07
*** portdirect has quit IRC01:12
*** mordred has quit IRC01:14
*** hberaud has quit IRC01:14
*** mordred has joined #openstack-release01:15
*** portdirect has joined #openstack-release01:15
*** hberaud has joined #openstack-release01:16
*** ekcs has quit IRC01:47
*** tetsuro has quit IRC01:56
*** tetsuro has joined #openstack-release01:56
*** tetsuro has quit IRC03:13
*** tetsuro has joined #openstack-release03:22
*** tetsuro_ has joined #openstack-release03:28
*** tetsuro has quit IRC03:30
*** tetsuro_ has quit IRC03:33
*** tetsuro has joined #openstack-release03:34
*** tetsuro_ has joined #openstack-release03:47
*** tetsuro has quit IRC03:48
*** ykarel|away is now known as ykarel03:51
*** raukadah is now known as chandankumar04:11
*** AJaeger has joined #openstack-release05:02
AJaegerrelease team, please merge to update jobs05:02
*** evrardjp has quit IRC05:34
*** evrardjp has joined #openstack-release05:34
*** ekcs has joined #openstack-release05:38
*** ekcs has quit IRC05:43
*** ricolin has quit IRC06:15
*** ricolin has joined #openstack-release07:21
*** lpetrut has joined #openstack-release07:39
*** rpittau|afk is now known as rpittau07:50
*** slaweq has joined #openstack-release08:00
rakhmerovhi, can we backport patches with changes in setup.cfg?08:05
rakhmerove.g. changes in stevedore entry points08:05
rakhmerovttx: hi, may be you can advise08:05
AJaegerrakhmerov: could those break existing code?08:13
rakhmerovwell, no08:13
AJaegerrakhmerov: I mean: Adding lines is fine, changing sounds like breaking the API08:13
rakhmerovthat's what I'm trying to understand08:13
rakhmerovlet me show you...08:13
rakhmerovI understand that it's not desirable to backport such patches, but I'm in the middle of making changes that I want to backport on top of this patch08:14
rakhmerovit will be easier for me to finish them if this patch can also be backported08:15
rakhmerovI'm changing code, yes, and setup.cfg so that a project needs to be reinstalled anyway08:15
rakhmerovbut I don't break API, main config file, DB schema etc.08:15
AJaegerrakhmerov: those lines look fine in that file - but to me it looks like it shows you break the API elsewhere.08:17
rakhmerovAPI? Why?08:17
AJaegerrakhmerov: I hear you - let's see what others say, might want to talk with stable team as well08:17
rakhmerovno, those python changes are not part of the API08:17
rakhmerovAJaeger: ok08:17
AJaegerrakhmerov: you might be right - I don't know the code good enough. It just smells like API change ;)08:18
rakhmerovno-no, I reassure you :)08:18
rakhmerovAJaeger: also wrote this question to #openstack-stable08:25
*** tetsuro_ has quit IRC08:26
*** tosky has joined #openstack-release08:27
*** ykarel is now known as ykarel|lunch08:47
*** tetsuro has joined #openstack-release08:51
*** ykarel|lunch is now known as ykarel08:55
*** e0ne has joined #openstack-release08:59
*** e0ne has quit IRC08:59
*** e0ne has joined #openstack-release09:00
*** tetsuro has quit IRC09:03
ttxAJaeger: done10:05
AJaegerthanks, ttx!10:06
ttxrakhmerov: It's more of a stable question, but we have limited activity in #openstack-stable those days10:08
ttxrakhmerov: it's definitely a grey area, I think it would be worth raising it on the ML to get wider feedback10:08
ttxMy personal 2c would be: we avoid refactoring in stable patches, especially if that refactoring ends up affecting public function names, and even more if those are exposed by setup.cfg. But I guess it's a trade-off with how important it is to backport the patch that lives on top of that refactoring... and how painful it would be to adapt that patch so that it would live on top of non-refactored10:11
rakhmerovttx: ok, got it10:12
rakhmerovI'll try to avoid backporting it I guess10:12
ttxI can think of a few examples where a security patch lived on top of lightly-refactored code that was affectign public, but rarely-used functions... and that was deemed a good trade-off10:13
ttxBut here it sounds like a bit too far on the "user-visible change to facilitate unrelated backporting" side10:14
openstackgerritMerged openstack/releases master: Update docs publishing job
*** jbadiapa has joined #openstack-release10:31
openstackgerritMerged openstack/releases master: Release Adjutant 0.5.0
*** haleyb|away has quit IRC11:15
*** ykarel is now known as ykarel|afk11:16
*** N3l1x has joined #openstack-release11:30
*** ykarel|afk is now known as ykarel11:42
*** rpittau is now known as rpittau|bbl11:43
evrardjprakhmerov: (and ttx) my opinion is that it's indeed in grey area -- though here the functions exposed aren't changed themselves, only their implementation in setup.cfg. So for me, this is only a refactor, not an interface change. An interface change in stable is a no-go by default (basically unless no choice). The refactor in a stable branch is, for me, up to the project, but there is probably not many reasons to want11:49
evrardjp this, except to backport other bugfixes later in an easier fashion. In any case, that should be documented on _why_ it is done .11:49
*** dtantsur|afk is now known as dtantsur12:06
*** ricolin has quit IRC12:14
*** adriant has quit IRC12:53
*** adriant has joined #openstack-release12:54
*** rpittau|bbl is now known as rpittau13:00
*** ykarel is now known as ykarel|afk13:22
*** rpittau is now known as rpittau|afk13:57
*** mlavalle has joined #openstack-release14:00
*** mgoddard has joined #openstack-release14:08
mgoddardHello, I'd like to release some kayobe branches that existed prior to becoming an official deliverable.14:08
mgoddardIs there someone that can help with that?14:08
mgoddardI'd like 6.1.0 on stable/stein, and 5.1.0 on stable/rocky14:09
openstackgerritMerged openstack/releases master: Release os-brick 3.0.0
smcginnismgoddard: Pre-official deliverables would have had to have been done directly by the team.14:33
smcginnismgoddard: Not sure about the ACLs, but you would just need to push a tag on those branches.14:33
mgoddardhmm, I assumed it wouldn't work, but let me try14:34
smcginnisttx: Do we have any similar situations in the past for that? ^14:35
mgoddard ! [remote rejected] 6.1.0 -> 6.1.0 (prohibited by Gerrit)14:35
*** ykarel|afk is now known as ykarel15:34
ttxNo, branches need to be created manually15:52
ttxhmm, sorry let me have a look15:52
ttxAh yeah, the issue is that you are limited by the ACL. We can probably push those mnanually for you15:53
ttxIIRC members of the "Release Managers" group should be able to push the tag, doublechecknig15:54
ttxor infra15:55
ttxsmcginnis: so we should push the signed tag, as the ACL is not branch-specific. Iguess we could make it branch-specific, but pushing it ourselves is simpler15:55
openstackgerritKristi Nikolla proposed openstack/releases master: stable/rocky releases for keystone deliverables
*** lpetrut has quit IRC16:09
smcginnismgoddard: Looks like I can do that. To confirm before I do anything, that would be the following tags:16:20
smcginnistag 5.1.0 on stable/rocky, commit 755bc38ce1d4867855dae0cdee81269a1e14edf216:20
smcginnistag 6.1.0 on stable/stein, commit 26c8abc6642af90e1436b68c4fedad27c40f7cae16:21
mgoddardsmcginnis: +116:21
smcginnisOK, I'll give it a shot now.16:21
ttxFeels like 2014 all over again16:25
*** e0ne has quit IRC16:27
openstackgerritMerged openstack/releases master: Release ussuri beta1 of neutron and related projects
smcginnismgoddard: Post tagging release jobs are running.16:34
mgoddardgreat, thanks smcginnis!16:36
smcginnisNo problem!16:39
*** ekcs has joined #openstack-release16:39
smcginnisttx: Did you want me to put the proposed release schedule the ML?16:39
smcginnisI think in the past we just decided, but I could be forgetting.16:40
*** ekcs has quit IRC16:43
ttxsmcginnis: we decide, but it's good to open for comments, just in case we missed something obvious16:50
openstackgerritKristi Nikolla proposed openstack/releases master: stable/rocky releases for keystone deliverables
ttxsmcginnis: see
*** dtantsur is now known as dtantsur|afk17:16
*** tosky has quit IRC17:23
*** evrardjp has quit IRC17:34
*** evrardjp has joined #openstack-release17:35
*** ekcs has joined #openstack-release17:53
*** ekcs has quit IRC17:53
openstackgerritAndreas Jaeger proposed openstack/releases master: Release openstack-doc-tools, os-api-ref,openstackdocstheme
*** diablo_rojo has joined #openstack-release17:59
knikollasmcginnis: i'm not quite sure why is failing on the sdist build.18:03
*** e0ne has joined #openstack-release18:08
smcginnisknikolla: Ah, it's a reno issue we've seen elsewhere.18:15
smcginnisknikolla: You will need to do something like this:
smcginnisAJaeger: Do you want to pick up that just merged URL fix for the openstackdocstheme release?18:17
AJaegersmcginnis: yes, didn't I include it?18:17
smcginnisAJaeger: Oh, maybe. I was just looking at the timing, and I didn't think it merged until after you had proposed the patch.18:18
AJaegersmcginnis: I was lucky it just worked ;)18:18
smcginnisSorry, I actually looked at the list-changes output, and you did indeed get it. I should have looked first before saying anything. :)18:18
AJaegersmcginnis: I was expecting to recheck because of the timing - and then it passed, so wondered already whether I screwed up...18:19
smcginnisVery good timing. ;)18:20
AJaegersmcginnis: thanks for double checking!18:20
knikollasmcginnis: would that patch get an exception to rule 4 or does it need to go through master->train->stein->rocky in succession?
smcginnisknikolla: Since it is just a release note configuration change, I do not think it needs to go step by step. Should be able to propose and merge it to all right away.18:21
*** ykarel is now known as ykarel|away18:24
*** ekcs has joined #openstack-release18:25
openstackgerritMerged openstack/releases master: Release openstack-doc-tools, os-api-ref,openstackdocstheme
*** openstackstatus has quit IRC18:33
*** diablo_rojo has quit IRC18:36
*** e0ne has quit IRC18:37
*** diablo_rojo has joined #openstack-release18:37
mlavallediablo_rojo: hey, for this year, are we only going to have this event:
diablo_rojo_phonmlavalle: no, there will be a summit + ptg in October as well18:53
mlavallediablo_rojo_phon: do we know where?18:53
mlavallesorry to bother, but I have to make a budget18:53
diablo_rojo_phonmlavalle: Berlin. It was in the last foundation newsletter :)18:54
diablo_rojo_phonNo problem at all!18:54
knikollasmcginnis: Thanks for all the help! I tried that locally and it seemed to work
mlavallediablo_rojo_phon: should we expect discount codes for code contributors? If yes, do you know how much would we pay?19:00
*** ekcs has quit IRC19:09
*** ianychoi_ has joined #openstack-release19:15
diablo_rojo_phonmlavalle: I don't think it need decided yet but I think they were leaning towards not doing discounts this time?19:16
diablo_rojo_phonI would email Kendall Waters about that one.19:17
*** ianychoi has quit IRC19:18
*** chandankumar is now known as raukadah19:27
*** ekcs has joined #openstack-release19:31
*** diablo_rojo has quit IRC19:37
*** diablo_rojo has joined #openstack-release19:37
*** N3l1x has quit IRC19:51
*** ekcs has quit IRC19:54
mlavallediablo_rojo_phon: ah, ok, I'll ask your tocaya20:10
*** AJaeger has left #openstack-release20:12
*** e0ne has joined #openstack-release20:15
diablo_rojo_phonmlavalle: no problem!20:27
*** ekcs has joined #openstack-release20:28
*** ekcs has quit IRC20:38
*** ekcs has joined #openstack-release20:41
*** e0ne has quit IRC21:38
*** e0ne has joined #openstack-release21:39
*** e0ne has quit IRC22:27
*** ekcs has quit IRC22:44
*** armstrong has joined #openstack-release22:44
*** ekcs has joined #openstack-release22:47
*** dhellmann_ has joined #openstack-release23:05
*** dhellmann has quit IRC23:06
*** dhellmann_ is now known as dhellmann23:06
*** dave-mccowan has quit IRC23:11
*** ekcs has quit IRC23:13
*** ekcs has joined #openstack-release23:13
*** dave-mccowan has joined #openstack-release23:29
*** mlavalle has quit IRC23:29
*** jbadiapa has quit IRC23:48

Generated by 2.15.3 by Marius Gedminas - find it at!