Thursday, 2019-12-05

*** tosky has quit IRC00:10
*** goldyfruit_ has joined #openstack-sdks01:20
*** ricolin has joined #openstack-sdks02:11
*** slaweq has joined #openstack-sdks02:29
*** slaweq has quit IRC02:35
*** slaweq has joined #openstack-sdks02:51
*** slaweq has quit IRC02:56
*** slaweq has joined #openstack-sdks03:21
*** slaweq has quit IRC03:26
*** slaweq has joined #openstack-sdks03:34
*** slaweq has quit IRC03:39
*** slaweq has joined #openstack-sdks03:58
*** slaweq has quit IRC04:03
*** slaweq has joined #openstack-sdks04:22
*** slaweq has quit IRC04:27
*** slaweq has joined #openstack-sdks04:33
*** slaweq has quit IRC04:38
*** slaweq has joined #openstack-sdks05:02
*** slaweq has quit IRC05:07
*** slaweq has joined #openstack-sdks05:17
*** slaweq has quit IRC05:22
*** slaweq has joined #openstack-sdks05:25
*** slaweq has quit IRC05:30
*** slaweq has joined #openstack-sdks05:45
*** slaweq has quit IRC05:54
*** slaweq has joined #openstack-sdks05:58
*** slaweq has quit IRC06:03
*** Luzi has joined #openstack-sdks06:03
*** slaweq has joined #openstack-sdks06:11
*** slaweq has quit IRC06:18
*** ralonsoh has joined #openstack-sdks06:58
*** gtema has joined #openstack-sdks07:47
*** gtema has quit IRC08:10
*** mattoliverau has quit IRC08:34
*** tosky has joined #openstack-sdks08:36
*** mattoliverau has joined #openstack-sdks08:38
*** slaweq has joined #openstack-sdks08:47
*** jistr has joined #openstack-sdks08:48
*** slaweq_ has joined #openstack-sdks08:50
*** jpena|off is now known as jpena08:50
*** jpich has joined #openstack-sdks08:51
*** slaweq has quit IRC08:52
*** slaweq_ has quit IRC08:59
*** jangutter has joined #openstack-sdks09:09
*** gtema has joined #openstack-sdks09:22
*** mattoliverau has quit IRC09:54
*** jpich has quit IRC09:54
*** ricolin has quit IRC09:54
*** goldyfruit_ has quit IRC09:54
*** dmellado has quit IRC09:54
*** jangutter has quit IRC09:54
*** mugsie has quit IRC09:54
*** fyx has quit IRC09:54
*** ccamel has quit IRC09:54
*** TheJulia has quit IRC09:54
*** mnasiadka has quit IRC09:54
*** frickler has quit IRC09:54
*** masayukig has quit IRC09:54
*** amito has quit IRC09:54
*** openstackgerrit has quit IRC09:54
*** jpena has quit IRC09:54
*** mtreinish has quit IRC09:54
*** brtknr has quit IRC09:54
*** evrardjp has quit IRC09:54
*** gtema has quit IRC09:54
*** dasp has quit IRC09:54
*** jkulik has quit IRC09:54
*** cmurphy has quit IRC09:54
*** logan- has quit IRC09:54
*** dtantsur|afk has quit IRC09:54
*** stingrayza has quit IRC09:54
*** nicholas has quit IRC09:54
*** christi3k has quit IRC09:54
*** amotoki has quit IRC09:54
*** knikolla has quit IRC09:54
*** tosky has quit IRC09:54
*** ralonsoh has quit IRC09:54
*** Luzi has quit IRC09:54
*** mgariepy has quit IRC09:54
*** yolanda has quit IRC09:54
*** factor has quit IRC09:54
*** Jeffrey4l has quit IRC09:54
*** tbarron has quit IRC09:54
*** irclogbot_2 has quit IRC09:55
*** corvus has quit IRC09:55
*** ianw has quit IRC09:55
*** efried has quit IRC09:55
*** eandersson has quit IRC09:55
*** stephenfin has quit IRC09:55
*** aspiers has quit IRC09:55
*** larsks has quit IRC09:55
*** jroll has quit IRC09:55
*** bverschueren has quit IRC09:55
*** gouthamr has quit IRC09:55
*** sshnaidm has quit IRC09:55
*** lxkong has quit IRC09:55
*** gundalow has quit IRC09:55
*** mmedvede has quit IRC09:55
*** gmann has quit IRC09:55
*** dustinc has quit IRC09:55
*** elmiko has quit IRC09:55
*** Blinkiz has quit IRC09:55
*** jistr has quit IRC09:55
*** gryf has quit IRC09:55
*** adriant has quit IRC09:55
*** tonyb has quit IRC09:55
*** tobiash has quit IRC09:55
*** Qiming has quit IRC09:55
*** purplerbot has quit IRC09:55
*** tobberydberg has quit IRC09:55
*** timburke has quit IRC09:55
*** ebbex has quit IRC09:55
*** lastmikoi has quit IRC09:55
*** dtruong has quit IRC09:55
*** johnsom has quit IRC09:55
*** Horrorcat has quit IRC09:55
*** __ministry has quit IRC09:55
*** vishakha has quit IRC09:55
*** mgagne has quit IRC09:55
*** dtroyer has quit IRC09:55
*** lifeless has quit IRC09:55
*** guilhermesp has quit IRC09:55
*** kmalloc has quit IRC09:55
*** odyssey4me has quit IRC09:55
*** mordred has quit IRC09:55
*** vdrok has quit IRC09:55
*** mnaser has quit IRC09:55
*** rm_work has quit IRC09:55
*** melwitt has quit IRC09:55
*** Hazelesque has quit IRC09:55
*** Shrews has quit IRC09:55
*** mattoliverau has joined #openstack-sdks10:04
*** dmellado has joined #openstack-sdks10:04
*** goldyfruit_ has joined #openstack-sdks10:04
*** ricolin has joined #openstack-sdks10:04
*** jpich has joined #openstack-sdks10:04
*** masayukig has joined #openstack-sdks10:04
*** amito has joined #openstack-sdks10:04
*** frickler has joined #openstack-sdks10:04
*** mnasiadka has joined #openstack-sdks10:04
*** TheJulia has joined #openstack-sdks10:04
*** ccamel has joined #openstack-sdks10:04
*** fyx has joined #openstack-sdks10:04
*** mugsie has joined #openstack-sdks10:04
*** jangutter has joined #openstack-sdks10:04
*** evrardjp has joined #openstack-sdks10:04
*** brtknr has joined #openstack-sdks10:04
*** mtreinish has joined #openstack-sdks10:04
*** jpena has joined #openstack-sdks10:04
*** openstackgerrit has joined #openstack-sdks10:04
*** jistr has joined #openstack-sdks10:04
*** tosky has joined #openstack-sdks10:04
*** ralonsoh has joined #openstack-sdks10:04
*** Luzi has joined #openstack-sdks10:04
*** mgariepy has joined #openstack-sdks10:04
*** gundalow has joined #openstack-sdks10:04
*** lxkong has joined #openstack-sdks10:04
*** odyssey4me has joined #openstack-sdks10:04
*** yolanda has joined #openstack-sdks10:04
*** sshnaidm has joined #openstack-sdks10:04
*** elmiko has joined #openstack-sdks10:04
*** Blinkiz has joined #openstack-sdks10:04
*** factor has joined #openstack-sdks10:04
*** gryf has joined #openstack-sdks10:04
*** Horrorcat has joined #openstack-sdks10:04
*** Jeffrey4l has joined #openstack-sdks10:04
*** jroll has joined #openstack-sdks10:04
*** tbarron has joined #openstack-sdks10:04
*** lastmikoi has joined #openstack-sdks10:04
*** __ministry has joined #openstack-sdks10:04
*** vishakha has joined #openstack-sdks10:04
*** bverschueren has joined #openstack-sdks10:04
*** gouthamr has joined #openstack-sdks10:04
*** adriant has joined #openstack-sdks10:04
*** tonyb has joined #openstack-sdks10:04
*** irclogbot_2 has joined #openstack-sdks10:04
*** tobiash has joined #openstack-sdks10:04
*** corvus has joined #openstack-sdks10:04
*** mgagne has joined #openstack-sdks10:04
*** dtroyer has joined #openstack-sdks10:04
*** ianw has joined #openstack-sdks10:04
*** eandersson has joined #openstack-sdks10:04
*** lifeless has joined #openstack-sdks10:04
*** efried has joined #openstack-sdks10:04
*** Qiming has joined #openstack-sdks10:04
*** kmalloc has joined #openstack-sdks10:04
*** guilhermesp has joined #openstack-sdks10:04
*** purplerbot has joined #openstack-sdks10:04
*** mordred has joined #openstack-sdks10:04
*** stephenfin has joined #openstack-sdks10:04
*** aspiers has joined #openstack-sdks10:04
*** dtruong has joined #openstack-sdks10:04
*** larsks has joined #openstack-sdks10:04
*** tobberydberg has joined #openstack-sdks10:04
*** mmedvede has joined #openstack-sdks10:04
*** timburke has joined #openstack-sdks10:04
*** gmann has joined #openstack-sdks10:04
*** dustinc has joined #openstack-sdks10:04
*** johnsom has joined #openstack-sdks10:04
*** ebbex has joined #openstack-sdks10:04
*** Shrews has joined #openstack-sdks10:04
*** Hazelesque has joined #openstack-sdks10:04
*** melwitt has joined #openstack-sdks10:04
*** rm_work has joined #openstack-sdks10:04
*** mnaser has joined #openstack-sdks10:04
*** vdrok has joined #openstack-sdks10:04
*** gtema has joined #openstack-sdks10:04
*** dasp has joined #openstack-sdks10:04
*** jkulik has joined #openstack-sdks10:04
*** cmurphy has joined #openstack-sdks10:04
*** logan- has joined #openstack-sdks10:04
*** dtantsur|afk has joined #openstack-sdks10:04
*** stingrayza has joined #openstack-sdks10:04
*** nicholas has joined #openstack-sdks10:04
*** christi3k has joined #openstack-sdks10:04
*** amotoki has joined #openstack-sdks10:04
*** knikolla has joined #openstack-sdks10:04
*** logan- has quit IRC10:05
*** dtantsur|afk is now known as dtantsur10:06
*** logan- has joined #openstack-sdks10:06
openstackgerritVishakha Agarwal proposed openstack/python-openstackclient master: WIP Adding options to user cli
*** jpena is now known as jpena|lunch12:22
*** enriquetaso has joined #openstack-sdks12:25
*** enriquetaso has quit IRC12:42
*** jpena|lunch is now known as jpena13:14
*** goldyfruit_ has quit IRC13:20
*** mriedem has joined #openstack-sdks13:54
mordredrm_work: ossum13:59
*** enriquetaso has joined #openstack-sdks14:24
*** Luzi has quit IRC14:28
*** goldyfruit_ has joined #openstack-sdks14:48
*** goldyfruit___ has joined #openstack-sdks15:00
*** goldyfruit_ has quit IRC15:02
*** gtema has quit IRC15:16
*** gtema has joined #openstack-sdks15:22
*** gtema has quit IRC15:29
openstackgerritMerged openstack/shade master: Add source and contributing link in readme
*** pabelanger has joined #openstack-sdks15:53
*** d0ugal has joined #openstack-sdks15:54
*** ekultails has joined #openstack-sdks15:54
elmikodtantsur: is the ironic thing happening today?15:55
dtantsurelmiko: ironic things happen every day15:56
*** mgoddard has joined #openstack-sdks15:57
dtantsurit's not ironic-specific, more of adopting all openstack ansible modules (and re-shaping their API for consistency - hence API SIG involvement)15:57
elmikoahh, gotcha15:57
*** cloudnull has joined #openstack-sdks15:58
elmikois ironic things happening everyday also ironic?15:58
dtantsurironic things are ironically also ironic15:58
cloudnullisn't it ironic, don't you think?15:59
sshnaidmdtantsur, can you run the mtg bot here?15:59
dtantsurwe haven't done it for a while, I guess I can16:00
elmikoAPI SIG office hour now open16:00
elmikoi'll let you do your bot thingie16:00
dtantsur#startmeeting API-SIG16:00
openstackMeeting started Thu Dec  5 16:00:31 2019 UTC and is due to finish in 60 minutes.  The chair is dtantsur. Information about MeetBot at
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.16:00
*** openstack changes topic to " (Meeting topic: API-SIG)"16:00
openstackThe meeting name has been set to 'api_sig'16:00
sshnaidmdtantsur, cool, if you can chair me too please :)16:00
dtantsurtoday we're a bit hijacked by Ansible SIG and the ongoing effort of bringing OpenStack ansible modules under our wing16:01
dtantsur#chair sshnaidm elmiko16:01
openstackCurrent chairs: dtantsur elmiko sshnaidm16:01
*** gtema has joined #openstack-sdks16:01
sshnaidmok, so thanks to SDK folks for the channel time :)16:01
dtantsura big part of it will be re-designing the modules for consistency, hence our (API SIG) involvement16:01
dtantsursshnaidm: take it from here :)16:01
sshnaidmdtantsur, right16:02
sshnaidmplease use the etherpad:
sshnaidmwe have update from gundalow from Ansible team16:02
* gundalow waves16:02
sshnaidmit's written in the bottom of etherpad16:02
sshnaidmand I think we have a few decisions to make16:02
sshnaidm1. what is namespace name of openstack collections16:03
* gundalow is available if there are any questions now (or later)16:03
sshnaidmgundalow, great16:03
elmikoi just want to state for the record, that it is currently not within the goals of the api-sig to define guidance for these type of apis. /but/ with that said, it is within our goals to facilitate and promote these types of discussions.16:03
sshnaidmelmiko, thanks16:03
dtantsurwell, our *actual* goals is to spend some time joking on each other16:04
dtantsursoooo... :D16:04
elmikohaha, too true16:04
* cloudnull reengages with the meeting 16:04
sshnaidmso, can we use "openstack." namespace?16:05
gundalowSo goal wise, all (will nearly all) modules/plugins will be removed from ansible/ansible before Ansible 2.10 ships. Getting a head of the curve and thinking about this now is good think16:05
gtemais it official target of Ansible to remove modules from core?16:05
dtantsursshnaidm: my only concern is the fact that the collection won't be strictly "official"16:05
sshnaidmgundalow, so no more community modules in 2.10?16:05
ekultailsWhat about modules that won't be able to find a new home?16:06
sshnaidmdtantsur, yeah, these licenses differences, that's why raised this16:06
gundalowansible/ansible:stable-2.10 will only contain
* jistr waves16:06
gundalowekultails: for "homeless modules" that are currently in ansible/ansible they may go in the "catch-all" (aka dumping ground) repo & Collection16:07
dtantsuransible's stackforge :)16:07
sshnaidmso we're on time..16:07
jistrsshnaidm: +1 for using openstack. namespace for the collection, since it's supposed to be "the official thing"16:08
ekultailsgundalow: Okay, good to know, thanks.16:08
gtemaand what about collection name?16:08
jistropenstack.openstack? :)16:08
gtematoo weird16:08
gundalow might have some useful background16:08
dtantsuropenstack.<service type>?16:08
dtantsurotherwise why do we need both?16:08
sshnaidmgundalow, well, I think nobody should take "openstack" namespace except Openstack..16:08
ekultailsNot openstack.ansible, that could get confusing with OpenStack-Ansible. :-P16:08
jistrhaving both is a requirement by Ansible galaxy IIUC16:09
dtantsur#link some helpful background16:09
*** dave-mccowan has joined #openstack-sdks16:09
gtemafor my org I ended up with "namespace.core" - for the core modules16:09
gundalowWhat else could go under `openstack.`?16:09
dtantsurgundalow: we can have a collection per service16:09
gundalow(forgive me, I'm not familiar with all the parts of OpenStack)16:09
dtantsuralthough it will probably complicate everything16:09
jistryea i also thing that would complicate it...16:09
dtantsurgtema: we no longer have an official concept of core modules16:09
gtemaI don not think collection per service make sense, more a one collection with "core" modules, and perhaps additional ones for some "extra-weird" stuff16:09
dtantsurwe could have openstack.openstack and openstack.unofficial16:10
sshnaidmwhat is "core" modules?16:10
dtantsurlet's avoid the term "core", it's a different thing for everyone16:10
jistrgtema, dtantsur: +1 i think that scenario might come up perhaps16:10
gtemawell, everything what we have now16:10
gtemaand i.e. for some weird provisioning of triple-o infra - there might be openstack.tripleo16:10
sshnaidmeverything we have is kinda casual..16:10
sshnaidmit's just what people could made quickly16:11
jistrgtema: +1. So for the ones that are getting imported from Ansible, i'd go openstack.openstack or openstack.core or something like that16:11
sshnaidmand I think it will look very different in the end16:11
gtemayeah, but I think openstack.core.server looks better than openstack.openstack.server or openstack.compute.server16:11
sshnaidmI'd vote for openstack.[service/project] names16:12
dtantsurI think for many people 'core' may be nova+cinder+neutron+glance+keystone16:12
dtantsurnot things like ironic16:12
sshnaidmgtema, openstack.compute.servers is actually as in SDK16:12
sshnaidmand we may align with SDK here16:12
jistrsshnaidm: hmm that does look good indeed16:12
gtemaI do not want to end up needing to do "ansible-galaxy collection install nova"16:12
gtemaand neutron and ...16:12
jistrwe'd force people to fetch multiple collections though... maybe not much of a problem16:13
jistrbut i wonder about code sharing16:13
ekultailsWould there be a way to have a meta collection that installs all relevant collections?16:13
pabelangerI would recommend openstack.openstack as the collection then you call openstack.openstack.os_nova, or maybe even community.openstack.os_nova16:13
jistrwill openstack.<service> pattern complicate *writing* those collections in case we want to share code?16:13
dtantsurwe could avoid the problem with code sharing by putting EVERYTHING in openstacksdk16:13
sshnaidmjistr, as we were talking previously there should be definitely common part, like auth for example16:13
pabelangeropentack.openstacksdk.os_nova would work too16:14
dtantsurare we keeping the os_ prefix?16:14
sshnaidmopenstack.compute,, openstack.tripleo - can it work?16:14
pabelangerbut, i think you still want os_nova / os_neutron, otherwie, you will break a ton of playbooks16:14
gundalowFYI you can use with a scenario file to pull the contents out of ansible/ansible into the correct structure16:14
dtantsurI guess it was only useful before namespacing?16:14
sshnaidmI dont' think need os_ now, it's a dup16:15
gtemawhile "openstacksdk" forever, what should it stand for?16:15
pabelangerfor ansible/ansible migration, they should all be migrated to single repo16:15
pabelangerand namespace16:15
gtemaos_ should be definitely dropped16:15
gundalowIf you change the module names, you may break backwards compatibility between 2.9 and 2.10 unless you put in aliasing (symlink)16:15
pabelangergtema: that is a huge breaking change16:15
dtantsurpabelanger: we're considering some breaking changes anyway, like unifying authentication16:15
pabelangerif you do, openstack.sdk.os_nova, you could still reference os_nova in your playbook16:16
dtantsurgundalow: would plain os_nova (without openstack.<something>) work after the migration?16:16
dtantsurmmm, I see16:16
gtemahe? is the migration from ansible to collection not a breaking change by itself?16:16
dtantsurit sounds like it's not THAT breaking16:16
gtemayou anyway need to change your playbooks16:16
dtantsuror rather: it doesn't break the playbooks16:16
pabelangerI'd say no, you can keep same playbooks for most part16:16
gtemait does16:16
gundalowgtema: There is the idea of a "Community version of Ansible" that will ship with all the collections that make up what's in Ansible 2.916:16
gtemayou need to refer to modules from collection or import collection16:17
dtantsurI'd seriously regret if we couldn't unify the auth though16:17
dtantsur(and s/os_ironic/os_baremetal/ )16:17
gtemagundalow, but you anyway need to modify playbook16:17
sshnaidmmaybe we can keep migrated ones and develop their replacement aside with new names..?16:17
pabelangerIMO, this goal for migration out of ansible/ansible, would be to use migration tool from ansible-core, then keep naming schema of modules the same16:17
gtemaor "community version" will do some extra magic?16:17
pabelangerthen do deprecation process on os_nova, os_neutron16:18
gundalowWe may add some magic during Ansible 2.10 development that means if you are using the Ansible Community Distribution that you wouldn't need to change your playbooks - Though the exact details of this haven't been worked out uet16:18
pabelangersshnaidm: yah, that would be fine16:18
dtantsurrealistically, we should probably do what sshnaidm proposes16:18
dtantsurI'm not sure we're going to have enough cycles to simultaneously refactor things16:18
dtantsurso maybe keep os_ prefix to designate old modules16:19
sshnaidmwe'll just advertise new names for a few versions and then will remove old ones I think16:19
dtantsurand drop it for new ones16:19
sshnaidmdtantsur, ya16:19
dtantsuros_ironic_node is old, baremetal_node is new16:19
gtemaperhaps better name them properly now and make alias16:19
gundalowFYI Ansible's deprecation cycle has been 4 major versions, so two years16:19
dtantsurgtema: not everything will be a mere rename16:19
dtantsurgundalow: is it the same for collections?16:19
jistrgtema: separating will allow developing new ones in a common way without worrying about breaking existing use16:19
gtemaI know we are talking about more than simple rename16:20
dtantsurgundalow: to generalize my question: is there an expected release cadence?16:20
gundalowdtantsur: unsure16:20
gtemabut if we do rename and alias16:20
gtemawe can develop some standard, follow it16:20
gundalowdtantsur: at least twice a year (so new Ansible Engine releases can pick up the new contents)16:20
gundalowI'd hope that over time we can get quicker at doing releases16:20
gtemaand be consistent in the new modules for those 4 cycles16:20
dtantsurokay, so synchronized with ansible releases?16:20
pabelangerI would go as far, until code is deleted in ansible/ansible (which date is up in the air), we continusoly run migration sync nightly, and make new collection repo read-only, then once ansible/ansible code is deleted, we unfreeze openstack collection16:20
gtemaotherwise we have a mix of different patterns16:21
pabelangerthis is our plan for ansible network team16:21
dtantsurpabelanger: I'd prefer to freeze only the existing modules, but allow new development16:21
gundalowSo, for a moment ignoring refactoring. I don't think you want to break Playbooks that work in Ansible 2.916:21
sshnaidmpabelanger, +1 to dtantsur16:21
sshnaidmpabelanger, we'll have old ones for usage, but the main goal is new ones..16:22
pabelangersshnaidm: dtantsur: that is fine, but keep in mind, if a nightly sync happens, like proposed bot, we want to avoid conflicts16:22
dtantsurright, so make sure not to modify the same files16:22
pabelangerfor me, we need to keep old ones for 2 years16:22
pabelangereven in a collection16:22
dtantsurlike, set up gerrit to outright reject changes to os_*.py until we unfreeze16:22
dtantsurs/gerrit/zuul jobs/16:22
gundalowOnce the collection is up and running we can `git rm` the files from ansible/ansible16:23
sshnaidmpabelanger, I think we're good with that16:23
pabelangeryah, once migration happens, if everything is sync'd we can push on git rm on ansible/ansible16:23
pabelangerthen freeze is over16:23
pabelangerbut, that sync should be a zuul jobs16:23
dtantsur#agreed Import the existing modules as they are and freeze them. Keep syncing until ansible removes them from tree.16:23
pabelangerI'm happy to share job for that, as we do it16:24
sshnaidm#action to collaborate with pabelanger to create sync zuul jobs16:24
gtemafor how long do we expect to have this freeze?16:24
sshnaidmpabelanger, thanks, it'd be great16:24
sshnaidmgtema, to keep old modules you mean?16:24
dtantsursshnaidm: #actions requires <who>16:24
sshnaidm#action sshnaidm to collaborate with pabelanger to create sync zuul jobs16:25
pabelangergtema: freeze should be as long as we all agree, final sync looks good and we can rm -rf ansible/ansible16:25
pabelangerwe should be able to drive delete date16:25
dtantsurgtema: until they are removed from ansible/ansible. then it's bugfix-only mode.16:25
*** jangutter has quit IRC16:25
pabelangerthat means, getting content uploaded into galaxy and some sort of pre-release versioning16:25
gtemaok then. I just wanted to avoid freeze, which will keep us ourselves with tied hands16:26
gundalowFYI for bug fixes for ansible/ansible:stable-2.9, once you've gitrm'd from devel  you can create a PR directly against that branch, rather than `git cherry-pick -x` from devel into stable-2.916:26
sshnaidmand these bugfixes should be done in openstack repos since moving, right?16:26
gundalowWhich may mean one fewer things to keep moving16:26
gundalowOnce the collection is done, I'd expect bug fixes do be done in the collection first. Then optionally made to ansible/ansible:stable-2.9. Realistically I expect fewer backports to ansible/ansible once Collections are up and running16:27
sshnaidmgundalow, lemme understand, 2.9 backports via ansible github, all new - via our repos, right?16:27
dtantsurhmm, that's the opposite direction16:28
gtema"main dev on our side"16:28
gtemaand "port" fixes back to ansible-2.916:28
pabelangerthat seem correct to me16:29
sshnaidmgtema, I think we shouldn't port there, maybe only some critical fixes16:29
gtemathat's exactly about critical things, but I do not expect those honestly16:29
pabelangerit raises the question of maybe stable backports team?16:29
pabelangerwhich can handle sync to github16:30
gtemaI don't think that just for 2.9 it make sense to have this overhead16:30
gundalowCurrent backport docs:
gundalowBefore Collections: 1) PR to `devel`, 2) wait for that to be merged.  3) For bug fixes only Create backport PR to `stable-2.x`16:30
sshnaidmI don't think we should backport anything to 2.9 at all, it looks working as is, and no much activity was there16:30
daspCan I get some reviews on and ? (new qos_policy modules for Ansible)16:30
pabelangersshnaidm: we can't say that however, as ansible/ansible still maintans that16:31
gundalowUnless I see a critical bug, I'm not going to be backporting anything to stable-2.9 after we've gitrm'd from devel16:31
gtemadasp - wrong channel. We are deciding to drop that at all ;-)16:31
sshnaidmpabelanger, I think we agree it should be something exceptional16:31
dtantsurokay, so do we even need a constant sync then?16:31
sshnaidmgtema, :D16:31
gundalowand "critical bug" is subjective, maybe you'd use that to your advantage :P16:32
dtantsuror just copy everything once and apply some policies by humans?16:32
sshnaidmdtantsur, I'd go for that for simplifying things16:32
gundalowMaybe I could give an example of what I'm doing with Grafana16:32
gundalowsshnaidm: ++ KEEP THINGS SINGLE16:32
gtemaI think: import, github freeze, collection release, github rm16:32
sshnaidmsingle and simple :)16:32
gtemagithub = ansible upstream16:33
sshnaidmso no need to have sync jobs as well?16:33
sshnaidmpabelanger, ^16:33
gtemaif we handle import and first release in few days - I would say no16:33
pabelangerwell, need some import process to gerrit16:33
pabelangerwhich, usually isn't manual16:34
pabelangerso we need something automated16:34
pabelangeror ask infra what they want to do16:34
dtantsurwell, we did a manual import for ironic-tempest-plugin16:34
dtantsurit was a bit.. interesting16:34
gtemawell, we have already changes in gerrit, so just more-or-less merge and release it16:34
gundalowFor grafana (own repe & collection).16:34
gundalowI've create a new repo and the maintainers are doing there work there (CI is up, they are merging stuff)16:34
gundalowOnce we've got publishing to Galaxy done and a final check we will see if there have been any other changes to grafana code in ansible/ansible, copy those across16:34
gundalowthen gitrm & add `migrated_to` in BOTMETA.yml16:34
gundalowThen do $something with open grafana issues & PRs16:34
pabelangerright, not out of the question but we can ask infra16:34
dtantsurwe extracted a git subtree from ironic, then asked infra adminds to push it :)16:34
dtantsurgundalow: sounds right and simple16:35
gundalowI'd avoid having to keep too many things in progres, and even fewer things in sync16:35
gundalowSo maybe create a test collection, play with it, check it works. then do the move for real16:35
gtematherefore if we here (mostly the only ones approving changes in ansible) agree not to review anything else during releasing collection - we are fine16:36
sshnaidm(and I think we have some job that testing modules, it runs on github changes)16:36
sshnaidmgtema, I think we can agree here16:36
gundalowThere is an OpenStack specific Zuul job that runs against OpenStack PRs in ansible/ansible16:36
sshnaidmgundalow, yeah, exactly, we can import it for collections repo16:37
sshnaidmas a start16:37
gtemanow with tests: from what I have experienced as of now with collections - you MUST modify plybooks16:37
dtantsur#agreed No periodic sync, just one-time migration with temporary freezes on both sides (ending with `git rm` on the ansible side)16:37
gundalowgtema: ansible-test (in 2.9) has been updated so it should work without FQCN (Fully Qualified Collection Name)16:37
gtemagundalow, ah ok. Nice to know16:38
gundalowdtantsur: Please add + add `migrated_to: openstack.FOO` in `.github/BOTMETA.yml`16:38
gundalowIn the same PR16:38
openstackRemoving item from minutes: #agreed No periodic sync, just one-time migration with temporary freezes on both sides (ending with `git rm` on the ansible side)16:38
gtemabut then we need to change our jobs to use "ansible-test" instead of "ansible-playbook". Right?16:38
dtantsur#agreed No periodic sync, just one-time migration with temporary freezes on both sides (ending with `git rm` and `migrated_to: openstack.FOO` on the ansible side)16:38
dtantsurgundalow: like ^^?16:38
sshnaidmgtema, ansible-test is just a different tool afaiu16:39
sshnaidmwith sanity checks etc16:39
gtemayeah, looks like that16:39
sshnaidmmaybe let's set a time for the moving?16:39
sshnaidmwhen can we do it?16:39
sshnaidmgundalow, any restrictions..?16:40
gundalowdtantsur: perfect, thank you16:40
gtemagundalow, what do I need to do not to use FQCN?16:40
dtantsurprobably not right before winter holidays :)16:40
dtantsuror are we in time pressure?16:40
sshnaidmdtantsur, yeah, before xmass of after :)16:40
gundalowI'd suggest you continue to use `ansible-test sanity` that will help spot functional problems16:40
dtantsurI'm here next week, then out till January16:41
gundalowrealistically I'd say play now (do a trial move, get CI working, define merging/+1, etc), move in Jan16:41
*** slaweq_ has joined #openstack-sdks16:41
gtemagundalow, our tests run "ansible-playbook ..."16:41
sshnaidmif no time pressure, let's do after holidays16:41
gundalowJan is fine for me16:41
dtantsurany specific dates?16:42
dtantsurright the week of 6th or?16:42
sshnaidmdtantsur, we'll need to sync maybe16:42
gundalowI'm back on the 13th Jan16:42
*** openstackgerrit has quit IRC16:42
sshnaidmpabelanger, are you available in Jan?16:43
*** goldyfruit_ has joined #openstack-sdks16:43
sshnaidmdtantsur, I think we can start at 13 then16:43
dtantsurlet's put it this way16:44
sshnaidmdtantsur, or at least to decide to postpone it then16:44
dtantsur#agreed Do the migration in the middle of January 202016:44
gtemagundalow, any change to "survive" with "ansible-playbook our_tests.yml"?16:44
gundalowgtema: So continue to use `ansible-playbook` directly for integration tests if that's how you currently do things 2) Please look at using `ansible-test sanity` in your new repo (currently this is run via Shippable on ansible/ansible PRs), this *will* catch valid issues that your integration tests may not16:44
sshnaidmok, let's get back to agenda then16:45
sshnaidmare we keeping them GPL?16:45
gundalowgtema: unsure, if you speak to mattclay in #ansible-devel he might be able to give some ideas16:45
gtemabut I mean ansible-playbook will fail when there is "git rm"16:45
dtantsurcan we practically re-license them?16:45
sshnaidmor we start develop new in Apache2?16:45
dtantsurthen we'll need clean-room development (without looking at old ones)16:45
*** jpena is now known as jpena|brb16:45
gtemagundalow, ok, will ask in #ansible-devel, but this is important to verify the whole migration16:46
gtemadtantsur: without looking will be very hard16:46
*** goldyfruit___ has quit IRC16:46
dtantsurexactly, but that's the condition of re-licensing via rewriting16:46
gtemaand we repeat all the issues we were fixing for ages16:46
dtantsurotherwise we need an agreement from every contributor whose code is still there16:47
gtemaI know, and this is not what I would like to do16:47
dtantsurI'd keep shared bits permissive if possible, but the modules themselves probably have to be GPL16:47
gtemathis is terrible, even if one has changed a typo16:47
sshnaidmare there licenses experts here? :)16:48
pabelangersshnaidm: yes16:48
* dtantsur wow16:48
sshnaidmcan we write new modules in different licenses?16:48
sshnaidmI mean not in GPL16:48
sshnaidmthe goal is to deprecate old ones16:48
gundalowgit log --pretty=format:"%an%x09"  lib/ansible/modules/cloud/openstack | sort | uniq | wc -l ----> 15716:49
gundalow(may be some duplicates, though that's a lot of people to get sign off from)16:49
sshnaidmshould we ask all 155 people to agree?16:50
gundalowIANAL, though I if one person doesn't respond you can't do it16:50
sshnaidmI don't think it's doable at all16:50
sshnaidmgundalow, but it's for re-licensing, I'm talking about new modules develop16:51
sshnaidmI think it's kinda gray area here..16:51
dtantsurit's grey indeed. how to distinguish new development from copy-pasting in case of very similar code?16:51
sshnaidmno idea, need license experts opinion :)16:51
sshnaidmalthough I hope it won't be copy paste16:52
dtantsurIANAL, only relaying what I've previously heard16:52
sshnaidmwe don't need to copy paste, we just have them available..16:52
sshnaidmmnaser mentioned some license-... maillist, maybe worth to ask there16:53
mnaserlegal-discuss :)16:53
sshnaidmmnaser, yes! :)16:53
mnaserand yes i think you need to get all 155 people to agree AFAIK. but im not a license expert16:54
dtantsurthe problem as I understand it, if you're familiar with old code, you won't be able to claim you wrote the new code without taking from it.16:54
*** ricolin has quit IRC16:54
sshnaidmok, but modules that don't have a parallel I think we can start with Apache216:54
dtantsurI mean, legally claim, we all here believe you :)16:54
dtantsurprobably? I'd really bring it to legal-discuss16:54
gtemaok, time is running out. Summary?16:54
dtantsurgtema: we had a few #agreed16:55
gtemaI would also vote for raising16:55
gtemain legal-discuss16:55
dtantsur#action sshnaidm To raise the licensing topic on legal-discuss16:55
sshnaidm#action to figure out which license to use for new developed modules in openstack collection,16:55
dtantsurcorrect? :)16:55
sshnaidmdtantsur, ack16:55
dtantsuryou forgot <who>16:55
openstackRemoving item from minutes: #action to figure out which license to use for new developed modules in openstack collection,16:55
dtantsurokay, now we have one item16:55
dtantsuranything else?16:55
*** openstackgerrit has joined #openstack-sdks16:56
openstackgerritMerged openstack/openstacksdk master: tox: Keeping going with docs
sshnaidmwanted to ask about versioning of modules16:56
sshnaidmI don't think we need branches as openstack releases, right?16:56
sshnaidmshould we just keep them compatible as much as possible? in case we have differences in openstacksdk for example16:56
sshnaidmany strong opinions about it?16:57
dtantsurI guess we need to match ansible?16:57
gtemaI think current version checking in modules is sufficient. So no need for openstack releases from my POV16:57
sshnaidmdtantsur, why?16:58
dtantsurto allow combining them as gundalow mentioned?16:58
*** dave-mccowan has quit IRC16:58
dtantsurSDK has stable branches, OSC too (?)16:58
sshnaidmdtantsur, sorry, I missed it, to combine?16:58
dtantsurit feels like the modules could16:58
gtemaat least "API" for the collections structure need to be kept in sync with Ansible16:58
sshnaidmgtema, yeah, that's right16:58
dtantsursshnaidm: a distribution of ansible with all collections16:59
sshnaidmyeah, I think we agree on keeping them compatible with ansible16:59
sshnaidmotherwise it won't work :)16:59
dtantsurexactly :)17:00
dtantsurI guess we need to decide about stable branches17:00
dtantsurprobably next time17:00
sshnaidm#agreed to keep modules compatible with Ansible collections API17:00
sshnaidmdtantsur, yep17:00
sshnaidmok, thanks all for participation!17:00
sshnaidmnext Thu same time same place17:00
dtantsurthanks, and thanks elmiko for tolerating us!17:00
*** openstack changes topic to "Bug tracker for SDK and OSC is now at"17:00
gtemaotherwise it is too silent :D17:00
openstackMeeting ended Thu Dec  5 17:00:56 2019 UTC.  Information about MeetBot at . (v 0.1.4)17:00
openstackMinutes (text):
sshnaidmyeah, we fixed that)17:01
*** slaweq_ has quit IRC17:02
elmikodtantsur: gladly =)17:02
gundalowGood to chat everybody, please do shout out if there is anything else17:04
gtemathanks gundalow17:04
*** gtema has quit IRC17:09
*** enriquetaso has quit IRC17:10
*** goldyfruit___ has joined #openstack-sdks17:12
*** goldyfruit_ has quit IRC17:15
*** dtantsur is now known as dtantsur|afk17:19
*** jpich has quit IRC17:19
*** lastmikoi has quit IRC17:24
*** jpena|brb is now known as jpena17:25
*** lastmikoi has joined #openstack-sdks17:28
gundalowdtantsur|afk: sshnaidm another option could be `redhat.openstack`. I think this comes down to how many OpenStack Collections you think there maybe17:36
sshnaidmgundalow, mm.. but our projects are community wide, not limited to redhat17:39
*** enriquetaso has joined #openstack-sdks17:46
gundalowsshnaidm: nod, that's a fair point17:47
mordredyeah - I would hope we only have one openstack collection18:04
mordredalso - sorry I missed the meeting - I have plague today18:04
*** jpena is now known as jpena|off18:18
*** enriquetaso has quit IRC18:23
elmikohope you feel better soon mordred =)18:26
mordredelmiko: thanks! nothing says the holidays like laying around groaning18:26
sshnaidmgundalow, is there somewhere written about plans to move modules in 2.10? The roadmap looks a little bit empty
*** ralonsoh has quit IRC18:37
*** dave-mccowan has joined #openstack-sdks18:39
gundalowsshnaidm: we are working on comma at the moment. I expect it to be out in the next week or so18:42
elmikomordred: heh, totally18:59
ekultailsAnsible 2.10 should really be 3.0 with all of this restructuring going on. Just my two cents.19:08
*** gtema has joined #openstack-sdks19:10
pabelangerI'd be interested in how the collection release process will work19:11
pabelangerwhat will be the version number of this first collection release, targetted for 2.1019:12
*** gtema has quit IRC19:15
*** dave-mccowan has quit IRC19:36
*** enriquetaso has joined #openstack-sdks19:41
ekultailsI can't imagine every Collection aligning with Ansible's life cycle. I guess that is both a pro and a con that a Collection could push out updates independent of Ansible.19:44
*** enriquetaso has quit IRC20:39
*** dave-mccowan has joined #openstack-sdks21:13
*** stephenfin has quit IRC21:50
*** dasp_ has joined #openstack-sdks21:52
*** dasp has quit IRC21:55
*** stephenfin has joined #openstack-sdks21:57
*** goldyfruit___ has quit IRC22:05
*** ekultails has quit IRC22:26
*** sshnaidm is now known as sshnaidm|afk23:24
*** tosky has quit IRC23:38

Generated by 2.15.3 by Marius Gedminas - find it at!