Monday, 2020-03-02

*** dasp has joined #openstack-sdks00:16
*** dasp_ has quit IRC00:18
*** factor has joined #openstack-sdks01:00
*** tkajinam has quit IRC01:24
*** tkajinam has joined #openstack-sdks01:25
*** slaweq has joined #openstack-sdks04:17
*** slaweq has quit IRC04:22
*** slaweq has joined #openstack-sdks05:20
*** dave-mccowan has quit IRC05:24
*** slaweq has quit IRC05:24
*** evrardjp has quit IRC05:35
*** evrardjp has joined #openstack-sdks05:35
*** slaweq has joined #openstack-sdks05:35
*** slaweq has quit IRC05:40
*** yolanda has joined #openstack-sdks06:47
*** iurygregory has joined #openstack-sdks07:28
*** slaweq has joined #openstack-sdks07:53
*** tosky has joined #openstack-sdks08:23
*** jpena|off is now known as jpena08:23
*** tkajinam has quit IRC08:24
*** mgoddard has quit IRC08:27
*** mgoddard has joined #openstack-sdks08:33
*** ralonsoh has joined #openstack-sdks08:34
*** jpich has joined #openstack-sdks08:59
*** dmellado has quit IRC09:23
*** dmellado has joined #openstack-sdks09:26
*** dmellado has quit IRC09:33
*** dmellado has joined #openstack-sdks09:35
*** tkajinam has joined #openstack-sdks09:39
*** sshnaidm is now known as sshnaidm|pto09:59
*** dtantsur|afk is now known as dtantsur10:08
*** tkajinam has quit IRC10:18
*** sshnaidm|pto has quit IRC10:26
*** sshnaidm has joined #openstack-sdks10:42
*** sshnaidm_ has joined #openstack-sdks11:12
*** sshnaidm has quit IRC11:12
*** sshnaidm_ is now known as sshnaidm|pto11:15
*** sshnaidm|pto has quit IRC11:23
*** sshnaidm has joined #openstack-sdks11:23
*** sshnaidm is now known as sshnaidm|pto11:24
*** stingrayza has quit IRC11:28
*** stingrayza has joined #openstack-sdks11:34
*** tkajinam has joined #openstack-sdks11:47
*** Luzi has joined #openstack-sdks11:53
*** yolanda has quit IRC12:10
*** yolanda has joined #openstack-sdks12:11
*** jpena is now known as jpena|lunch12:36
*** tkajinam has quit IRC12:45
*** dave-mccowan has joined #openstack-sdks12:47
*** larsks has joined #openstack-sdks12:53
larsksHow does one pass all_tenants=true to the list_volumes method in python-openstacksdk?12:53
*** jpich has quit IRC13:06
*** jpich has joined #openstack-sdks13:06
*** enriquetaso has joined #openstack-sdks13:16
openstackgerritMerged openstack/ansible-collections-openstack master: Remove old artifacts when building new ones
openstackgerritMerged openstack/ansible-collections-openstack master: Run flake8 in linters
openstackgerritMerged openstack/ansible-collections-openstack master: Fix F841 and remove exclusion
openstackgerritMerged openstack/ansible-collections-openstack master: Remove F403 and F405 exclusions
openstackgerritMerged openstack/ansible-collections-openstack master: Fix E128 and remove exclusion
openstackgerritMerged openstack/ansible-collections-openstack master: Fix W504 and remove exclusion
openstackgerritMerged openstack/ansible-collections-openstack master: Fix F401 and remove exclusion
*** jpena|lunch is now known as jpena13:32
*** tosky_ has joined #openstack-sdks13:47
*** tosky has quit IRC13:47
*** tosky_ is now known as tosky13:47
*** jpich has quit IRC14:09
*** jpich has joined #openstack-sdks14:10
*** jawad_axd has joined #openstack-sdks14:33
*** jawad_axd has quit IRC14:37
*** Luzi has quit IRC15:01
*** __ministry has quit IRC15:15
*** __ministry has joined #openstack-sdks15:16
dtantsurmordred: Oo15:38
mordreddtantsur: uhm15:38
brtknrmordred: since when did openstacksdk start to replace uuid keys with id?15:39
brtknrdtantsur: ^15:39
brtknrfor get requests15:39
*** sshnaidm|pto has quit IRC15:41
mordredbrtknr: since always. the REST interactions should use uuid as the key name if it needs to, and uuid _should_ be available as an alias - but it's a primary design of SDK that every object returned has an "id" field that is the id of the object15:43
mordreddtantsur: I have also responded to the thread, possibly a bit tersely15:44
dtantsurcool, that's what I expected :)15:45
*** sshnaidm|pto has joined #openstack-sdks15:55
*** dmellado has quit IRC15:57
mordreddtantsur: if you get bored, I left a disagreement on - I think we should probably come to an agreement amongst ourselves15:59
*** dmellado has joined #openstack-sdks16:00
dtantsurmordred: I like the idea of a global executor and more or less okay with defaulting to threads as long as it can be completely overridden16:09
toskyjust to bring the discussion in the right place, one of the blocker raised by the cinder people was and is
toskyand the lack of updates on it16:10
toskyis there any update on it?16:10
mordreddtantsur: cool!16:11
mordredtosky: looking16:12
mordredtosky: 3 things. a) I'd still prefer migrating to sdk there, since microversion support is solid and we use it in other places and all. HOWEVER b) I don't think we should block forward progress on that, we can certainly migrate later c) that patch should probably be fixed to pass tests - all of the previous blockers that existed at the time of that patch originally should not exist anymore16:15
*** dmellado has quit IRC16:18
toskymordred: that's good to hear (all of it) - could you please comment on the patch? :)16:18
brtknrmordred: so uuid not being available too is possibly a bug?16:19
*** dmellado has joined #openstack-sdks16:20
mordredbrtknr: yeah, possibly so. where are you seeing the issue?16:20
brtknrfor os_coe_cluster and os_coe_cluster_template modules, only id is returned, not uuid16:20
brtknrlooks like the corresponding ansible module expects to see uuid field which leads to issues with idempotency16:21
mordredbrtknr: that's super weird :)16:21
brtknrsorry the get_coe_cluster and get_coe_cluster_template functions16:21
mordredbrtknr: oh - that's even weirder. that's teh cloud layer, becuase it's for ansible. I'm not sure that's _ever_ returned uuid - so I have no idea how those ansible modules worked? looking further though .,..16:22
brtknrmordred: wait, only get_coe_cluster has this problem16:23
brtknrget_coe_cluster_template seems to return uuid as an alias16:23
mordredbrtknr: yes - I agree with you16:24
mordredcluster template adds it ba k16:24
brtknrmordred: can you show me the code?16:24
brtknrmordred: dont worry, i will try to find it myself, i should stop being lazy16:25
mordredbrtknr: is where cluster_template adds it back if strict == False16:25
mordredand you can see that doens't do a similar thing16:25
brtknrwho'd like to propose a patch?16:26
brtknrI'm happy to if people are busy :)16:26
brtknrActually I am not sure what this strict_mode is about16:28
brtknrfound the reno:
mordredbrtknr: the idea was that we'd have a consistent interface peopel could count on - as well as fallback values16:35
mordredhowever - that concept pre-dates the merge of shade and sdk. with the sdk objects we have the opportunity to define aliases - and that seems to be a better long-term experience for users anyway16:36
mordredwe've been working on shifting the shade layer to using the sdk resource objects and getting rid of that normalize layer - but havne't made it to the coe methods yet16:37
openstackgerritRodolfo Alonso Hernandez proposed openstack/openstacksdk master: Include "fields" to "SecurityGroup" query parameters
mordredbrtknr: if you have a sec, you could probably just cargo-cult in an if not strict_mode: add back uuid patch to keep it simple for the time being - and we can see about migrating the methods as a followup16:37
larsksmordred: Do you know if there is a particular reason that volume.volumes(details=True) doesn't return the user_id information from the remote response?16:39
openstackgerritBharat Kunwar proposed openstack/openstacksdk master: Return uuid alias for coe_cluster in non strict mode
brtknrmordred: ^16:41
mordredlarsks: looking16:41
mordredbrtknr: looks great. it's possible a test mock will need to get updated as a result - but let's see how it goes16:42
openstackgerritRodolfo Alonso Hernandez proposed openstack/python-openstackclient master: Add "fields" parameter to ListSecurityGroup query
mordredlarsks: there is no mention of user_id in openstack/block_storage/v3/ - so I think it's purely that it's missing and not anything on purpose16:43
larsksmordred: thanks. I was just curious if the omission was intentional or not. I'll submit a change.16:43
brtknrmordred: locally, all my tox -e py36 failures are related to cloud devstack not found as I export OS_CLOUD=devstack to my env vars16:45
mordredbrtknr: hah. nice16:46
mordredbrtknr: well - let's see what zuul tells us :)16:46
brtknrmordred: also TIL what cargo cult means16:46
mordredbrtknr: :)16:46
* mordred is going to grab the patch and run it just to check16:46
*** dmellado has quit IRC16:46
brtknrIf this patch goes through, it will void this patch:
brtknr  py36: commands succeeded16:48
brtknr  congratulations :)16:48
brtknrafter unsetting OS_CLOUD16:48
*** dmellado has joined #openstack-sdks16:49
mordredbrtknr: I think we should land that patch too - it describes the current state - and somone might have those modules and _not_ latest sdk - and the id codepath you did is always going to work16:49
mordredwhen the uuid codepath lands in sdk it'll improve things for people and then we can add a doc that says uuid will be an alias in the return value if sdk is new enough16:50
brtknrhmm this seems to highlight an underlying problem, why is tox -e py36 fail when I export OS_CLOUD=16:50
brtknrmordred: sounds good to me16:50
brtknrI will revert the change to the doc16:50
brtknrbtw how the hell do you use ansible-collection-openstack?16:50
brtknrIts not available on galaxy yet so cant use ansible-galaxy to install it16:51
smcginnisbrtknr: I haven't been following the conversation here, but sound like the OS_CLOUD issue might be a side effect of this:
mordredbrtknr: no - don't revert it - I think it's solid16:51
mordredbrtknr: it's the right thing16:52
mordredbrtknr: you need to build a tarball from source then you can install the tarball16:52
mordredwe're working on getting galaxy pub working16:52
mordredbrtknr: I recommend using tox -ebuild to make the tarball16:52
mordredotherwise it's going to have a sad16:52
brtknr revert line 190 here:
mordredno - don't revert that16:53
mordredyour comment there is correct16:53
brtknrAh okay fair enough16:53
mordredthat _is_ the situation most users will encounter - they won't see uuid unless they install an as-yet-unreleased sdk :)16:53
brtknrmordred: is it correct that when a cluster gets *created* then uuid is returned?16:54
brtknrshouldn't the uuid get normalised in the same way for create requests too?16:54
mordredit shouldn't - it should be the same data structure all the time16:54
brtknrok so thats another bummer, it returns uuid at present16:55
brtknrwhen cluster get created16:55
mordredwow really?16:55
* mordred cries16:55
brtknryes sorry16:56
mordredso that means we're not running normalize on the result of create - which is sadmaking. maybe we need to do a workaround in the ansible module itself16:56
mordredto grab the id in both cases with a get('id', get('uuid') - then make sure both id and uuid are set in the dict before we return to teh ansible user16:56
mordredcause otherwise this is just a mess for the user16:57
brtknrthat is the premise of the patch16:58
brtknrits what i am already proposing we do with that patch16:58
mordredcool. then yes - I think we should land that patch asap17:00
mordredShrews: ^^ feel like an easy +A? I know you like patches with no context17:00
Shrewsso much17:00
mordredbrtknr: oh - actually - one more thing...17:00
*** dmellado has quit IRC17:01
* brtknr stops pauses stuffing his bag with his belongings17:01
* brtknr pauses stuffing his bag with his belongings17:01
mordredbrtknr: real quick ... or I can update the patch if you're packing up17:01
mordredbrtknr: (left inline comment)17:01
Shrewsmordred: let's pretend i haven't read the backscroll yet. is 709734 the thing you referred to?17:02
Shrewsthat looks doubtful. /me just reads bs17:03
mordredShrews: yes. and the tl;dr is that normalization of cluster_template dicts is blown and we forgot to add the 'uuid' field back to the dict, but the ansible docs and module refer to uuid not id17:03
mordredthere's an sdk fix to add uuid - but of course latest module doesn't mean latest sdk - so there's also a patch to the module to work around the fact that it might not get uuid back - but a user might be expecting it to exist because of the docs saying it's there17:04
brtknrmordred: done17:07
openstackgerritBharat Kunwar proposed openstack/ansible-collections-openstack master: os_coe_cluster: Retrive id/uuid correctly
*** dmellado has joined #openstack-sdks17:07
mordredbrtknr: thanks!17:08
brtknris cluster['id'] = cluster['uuid'] = cluster_id  okay?17:08
Shrewsshouldn't the changelog fragment at least mention the module being fixed? not sure that that is clear when just reading the resulting log17:08
brtknrmordred: or is that bad practice?17:08
brtknrShrews: done17:09
openstackgerritBharat Kunwar proposed openstack/ansible-collections-openstack master: os_coe_cluster: Retrive id/uuid correctly
mordredbrtknr: I think it's fine in this case - not sure if pep8 will agree or not :)17:09
brtknrI'm running pep8 locally17:10
brtknrband it succeeds17:10
brtknrThanks folks17:10
openstackgerritLars Kellogg-Stedman proposed openstack/openstacksdk master: Include user_id attribute in volume information
mordredwoot. thanks brtknr17:11
mordredlarsks: thanks17:11
mordredI think when that and the coe patch above land we should cut a bugfix release17:12
brtknrwould be nice to fix the create case too on openstacksdk side for consistency?17:12
Shrewsmordred: you'll need to reapply your +2 to 70973417:13
mordredbrtknr: yeah, it would17:13
mordredShrews: done. thanks17:13
mordredbrtknr: if you want to do it - it would be adding a normalize call in openstack/cloud/ in create_cluster_template (and create_cluster) -17:14
mordredbrtknr: so like return self._normalize_cluster_template(cluster_template)17:15
*** jpich has quit IRC17:18
*** dmellado has quit IRC17:19
*** dmellado has joined #openstack-sdks17:25
*** stingrayza has quit IRC17:31
*** evrardjp has quit IRC17:35
*** evrardjp has joined #openstack-sdks17:35
*** jpena is now known as jpena|off17:39
openstackgerritAdam Harwell proposed openstack/python-openstackclient master: Switch image to use SDK
openstackgerritAdam Harwell proposed openstack/python-openstackclient master: Complete switch from glanceclient to SDK for image service
*** efried has quit IRC17:40
*** efried has joined #openstack-sdks17:42
mordredrm_work: \o/17:48
rm_work... i believe i still owe something with ospurge too <_<17:48
rm_workmy volunteer backlog with SDK/OSC is high T_T17:48
mordredrm_work: well -
rm_workyes, but i believe that was superseded by the decisions we made in the meetup in Shanghai :D17:50
rm_work(or is this something else?)17:51
mordredI think this is still aiming at that - but it's entirely possible it's going down the wrong road17:51
rm_workI believe what we said was "just drop-in the ospurge codebase into SDK"17:52
mordredoh right17:52
rm_workand then drop the mic and walk away :D17:52
mordredand I think we keep getting busy ...17:52
mordredstupid busy17:53
mordredok - let me wrap my head around that whole thing this wweek17:53
rm_workmy current quest is fixing octavia's AZ handling and compatibility with nova/neutron routed-network stuff, which requires also ... fixing nova/neutron routed-network stuff :D17:53
mordredthis is what we get for gtema not being able to come to shanghai - and has nothing to do with me forgetting things at all17:53
rm_workwell, it's also not nearly a top priority17:53
rm_workwhich is the other reason it's been sidelined thus far for me17:53
mordredand oh goodie - it sounds like you have a wonderful rabbit hole to go down with routed networks :)17:54
rm_workugh yes17:54
rm_worki was hoping i could take the easy route but i was denied :D17:54
mordredwhere would the fun in that be?17:55
* dtroyer peeks in to see if vegetables will get thrown this direction…17:56
dtroyerdamn, you guys have been busy17:56
mordredit's a dtroyer!17:56
mordredssssh. it's a sekrit17:56
dtroyeroh good, don't let him out of the asylum17:56
mordredI hear he's bringing vegetables17:56
dtroyerhe hates vegetables, which is why being vegetarian for 8 years was, um, interesting17:57
mordredmmm. eggs17:57
mordredcould I have some eggs with those eggs? maybe with a side of eggs?17:58
dtroyerSo I think it is time to acknowledge the clear truth that I'm not going to continue to be working on OSC, not OpenStack for that matter.  I owe an apology and major thanks to everyone who has kept the lights on.17:59
mordreddtroyer: that is sad panda - but thanks for your many years of keeping the lights on yourself!17:59
dtroyerI don't have a new job yet, but nothing I am looking at includes room here, unfortunately17:59
rm_workis that because you're explicitly not looking at stuff that's related, because you need to GTFO for your sanity? :D18:06
rm_workor just coincidental18:06
dtroyermore coincidental.  dis-investment in OpenStack is a real thing and more than just Intel18:07
rm_workI feel like it's just ... shifting18:08
dtroyerbut it may be time, too.  I almost made it to Nova's 10th birthday, but that's a long time in one place18:08
rm_workbut yes18:08
mordreddtroyer: tell me about it - I have certainly never worked 10 years on anything else18:12
*** iurygregory has quit IRC18:19
rm_workme either!18:21
* rm_work is now 10 years out of college18:21
rm_workI think I'm going on 7 years on openstack now tho :)18:22
dtantsurtosky: dumped some microversion microthoughts to the patch :)18:24
mordredShrews: is the sdk-side of that uuid thing from earlier18:25
dtantsurmordred: will it be too much to ask for a unit test on ^^^?18:26
dtantsurI guess it's a question for brtknr really18:26
rm_workoh hey, good place for this since i'm already here:
rm_workneed that for senlin ASAP, and will need that to do a conversion for OSC to SDK18:27
dtantsurrm_work: does this resource even handle microversions? otherwise you'll always see None there18:27
rm_workwhich, the one i just linked?18:28
rm_workor the glance thing18:28
dtantsurrm_work: the one you just linked18:29
rm_workno microversioned involved18:30
rm_workthere's just a note in the commit message about which octavia version it appears in18:30
dtantsurmmm, so 2.14 is not a microversion18:30
rm_workthere's no code around verifying that (though the octavia API will let you see what version it's on, yes)18:30
rm_workbut it's not strictly microversions18:30
mordredso - it'll be there if octavia supports it - and none otherwise yeah?18:30
dtantsurah, similar to what ironic-inspector does, information-only microversions18:31
rm_workerr, in a get? it'd be missing in a GET if the octavia version is older18:32
mordredkinda like what glance does18:32
rm_worknon-present on the octavia side, illustrated by None on the SDK object side18:32
Shrewsmordred: should there be a non-strict-mode test to validate that? We have an entire TestNormalize/TestStrictNormalize suite of tests for such things currently18:33
rm_workand I guess a 400 if passed in fromt he SDK on a create18:33
johnsomdtroyer Thanks for all of your excellent work on OSC. Good luck on your next adventure.18:33
Shrewsbut apparently for not all things  :/18:33
mordredShrews: yeah, probably. I think you and dtantsur had the same thought18:33
mordredit's almost like you value tests or something18:33
dtantsurrm_work: +A then18:34
rm_workwhaaaaaaat, tests are important? who knew :D18:34
Shrewsmordred: we're such silly people folk18:34
dtantsurcan confirm ^^18:34
rm_workOh my, yes.</farnsworthy>18:34
*** KeithMnemonic has joined #openstack-sdks18:35
openstackgerritMerged openstack/ansible-collections-openstack master: os_coe_cluster: Retrive id/uuid correctly
KeithMnemonicmordred: so is there anyway to get this merged or is it too late now?
*** ralonsoh has quit IRC18:42
*** dtantsur is now known as dtantsur|afk19:04
mordredKeithMnemonic: there you go19:26
KeithMnemonicthank you!!19:28
KeithMnemonicmordred: while you are in the giving mood, care to also help on this one for train?
mordredKeithMnemonic: done19:41
openstackgerritMerged openstack/openstacksdk master: Add availability_zone param to load balancer
*** tobiash has quit IRC21:13
*** tobiash has joined #openstack-sdks21:15
openstackgerritMerged openstack/python-openstackclient stable/train: Fix openstack server list --deleted --marker option
KeithMnemoniclooking for review on this backport now that the train branch merged. thank you
*** slaweq has quit IRC22:09
*** slaweq has joined #openstack-sdks22:11
*** slaweq has quit IRC22:16
*** tkajinam has joined #openstack-sdks22:28
*** slaweq has joined #openstack-sdks22:33
*** slaweq has quit IRC22:39
*** slaweq has joined #openstack-sdks22:41
*** slaweq has quit IRC22:46
*** slaweq has joined #openstack-sdks23:11
*** slaweq has quit IRC23:16
*** enriquetaso has quit IRC23:25
*** tosky has quit IRC23:57

Generated by 2.15.3 by Marius Gedminas - find it at!