Wednesday, 2021-01-13

*** tosky has quit IRC00:01
*** holser has quit IRC00:35
*** LinPeiWen has joined #openstack-sdks00:40
*** LinPeiWen has quit IRC01:08
*** khomesh24 has joined #openstack-sdks01:09
*** LinPeiWen94 has joined #openstack-sdks01:19
*** tkajinam has quit IRC01:41
*** tkajinam has joined #openstack-sdks01:42
*** tkajinam has quit IRC02:09
*** tkajinam has joined #openstack-sdks02:10
openstackgerritwushiming proposed openstack/openstacksdk master: Drop lower-constraints job
*** enriquetaso has joined #openstack-sdks03:02
*** ysandeep|away is now known as ysandeep03:24
*** ricolin_ has joined #openstack-sdks03:39
*** udesale has joined #openstack-sdks03:48
*** ricolin_ has quit IRC03:54
*** enriquetaso has quit IRC04:15
*** evrardjp has quit IRC05:20
*** evrardjp has joined #openstack-sdks05:24
*** evrardjp has quit IRC05:33
*** evrardjp has joined #openstack-sdks05:35
*** nikparasyr has joined #openstack-sdks06:05
*** ralonsoh has joined #openstack-sdks07:19
*** gtema has joined #openstack-sdks07:31
*** gtema has quit IRC07:43
*** BakaKuna has joined #openstack-sdks07:45
*** openstackgerrit has quit IRC07:47
*** gtema has joined #openstack-sdks07:51
*** nightmare_unreal has joined #openstack-sdks07:53
*** gtema has quit IRC07:55
*** slaweq has joined #openstack-sdks07:59
*** diablo_rojo__ has quit IRC08:01
*** slaweq has quit IRC08:04
*** slaweq has joined #openstack-sdks08:10
*** openstackgerrit has joined #openstack-sdks08:10
openstackgerritTakashi Kajinami proposed openstack/python-openstackclient master: Volume type: Update is_public attribute
*** gtema has joined #openstack-sdks08:14
*** rpittau|afk is now known as rpittau08:25
*** holser has joined #openstack-sdks08:34
*** tosky has joined #openstack-sdks08:39
*** jpich has joined #openstack-sdks09:06
*** dasp_ has quit IRC09:16
*** dasp has joined #openstack-sdks09:17
stephenfingtema: Are you okay with me self-approving OSC compute-related patches that have +1s from nova cores?09:19
stephenfincool, thanks09:20
gtemathank you for taking care of that also09:20
stephenfinno problem at all09:20
stephenfingtema: Oh, I do have one non-nova gate fix patch that needs attention though /o\
stephenfin(python-karborclient was removed from tracking in openstack/requirements so the requirements job is failing)09:23
gtemaI thought I was approving it, weird09:23
gtemacan you please have a look at
gtemaI want to cut SDK release soon09:25
stephenfinsure, looking now09:25
gtemaand complete this hypervisor patch09:25
*** jpich has quit IRC09:34
*** jpich has joined #openstack-sdks09:34
*** BakaKuna has quit IRC09:37
stephenfingtema: Done10:01
gtemacool, thks10:01
stephenfinIt's a -1 btw. I think some of the methods from the TagMixin don't work. Apologies if I've misunderstood something10:02
gtemano problem. For image I (I guess) it was not a complete TagMixin (some methods are also not existing)10:03
gtemaneed to think whether to override functions to raise NotImplemented or what else10:03
stephenfinMaybe we could split it into two mixins?10:03
gtemawould affect lot's of other services10:03
stephenfinOh yeah, that's mixin is pretty widely spread10:04
stephenfinCould split the query and body parameters part of the mixin into a new e.g. TagBodyMixin and have the existing TagMixin subclass that and add the methods. You could then use TagBodyMixin here10:06
stephenfinWe could audit existing uses of TagMixin to make sure they all make sense10:06
stephenfinlater, that is10:06
gtemaexcept of image others seem to implement tagging properly10:06
stephenfinYeah, I know nova does that properly10:07
stephenfinat least10:07
gtemanetwork also10:07
stephenfinWould a separate subclass work so?  TagBodyMixin isn't a great name. Maybe MinimalTagMixin or something better again10:08
gtemayou propose separate query params and attributes from methods?10:08
openstackgerritStephen Finucane proposed openstack/openstacksdk master: Demonstrate separate tag mixin
stephenfinyeah, like that ^10:09
stephenfintbc I still don't like the name 'MinimalTagMixin' but I can't think of anything better right now, but you get the idea10:10
gtemaI would say in that case it is easier to manually use query params and attribute without really using mixin10:10
stephenfinThat's fair. It doesn't really give us a lot10:10
openstackgerritArtem Goncharov proposed openstack/openstacksdk master: Add query parameters to listing heat stacks
stephenfingtema: Do we add release notes for this kind of change?10:35
gtemanot really10:35
stephenfinack, +2 so10:35
stephenfinand +W10:35
gtemawhat should we do with hypervisor patch and planned (2.88) changes?10:39
stephenfinwell they're no longer planned - they're now merged10:40
stephenfinso we could merge this as-is and then do a follow-up for 2.8810:40
stephenfinor fix now10:40
stephenfinwhatever suits you10:40
*** dtantsur|afk is now known as dtantsur10:41
gtemaidea was to prepare SDK to finilaze OSC for hypervisors10:41
gtemaso I think we can then spend some more time making SDK proper10:41
gtemaI do not really like dropped functions10:41
gtemaand uptime handling becomes really terrbile10:42
gtemaI think we need to orient on 2.88 and drop things from OSC that are not there anymore10:44
gtemaso complete disk and memory info about hypervisor disappears, right?10:45
stephenfinyes, all that info is gone10:45
gtemabut we show it in OSC now10:45
gtemaI would really just drop them from SDK/OSC not to have this mess10:46
stephenfinyeah, we'll just have to drop those fields10:46
stephenfinthat's fine by me10:46
stephenfinit's bad info10:46
gtemastatistics go away also?10:47
stephenfinyes, for the same reason (bad info)10:47
gtemadropping code is fun from that perspective10:47
gtemahow do we want to treat uptime: fetch automatically if under 2.88 or leave it optionally to user?10:50
stephenfinI think leave it optional10:51
gtemaok, agree10:51
*** ysandeep is now known as ysandeep|afk11:04
*** udesale has quit IRC11:07
*** BakaKuna has joined #openstack-sdks11:16
*** udesale has joined #openstack-sdks11:17
openstackgerritStephen Finucane proposed openstack/os-client-config stable/train: Uncap jsonschema
*** sshnaidm|afk is now known as sshnaidm|ruck11:19
stephenfindtantsur: gtema: yet another pip 20.3 fix there ^11:19
stephenfinwe can recheck once that merges11:20
openstackgerritStephen Finucane proposed openstack/os-client-config stable/stein: Uncap jsonschema
openstackgerritArtem Goncharov proposed openstack/openstacksdk master: Complete compute.hypervisor functions
openstackgerritTakashi Kajinami proposed openstack/python-openstackclient master: Volume type: Update is_public attribute
openstackgerritArtem Goncharov proposed openstack/openstacksdk master: Complete compute.hypervisor functions
*** mgariepy has quit IRC12:04
*** jpich has quit IRC12:19
*** jpich has joined #openstack-sdks12:19
*** jpich has quit IRC12:40
*** jpich has joined #openstack-sdks12:41
*** brinzhang has quit IRC13:11
*** brinzhang has joined #openstack-sdks13:11
openstackgerritJakob Meng proposed openstack/ansible-collections-openstack master: Fix volume_info result for SDK < 0.19
*** mgariepy has joined #openstack-sdks13:21
*** udesale has quit IRC13:25
*** ysandeep|afk is now known as ysandeep13:25
*** Hybrid512 has joined #openstack-sdks13:34
*** jpich has quit IRC13:34
*** jpich has joined #openstack-sdks13:35
*** nikparasyr has left #openstack-sdks13:41
*** whoami-rajat__ has joined #openstack-sdks13:46
*** enriquetaso has joined #openstack-sdks14:01
*** udesale has joined #openstack-sdks14:04
gtemadtantsur, stephenfin, slaweq, what do you think of dropping support for compute networking from SDK now. I am reworking cloud layer networking functions and nova there is disturbing a lot15:00
dtantsurhas it been dropped from nova already?15:00
gtemawe passed Ussuri (where Nova dropped support for it)15:00
dtantsurso it affects Train and older (potentially)?15:01
*** udesale has quit IRC15:01
gtemaseems to be so15:01
gtemaI am not sure somebody was recently using it in cloud layer at all15:01
gtema"These will fail with a 404 starting from microversion 2.36. They were removed in the 21.0.0 (Ussuri) release."15:02
gtemamordred, not sure whether you are here, if yes - ^^15:04
mordredI'm pretty sure we still use nova-network calls for rackspace15:06
mordredI mean - it's not actually calls ... but the server network info dict is the nova version iirc15:06
gtemareally? Oh that's sad15:06
mordredbut yeah - I'm not sure who out there might still be running an older cloud, but the theory at least was that things should still work15:07
mordredTHAT SAID15:07
gtemathat's around 1500 loc, that nobody else is really using15:07
mordredyeah - there's a bunch in here that's essentially in support of rackspace that I kept wanting to come up with a plan for dealing with better15:08
gtemaperhaps we can move it similarly to what I do for my cloud with profile?15:08
mordredit made more sense when there were two clouds and one of them was rackspace15:08
slaweqgtema: I'm ok with that, I think nova already removed all of that15:08
mordredyeah - that's sort of what I was thinking15:08
mordredmake a rackspace plugin essentially15:09
mordredexcept - doing the work for that would be tricky - especially because of the cloud-layer fallback logic15:09
gtemamordred: that's kind of initialization hack that I do15:09
gtemawhen the package is present and in clouds.yaml there is profile: otc - it get's injected and override some things15:10
mordredit might be more work to engiineer a good plugin layer that can handle the current use cases than to just maintian the existing code15:10
mordredyeah. trickiest part is not breaking the opendev nodepool while doing the work :)15:10
gtemaat least according to nova/os-floating-ip seem to be the only used stuff15:12
gtemanetworking comes from network15:12
gtema(whether it all reliable source of truth)15:12
mordredoh - you're mostly talking about the actual network calls, like list networks, list routers, etc?15:12
gtemapretty much everything in cloud15:13
mordredthe nova-network related paths in cloud layer that are important are the things in meta - like get_server_external_ipv4 and the things its calls15:13
gtemadisturbing thing is that normally we would invoke network.find_subnet, but currently we mostly go through complete list15:13
mordredI don't think any of that actually calls nova net endpoints15:13
mordredyah - we do full lists so that the caching works better and it's not a thundering herd in nodepool15:14
mordredone thread calling list subnets once and caching it and 1000 threads doing a lookup in the list rather than 1000 threads doing find_subnet15:15
mordredwhich for some of the thigns (liek servers) it could totally work to just move into the nodepool openstack driver - but for the stuff in meta it's called behind the scenes - so it's hard to defer the batching logic to nodepool15:16
mordredyeah. sorry - this is the area where the most complex use cases lurk15:16
gtemawhich meta exactly you mean?15:17
mordredbasically anything that touches is a potential nightmare to change15:17
gtemaah, yes15:17
gtemaokay. Then it is clear that we can't drop it without leaving backup for Rack (nodepool)15:18
mordredthat said - it almost certainly _should_ get reworked - it's carrying too much complexity on its shoulders. but - it's carrying too much complexity on its shoulders so reworking it without major breakages will be "fun"15:18
gtemamakes my night complex15:18
* mordred hands gtema a few beers15:18
gtemayeah, I liked US beers15:19
mordredremember when we could sit down an have a beer in the same room?15:23
gtemapuh, when can this happen again15:23
mordredyou can visit new orleans and we can have beers outside - but then you might not be allowed back in germany15:25
gtemayupp ;-)15:25
gtemalisting functions in cloud layer already return empty lists for nova case15:26
*** ricolin has quit IRC15:38
*** ricolin has joined #openstack-sdks15:47
gtemaI doubt "filters" in calls like get_subnet where name_or_id is always present are making any sense15:49
gtemaam I missing something?15:50
openstackgerritLee Yarwood proposed openstack/openstacksdk stable/train: Fix aggregate functional test for id restriction
mordredfilters can additionally filter - for cases where the only thing that's passed to name_or_id is name?15:52
mordred(subnet names are not guaranteed to be unique iirc)15:52
gtemaaaah, right15:52
dtantsurbeers with people.. I seem to be vaguely remembering..15:53
openstackgerritStephen Finucane proposed openstack/os-client-config stable/train: Uncap jsonschema
*** ysandeep is now known as ysandeep|out16:12
*** khomesh24 has quit IRC16:16
*** mgariepy has quit IRC16:20
*** khomesh24 has joined #openstack-sdks16:31
*** slaweq has quit IRC16:34
*** slaweq has joined #openstack-sdks16:36
*** tosky has quit IRC16:41
*** tosky has joined #openstack-sdks16:42
*** khomesh24 has quit IRC17:08
*** mgariepy has joined #openstack-sdks17:20
*** jpich has quit IRC17:22
*** Hybrid512 has quit IRC17:27
*** sshnaidm|ruck is now known as sshnaidm|afk17:30
*** gtema has quit IRC17:46
stephenfindtantsur: If you're still about, could you push this through? gtema approved earlier but there were more issues
stephenfinCI is green now17:50
* dtantsur looks17:50
*** rpittau is now known as rpittau|afk18:05
*** gtema has joined #openstack-sdks18:08
*** ralonsoh has quit IRC18:20
*** dtantsur is now known as dtantsur|afk18:43
openstackgerritArtem Goncharov proposed openstack/openstacksdk master: Switch networking function in cloud layer to proxy
*** gtema has quit IRC18:59
openstackgerritMerged openstack/os-client-config stable/train: Uncap jsonschema
*** whoami-rajat__ has quit IRC19:55
*** nightmare_unreal has quit IRC20:16
*** slaweq has quit IRC20:41
*** BakaKuna has quit IRC22:52
*** brinzhang_ has joined #openstack-sdks23:02
*** brinzhang has quit IRC23:05
rm_workanyone here familiar with gophercloud + openstack? would love to snipe someone for just a minute to ask a couple questions about passing mTLS data with connections23:38
mordredrm_work: I know that it exists and people use it23:55

Generated by 2.17.2 by Marius Gedminas - find it at!