Earlier  
Later
New Search
Posted Nick Remark
#openstack-sdks - 2019-01-03
16:01:50 elmiko dtantsur: quiet and relaxing =)
16:02:11 elmiko how about for you?
16:02:27 dtantsur yeah, ditto :) a bit of travelling around, a bit of calm drinking at home :)
16:02:41 elmiko nice, i'm all for calm drinking ;)
16:02:47 dtantsur major version bump to /dtantsur/v3.0 :)
16:02:57 elmiko \o/ congrats!
16:03:10 dtantsur thnx :) I hope it's a good thing :D
16:03:18 elmiko haha, ++
16:05:02 dtantsur so, wanna have a discussion or we better just relax (and wait for someone)? :)
16:05:49 elmiko depends on the conversation
16:06:05 elmiko looks like maybe edleafe is not here
16:07:36 elmiko dtantsur: did you want to have a discussion about sig stuff?
16:08:11 dtantsur well, I have some API related thoughts.. or we can talk some SIG business (API vs SDK)
16:08:46 elmiko i'm game for either, might have any conclusive answers but we can talk it through
16:08:54 elmiko *might not
16:12:05 dtantsur so, we have /v1/nodes. Now I'm introducing a resource that has 1-1 mapping with nodes. Do we have any opinion on it being /v1/nodes/<node>/something, /v1/something (with a node_uuid), both, neither?
16:12:52 elmiko cdent: you around? i feel like you might have some good thoughts about ^^
16:13:00 cdent oh hi
16:13:29 elmiko i don't think we have an official opinion, i think my personal preference would be to create the route hierarchy if it makes sense to decompose the components like that
16:14:01 elmiko i would think something like /v1/nodes/<node id>/something
16:14:28 cdent it depends on the meaning of something. If <something> is a part of a (a sub thing) node, then yeah, hierarchy
16:14:55 cdent but if something is simply something that just happens to have a node id in it, maybe not
16:15:32 cdent I tend to like sub resources, where there is a notiion of containership
16:15:47 elmiko good point about the something containing the node id within it
16:16:26 elmiko dtantsur: will these "somethings" only ever be accessed in relation to the node they are associated with or can they be operated on independently?
16:17:24 dtantsur so, we're talking about a semi-virtual "deployment" resource for a node. it may not be even completely independent on the database level.
16:17:56 dtantsur I'm not sure how much sense it makes to talk about a deployment without the node
16:18:39 elmiko i guess would /v1/deployments/<id> make any sense?
16:24:23 cdent /v1/nodes/<node id>/deployments/<deployment id> sounds vaguely correct, or maybe deployment$ if there's only ever one
16:24:53 cdent presumably this thing only exists when something is in a state of deployed?
16:37:47 mordred I agree with everything all of you have said
16:40:07 elmiko ++ more cookies for mordred XD
16:40:49 cdent elmiko: on things, I went back to healy's a bit before xmas and got the cider that was aged in bourbon casks. double plus good
16:41:12 mordred mmm. cookie
16:41:13 elmiko yay!
16:41:21 elmiko cdent: that place is too much fun
16:41:50 cdent met a nice cat
16:42:24 elmiko some friends spent the holidays in cornwall, they knew i had went to healy's and sent me a photo of them all drinking pints of the rattler. it was awesome
16:42:51 cdent excellent
16:43:06 elmiko hmm, so pints of cider + meeting a nice cat, sounds like a win!
16:43:20 cdent we did not have the most inspiring weather for "holidays in cornwall"
16:43:24 cdent was damp and mild
16:43:31 elmiko bummer =(
16:43:32 cdent which is just no fun in either direction
16:43:47 elmiko right, plus i imagine it kills the beach trips
16:46:26 cdent been in the sea three times in the last week or so
16:47:23 elmiko sounds chilly!
17:01:13 elmiko i'm stepping out for some lunch, dtantsur i know it's late over there but if you want to talk more i should be around
17:01:29 dtantsur sure, I forgot to say thanks all :)
17:01:36 elmiko np =)
17:01:37 dtantsur I think I'm good for now
17:01:46 elmiko awesome, good luck!
17:08:56 dtantsur \o
#openstack-sdks - 2019-01-04
03:38:13 openstackgerrit Martin Chlumsky proposed openstack/python-openstackclient master: Allow "server migrate" (not live) to take "--host" option https://review.openstack.org/628334
14:18:39 mordred cdent: it occurs to me that you should have chosen "placemat" as the codename for placement - because it wouldn't have made typing terrible at all
14:19:03 cdent i know how that can be
14:25:56 mordred cdent: it's all his fault isn't it?
14:26:20 cdent meh, he had a good idea and then we all ran with it and ruined it
16:01:11 elmiko mordred: LOL ++ for placemat
16:21:45 dtantsur ++
16:42:57 openstackgerrit weizj proposed openstack/os-api-ref master: Update hacking version to latest https://review.openstack.org/628593
16:43:00 openstackgerrit weizj proposed openstack/os-client-config master: Update hacking version to latest https://review.openstack.org/628597
16:43:03 openstackgerrit weizj proposed openstack/os-service-types master: Update hacking version to latest https://review.openstack.org/628602
16:57:02 elmiko placements have sufficient resources this year.
16:57:12 elmiko cdent: i loved that line!
16:57:32 cdent i figured something was required
16:57:40 elmiko ++
#openstack-sdks - 2019-01-06
10:12:55 openstackgerrit weizj proposed openstack/os-api-ref master: Update hacking version https://review.openstack.org/628593
10:51:27 openstackgerrit weizj proposed openstack/os-service-types master: Update hacking version https://review.openstack.org/628602
11:06:20 openstackgerrit weizj proposed openstack/os-service-types master: Update hacking version https://review.openstack.org/628602
16:12:45 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Compute location properly in server https://review.openstack.org/626380
16:13:04 mordred dtantsur|afk: I know how much you like patches that touch guts ^^
18:35:59 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Compute location properly in server https://review.openstack.org/626380
#openstack-sdks - 2019-01-07
05:08:19 openstackgerrit Clint 'SpamapS' Byrum proposed openstack-infra/shade master: Split parser creation and parser for inventory https://review.openstack.org/628859
09:13:20 openstackgerrit Merged openstack/os-service-types master: Update hacking version https://review.openstack.org/628602
10:09:32 dtantsur mordred: of course :)
11:40:38 openstackgerrit Clint 'SpamapS' Byrum proposed openstack-infra/shade master: Split parser creation and parser for inventory https://review.openstack.org/628859
12:11:12 openstackgerrit Johannes Kulik proposed openstack/python-openstackclient master: Add floating IP filter to floating IP list command https://review.openstack.org/628941
12:56:02 openstackgerrit Johannes Kulik proposed openstack/openstacksdk master: Enable pagination for Floating IPs https://review.openstack.org/628950
13:48:12 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Compute location properly in server https://review.openstack.org/626380
13:48:13 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Skip block-storage tests when service is not found https://review.openstack.org/628969
14:11:35 openstackgerrit Johannes Kulik proposed openstack/openstacksdk master: Fix pagination key detection https://review.openstack.org/628977
14:12:13 openstackgerrit Merged openstack-infra/shade master: Split parser creation and parser for inventory https://review.openstack.org/628859
14:24:37 jkulik mordred, regarding your comment on https://review.openstack.org/#/c/628950/1 There are a lot of other functions setting paginated=False and thus a lot of tests are failing when changing the default. Should these functions also be changed?
14:25:39 jkulik If I understood you correctly, it should not be necessary for anyone to set the paginated parameter anymore.
14:26:38 mordred jkulik: yes - that's the intent - the parameter should really only be needed as a safety valve in case some service implements pagination in so broken a way that it needs to be turned off
14:27:38 mordred jkulik: so yeah - changing those functions would be awesome - but if the carnage becomes too great, we can also just land your first patch and do the cleanup as a followup
14:28:07 jkulik Should be fine. Thanks for the guidance.
14:45:47 mordred jkulik: thank you!
15:41:49 openstackgerrit Sorin Sbarnea proposed openstack/openstacksdk master: DNM: testing cp --reflink https://review.openstack.org/629002
15:44:13 dmsimard dtroyer: not sure if you noticed the ping last week about osc-lib: http://eavesdrop.openstack.org/irclogs/%23openstack-sdks/%23openstack-sdks.2019-01-02.log.html#t2019-01-02T20:49:17
17:07:57 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Skip block-storage tests when service is not found https://review.openstack.org/628969
17:07:58 openstackgerrit Monty Taylor proposed openstack/openstacksdk master: Compute location properly in server https://review.openstack.org/626380
18:05:59 dtroyer dmsimard: I've been away from everything for two weeks, just getting through things now…
18:16:01 dmsimard dtroyer: expected as much, hence the re-ping -- no worries :D
18:16:03 dtroyer cdent: mmmmmm, thanks!
18:16:16 dtroyer cdent: Hoping you had some enjoyable time off
18:17:18 cdent dtroyer: thanks, was nice, but not long enough
18:17:39 dmsimard cdent: I would've taken an extra week for sure
18:18:17 dtroyer we took thre days last week and hid from everyone, that was our vacation-from-vacationing :)

Earlier   Later