| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2018-05-23 | |||
| 10:45:09 | openstackgerrit | tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386 | |
| 15:37:55 | frickler | mordred: seems this bug still affects ocata+pike stable branches, would you have time to look into backporting? https://storyboard.openstack.org/#!/story/1733258 | |
| 15:38:19 | frickler | side note: would be interesting to discuss how we should track backports in storyboard | |
| #openstack-sdks - 2018-05-24 | |||
| 00:41:30 | openstackgerrit | Merged openstack/openstacksdk master: Added few image properties to Image class https://review.openstack.org/544049 | |
| 00:41:30 | openstackgerrit | Merged openstack/openstacksdk master: Decode additional heat files https://review.openstack.org/567648 | |
| 05:54:23 | openstackgerrit | tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386 | |
| 11:37:13 | openstackgerrit | tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386 | |
| 11:54:24 | openstackgerrit | tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386 | |
| 12:00:16 | dtantsur | mordred: hey! is my impression right that things exposed through proxies in openstacksdk (connection.compute.* etc) are a bit more low-level than things on connection itself? | |
| 12:01:19 | gtema | dtansur: yes, you are right. Sometimes even much more low-level. They are basically the "basic rest calls" wrappers mostly | |
| 12:03:40 | dtantsur | I see, it makes sense | |
| 12:03:49 | dtantsur | so, this is intended, right? | |
| 12:04:33 | gtema | well, yes. The connection methods are then giving a higher level of abstraction. So to say "function" | |
| 12:09:34 | openstackgerrit | tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386 | |
| 12:29:34 | gtema | hey SDK/shade/ansible cores: I've found out, that May 25. is the community freeze for ansible 2.6 (https://docs.ansible.com/ansible/devel/roadmap/ROADMAP_2_6.html). If we want to land shade retirement (see https://github.com/ansible/ansible/pull/40532) - it's really time to review the PR. Otherwise we might be late for 2.6 | |
| 15:38:47 | mordred | gtema: awesome! I just started looking at it last night - looks great - I should be ale to actually look today | |
| 15:46:07 | gtema | mordred: great. I just found, that in the meanwhile new module was added: os_server_metadata. I am fighting with rebasing. If it doesn't work we can still merge PR and then additional PR to switch new module (is not a conflicting change) | |
| 15:53:07 | mordred | gtema: yah - I think merge one and then do an additional is fine if the conflicts suck too much | |
| 15:53:53 | dtantsur | cdent: hey, is there a meeting today? | |
| 15:54:54 | cdent | dtantsur: I reckon not. I also reckon we forgot to announce that there's not. If you're inclined to show up in case any one else does, that would probably be useful | |
| 15:55:08 | mordred | gtema: there's one more thing we should do that's an issue - I think someone maybe started looking at it already ... (also, I can hopefully help a bit -today isn't as busy) | |
| 15:55:24 | dtantsur | cdent: I can be here, but if everyone else is on the Forum... | |
| 15:57:30 | mordred | gtema: which is any place using cloud.nova_client we need to replace those calls since sdk doesn't have novaclient connnections | |
| 15:57:55 | mordred | it's mostly going to be lib/ansible/modules/cloud/openstack/os_server_action.py | |
| 15:59:21 | mordred | I think we need to add a few calls to sdk - but in the mean time the action calls are pretty easy rest calls, so we can just use the rest passthrough and then in 2.6 replace the rest calls with proper sdk method calls | |
| 15:59:35 | gtema | mordred: ooh, yes | |
| 15:59:52 | mordred | oh - nevermind: https://github.com/ansible/ansible/pull/40462 | |
| 15:59:53 | gtema | so let's decide what to do now | |
| 16:00:45 | gtema | ok, hopefully would be no merge conflict | |
| 16:01:48 | mordred | ah - that was just one of the files - still need to do the actions one - lemme see if I can crank out a patch real quick | |
| 16:02:00 | gtema | ok | |
| 16:06:38 | mordred | gtema: http://paste.openstack.org/show/721908 it probably needs to go into your patch, since we need the code to use the sdk and not just shade for it to work properly | |
| 16:07:17 | mordred | gtema: that patch is the first one - but it should e pretty easy to do the rest - the url call is the same, https://developer.openstack.org/api-ref/compute/#servers-run-an-action-servers-action has the payloads needed for the others | |
| 16:07:40 | mordred | gtema: (I can make the rest if you want - whichever you think it easier) | |
| 16:09:35 | gtema | mordred: please make them, since it's an evening here in Germany and my wife will shortly start blaming ;-) | |
| 16:11:55 | mordred | :) | |
| 16:12:36 | mordred | ok. I think maybe let's just land your PR now - and we can do the nova_client cleanup as a followup as a bugfix | |
| 16:12:37 | gtema | I assume for servers.start it's json={'os-start' | |
| 16:12:55 | mordred | http://paste.openstack.org/show/721909 | |
| 16:13:03 | mordred | gtema: I went ahead and did the whole set ^^ | |
| 16:13:21 | gtema | great, PR will be updated in couple of minutes | |
| 16:13:33 | mordred | sweet. thanks - and thanks for doing the PR - it looks great! | |
| 16:13:44 | gtema | thanks | |
| 16:17:14 | gtema | mordred: it's there | |
| 16:19:36 | gtema | mordred: thanks for helping | |
| 16:31:24 | mordred | gtema: found one more if you're still around: http://paste.openstack.org/show/721911 | |
| 16:31:58 | gtema | yes, still here | |
| 16:37:31 | gtema | mordred: ok, rebase completed and new module os_server_metadata is also part of the change | |
| 16:37:57 | mordred | sweet. thanks! | |
| 16:41:26 | gtema | mordred: another nova_client usage: os_server: L543 - remove_floating_ip | |
| 16:42:32 | mordred | gtema: yah - there's a pr up for that one | |
| 16:42:46 | mordred | gtema: I'm gonna see if I can get all of these in today | |
| 16:43:18 | gtema | ok | |
| 16:43:53 | gtema | right, missed that already. Code-search | |
| 16:46:14 | gtema | mordred: do you know if one can "recheck" in github ansible - 3 last jobs were cancelled due to the following changes and now it stucked | |
| 17:06:34 | gtema | mordred: I am off for today. If further changes are necessary - just leave a note. I will cover them tomorrow morning ASAP. | |
| #openstack-sdks - 2018-05-25 | |||
| 00:09:03 | harlowja | did shade ever release with the fix to http://paste.openstack.org/show/721925/ ? | |
| 00:10:09 | harlowja | https://review.openstack.org/#/c/564861/ i think was the fix | |
| 00:10:14 | harlowja | but doesn't ever seem to have gotten released? | |
| 01:29:00 | openstackgerrit | tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386 | |
| 02:45:51 | openstackgerrit | tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386 | |
| 06:18:22 | openstackgerrit | tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386 | |
| 08:55:39 | openstackgerrit | Dmitry Tantsur proposed openstack/openstacksdk master: Several improvements to resource.wait_for_status https://review.openstack.org/570538 | |
| 09:16:48 | openstackgerrit | Dmitry Tantsur proposed openstack/openstacksdk master: Several improvements to resource.wait_for_status https://review.openstack.org/570538 | |
| 12:00:54 | openstackgerrit | Dmitry Tantsur proposed openstack/openstacksdk master: Several improvements to resource.wait_for_status https://review.openstack.org/570538 | |
| 14:34:07 | gtema | mordred: any progress on https://github.com/ansible/ansible/pull/40532? ansible-dev is already buzzing around todays dev-freeze | |
| 16:10:43 | openstackgerrit | Harald Jensås proposed openstack/python-openstackclient master: Allow setting network-segment on subnet update https://review.openstack.org/523974 | |
| 17:18:24 | openstackgerrit | Artem Goncharov proposed openstack/openstacksdk master: allow passing ansible variables to ansible tests https://review.openstack.org/570600 | |
| 17:20:45 | mordred | gtema: up and working on it now | |
| 17:31:55 | gtema | mordred: aren't sdk.OpenStackCloudException sdk.exceptions.OpenStackCloudException in ansible PR? | |
| 17:34:57 | mordred | gtema: wait - are they? I just tried importng locally | |
| 17:35:01 | mordred | gtema: >>> import openstack | |
| 17:35:03 | mordred | >>> openstack.OpenStackCloudException | |
| 17:35:08 | mordred | AttributeError: module 'openstack' has no attribute 'OpenStackCloudException' | |
| 17:35:30 | mordred | gtema: it's possible I'm not using my brain properly though | |
| 17:35:37 | gtema | exactly because of that I did that | |
| 17:36:00 | gtema | I tried also small POC and sdk.OpenStackCloudException was not working | |
| 17:38:12 | gtema | I was verifying by testing os_auth and trying a bad credentials | |
| 17:38:56 | mordred | gtema: yah ... so the ansible modules shoul duse sdk.exceptions.OpenStackCloudException yes? | |
| 17:39:26 | gtema | mordred: yupp | |
| 17:40:38 | mordred | cool - do you have time to fix the places that are using sdk.OpenStackCloudException ? Or we could just do it as a followup if you don't | |
| 17:40:59 | gtema | where have you found direct use? | |
| 17:41:36 | gtema | I see, os_itonic | |
| 17:41:38 | gtema | noing now | |
| 17:41:42 | mordred | gtema: I left inline comments on the PR - if you search for 'sdk.Open' it'll show them | |
| 17:41:44 | mordred | sweet | |
| 17:43:23 | gtema | mordred: went out | |
| 17:46:16 | mordred | gtema: crap. sorry - I just landed a pr that made a conflict with yours. can you rebase? | |
| 17:47:09 | gtema | yes, I see it already | |
| 17:47:12 | gtema | working | |
| 17:48:27 | gtema | so I should just remove this _get_aloowed_project, right? | |
| 17:48:47 | mordred | yah | |
| 17:51:58 | gtema | mordred: ready | |
| 17:59:45 | mordred | \o/ | |
| 18:03:31 | gtema | mordred: crap, again shippable stucked | |
| 18:04:16 | mordred | gtema: yah - I just asked how to re-trigger it | |
| 18:04:45 | gtema | as I know either admin retriggers it or close-open PR | |
| 18:04:54 | mordred | ugh | |
| 18:06:27 | gtema | mordred: should I close-open? | |
| 18:06:56 | mordred | gtema: yah - let's try that | |