| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2018-05-30 | |||
| 14:52:41 | openstackgerrit | Merged openstack/cliff master: update cmd2 dependency to handle py3 only versions https://review.openstack.org/571044 | |
| 15:34:48 | mordred | haleyb, frickler: well, it's not a neutronclient regression since osc doesn't use neutronclient - but yeah, we should make sure that openstacksdk is throwing an exception that actually has the server-side error message (there is code in sdk to do this, but since error content isn't consistent it might not be doing the right thing for neutron) | |
| 15:37:25 | haleyb | mordred: so does that look like something in osc, or the sdk, or ? | |
| 15:38:17 | openstackgerrit | Lance Bragstad proposed openstack/python-openstackclient master: Add system role functionality https://review.openstack.org/524416 | |
| 15:39:14 | mordred | haleyb: not 100% sure - it'll take some investigation to be sure- just wanted to bring up that it's possible the sdk isn't giving osc enough data, so improvements might need to be made in either or both | |
| 16:02:00 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Add ansible functional tests on stable-2.6 https://review.openstack.org/571240 | |
| 16:10:54 | frickler | mordred: it's a regression for users moving from neutron cli to OSC, not a regression within osc itself, but still very important I think. if you have a better term for that kind of situation, feel free to propose a different tag | |
| 16:11:04 | mordred | ah- yes | |
| 16:11:20 | mordred | that makes sense to me - I thought the earlier suggestion was that it was a regression in neutronclient | |
| 16:11:37 | lbragstad | o/ curious if anyone would be interested in reviewing a python-openstackclient patch https://review.openstack.org/#/c/524416/5 | |
| 17:09:03 | openstackgerrit | Merged openstack/openstacksdk master: Add ansible functional tests https://review.openstack.org/570678 | |
| 17:34:59 | openstackgerrit | Maksim Malchuk proposed openstack/cliff master: Don't install cmd2>=0.9 https://review.openstack.org/570817 | |
| 19:56:26 | openstackgerrit | boden proposed openstack/python-openstackclient master: DNM: test updated segment API description https://review.openstack.org/571293 | |
| #openstack-sdks - 2018-05-31 | |||
| 06:59:48 | Luzi | Hej, I would like to contribute to osc. But I am new and a little bit confused: where can I submit a blueprint? Is it still Launchpad or is it Storyboard? | |
| 07:42:55 | openstackgerrit | Jens Harbott (frickler) proposed openstack/python-openstackclient master: Fix lower-constraints.txt https://review.openstack.org/571397 | |
| 11:30:54 | frickler | dtroyer: amotoki: dhellmann: https://review.openstack.org/571397 fixes the reqs check for e.g. https://review.openstack.org/524416 . and then I still have some patches waiting for multiple months now, 516701, 558824, 559742, 540633 | |
| 13:00:37 | elmiko | edleafe dtantsur|afk heads up, i have a conflict today and won't be able to make the meeting | |
| 13:03:01 | edleafe | elmiko: guess who's getting assigned all the action items! | |
| 13:04:35 | Luzi | edleafe: where is the meeting held? | |
| 13:05:35 | elmiko | edleafe: sounds fair | |
| 13:05:43 | edleafe | Luzi: in #openstack-meeting-3 at 1600UTC | |
| 13:06:10 | Luzi | edleafe: thank you | |
| 13:16:49 | Luzi | edleafe: may I ask, how to contribute to osc? I was at the summit last week. I attended the upstream institute and several onboarding sessions. But it seems to be different for OSC. | |
| 13:17:45 | Luzi | I would like to write a spec (or something similar). Where can I do this? | |
| 13:23:29 | frickler | Luzi: sadly there isn't much activity in osc currently. I'd suggest you start simple and just write up your plan on an etherpad (just create one on etherpad.openstack.org) and then post the link here | |
| 13:23:57 | Luzi | frickler: ok I will do that. Tank you :-) | |
| 13:25:22 | frickler | Luzi: I'm assuming that you did find https://blueprints.launchpad.net/python-openstackclient and checked for possible duplicates? though I expect that list to be pretty outdated | |
| 13:26:32 | Luzi | frickler: yes, i have checked this | |
| 13:33:50 | frickler | Luzi: another approach to get started on OSC would be to take a look at the current bug reports/stories at https://storyboard.openstack.org/#!/project/975 and see if you find something that you would like to tackle. again I admit that that list is pretty long sadly | |
| 14:05:10 | Luzi | hi everyone: as frickler proposed, i've created an etherpad: https://etherpad.openstack.org/p/osc-included_image_signing | |
| 14:45:09 | openstackgerrit | Merged openstack/openstacksdk master: Modify the error message when unsetting gateway and setting FIP https://review.openstack.org/571101 | |
| 14:45:10 | openstackgerrit | Merged openstack/openstacksdk master: Add 'port_details' to Floating IP https://review.openstack.org/533811 | |
| 14:57:29 | openstackgerrit | Merged openstack-infra/shade master: Update ansible test job to run against stable-2.5 https://review.openstack.org/570680 | |
| 16:17:53 | openstackgerrit | Merged openstack/python-openstackclient master: Fix lower-constraints.txt https://review.openstack.org/571397 | |
| 16:42:52 | openstackgerrit | Doug Hellmann proposed openstack/cliff master: support cmd2 0.9.1 in interactive mode https://review.openstack.org/571524 | |
| 20:12:38 | openstackgerrit | Lance Bragstad proposed openstack/python-openstackclient master: Add system role functionality https://review.openstack.org/524416 | |
| #openstack-sdks - 2018-06-01 | |||
| 04:00:22 | openstackgerrit | yanpuqing proposed openstack/openstacksdk master: Modify the unhelpful error message when delete network https://review.openstack.org/571629 | |
| 06:51:38 | openstackgerrit | yanpuqing proposed openstack/openstacksdk master: Modify the unhelpful error message when delete network https://review.openstack.org/571629 | |
| 09:50:24 | odyssey4me | folks, I have some ansible 2.5 patches related to openstack integration ready to merge for anyone who has that capability: https://github.com/ansible/ansible/pull/40766 / https://github.com/ansible/ansible/pull/40763 / https://github.com/ansible/ansible/pull/40765 / https://github.com/ansible/ansible/pull/40841 | |
| 09:51:05 | odyssey4me | not sure if 2.4 will accept documentation patches, but I have these up too: https://github.com/ansible/ansible/pull/40842 / https://github.com/ansible/ansible/pull/40843 | |
| 10:50:51 | slaweq | hi OSC cores, can You take a look at https://review.openstack.org/#/c/567620/ ? | |
| 10:51:20 | slaweq | and https://review.openstack.org/#/c/567828/ (this one has already one +2 even) | |
| 12:10:43 | openstackgerrit | Jens Harbott (frickler) proposed openstack/python-openstackclient master: Add system role functionality https://review.openstack.org/524416 | |
| 12:48:37 | openstackgerrit | Jens Harbott (frickler) proposed openstack/python-openstackclient master: Add system role functionality https://review.openstack.org/524416 | |
| 13:30:11 | openstackgerrit | Artem Goncharov proposed openstack/openstacksdk master: Add ansible functional tests on stable-2.6 https://review.openstack.org/571240 | |
| 21:00:23 | openstackgerrit | Merged openstack/python-openstackclient master: Make max_burst_kbps option as optional for bw limit QoS rule https://review.openstack.org/567828 | |
| 21:00:25 | openstackgerrit | Merged openstack/python-openstackclient master: Add cliff project link https://review.openstack.org/570150 | |
| 21:15:45 | openstackgerrit | Merged openstack/python-openstackclient master: Prevent "server migrate --wait" from hanging https://review.openstack.org/567289 | |
| #openstack-sdks - 2018-06-03 | |||
| 16:14:09 | openstackgerrit | Merged openstack/openstacksdk master: Modify the unhelpful error message when delete network https://review.openstack.org/571629 | |
| 17:09:22 | openstackgerrit | Merged openstack/openstacksdk master: rename vars to defaults to allow overriding in ansible tests https://review.openstack.org/570751 | |
| 20:55:19 | openstackgerrit | Merged openstack/os-api-ref master: Raise warning if top-level value is not a dict https://review.openstack.org/563178 | |
| #openstack-sdks - 2018-06-04 | |||
| 15:31:53 | vadim | Using novaclient or os_client_config I can filter compute servises like http://paste.openstack.org/show/722663/ Can I filter compute.services() using openstacksdk? | |
| 20:09:22 | openstackgerrit | Lance Bragstad proposed openstack/python-openstackclient master: Add system role functionality https://review.openstack.org/524416 | |
| #openstack-sdks - 2018-06-05 | |||
| 00:08:28 | openstackgerrit | Merged openstack/openstacksdk master: Add ansible functional tests on stable-2.6 https://review.openstack.org/571240 | |
| 01:27:19 | openstackgerrit | wangxiyuan proposed openstack/python-openstackclient master: [WIP] Add unified limit CLIs https://review.openstack.org/572238 | |
| 02:11:24 | openstackgerrit | chenyb4 proposed openstack/openstacksdk master: Add compute server migrate support host parameter https://review.openstack.org/572240 | |
| 13:08:55 | openstackgerrit | Chuck Short proposed openstack/osc-lib master: Switch to using stestr https://review.openstack.org/572380 | |
| 22:12:43 | cdent | elmiko, dtantsur|afk : neither edleafe nor I are likely to be available on thursday for meeting. Worth doing or wanna skip? | |
| 22:19:26 | elmiko | cdent: i will be around, but i'm fine with cancelling this week. as a heads up, i will be ok next week. | |
| 22:19:35 | elmiko | s/ok/out/ | |
| 22:19:57 | cdent | please be ok next week too | |
| 22:20:02 | cdent | being ok is good | |
| 22:20:44 | cdent | unless you don't feel like it, maybe run it and see who shows up? | |
| 22:21:01 | cdent | I may be on a train by then (travelling home on thursday), so might be able to join | |
| 23:31:34 | elmiko | cdent: ok, we can just wing it. no worries | |
| 23:31:53 | elmiko | and i will definitely be ok as well, i will be on pto in bavaria =D | |
| 23:35:23 | cdent | nice | |
| #openstack-sdks - 2018-06-06 | |||
| 03:53:49 | openstackgerrit | tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386 | |
| 03:56:01 | openstackgerrit | tianhui proposed openstack/python-openstackclient master: Compute: Add tags support for server https://review.openstack.org/569386 | |
| 06:22:36 | openstackgerrit | chenge proposed openstack/python-openstackclient master: Replace Chinese quotes to English quotes https://review.openstack.org/572679 | |
| 08:21:16 | openstackgerrit | Catalin Popescu proposed openstack/openstacksdk master: Repair evacuate and migrate https://review.openstack.org/572693 | |
| 09:56:58 | openstackgerrit | Harald Jensås proposed openstack/python-openstackclient master: Allow setting network-segment on subnet update https://review.openstack.org/523974 | |
| 13:40:00 | openstackgerrit | Chen proposed openstack/python-openstackclient master: Fix a tyo https://review.openstack.org/572771 | |
| 14:44:15 | openstackgerrit | Jens Harbott (frickler) proposed openstack-infra/shade master: WIP: Test job based on devstack-new-neutron https://review.openstack.org/552459 | |
| 18:01:23 | mordred | dtantsur: I like https://review.openstack.org/#/c/570934 - but just as a quick fyi, we will still need the retry logic where it sits in sdk so that client-side rate-limiting works proparly | |
| 18:01:47 | dtantsur | mordred: hmmm | |
| 18:02:02 | dtantsur | cannot we do this rate limiting in Session as well? | |
| 18:02:17 | dtantsur | it feels more like a property of the HTTP client implementation rather than something specific to the SDK.. | |
| 18:02:25 | mordred | dtantsur: we can - but it's a much larger thing that kmalloc and I have each failed at doing at least once | |
| 18:02:26 | dtantsur | and I can imagine not only openstacksdk using it | |
| 18:02:53 | mordred | dtantsur: I still think we should do it - just wanted to warn we wouldn't be able to immediately make use of the ksa patch in sdk until we do | |
| 18:03:01 | dtantsur | mordred: do we have this rate-limiting now? it seems to me that at least baremetal requests just go through.. | |
| 18:04:16 | mordred | dtantsur: well... we have support for it, but the class that implements it is currently sitting in nodepool. I've been planning on moving the nodepool rate-limiting taskmanager into the sdk tree so that the whole picture is present (this plays in to the caching story too) | |
| 18:04:44 | mordred | dtantsur: so - it's a thing we support, but you woulnd't know it by looking at the code that's there - which needs to be fixed | |
| 18:04:56 | dtantsur | much complexity not wow | |
| 18:05:20 | dtantsur | I guess "just switch from ironicclient to openstacksdk" is not as easy as it seemed initially :) | |
| 18:05:36 | mordred | dtantsur: sure it is :) | |
| 18:06:02 | mordred | dtantsur: http://git.openstack.org/cgit/openstack-infra/nodepool/tree/nodepool/task_manager.py#n65 is the TaskManager that needs to be added to the openstacksdk tree - nodepool currently passes it in to the sdk connection constructor | |
| 18:06:05 | dtantsur | I wonder if it's a problem for other services, but ironic better has retries | |
| 18:06:08 | mordred | everyone else gets the direct-action taskmanager | |
| 18:06:26 | dtantsur | aha, taskmanagers is something I haven't figured out yet, to be honest | |
| 18:07:04 | mordred | you can mostly ignore them - until this - the one that's on by default just executes the calls when you make them | |
| 18:09:42 | dtantsur | mordred: do these work only in the former shade bits or in the other half as well? | |
| 18:09:51 | dtantsur | since connection.baremetal things seem to just use the session | |
| 18:09:59 | dtantsur | (unless I'm completely clueless, which is not unlikely) | |
| 18:10:00 | mordred | the other half - the way it's connected through is that it happens on the adapter calls | |
| 18:10:46 | mordred | dtantsur: this next thing I'm going to say will make you table flip - but the sessions laying around are all actually adapters - we need a gloal search-and-replace | |
| 18:11:37 | mordred | but - the Adapter created for each service by the sdk puts submits every external rest call into the taskmanager (which by default just executes it right then) - so if you did use a different taskmanager it would apply to any calls made on the adapter | |
| 18:12:05 | mordred | :) | |
| 18:12:13 | dtantsur | okay, so MAAAGIIIC happens, and it just works somewhere inside, right? | |