| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2017-01-04 | |||
| 17:46:42 | openstackgerrit | Merged openstack/python-openstackclient: Show team and repo badges on README https://review.openstack.org/403362 | |
| 17:56:08 | openstackgerrit | yuval shalev proposed openstack/python-openstacksdk: Added project role assignment https://review.openstack.org/416598 | |
| 17:57:45 | openstackgerrit | Dean Troyer proposed openstack/osc-lib: Include OSC additions https://review.openstack.org/416693 | |
| 18:14:13 | openstackgerrit | Steve Martinelli proposed openstack/python-openstackclient: Fixes image api URL endpoint for certain scenario https://review.openstack.org/415301 | |
| 18:26:15 | openstackgerrit | Jordan Pittier proposed openstack/python-openstackclient: Add support for Glance 'update image members' feature https://review.openstack.org/410848 | |
| 18:27:53 | stevemar | dtroyer: bah, that image api one needs a rel note i think | |
| 18:54:38 | stevemar | dtroyer: i don;t think https://review.openstack.org/#/c/371709/ is quite ready yet, i think knikolla was looking for help on adding options to the CLI | |
| 18:59:46 | madorn | anyone know if we are planning to add support for modifying swift headers (not just metadata) on swift containers with python-openstackclient? what about read/write acl's ? | |
| 19:03:07 | stevemar | madorn: its not on our radar, but we wouldn't be against it if someone tossed up a patch | |
| 19:03:13 | stevemar | *hint hint* ;) | |
| 19:07:35 | madorn | stevemar: cool!! | |
| 19:36:21 | dtroyer | stevemar: re 371709, is the osc-lib portion going to change? If it likely is not, we should merge it to get it in the release before the freeze | |
| 19:37:21 | dtroyer | madorn: we'd totally love to have it done, it just isn't at the top of the queue. If you do tackle that, let's discuss what the CLI options look like before you get too far down the road and would have to do a bunch of rework... | |
| 19:38:00 | stevemar | dtroyer: not sure how it's all structured *shrugs* | |
| 19:39:02 | dtroyer | ok, I'm just trying to look at what we're going to be handcuffed with for 4-6 weeks again | |
| 22:14:29 | openstackgerrit | Dean Troyer proposed openstack/osc-lib: Include OSC additions 2 https://review.openstack.org/416747 | |
| 22:21:06 | meteorfox | Hi everyone. I'm not sure if this is a known issue. I'm using python-openstackclient==3.6.0 and it seems 'openstack subnet show <name_or_id>' command is broken. It returns "NoneType object is not iterable" as far as I can see in the debug, Neutron returns a correct response | |
| 22:39:30 | dtroyer | meteorfox: can you check the version of openstacksdk you have installed? it should be 0.9.10 | |
| 22:42:24 | meteorfox | it returned '0.9.12' | |
| 22:48:17 | meteorfox | I downgraded that package to 0.9.10 and at least that command works | |
| 23:05:25 | openstackgerrit | Merged openstack/python-openstackclient: Fixes image api URL endpoint for certain scenario https://review.openstack.org/415301 | |
| 23:08:23 | dtroyer | meteorfox: oh, new release… probably didn't fix the problems that caused 0.9.11 to be blacklisted… | |
| 23:29:13 | openstackgerrit | Merged openstack/cliff: Remove tox environments that no longer work https://review.openstack.org/416360 | |
| #openstack-sdks - 2017-01-05 | |||
| 00:38:31 | stevemar | edleafe and other API working group folks -- question for y'all, whats the correct behaviour when requesting a list of resources with a query that is not supported? should it error out or just ignore the unsupported query? I looked at http://specs.openstack.org/openstack/api-wg/guidelines/pagination_filter_sort.html but couldn't find an answer at first glance | |
| 01:50:30 | openstackgerrit | Huanxuan Ao proposed openstack/python-openstackclient: Add "encryption-*" options in "volume type set/unset/create" https://review.openstack.org/414197 | |
| 03:32:47 | openstackgerrit | Merged openstack/python-openstackclient: Update the description format https://review.openstack.org/416485 | |
| 06:30:55 | openstackgerrit | jiahui.qiang proposed openstack/python-openstackclient: Fix filter error in os volume list https://review.openstack.org/416852 | |
| 06:49:13 | openstackgerrit | zhiyong.dai proposed openstack/python-openstackclient: Add unscoped-saml.rst https://review.openstack.org/416861 | |
| 11:39:33 | openstackgerrit | Jens Rosenboom proposed openstack/python-openstackclient: Fix creating a private flavor with ID auto https://review.openstack.org/416940 | |
| 12:50:21 | openstackgerrit | Huanxuan Ao proposed openstack/python-openstackclient: Add "encryption-*" options in volume type commands https://review.openstack.org/414197 | |
| 12:57:31 | openstackgerrit | Huanxuan Ao proposed openstack/python-openstackclient: Add "encryption-*" options in volume type commands https://review.openstack.org/414197 | |
| 12:59:27 | stevemar | huanxuan: dtroyer: meeting soon or in the evening? i never remember | |
| 12:59:50 | dtroyer | afternoon today | |
| 13:00:11 | dtroyer | 1:00 US Central, 2:00 US Eastern | |
| 13:00:36 | dtroyer | stevemar: I may be in and out today, getting Google Fiber installed (in the snow!) | |
| 13:01:19 | stevemar | dtroyer: you get snow down there? | |
| 13:01:44 | dtroyer | some years up to 10-12" | |
| 13:01:48 | dtroyer | last night just 2" | |
| 13:01:54 | stevemar | go figure | |
| 13:01:55 | dtroyer | driveway is already clear | |
| 13:02:04 | stevemar | clear it with a broom? :P | |
| 13:02:23 | dtroyer | leaf blower! | |
| 13:02:27 | dtroyer | (I wish) | |
| 13:02:47 | dtroyer | all of the schools around us are closed, except my wife's district. | |
| 13:02:59 | stevemar | d'oh! | |
| 13:03:12 | dtroyer | you guys would laugh at how quickly most of these districts shut down | |
| 13:03:29 | stevemar | TIL KC and Missouri are further north than NC | |
| 13:03:51 | stevemar | i never remember where half the states are, too many | |
| 13:03:59 | dtroyer | we also don't have a large body of heat-stabilizing water within 1000 miles :) | |
| 13:04:16 | stevemar | aye | |
| 13:04:35 | stevemar | water does do funny things like that | |
| 13:04:43 | dtroyer | Canada sends us cold stuff, Mexico sends us hot stuff, Pacific sends is wet stuff | |
| 13:04:59 | dtroyer | Atlantic just takes and takes and takes :) | |
| 13:05:03 | stevemar | hehe | |
| 13:05:18 | stevemar | thats one way of looking at it | |
| 13:05:19 | dtroyer | I'm sure Topol would argue with that when the hurricanes come though | |
| 13:05:34 | dtroyer | that's when the Atlantic gives it all back | |
| 13:06:08 | dtroyer | With that we conclude today's Weather Theory 101 | |
| 13:19:20 | stevemar | :) | |
| 13:50:12 | openstackgerrit | Dean Troyer proposed openstack/python-openstackclient: Functional tests - port https://review.openstack.org/416980 | |
| 13:52:40 | Dinesh_Bhor | briancurtin: Hi, Thanks for adding sort_key support for compute servers. I am curious to understand how I can pass multiple sort_keys like "GET /v2.1/servers/detail?sort_dir=desc&sort_dir=desc&sort_key=display_name&sort_key=vm_state" | |
| 13:53:15 | Dinesh_Bhor | briancurtin: I am using a small program similar to the mentioned here: https://pypi.python.org/pypi/openstacksdk | |
| 13:55:27 | Dinesh_Bhor | briancurtin: I am calling servers like "conn.compute.servers()" | |
| 14:09:00 | edleafe | stevemar: just saw your question. We generally recommend a 400 with a message that explains the problem | |
| 14:29:46 | sigmavirus | cdent: is there an API-WG meeting today? | |
| 14:30:15 | sigmavirus | etoews: ^ | |
| 14:30:22 | cdent | sigmavirus: yeah, in about 1.5 hours | |
| 14:30:29 | sigmavirus | cdent: excellent. Thank you! | |
| 14:38:47 | briancurtin | Dinesh_Bhor: is that supported by the server? | |
| 14:40:55 | Dinesh_Bhor | briancurtin: yes | |
| 14:41:14 | briancurtin | i can’t imagine they accept more than one sort_dir though, right? | |
| 14:43:10 | briancurtin | eh, either way, you can’t do that right now. it’s a dictionary, or rather there’s a small facade in front of a dictionary, but in the end query parameters are a dictionary | |
| 14:43:12 | Dinesh_Bhor | briancurtin: not sure, for my program if I pass conn.compute.servers(sort_key=['name', 'created_at']) like this then from openstacksdk it passes multiple sort keys successfully | |
| 14:44:08 | Dinesh_Bhor | briancurtin: "GET /v2.1/servers/detail?sort_key=name&sort_key=created_at" | |
| 14:44:11 | briancurtin | weird that it works like that, but if it does, then stick with it | |
| 14:45:30 | briancurtin | ahh, maybe that’s a feature of the requests library though | |
| 14:47:53 | Dinesh_Bhor | briancurtin: yes, Actually I am adding sort_keys support for 'masakariclient' which internally uses openstacksdk. While adding that I have now accepted list in the 'sort_key' from user and passed that to the masakari server directly and it works | |
| 14:51:23 | briancurtin | Dinesh_Bhor: it turns out that it is indeed a behavior of the requests library: http://stackoverflow.com/questions/22974772/querystring-array-parameters-in-python-using-requests#comment60268207_23347265 | |
| 14:53:41 | Dinesh_Bhor | briancurtin: Aah, ok thanks, so should we stick to this only? | |
| 14:54:13 | briancurtin | Dinesh_Bhor: yeah, if you do sort_key=[list, of, things] it will be fine | |
| 14:55:21 | Dinesh_Bhor | briancurtin: ok, thanks | |
| 14:57:03 | Dinesh_Bhor | briancurtin: bdw I have one more doubt if you have time then | |
| 14:58:02 | briancurtin | Dinesh_Bhor: sure, what’s up | |
| 14:58:15 | Dinesh_Bhor | briancurtin: If I pass any invalid sort_key to the servers then the nova-api returns the proper 400 but at the openstacksdk side I can just see "HttpException: Bad Request" | |
| 14:58:56 | Dinesh_Bhor | briancurtin: But at nova-api it has returned exception with message: Returning 400 to user: Sort key supplied was not valid. | |
| 14:59:35 | Dinesh_Bhor | briancurtin: conn.compute.servers(sort_key=['nadame', 'created_at']) | |
| 14:59:38 | openstackgerrit | Jens Rosenboom proposed openstack/python-openstackclient: Fix creating a private flavor with ID auto https://review.openstack.org/416940 | |
| 15:01:20 | Dinesh_Bhor | briancurtin: It looks like here https://github.com/openstack/python-openstacksdk/blob/master/openstack/session.py#L74 we are not getting anything in the message as well as in details | |
| 15:02:48 | briancurtin | yeah, a lot of projects do a lot of different/bad/wrong things with what they return. it’s probably best if you report a bug to https://bugs.launchpad.net/python-openstacksdk and myself or someone can take a look at it | |
| 15:03:25 | Dinesh_Bhor | briancurtin: ok, Can I take up this task? | |
| 15:03:40 | briancurtin | Dinesh_Bhor: sure, if you’d like to | |
| 15:04:08 | Dinesh_Bhor | briancurtin: ok, thanks I will push my patch soon | |
| 15:09:17 | briancurtin | Dinesh_Bhor: take a look at https://review.openstack.org/#/c/386440/ which did something about this same issue seen in Neutron, but it was reverted due to causing other issues. Qiming may know more about the other issues, i can’t remember what they were | |
| 15:13:45 | Dinesh_Bhor | briancurtin: ok, thanks I will take a look at it | |
| 15:15:26 | Qiming | Dinesh_Bhor, one thing worth a try is to wrap line 77 in that patch into a try...except statement | |
| 15:15:56 | Qiming | not all exceptions are returning things that can be converted to a JSON | |
| 15:16:33 | Qiming | that patch was reverted because it is over simplifying the situation | |
| 15:17:15 | Dinesh_Bhor | Qiming: can we do something like this: https://github.com/openstack/python-glanceclient/blob/master/glanceclient/exc.py#L156 | |
| 15:17:55 | Qiming | looks nicer | |
| 15:19:09 | Qiming | just I'm not sure if there will always be a content-type header when an exception is raised | |