| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2018-03-08 | |||
| 06:07:47 | adriant | I'm trying to switch all my code to the SDK client, hence why I keep finding all kind of weird little bugs :P | |
| 06:09:48 | adriant | that said, eandersson, you should be able to still pass in auth variables rather than an authenticator, it's just that the variables for auth need to be in an auth param. | |
| 06:11:09 | eandersson | Do you mean instead of generic.Token? | |
| 06:12:09 | adriant | lemme try something and send you code snippet to try | |
| 06:12:25 | eandersson | The problem is this code https://github.com/openstack/python-openstacksdk/blob/0.11.3/openstack/profile.py#L50 | |
| 06:12:53 | eandersson | https://github.com/openstack/python-openstacksdk/blob/0.11.3/openstack/connection.py#L277 | |
| 06:13:01 | eandersson | I would appreciate that adriant | |
| 06:18:21 | adriant | eandersson: hmmm, i think you're right, until that profile logic goes away, you have to pass in the authenticator | |
| 06:19:02 | eandersson | Wish that portion would have been kept for backwards compatibility | |
| 06:35:39 | openstackgerrit | Olivier Bourdon proposed openstack-infra/shade master: Add extra failure codes to bad request exception https://review.openstack.org/550107 | |
| 08:24:06 | openstackgerrit | Tobias Brox proposed openstack-infra/shade master: Add subnetpool_name_or_id parameter to create_subnet https://review.openstack.org/548722 | |
| 09:55:54 | openstackgerrit | Olivier Bourdon proposed openstack-infra/shade master: Add proper return value for validate_node https://review.openstack.org/549981 | |
| 11:17:33 | hwoarang | hello. we are seeing the following problems with queens https://bugs.launchpad.net/python-openstacksdk/+bug/1754302 not sure if anyone has any suggestion besides downgrading the openstacksdk component :) | |
| 11:17:34 | openstack | Launchpad bug 1754302 in OpenStack SDK "openstacksdk-0.12.0 breaks with self-signed certificates" [Undecided,New] | |
| 11:30:01 | openstackgerrit | Lingxian Kong proposed openstack/service-types-authority master: Add qinling project https://review.openstack.org/550787 | |
| 12:15:11 | openstackgerrit | James Denton proposed openstack/python-openstacksdk master: Adds prefixlen to the request body when creating subnets https://review.openstack.org/550558 | |
| 12:41:30 | cdent | dtantsur, elmiko, edleafe: can one of you run the meeting today please? I've cut my thumb somewhat badly (but not catastrophically) and typing is slow | |
| 12:42:08 | dtantsur | cdent: ugh, get better :( yes, I can | |
| 12:42:43 | cdent | thanks | |
| 12:53:55 | mordred | adriant: oh wow. I see much scrollback | |
| 13:02:05 | elmiko | cdent: sorry to hear about the thumb =( | |
| 13:02:27 | elmiko | cdent dtantsur edleafe , i will most likely miss the meeting today, i've got a conflict internally here | |
| 13:31:39 | edleafe | cdent: I'll run it | |
| 13:46:00 | openstackgerrit | Monty Taylor proposed openstack/keystoneauth master: Remove tox_install.sh and align with constraints consumption https://review.openstack.org/550837 | |
| 14:01:54 | openstackgerrit | Monty Taylor proposed openstack/keystoneauth master: Remove tox_install.sh and align with constraints consumption https://review.openstack.org/550837 | |
| 16:48:03 | edleafe | dtantsur: cdent: redy for review https://etherpad.openstack.org/p/api-sig-newsletter | |
| 16:48:10 | edleafe | geez | |
| 16:48:12 | edleafe | *ready | |
| 16:48:47 | dtantsur | short and nice, lgtm :) | |
| 16:51:08 | cdent | edleafe: you wanna add the cache-control review? | |
| 17:01:04 | edleafe | cdent: ah, good catch | |
| 17:03:18 | edleafe | added and shipped | |
| 17:08:46 | cdent | woot | |
| 17:57:38 | lbragstad | o/ curious if anyone from the api sig/wg would be interesting in sharing their thoughts on http://lists.openstack.org/pipermail/openstack-dev/2018-March/128027.html ? | |
| 17:57:44 | lbragstad | interested* | |
| 17:58:20 | lbragstad | the team is curious about where we should meet usability with the consistency of the rest of the API | |
| 17:58:43 | cdent | lbragstad: will put that in my queue for tomorrow. edleafe, dtantsur|afk ^ | |
| 17:59:15 | lbragstad | thanks cdent | |
| 17:59:46 | cdent | would today but I got melty brain | |
| 17:59:56 | lbragstad | that's understandable | |
| 18:00:25 | lbragstad | s/coffee/beer/ if symptoms persist :) | |
| 18:02:40 | cdent | :) | |
| 18:28:47 | edleafe | wait - brains aren't supposed to be melty? | |
| 19:42:48 | mordred | edleafe: oy. I've been doing it all wrong | |
| 21:41:13 | adriant | mordred: I found a bug, although this one is easy: https://review.openstack.org/#/c/550700/ | |
| 21:41:38 | adriant | I also took the chance to check all our resources for identity and make sure the query filters matched. They did, just this one was wrong. | |
| 21:42:34 | adriant | lbragstad: ^ I you fellows are evil for "user.id" as the filter rather than "user_id" :P | |
| 21:42:57 | adriant | and I think it's the only keystone API to do that. :( | |
| 21:43:43 | lbragstad | hmmmmmm | |
| 21:43:56 | lbragstad | i might have to dig into why that is | |
| 21:45:47 | adriant | lbragstad: https://developer.openstack.org/api-ref/identity/v3/#id595 this one | |
| 21:46:25 | adriant | I always used the keystoneclient so never realised until now when trying to switch to the sdk that the API was weird :P | |
| 21:47:25 | adriant | the keystoneclient handles it here: https://github.com/openstack/python-keystoneclient/blob/master/keystoneclient/v3/role_assignments.py#L101-L122 | |
| 22:18:47 | mordred | adriant: that's ... that's just amazing | |
| 22:19:18 | lbragstad | "you had *one* job!" | |
| 22:19:25 | mordred | :) | |
| 22:19:36 | mordred | adriant: btw - I'm thrilled that your patch works | |
| 22:19:54 | mordred | adriant: it's almost like that system working as designed! | |
| 22:20:32 | adriant | lbragstad: at least adding the '_' variants to keystone should be easy | |
| 22:20:43 | adriant | then we'll have both! :P | |
| 22:20:53 | lbragstad | yeah - that's about the best we could do | |
| 22:21:09 | lbragstad | adriant: would you mind opening a bug? we should be able to target something like that to office hours | |
| 22:21:21 | adriant | will do | |
| 22:21:28 | lbragstad | ty sir | |
| 22:21:52 | adriant | lbragstad: also, I swear I'll find time next week to write up that blasted token/mfa spec... | |
| 22:21:54 | adriant | I swear... | |
| 22:22:31 | openstackgerrit | Monty Taylor proposed openstack/python-openstacksdk master: Allow not resolving outputs on get stacks https://review.openstack.org/548595 | |
| 22:23:36 | adriant | and mordred, if briancurtin was responsible for those query mapping things, then he deserves thanks :P | |
| 22:23:52 | adriant | It's nice being able to make this stuff just work | |
| 22:23:58 | mordred | ++ | |
| 22:26:59 | adriant | now back to writing audit scripts... We rolled out MFA to our cloud this week, and now as a follow up action I'm writing monitoring tools for: "complain if anyone other than this list has the 'admin' role", "complain if those with admin and not on the exempt list don't have MFA enabled" | |
| 23:08:00 | openstackgerrit | Merged openstack-infra/shade master: Add extra failure codes to bad request exception https://review.openstack.org/550107 | |
| 23:09:15 | openstackgerrit | Merged openstack-infra/shade master: Add proper return value for validate_node https://review.openstack.org/549981 | |
| 23:11:22 | olivierbourdon38 | thx to all involved for the reviews and merges of the shade patchsets | |
| 23:28:04 | openstackgerrit | Merged openstack/python-openstacksdk master: Run os-client-config tests on sdk changes https://review.openstack.org/549474 | |
| 23:34:11 | mordred | olivierbourdon38: thanks for thepatches! | |
| #openstack-sdks - 2018-03-09 | |||
| 05:43:26 | openstackgerrit | James Denton proposed openstack/python-openstacksdk master: Adds prefixlen to the request body when creating subnets https://review.openstack.org/550558 | |
| 07:50:01 | openstackgerrit | rarm nagalingam proposed openstack-infra/shade master: Adds toggle port security on network create https://review.openstack.org/550646 | |
| 09:55:13 | openstackgerrit | rarm nagalingam proposed openstack-infra/shade master: Adds toggle port security on network create https://review.openstack.org/550646 | |
| 09:59:06 | openstackgerrit | rarm nagalingam proposed openstack-infra/shade master: Adds toggle port security on network create https://review.openstack.org/550646 | |
| 10:05:22 | openstackgerrit | rarm nagalingam proposed openstack-infra/shade master: Adds toggle port security on network create https://review.openstack.org/550646 | |
| 11:35:09 | amotoki | mordred: I see you in the etherpad of ops meetup tokyo. I added some note based on conversations after the session. It looks like a wrong approach which tries to extend/inherit SDK class. I suggest him to propose masakari API to OpenStackSDK :) | |
| 12:25:48 | mordred | amotoki: yes please! | |
| 12:55:28 | openstackgerrit | James Denton proposed openstack/python-openstacksdk master: Adds prefixlen to the request body when creating subnets https://review.openstack.org/550558 | |
| 13:04:31 | frickler | when running any command without authentication I'm seeing "AttributeError: 'module' object has no attribute 'OpenStackConfigException'" instead of a proper error, is this a known issue? | |
| 13:31:19 | openstackgerrit | Merged openstack/python-openstacksdk master: Allow not resolving outputs on get stacks https://review.openstack.org/548595 | |
| 17:06:46 | openstackgerrit | Merged openstack/python-openstacksdk master: Update the invalid url in pages https://review.openstack.org/548149 | |
| 17:25:22 | openstackgerrit | Merged openstack/python-openstackclient master: Update links in README https://review.openstack.org/549159 | |
| 17:49:32 | cdent | lbragstad: I'm trying to parse the email and irc log you point api-sig folk at (about unified limits) and I think I'm missing a critical piece: What is the actual question? | |
| 17:50:17 | cdent | Or if not question, then problem? | |
| 17:50:21 | cdent | I'm lacking context | |
| 17:50:24 | lbragstad | cdent: my question specifically | |
| 17:50:51 | lbragstad | is does it make sense to support a PUT API in that way when it's not consistent with any other parts of keystone's API | |
| 17:50:53 | cdent | the PUT of multple limits, if getting rid of that is okay? | |
| 17:51:13 | lbragstad | well - it's a PUT, with a list of limits, and an ID in each limit | |
| 17:51:17 | lbragstad | which seems weird | |
| 17:51:28 | cdent | yes, that is weird | |
| 17:51:40 | lbragstad | i can see the batch update case | |
| 17:51:45 | cdent | and the reason it got in was because of the desire to do batch changes? | |
| 17:51:49 | lbragstad | but the batch create seems more critical | |