| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2017-10-27 | |||
| 09:45:40 | ldl | return True | |
| 09:45:40 | ldl | if resp.status_code == 201: | |
| 09:46:01 | ldl | in openstacksdk, the session is not request.session. alright? | |
| 09:54:17 | ldl | Hi friends, who can help me with this? | |
| 09:54:18 | ldl | https://ask.openstack.org/en/question/110877/whats-the-session-in-def-validate_user_has_roleself-session-user-role-in-openstacksdk/ | |
| 10:18:20 | umbSublime | ldl, I think it should be yes, the Session object in openstacksdk inherits from keystoneauth Session object which does have a head method. https://github.com/openstack/python-openstacksdk/blob/master/openstack/session.py https://github.com/openstack/keystoneauth/blob/master/keystoneauth1/session.py | |
| 10:19:13 | ldl | thanks | |
| 10:21:07 | ldl | you mean, I should create a openstacksdk session instance, then pass into the method? | |
| 10:22:08 | umbSublime | no, the Connection object has a session atribute which is a Session object, so passing connection.session should be fine IMO https://github.com/openstack/python-openstacksdk/blob/master/openstack/connection.py#L172 | |
| 10:24:14 | ldl | :umbSublime thank you very much. | |
| 10:33:24 | umbSublime | np | |
| 10:35:55 | ldl | i don'w know why I pass the admin.conn.session, i get failure with 404 | |
| 10:39:15 | umbSublime | can you manually call: admin.connection.session.head method? | |
| 10:41:03 | ldl | 2017-10-27 18:29:52,228 DEBUG: keystoneauth.session Request returned failure status: 404 | |
| 10:41:03 | ldl | : keystoneauth.session HEAD call to identity for http://controller:5000/v3/projects/3db5410e1e0d4f739f4524ac74794728/users/e888a5fa2a554a26be46bb1e5dd59ee9/roles/fe3a32f5476b4662931c5757c813febe used request id req-1a6367b6-3f7c-42d4-8504-f4fd0ff48101 | |
| 10:41:03 | ldl | The traceback is bellow: | |
| 10:41:57 | umbSublime | the 404 in this case appears to be coming from the API and is just passed down by the method call | |
| 10:42:43 | ldl | I use the admin create the new project, and the new user, and then pass the admin's conn.session | |
| 10:43:52 | umbSublime | TBH I'm no openstacksdk pro :p, I was just trying to help. Perhaps someone more knowledgeable can help :) | |
| 10:44:07 | umbSublime | what about the role_id you pass, is it 'valid' ? | |
| 10:44:28 | ldl | its `user` | |
| 10:44:31 | ldl | user role | |
| 10:45:41 | ldl | and its roleID is correct, you can see the link | |
| 10:45:46 | ldl | upper | |
| 10:46:15 | umbSublime | validate_user_has_role(self, session, user, role) --> you need a 'session' a 'user' (probably uuid) and a 'role' (probably uuid) | |
| 10:51:01 | ldl | all I put | |
| 10:51:49 | umbSublime | hmm really not sure then sorry :( | |
| 10:52:01 | ldl | its ok, I will check | |
| 10:52:05 | ldl | thank you, my bro. | |
| 10:53:22 | umbSublime | npnp | |
| 10:53:47 | umbSublime | if you idle here long enough I have no doubt someone will come to your rescue :) | |
| 10:55:11 | ldl | ok | |
| 11:57:37 | ldl | :umbSublime, I figured out, I should assign_role_to_user() first | |
| 13:17:56 | fungi | no idea who's had their coffee already, but i got this random question by e-mail and am wondering if it's maybe a faq: | |
| 13:19:25 | fungi | the user quotes the validate_user_has_role() function definition from identity/project.py noting they're not sure what the session parameter should be, and stating that they're receiving the following error... "'SessionStore' object has no attribute 'head'" | |
| 13:20:21 | fungi | trying to be helpful, but my knowledge of openstacksdk is virtually nonexistent | |
| 13:21:02 | fungi | and they've provided basically no context to their question besides the above | |
| 13:21:31 | fungi | worst case i'll suggest that they come in here or post to ask.openstack.org | |
| 13:55:14 | dtroyer | ldl asked that here a couple of hours ago, it looked like it was sorted out | |
| 14:26:23 | openstackgerrit | Dean Troyer proposed openstack/cliff master: Add OSC functional-tips job to check pipeline https://review.openstack.org/515725 | |
| 14:44:09 | openstackgerrit | Vladyslav Drok proposed openstack/osc-lib master: Allow to use the none auth plugin https://review.openstack.org/512699 | |
| 14:45:35 | openstackgerrit | Vladyslav Drok proposed openstack/keystoneauth master: Make none auth usable in CLI https://review.openstack.org/515730 | |
| 14:55:12 | openstackgerrit | Slawek Kaplonski proposed openstack/python-openstackclient master: Remove convertion of Neutron Quota to dict https://review.openstack.org/515735 | |
| 14:55:42 | openstackgerrit | Dean Troyer proposed openstack/python-openstackclient master: Add server rescue unit tests https://review.openstack.org/515736 | |
| 14:55:54 | openstackgerrit | Dean Troyer proposed openstack/python-openstackclient master: Add missing parameters on openstack server rescue https://review.openstack.org/513393 | |
| 15:00:01 | openstackgerrit | Vladyslav Drok proposed openstack/osc-lib master: Allow to use the none auth plugin https://review.openstack.org/512699 | |
| 15:02:29 | openstackgerrit | Vladyslav Drok proposed openstack/osc-lib master: Allow to use the none auth plugin https://review.openstack.org/512699 | |
| 15:03:50 | openstackgerrit | Vladyslav Drok proposed openstack/keystoneauth master: Make none auth usable in CLI https://review.openstack.org/515730 | |
| 15:05:00 | fungi | dtroyer: thanks, i'll just ignore the e-mail in that case if it looks from logs to have probably been the same individual | |
| 15:35:28 | openstackgerrit | Dean Troyer proposed openstack/osc-lib master: Consume the devstack functional jobs from OSC https://review.openstack.org/515503 | |
| 15:36:38 | openstackgerrit | Dean Troyer proposed openstack/osc-lib master: Consume the devstack functional jobs from OSC https://review.openstack.org/515503 | |
| 15:46:28 | openstackgerrit | Vladyslav Drok proposed openstack/keystoneauth master: Make none auth usable in CLI https://review.openstack.org/515730 | |
| 16:18:05 | kuzko | good day everyone ! is sir Mordred here? | |
| 16:19:54 | kuzko | mordred : I'm confronted to a limit in shade's behavior while listing a container's objects, I'm currently limited to 10000 objects returned, and it seems there are no kwargs to pass the limit argument to the lower levels... | |
| 17:43:24 | openstackgerrit | Dean Troyer proposed openstack/python-openstackclient master: Zuul job updates https://review.openstack.org/515794 | |
| 17:45:55 | openstackgerrit | Dean Troyer proposed openstack/python-openstackclient master: Add server rescue unit tests https://review.openstack.org/515736 | |
| 18:29:08 | openstackgerrit | Merged openstack/python-openstackclient master: Add missing parameters on openstack server rescue https://review.openstack.org/513393 | |
| 20:19:17 | openstackgerrit | Dean Troyer proposed openstack/python-openstackclient master: Add server rescue unit tests https://review.openstack.org/515736 | |
| 20:45:22 | openstackgerrit | Dean Troyer proposed openstack/cliff master: Consume the OSC unit-tips job https://review.openstack.org/515929 | |
| 21:49:50 | openstackgerrit | Dean Troyer proposed openstack/cliff master: Consume the OSC unit-tips job https://review.openstack.org/515929 | |
| 23:00:30 | openstackgerrit | Lin Yang proposed openstack/python-openstackclient master: Add python-rsdclient into plugin list https://review.openstack.org/511683 | |
| #openstack-sdks - 2017-10-28 | |||
| 01:47:01 | openstackgerrit | Merged openstack/python-openstackclient master: Add server rescue unit tests https://review.openstack.org/515736 | |
| 03:26:54 | openstackgerrit | Dean Troyer proposed openstack/cliff master: Consume the OSC unit-tips job https://review.openstack.org/515929 | |
| 13:36:07 | openstackgerrit | Dean Troyer proposed openstack/cliff master: Consume the OSC unit-tips job https://review.openstack.org/515929 | |
| #openstack-sdks - 2017-10-29 | |||
| 11:39:28 | openstackgerrit | James E. Blair proposed openstack/python-openstackclient master: Zuul: add file extension to playbook path https://review.openstack.org/516081 | |
| 16:41:52 | openstackgerrit | James E. Blair proposed openstack/openstackclient master: Zuul: add file extension to playbook path https://review.openstack.org/516113 | |
| 17:28:01 | openstackgerrit | Merged openstack/python-openstackclient master: Zuul: add file extension to playbook path https://review.openstack.org/516081 | |
| 17:43:33 | openstackgerrit | Slawek Kaplonski proposed openstack/python-openstackclient master: Remove unnecessary comment https://review.openstack.org/515735 | |
| #openstack-sdks - 2017-10-30 | |||
| 04:21:42 | openstackgerrit | Merged openstack/python-openstackclient master: Added AddNetwork command to server https://review.openstack.org/509209 | |
| 08:13:28 | GreenBlood | mordred: Hi, on the API there's a "default" limit of object listing which is 10 000, the openstack cli allows to override this limit but it seems shade has not such argument for the list_object() fn | |
| 08:25:40 | openstackgerrit | Dongcan Ye proposed openstack/python-openstackclient master: Network: Add supports default rbac target_project_id https://review.openstack.org/516211 | |
| 09:06:30 | umbSublime | ldl, yah that makes sense :p, I'm assuming the 404 basically meant "nope, this user does not have this role" | |
| 10:16:21 | frickler | mordred: this seems to be a regression in shade-1.23.0, with 1.22.0 I do not get that issue: https://github.com/ansible/ansible/issues/29129 . trying to find a fix now | |
| 13:46:09 | cdent | Dinesh_Bhor: let me make sure I understand the issue | |
| 13:46:19 | cdent | the top level cinder api accepts and optional param as valid | |
| 13:46:28 | Dinesh_Bhor | cdent: yes | |
| 13:46:37 | Dinesh_Bhor | cdent: It sends it to cinder-volume | |
| 13:46:42 | cdent | then it sends it on to cinder-volumes which uses it, and in that particular instance the parameter is not valid | |
| 13:46:54 | Dinesh_Bhor | cdent: yes | |
| 13:47:03 | cdent | theoretically, could some other cinder-volume handle it? | |
| 13:47:44 | cdent | (the answer to that doesn’t really matter, just curious) | |
| 13:48:08 | Dinesh_Bhor | cdent: yeah, I need to check this. But yes ut depends on the cinder-backend used | |
| 13:48:16 | Dinesh_Bhor | s/ut/it | |
| 13:48:22 | cdent | since the input, though valid initially, wasn’t appropriate eventually, it’s still should be a 400 | |
| 13:48:47 | cdent | it could potentially be a 409 (because the state of the “server” is not what is expected) | |
| 13:49:19 | edleafe | and yeah, that would be a good clarification for the API guidelines | |
| 13:49:40 | Dinesh_Bhor | cdent, edleafe: Okay, we should return the same InvalidInput(400) to user from cinder-api | |
| 13:49:58 | cdent | edleafe: what do you see being clarified, what’s the additional dimension? | |
| 13:50:21 | edleafe | cdent: the multi-layer thing | |
| 13:50:40 | edleafe | if it generally acceptable, but not for a particular implementation, it should be 400 | |
| 13:56:17 | Dinesh_Bhor | cdent, edleafe: thank you for the guidance :). | |
| 14:21:31 | cdent | edleafe, elmiko, dtantsur : https://etherpad.openstack.org/p/api-sig-sydney-forum | |
| 14:22:15 | dtantsur | cdent: do we need mordred to finish his poem for #2? | |
| 14:22:45 | cdent | like all great epics, it will be forever unfinished | |
| 14:23:57 | dtantsur | lol | |
| 15:49:24 | openstackgerrit | Dongcan Ye proposed openstack/python-openstackclient master: Network: Add supports default rbac target_project_id https://review.openstack.org/516211 | |
| 15:55:15 | openstackgerrit | Dongcan Ye proposed openstack/python-openstackclient master: Network: Add supports default rbac target_project_id https://review.openstack.org/516211 | |
| 17:54:20 | rabel | is openstack server migrate always performing a live migration? and if yes, how would i do a cold migration? | |
| 17:55:13 | rabel | and if so the --live parameter is really missleading since it suggests that it is there to switch between live and cold migration, but it really only is the target host | |
| 18:12:33 | dtroyer | rabel: forgive my lack of memory on this (excuse-of-the-day is 'jetlag') but I do know there are some inconsistencies in the migration command options, and there is at least one review in the queue to fix this up. IIRC some of it reflects the API and some of it is OSC's incompleteness in this area. | |