| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2018-10-30 | |||
| 13:53:28 | Shrews | hrm, our unit tests don't even test passing filter | |
| 13:56:57 | openstackgerrit | Monty Taylor proposed openstack/cliff master: Don't try to run issubclass on non-classes https://review.openstack.org/614218 | |
| 14:00:21 | mordred | Shrews: oh - totally. let's definitely add some tests | |
| 14:03:06 | openstackgerrit | David Shrewsbury proposed openstack/openstacksdk master: Filter ports in list_ports when batching is in effect https://review.openstack.org/614213 | |
| 14:03:11 | Shrews | mordred: ^^ unit test | |
| 14:03:37 | Shrews | mordred: double check me, but passes locally for me | |
| 14:05:02 | openstackgerrit | Monty Taylor proposed openstack/cliff master: Don't try to run issubclass on non-classes https://review.openstack.org/614218 | |
| 14:05:45 | Shrews | mordred: hrm, no. that passes on master when it shouldn't | |
| 14:05:46 | Shrews | :( | |
| 14:06:16 | Shrews | so i've done something wrong | |
| 14:11:24 | mordred | Shrews: that test tests that the query returns an unfilted list | |
| 14:12:04 | mordred | mock_neutron_port_list_rep is a list of ports that has non-down ports in it right? | |
| 14:12:11 | Shrews | mordred: it isn't asserting that the uri contains the filter params? | |
| 14:12:30 | Shrews | i thought assert_calls did that | |
| 14:13:42 | Shrews | i wasn't worried so much about the return data as much as i wanted to confirm the uri had any filters we passed | |
| 14:13:46 | mordred | Shrews: oh - i was looking at line 253 | |
| 14:14:09 | mordred | Shrews: I think you need to configure caching | |
| 14:14:27 | mordred | Shrews: because we only skip passing the query string if caching is enabled | |
| 14:14:53 | mordred | (although maybe leave that test to test that non-caching also works) | |
| 14:15:21 | Shrews | ah. do we have an example of that? | |
| 14:15:27 | mordred | looking for one now | |
| 14:16:28 | mordred | Shrews: openstack/tests/unit/cloud/test_caching.py is set up to do servers caching - we could add prts to openstack/tests/unit/fixtures/clouds/clouds_cache.yaml | |
| 14:16:49 | mordred | Shrews: and add a copy of the test you just added but without the query string | |
| 14:21:07 | Shrews | don't you need the query string?? | |
| 14:21:55 | mordred | well - in the caching version we'd expect it to not pass filters on the query string | |
| 14:22:34 | Shrews | oh! b/c we filter ourselves | |
| 14:23:38 | Shrews | mordred: you seem to have a handle on it, feel free to push that PS while I caffeinate more appropriately | |
| 14:24:52 | mordred | Shrews: ok. I'll do that one | |
| 14:25:40 | mordred | Shrews: OH | |
| 14:25:44 | mordred | Shrews: also - re: functional ... | |
| 14:25:58 | mordred | Shrews: when you create a port but it isn't attached to anything - its state is totally DOWN | |
| 14:26:10 | mordred | Shrews: see mock_neutron_port_create_rep in openstack/tests/unit/cloud/test_port.py | |
| 14:29:14 | Shrews | oh cool | |
| 14:35:33 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Filter ports in list_ports when batching is in effect https://review.openstack.org/614213 | |
| 14:35:37 | mordred | Shrews: how's that look? | |
| 14:40:02 | Shrews | mordred: yes, that fails for me on master | |
| 14:40:05 | Shrews | ++ | |
| 14:40:26 | mordred | Shrews: \o/ | |
| 14:40:47 | mordred | slaweq: feel in the mood to review a quick neutron related sdk patch? ^^ | |
| 14:42:56 | mordred | Shrews: I believe this same bug exists for servers and floating ips - followup patch coming | |
| 14:43:03 | Shrews | oh goodie | |
| 14:45:01 | slaweq | mordred: +W (I hope CI will be happy with it too) :) | |
| 14:45:08 | mordred | slaweq: thanks! | |
| 14:48:02 | openstackgerrit | Monty Taylor proposed openstack/openstacksdk master: Apply list filter fix to servers and floating ips too https://review.openstack.org/614236 | |
| 14:48:15 | mordred | Shrews: ^^ there it is applied to servers and floating ips too | |
| 15:02:05 | Shrews | ossum | |
| 15:02:56 | mordred | Shrews: I put a pause on the 0.19.0 release (which contained the other fix we needed for nodepool) so we can get this in there | |
| 15:04:15 | Shrews | ++ | |
| 15:22:41 | mordred | dtroyer: if you get a sec, could you take a peep at https://review.openstack.org/#/c/614218/ ? | |
| 15:38:58 | openstackgerrit | Sean McGinnis proposed openstack/openstacksdk master: Remove setup.py check from pep8 job https://review.openstack.org/614274 | |
| 15:43:02 | mordred | Shrews: if you have a sec, https://review.openstack.org/#/c/613976 and https://review.openstack.org/#/c/613978 are in response to a feedback we got yesterday from wouterdb | |
| 15:43:51 | mordred | Shrews: who had a usecase where he was creating and destroying a ton of Connection objects, and the lack of a close to shutdown underlying threadpool objects was causing a preponderance of threads | |
| 15:45:59 | Shrews | mordred: i totes want a be_awesome() method | |
| 15:52:52 | mordred | Shrews: we should make one | |
| 15:54:51 | mordred | gtema: looking at your story from earlier about caching and wait_for methods ... we really need to do some work in the caching layer | |
| 15:55:05 | gtema | mordred: agree | |
| 15:55:20 | gtema | just spend some time trying to figure out why it suddenly does not work properly | |
| 15:55:43 | gtema | mordred: can I help there? | |
| 15:58:49 | mordred | gtema: almost certainly! I think once we get this task manager updates finished it'll be a good time to work on it | |
| 15:59:07 | gtema | ok | |
| 16:00:21 | gtema | mordred: what about backup support in block-storage (https://review.openstack.org/#/c/591445/) | |
| 16:00:45 | openstackgerrit | Sean McGinnis proposed openstack/os-client-config master: Remove setup.py check from pep8 job https://review.openstack.org/614280 | |
| 16:01:46 | mordred | gtema: "Note that backup restore is not documented in the official API, but is implemented in the cinderclient" | |
| 16:01:49 | mordred | that's so awesome | |
| 16:02:11 | gtema | yes, I was clarifying it - it was a bug promised to be fixed in docs | |
| 16:02:53 | gtema | oh, it is even now fixed - will update commit message | |
| 16:04:07 | openstackgerrit | Artem Goncharov proposed openstack/openstacksdk master: implement block-storage backup resource https://review.openstack.org/591445 | |
| 16:04:10 | smcginnis | Which "official API" is Cinder backup not documented in? | |
| 16:04:32 | gtema | backup restore was not documented | |
| 16:04:36 | smcginnis | Where? | |
| 16:04:46 | gtema | https://developer.openstack.org/api-ref/block-storage/v2/index.html?expanded=restore-backup-detail#backups-backups | |
| 16:05:04 | gtema | but now it is documented | |
| 16:05:11 | mordred | \o/ | |
| 16:05:22 | smcginnis | Must have been a really old reference. | |
| 16:06:14 | gtema | from what I remember I was told there was a content error somewhere in sources, so that this method was not shown properly | |
| 16:06:30 | smcginnis | Ah, that could be then. | |
| 16:07:17 | smcginnis | OK, yeah, I fixed that just in August. https://review.openstack.org/#/c/592026/ | |
| 16:08:15 | gtema | yes, I was asking in maillist sometime back in august - my change was for this time forgotten and I rebased it yesterday | |
| 16:08:29 | gtema | mordred: thanks | |
| 16:08:35 | smcginnis | gtema: Cool, thanks! | |
| 16:09:01 | gtema | welcome | |
| 16:11:51 | dtroyer | mordred: +2… there is an isclass() call in OSC that has gone back and forth a time or two, I'm wondering if a similar solution would be appropriate there (without looking before typing of course) | |
| 17:13:38 | openstackgerrit | Merged openstack/openstacksdk master: Add close method to shutdown threadpool https://review.openstack.org/613976 | |
| 17:13:39 | openstackgerrit | Merged openstack/openstacksdk master: Make Connection a context manager https://review.openstack.org/613978 | |
| 17:50:30 | gyroknight_ | Hi, just wondering if anyone's had trouble with list_servers(all_projects=True) in openstack.connection.Connection? It seems to only list instances belonging to the authenticated user's project, even when the user is an admin | |
| 17:50:49 | gyroknight_ | I can't find any related tasks or stories in the StoryBoard | |
| 17:53:57 | mordred | gyroknight_: I feel like I saw a patch related to that recently | |
| 17:53:59 | mordred | looking | |
| 17:54:24 | gyroknight_ | The servers(all_projects=True) generator in the compute v2 proxy also doesn't seem to work. The generator creation seems to work fine, but if I try to iterate through it, Resource throws an error "openstack.exceptions.InvalidResourceQuery: Invalid query params: all_projects" | |
| 17:54:34 | mordred | oh - maybe not | |
| 17:55:01 | mordred | gyroknight_: cool (I mean, not cool, but good to know) - I've gotta run out for a sec, but will investigate in just a little bit | |
| 17:55:37 | gyroknight_ | Thanks | |
| 18:22:36 | gyroknight_ | Figured out the compute proxy, seems the documentation is incorrect, it should be all_tenants instead of all_projects | |
| 18:23:47 | gyroknight_ | I'm using the documentation here: https://docs.openstack.org/openstacksdk/latest/user/proxies/compute.html#openstack.compute.v2._proxy.Proxy | |
| 18:56:49 | mordred | gyroknight: we just recently merged a fix for that ... https://review.openstack.org/#/c/540390/ (like, 2 days ago and it'll be in there when we release later today) | |
| 18:57:42 | gyroknight | Awesome, thanks | |
| 18:59:32 | gyroknight | I figured out the workaround with list_servers(filters={'all_tenants': True}) in the meantime, but it'll be nice to have the alias later on | |
| 19:04:17 | mordred | ++ | |
| 19:06:30 | mordred | hrm. list_servers(filters={'all_tenants': True}) works but list_servers(all_projects=True) doesn't work? that's sad making ... | |
| 19:06:41 | mordred | gyroknight: do you have resource caching enabled? | |