| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-sdks - 2020-06-10 | |||
| 15:17:49 | openstackgerrit | Merged openstack/ansible-collections-openstack master: Add notes to README about deps and versions https://review.opendev.org/734557 | |
| 19:20:38 | openstackgerrit | Lance Bragstad proposed openstack/python-openstackclient stable/ussuri: Client should parse string to boolean for value 'is_domain' https://review.opendev.org/732397 | |
| 19:23:56 | openstackgerrit | Lance Bragstad proposed openstack/python-openstackclient stable/ussuri: Client should parse string to boolean for value 'is_domain' https://review.opendev.org/732397 | |
| 19:27:48 | openstackgerrit | Lance Bragstad proposed openstack/python-openstackclient stable/train: Client should parse string to boolean for value 'is_domain' https://review.opendev.org/732398 | |
| 19:30:26 | lbragstad | mordred ^ that's a pretty straight forward backport and we have the tests working against the train/ussuri branches | |
| 21:07:59 | openstackgerrit | Merged openstack/openstacksdk master: Do not clean keypairs in the project cleanup https://review.opendev.org/734019 | |
| 22:51:09 | openstackgerrit | Gabriel Ramirez proposed openstack/python-openstackclient master: Allow openstack flavor set to update flavor description using name https://review.opendev.org/733688 | |
| 22:51:28 | mordred | lbragstad: lgtm | |
| #openstack-sdks - 2020-06-11 | |||
| 00:10:31 | openstackgerrit | Monty Taylor proposed openstack/osc-lib master: Remove commandmanager subclass https://review.opendev.org/734000 | |
| 02:37:03 | openstackgerrit | Merged openstack/python-openstackclient stable/ussuri: Client should parse string to boolean for value 'is_domain' https://review.opendev.org/732397 | |
| 08:51:11 | openstackgerrit | Sagi Shnaidman proposed openstack/ansible-collections-openstack master: Add note about py3 in readme https://review.opendev.org/735081 | |
| 09:45:52 | openstackgerrit | Dirk Mueller proposed openstack/osc-lib stable/ussuri: Replace assertItemsEqual with assertCountEqual https://review.opendev.org/735090 | |
| 09:48:06 | openstackgerrit | Dirk Mueller proposed openstack/python-openstackclient stable/ussuri: Replace assertItemsEqual with assertCountEqual https://review.opendev.org/735091 | |
| 09:55:57 | openstackgerrit | Sagi Shnaidman proposed openstack/ansible-collections-openstack master: Fix ansible-devel sanity tests for deprecations https://review.opendev.org/735095 | |
| 12:18:20 | mordred | dtroyer, mnaser, dtantsur|afk, frickler: with the move of the group code to cliff, the osc-lib patch is now green: https://review.opendev.org/#/c/734000/ | |
| 12:34:54 | openstackgerrit | Sagi Shnaidman proposed openstack/ansible-collections-openstack master: Add note about py3 in readme https://review.opendev.org/735081 | |
| 12:48:12 | sshnaidm | hmm.. I've got a weird error while trying to use conn.compute.wait_for_server() - AttributeError: 'Munch' object has no attribute 'fetch' | |
| 12:48:25 | sshnaidm | anybody knows what is the issue? ^ | |
| 12:49:06 | openstackgerrit | jayaditya gupta proposed openstack/cliff master: Fix json output for security_groups property in "openstack server show -f json" command https://review.opendev.org/735128 | |
| 12:50:39 | nightmare_unreal | Patch : https://review.opendev.org/735128 | |
| 12:50:39 | nightmare_unreal | frickler mordred , This is how I am thinking to fix the json properties issue. This i probably not the right way so please comment on it when anyone gets time :) Thanks | |
| 12:57:16 | mordred | sshnaidm: sounds like you're using wait_for_server on something that's returning just the munch and not the Server resource - we haven't finished reworking the shade api to completely use the underlying sdk resource layer yet | |
| 12:57:35 | mordred | sshnaidm: so there are places still where the shade layer is still making direct rest calls | |
| 12:58:17 | jawad_axd | Hi all, Is it to possible to allocate cidr(subnet) automatically when creating network? I came acorss with get-me-network, but I dont really need router creation, just network with automatic cidr. Any pointers? | |
| 13:00:41 | sshnaidm | mordred, hmm.. maybe I don't understand something, but where is "resource" reinitialization in the loop? https://opendev.org/openstack/openstacksdk/src/branch/master/openstack/resource.py#L2035 | |
| 13:01:11 | sshnaidm | does it just repeat "resource = resource.fetch(session)" ? | |
| 13:01:39 | mordred | yes. iterate_timeout is a loop-until-timeout function | |
| 13:01:58 | mordred | sshnaidm: where are you calling that? | |
| 13:02:38 | mordred | sshnaidm: (I can almost certaily tell you how to do things in your code differently so it works, but what would be great is to take this as an opportunity to fix whatever gave you a server that doesn't work there) | |
| 13:03:16 | sshnaidm | mordred, I'm calling from ansible module: self.conn.compute.wait_for_server(os_server,status=_action_map[self.params['action']],wait=self.params['timeout']) | |
| 13:03:39 | mordred | sshnaidm: nod - and what did you call to get teh os_server object? | |
| 13:04:21 | sshnaidm | mordred, self.conn.get_server(self.params['server']) | |
| 13:04:34 | mordred | sshnaidm: wow | |
| 13:05:16 | mordred | oh - wait | |
| 13:05:19 | mordred | no, that's expected | |
| 13:05:24 | mordred | and we should fix it - but for now | |
| 13:05:34 | mordred | use self.conn.wait_for_server if you use self.conn.get_server | |
| 13:06:08 | mordred | (although that's not going to let you pass in a status) | |
| 13:06:37 | sshnaidm | yeah, server is <class 'munch.Munch'> | |
| 13:06:39 | mordred | sshnaidm: actually - for this case ... | |
| 13:06:57 | mordred | sshnaidm: try using self.conn.compute.get_server(self.params['server']) | |
| 13:06:58 | frickler | jawad_axd: with subnetpools creating a subnet is pretty simple, but still a dedicated action, I don't think there's a way around that | |
| 13:08:35 | mordred | sshnaidm: we need to finish https://review.opendev.org/#/c/630912 to fix the other thing, and that's important, but probably not a thing you want to wait on | |
| 13:08:40 | jawad_axd | @frickler Thanks. | |
| 13:09:26 | sshnaidm | mordred, hmm, it didn't find server with self.conn.compute.get_server(self.params['server']) | |
| 13:11:09 | sshnaidm | and w/o compute: TypeError: wait_for_server() got an unexpected keyword argument 'status' | |
| 13:11:31 | sshnaidm | I think I'll wait a little to introduce this function.. | |
| 13:11:49 | mordred | sshnaidm: yeah - there's no status parameter for self.conn.wait_for_server | |
| 13:12:37 | mordred | sshnaidm: one more thign to try (sorry, still on first coffee) | |
| 13:12:50 | mordred | sshnaidm: self.conn.compute.find_server(self.params['server']) | |
| 13:15:43 | sshnaidm | mordred, that seems working | |
| 13:21:34 | sshnaidm | mordred, <class 'openstack.compute.v2.server.Server'> - it seems to have all attributes as "None" | |
| 13:25:32 | sshnaidm | I've got only ID and name | |
| 13:35:09 | openstackgerrit | Merged openstack/ansible-collections-openstack master: Fix ansible-devel sanity tests for deprecations https://review.opendev.org/735095 | |
| 13:41:43 | mordred | sshnaidm: that resource doesn't seem to have been fetched | |
| 13:42:31 | sshnaidm | mordred, well, it got id and name | |
| 14:24:04 | openstackgerrit | Sagi Shnaidman proposed openstack/ansible-collections-openstack master: Add note about py3 in readme https://review.opendev.org/735081 | |
| 14:54:10 | openstackgerrit | Sagi Shnaidman proposed openstack/ansible-collections-openstack master: Add OpenstackModule to os_server_action https://review.opendev.org/721746 | |
| 14:54:40 | sshnaidm | mordred, ended up with two objects, "Server" for wait_for_server and "Munch" for all the rest ^ | |
| 14:55:27 | sshnaidm | if it's not ok, let's leave _wait function as it was: https://review.opendev.org/#/c/721746/14..15/plugins/modules/server_action.py | |
| 15:01:30 | mordred | sshnaidm: yeah - I think that's fine. I think we can simplify it in the future - but waiting on that is not valuable | |
| 15:44:43 | openstackgerrit | Merged openstack/python-openstackclient master: Remove os-client-config references https://review.opendev.org/734097 | |
| 17:14:22 | openstackgerrit | Merged openstack/python-openstackclient stable/train: Client should parse string to boolean for value 'is_domain' https://review.opendev.org/732398 | |
| 23:51:14 | openstackgerrit | Merged openstack/ansible-collections-openstack master: Add note about py3 in readme https://review.opendev.org/735081 | |
| #openstack-sdks - 2020-06-12 | |||
| 05:53:24 | frickler | is there a reason why "router create" doesn't allow additional settings like external-gateway? according to the api-ref that should be possible | |
| 08:34:55 | openstackgerrit | Merged openstack/openstacksdk master: NIT: Fix application credential https://review.opendev.org/734756 | |
| 13:11:51 | openstackgerrit | Artem Goncharov proposed openstack/openstacksdk master: Add support for filters into the project cleanup https://review.opendev.org/735336 | |
| 13:27:32 | mordred | frickler: I doubt it's on purpose | |
| 13:28:47 | mordred | gtema: if you get a chance this morning, could you look at https://review.opendev.org/#/c/734000/ - it's green now that we released cliff | |
| 13:29:19 | gtema | btw, just few hours ago noticed, that when we do now "openstack command list" - everything is doubled | |
| 13:30:05 | mordred | neat | |
| 13:33:02 | mordred | gtema: that is not true for me | |
| 13:33:13 | gtema | okay, pretty interesting | |
| 13:33:44 | gtema | every plugin is doubled for me and for my colleague | |
| 13:34:13 | gtema | lemme check again | |
| 13:34:24 | mordred | I'm using the container image with sha a865640d8bef | |
| 13:35:43 | gtema | hmm, you are right. Then something is corrupted in our env setup | |
| 13:36:03 | gtema | but it started really very recently | |
| 13:36:05 | mordred | cool. to double-check I pulled the latest image and it also still works | |
| 13:36:09 | gtema | okay, will dig further | |
| 13:36:37 | mordred | gtema: maybe it's something with the latest cliff change? (don't know how that would do it though) | |
| 13:37:00 | gtema | but funny that you exactly work in this area | |
| 13:37:09 | mordred | gtema: yeah | |
| 13:37:16 | gtema | it happens though only for our project, which adds plugins | |
| 13:37:30 | mordred | oh - wait! | |
| 13:37:32 | mordred | no | |
| 13:37:36 | mordred | it totally happens for me | |
| 13:41:00 | mordred | gtema: definitely caused by latest cliff | |
| 13:41:11 | mordred | let me see if applying the osc-lib patch fixes it | |
| 13:41:18 | gtema | okay | |
| 13:41:50 | mordred | yes | |
| 13:41:58 | mordred | https://review.opendev.org/#/c/734000/ fixes it | |
| 13:42:18 | mordred | I don't understand WHY - but since that wants to land anyway maybe I'm not too worried about it | |
| 13:42:21 | gtema | okay, great | |
| 13:42:50 | mordred | gtema: I went ahead and +A'd it. I'll cut a release as soon as that lands | |
| 13:42:58 | gtema | perfect | |
| 13:56:45 | vkmc | o/ | |
| 13:56:47 | vkmc | hi folks, we have a quick doubt on format | |
| 13:57:11 | vkmc | we want to comply with the human interface guidelines as much as possible | |
| 13:57:19 | vkmc | we ran into this https://docs.openstack.org/python-openstackclient/latest/contributor/command-options.html#required-options | |