Earlier  
Later
New Search
Posted Nick Remark
#openstack-sdks - 2019-06-07
03:49:32 shubham_potale I have one question regarding the url formation for the APIs like tacker, neutron apis for example openstack --debug network list the url for the above command is http://10.232.48.200:9696/v2.0/networks in above url where that v2.0 is appended is not clear to me. If we see openstack endpoint list we get network base url as http://10.232.48.200:9696/ for this part (/v2.0/networks) i found that /networks is app
03:50:41 shubham_potale if anyone knows about it, please let me know
04:40:28 efried mordred: Nova side looks good (except I needed to bump the ost l-c there too, which seems weird) so for my money we can cut a release any time. LMK if you want me to do the paperwork.
06:44:50 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Add access alias (aka) for the resource attributes https://review.opendev.org/663846
06:48:42 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Use Resource layer for next compute methods https://review.opendev.org/663064
06:50:16 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Use resource layer for compute flavors https://review.opendev.org/650903
09:27:53 openstackgerrit Dmitry Tantsur proposed openstack/openstacksdk master: baremetal: raise more specific ResourceFailure in wait_for_* methods https://review.opendev.org/663867
11:03:41 dtantsur folks, a trivial fix ^^^ when you have some time
11:10:07 mordred gtema: ^^ have a sec to look at that patch from dtantsur ? also - nice aka patch
11:10:41 gtema yeah - it found actual problem in the image checksum validation - hehe
11:13:17 mordred \o/
11:37:00 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Add access alias (aka) for the resource attributes https://review.opendev.org/663846
11:55:32 mordred gtema: if that passes tests, that looks pretty nice!
11:55:53 gtema thanks
11:56:13 gtema some time later probably rename "alias" to "alternate_attribute"
11:57:44 gtema it actually proved, that in the case of image the checksum validation would have never worked really
11:58:07 gtema since after fetch the checksum was in reality empty
11:59:44 gtema what's happening - already second change fails due to timeout in nodepool job
12:18:19 mordred uhoh
12:35:05 openstackgerrit Yves-Gwenael Bourhis proposed openstack/shade master: URL encode swift objects endpoints https://review.opendev.org/663922
12:39:34 openstackgerrit Yves-Gwenael Bourhis proposed openstack/openstacksdk master: URL encode swift objects endpoints https://review.opendev.org/663925
12:56:11 openstackgerrit Merged openstack/openstacksdk master: baremetal: raise more specific ResourceFailure in wait_for_* methods https://review.opendev.org/663867
13:14:30 efried mordred: We green for a release of sdk?
13:16:14 mordred efried: yup!
13:16:21 efried mordred: I'm proposing, stand by.
13:17:22 mordred remote: https://review.opendev.org/663939 Release 0.31.0 of openstacksdk
13:17:34 mordred efried: oh - crappit - didn't see that last one from you
13:17:38 efried mordred: https://review.opendev.org/663940
13:17:59 efried dangit, I woulda beat you if I hadn't sat there staring at the commit message.
13:18:08 mordred efried: I'll ack yours and abandon mine
13:18:13 mordred your commit message is better
13:18:13 efried either way
13:18:37 efried :)
13:25:47 openstackgerrit Merged openstack/openstacksdk master: Use Resource layer for compute AZ and Aggregates https://review.opendev.org/662738
14:53:19 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Add access alias (aka) for the resource attributes https://review.opendev.org/663846
15:09:37 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Rework statistics reporting https://review.opendev.org/659841
17:27:43 openstackgerrit Merged openstack/keystoneauth master: add a handler for unknown HTTP errors https://review.opendev.org/663132
18:13:11 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Rework statistics reporting https://review.opendev.org/659841
18:21:38 gtema mordred: still here?
20:09:59 efried mordred: FYI: https://review.opendev.org/656023 (green CI) is using 0.30.0
20:18:00 mordred efried: \o/
#openstack-sdks - 2019-06-08
06:31:14 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Add access alias (aka) for the resource attributes https://review.opendev.org/663846
08:46:33 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Add access alias (aka) for the resource attributes https://review.opendev.org/663846
08:48:16 mordred gtema: just a bad test?
08:48:24 gtema yes
08:48:52 mordred gtema: it's a truly lovely bug you found :)
08:49:09 gtema I replaced test_create_image_put_v2_wrong_checksum_delete with the test_create_image_put_v2_checksums I added in PS4
08:49:10 gtema sure
08:49:19 gtema I was like crazy debugging nodepool logs
08:49:32 gtema was not able to sleep pulling out my hair
08:49:38 gtema :D
08:50:14 gtema and that's all just from one line changed in resource.py
08:50:37 gtema namely line :1105
08:51:30 gtema with the flavors patch - nodepool seems not to like it either
08:51:42 gtema but I had no time to look in detail
08:52:01 mordred gtema: nodepool is here to keep you from sleeping ;)
08:52:10 gtema hehe
08:53:08 gtema and working on Saturday is also fun
08:53:21 gtema unless wife starts screaming ;-)
08:53:27 mordred so is waking up at 0300 for no good reason :)
08:53:43 gtema sure
10:22:44 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Use resource layer for compute flavors https://review.opendev.org/650903
#openstack-sdks - 2019-06-10
05:38:23 shubham_potale Hi All,
05:38:38 shubham_potale I have one question regarding the url formation of the APIs like tacker, neutron
05:40:02 shubham_potale for ex. for the command openstack --debug vnf list, url is GET http://10.232.48.200:9890/v1.0/vnfs in this url for this part (v1.0/vnfs) /vnfs is appened in tackerclient but where v1.0 is appended is not clear to me. if we see openstack endponit list base url for tacker is http://10.232.48.200:9890/. if tou know about it let me know
15:29:34 stephenfin dtroyer: Any reason we use 'openstack server resize --revert' and not 'openstack server resize revert', seeing as "resize revert" feels like an action?
16:09:52 efried mordred: https://review.opendev.org/#/c/664381/ bumps u-c for openstacksdk to 0.31.0, for your acking pleasure
16:14:54 efried oh, core in requirements are we? Even better.
16:26:57 openstackgerrit Stephen Finucane proposed openstack/python-openstackclient master: Add 'openstack server migrate [--resize|--confirm]' flags https://review.opendev.org/664393
17:41:30 mnaser ok, i'm not the only one ever to deal with this but
17:41:41 mnaser is there any reason why we couldnt add a cascade delete option to openstackclient
17:55:55 dtroyer stephenfin: I didn't do that because I didn't think about it. set/unset is a natural opposite pair, revert by itself might be too generic but we certainly could have a two word verb just as much as we have multi-word resources
17:56:28 dtroyer modulo if I was the one who originally mapped that command, I honestly don't remember, so if its wrong I prolly did :)
17:57:15 dtroyer mnaser: is that different from the purge command? I assume so but am not sure exactly how…
17:57:55 mnaser dtroyer: i think purge is meant to kill an entire project, where as this.. the idea is that i can give it a network and it can list all subnets, then list all ports a ttached tos ubnets, and delete from ports up to make the network deletable
17:58:00 mnaser unless purge exists and does just that..
17:58:48 dtroyer I have not used nor worked on purge, but yeah I do think it is for a whole project. so do we need to come up with a way to specify subsets of project-owned resources?
17:59:19 mnaser dtroyer: yeah well the idea is i'm adding a --cascade command and i guess it'll naively gather the resources we know about at least
18:00:41 dtroyer you are tying that to a network? would it make sense to also have subsets of that? I don't have enough Neutron in my head for it to be obvious where else to use that idea
18:01:22 mnaser dtroyer: correct, you would ask for a cascading network delete, and it would go, list all subnets, then list all ports attached to each subnet, delete all ports, and then now we'll be able to delete all subnets, then we'll be able to delete the network
18:01:41 mnaser and i'm adding a --force which will do it regardless, and a confirmation of what resource will be deleted before
18:02:24 dtroyer we'll need to think about output on a delete command, they don't do that now other than warnings and errors…
18:02:36 mnaser yeah i noticed that too..
18:03:02 mnaser i mean i can drop the --force thing too
18:03:08 mnaser it was mostly for my development too
18:03:43 mnaser its probably eaiser to do it without it for now
18:04:13 dtroyer —force seems useful if the API implements something for it. so would —dry-run but I think a way to see what would be deleted is necessary for —dry-run to actually be useful...
18:40:21 openstackgerrit Mohammed Naser proposed openstack/python-openstackclient master: port: drop host-id and device-id https://review.opendev.org/664421
19:02:55 openstackgerrit Mohammed Naser proposed openstack/python-openstackclient master: port: drop host-id and device-id https://review.opendev.org/664421
19:03:12 openstackgerrit Mohammed Naser proposed openstack/python-openstackclient master: port: drop host-id and device-id https://review.opendev.org/664421
19:05:35 openstackgerrit Mohammed Naser proposed openstack/python-openstackclient master: port: drop host-id and device-id https://review.opendev.org/664421
19:05:36 openstackgerrit Mohammed Naser proposed openstack/python-openstackclient master: network: add cascade delete https://review.opendev.org/664422
19:06:17 mnaser dtroyer: ^ first run without --dry-run
19:06:26 mnaser adding --dry-run should be fairly straightforward
#openstack-sdks - 2019-06-11
09:29:53 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Use resource layer for compute flavors https://review.opendev.org/650903
09:35:28 openstackgerrit Artem Goncharov proposed openstack/openstacksdk master: Use Resource layer for network SecurityGroups https://review.opendev.org/662724
11:27:06 mordred gtema: morning! we ran in to https://storyboard.openstack.org/#!/story/2005851 with nodepool and had to pin to a previous version
11:27:26 gtema morning mordred
11:27:37 mordred ah - I see you commented on it ... yeah, that header is present
11:27:40 gtema Is it possible to see a real query going there?

Earlier   Later