Earlier  
Later
New Search
Posted Nick Remark
#openstack-sdks - 2017-10-10
09:09:09 openstackgerrit Stephen Finucane proposed openstack/cliff master: Generate demoapp CLI refernece https://review.openstack.org/483134
09:14:09 openstackgerrit Stephen Finucane proposed openstack/cliff master: doc: Cleanup of demoapp doc https://review.openstack.org/510802
09:58:16 umbSublime is there currently a way to get a flavor's extra_specs from the SDK, I can't seem to find one :/.
11:32:39 openstackgerrit Merged openstack/cliff master: doc: Remove blank lines between term and definition https://review.openstack.org/485698
12:49:01 mordred umbSublime: no, openstacksdk does not have support for flavor extra_specs at the moment. shade does, so if you're blocked on it I'd suggest shade - there will be work to merge the two over the next cycle
12:49:07 mordred thingee: on-boarding room sounds great
12:52:17 umbSublime mordred, thank you, I'm no developer, but looking at the codebase for the SDK I was wondering if as a work-around for now I could just modify the Flavor object to add a property for extra_specs would just "work" https://github.com/openstack/python-openstacksdk/blob/79462437058e37b83b8153531b5078ddc5a1edb8/openstack/compute/v2/flavor.py#L56.
12:52:25 umbSublime Using shade would work indeed, but performance wise in my experience using the SDK directly is much faster.
12:53:24 umbSublime concerning you last bit, you mean if all goes well in next cycle there will be a merge of shade and openstacksdk ?
12:55:14 mordred umbSublime: yah - that's likely because of the current shade behavior to use list() and then client-side filtering for get() calls ...
12:56:01 mordred umbSublime: there is a constructor parameter now "use_direct_get" which switches behavior to using get() instead ... there is also, ironically enough, a flag to tell shade to NOT fetch extra_specs every time you look for a flavor :)
12:56:24 mordred (the current default behavior is to fetch a flavor's extra_specs when fetching a flavor - this was a bad choice)
12:57:58 mordred umbSublime: and yes - we've merged the teams just recently and we'll be working to move what is now shade into being some classes inside of sdk
12:58:01 umbSublime sort or like when using nova/openstack cli, it does 2 api calls, one to get flavor and another to get the extra_specs with no way to disable extra_specs
12:58:18 mordred umbSublime: unfortunately extra_specs are an extra REST call - so just adding an attribute there won't get you any data
12:59:38 umbSublime maybe another question, with the sdk is it possible to use the connection to do a "custom" api call ? if that's the case i could work-around this and deal with the api call json result myself
13:00:47 mordred umbSublime: yah - re: extra_specs ... if you add:
13:00:49 mordred shade:
13:00:51 mordred get_flavor_extra_specs: false
13:01:10 mordred to clouds.yaml (top level) it will turn off the shade behavior (in case that winds up being useful to you)
13:01:26 mordred umbSublime: not at the moment (re REST) - but I actually have a patch up to add that ability
13:01:57 mordred umbSublime: in part of the patch series aimed at laying the plumbing groundwork for integrating shade and sdk with each other
13:02:05 umbSublime awesome, then I guess I'll work-around my issue with shade for now and switch to using the REST call when it's ready. Thanks for the help
13:02:06 openstackgerrit Rodolfo Alonso Hernandez proposed openstack/python-openstacksdk master: Avoid default mutable values in arguments https://review.openstack.org/510870
13:02:40 mordred umbSublime: sure thing! hopefully in the not-too-distant-future we'll have answers for both side of this :)
13:03:15 umbSublime mordred, maybe I can do a "hack" to get the token from the connection object and just use requests module for my custom api call. Would that be possible ?
13:05:22 mordred umbSublime: I think instead just try using the session property on the Connection
13:05:40 mordred umbSublime: one sec ...
13:05:59 umbSublime ohh right, that's actually what i was refering too. Thanks again
13:12:30 openstackgerrit Rodolfo Alonso Hernandez proposed openstack/python-openstackclient master: Avoid default mutable values in arguments https://review.openstack.org/510876
15:10:25 openstackgerrit Stephen Finucane proposed openstack/cliff master: Generate demoapp CLI refernece https://review.openstack.org/483134
15:10:26 openstackgerrit Stephen Finucane proposed openstack/cliff master: doc: Cleanup of demoapp doc https://review.openstack.org/510802
15:24:53 openstackgerrit Stephen Finucane proposed openstack/cliff master: sphinxext: Warn if namespace or command pattern invalid https://review.openstack.org/510913
16:39:29 efried mordred cdent (sdague) Should we be adding a block-storage endpoint in devstack, since that be the official service type for cinder?
16:40:26 cdent efried: yes, it would be great if what’s in devstack matched sta
16:40:46 cdent but it’s likely/possible that could break some stuff
16:40:50 efried cdent Okay. I'm going to be poking in the dark a little bit here, but I'll shoot you the review once I've got it.
16:40:56 efried cdent Oh, I wasn't going to remove volumev3
16:40:57 cdent ✔
16:40:58 efried yet
16:41:45 mordred efried: yah- definitely adding a block-storage owuld be great
16:42:06 mordred efried: obvoiusly if you do, add the unversioned endpoint as block-storage, not the versoined :)
16:42:33 efried mordred Cool, on it. (Currently trying to figure out whether I can duplicate the service name, or if not, what I should call it)
16:44:23 mordred efried: service name carries no meaning - so it's fine to duplicate it
16:44:28 efried sweet
16:46:30 cdent <pic of service> this is not a service
16:50:43 efried cdent mordred This look right? https://review.openstack.org/510939
16:52:13 cdent efried: superficially glancing, yes
16:52:27 cdent it’ll be a great shiny day when we can get rid of these rest of those
16:53:48 efried cdent I got here from test failures on https://review.openstack.org/#/c/508345/ which I'm pretty sure are happening because there's no 'block-storage' entry in the catalog.
16:54:31 efried though I thought ksa was supposed to try the aliases, maybe?
16:54:32 cdent isn’t it supposed to fall back to aliases?
16:54:35 efried I'm confusing myself.
16:54:36 cdent jinx
16:54:39 cdent me too
16:54:57 efried and cinder is way messed up
17:13:29 mordred cdent, efried: alias fallback is the last missing thing we haven't yet added to ksa
17:13:39 efried ah
17:13:50 mordred cdent, efried: the patch is here: https://review.openstack.org/#/c/484341/ and it's on my list to finish as soon as zuul v3 rollout is done
17:14:06 cdent k3w!
18:47:32 briancurtin umbSublime: if you just want the token you can call https://developer.openstack.org/sdks/python/openstacksdk/users/connection.html#openstack.connection.Connection.authorize
18:48:17 briancurtin that’s used for more explicitly authorizing (if necessary, it’s not the usual way you auth), and that will return the auth token
22:10:48 openstackgerrit Jackie Truong proposed openstack/python-openstackclient master: Add trusted_image_certificates to server https://review.openstack.org/501926
23:20:07 openstackgerrit Merged openstack/cliff master: Alias exit to cmd2's quit command to exit interactive shell https://review.openstack.org/491217
#openstack-sdks - 2017-10-11
04:25:51 openstackgerrit Monty Taylor proposed openstack/keystoneauth master: Split request logging into four different loggers https://review.openstack.org/505764
06:51:53 ldl https://stackoverflow.com/questions/46681591/my-project-and-admin-panels-resource-can-not-be-retrieve-but-identity-resource
06:52:15 ldl Hi, friends, do you have encountered this issue?
08:14:16 umbSublime thanks briancurtin
08:50:21 slaweq_ hello briancurtin
08:50:41 slaweq_ briancurtin: some time ago You told me that You will check if You can release new version of OpenStack SDK
08:50:50 slaweq_ did You check this already?
08:50:59 slaweq_ do You know if it's possible to release it?
09:47:29 openstackgerrit Stephen Finucane proposed openstack/python-openstackclient master: Convert 'server' commands to use autoprogram https://review.openstack.org/458452
13:43:22 kuzko hello, I am using shade to try to create my instances, but I cannot seem to boot from a volume only, image seems to be a requirement
13:44:13 kuzko I am getting TypeError: create_server() missing 1 required positional argument: 'image'
13:51:32 mordred kuzko: one sec ... lemme check something
13:53:00 mordred kuzko: OH FOR THE LOVE OF ... wow. yah. that's a bug, I'm sorry. pass image=None and it should work - and I'll fix it in the code
13:53:31 kuzko ok, I've seen that placing image=None works.. but, now my volume isn't bootable ^^
13:54:00 mordred oh. well, that's less good - can you paste me your whole create_server call?
13:54:23 mordred this is basic feature so I'm *extra* unpleased with it not working for you
13:54:43 kuzko ok, I create a volume beforehand, maybe there is a switch to set it bootable I did not see.. hang on i'm pastebining it
13:55:13 kuzko don't sweat it, there is a workaround so that's not biggie...
13:56:25 kuzko https://pastebin.com/FrySUpSP there you go mordred
13:57:25 kuzko that's the correct one, remove image=None and that's what I had beforehand
14:01:15 mordred kuzko: sweet - thanks. that's SUPPOSED to create a bootable volume (the create_volume(image= ..)
14:02:21 mordred aha
14:02:41 mordred kuzko: there's a way to toggle the bootable flag on a volume but we don't expose it - I'll add that for you too
14:05:01 kuzko well that's nice, I used pip to fech the libray, should I update using pip or git clone?
14:05:15 kuzko *library...
14:10:01 mordred kuzko: I'm overdue for a release - so I'll try to get one cut tomorrow
14:24:02 umbSublime mordred, if you can add that feature for custom api call like we discussed yesterday in the next release (if it's ready that is) it would be awesome !!
14:25:46 mordred umbSublime: yes - I fully intend to get that in to the next release
14:26:18 umbSublime awesome!!
14:26:21 mordred umbSublime: well - next sdk release - not next shade release
14:26:45 umbSublime ohh right you guys were talking about shade, my bad :)
14:27:15 mordred umbSublime: well - hopefully that confusion will not persist for TOO much longer :)
14:27:21 umbSublime hehe
14:27:25 mordred but that certaily won't get done by tomorrow :)
14:27:41 umbSublime no worries, I have my work-around working :)
14:37:34 kuzko is shade supposed to become the SDK at a later date? ^^

Earlier   Later