*** karimb has quit IRC | 00:14 | |
*** pm90_ has quit IRC | 01:02 | |
*** britthouser has joined #openstack-sdks | 01:28 | |
*** britthou_ has joined #openstack-sdks | 01:30 | |
*** Qiming has joined #openstack-sdks | 01:30 | |
*** britthou_ has quit IRC | 01:32 | |
*** britthou_ has joined #openstack-sdks | 01:32 | |
*** britthouser has quit IRC | 01:34 | |
*** Yanyanhu has joined #openstack-sdks | 01:38 | |
*** britthou_ has quit IRC | 01:43 | |
*** Qiming_ has joined #openstack-sdks | 01:49 | |
*** Qiming has quit IRC | 01:53 | |
*** thrash is now known as thrash|g0ne | 02:04 | |
*** sigmavirus24_awa is now known as sigmavirus24 | 02:12 | |
greghaynes | mordred: is http://logs.openstack.org/23/183323/1/check/gate-os-client-config-python27/469a3f9/console.html#_2015-05-14_22_40_18_415 familiar to you at all? | 02:12 |
---|---|---|
greghaynes | mordred: seems unrelated to the change | 02:12 |
mordred | nope | 02:13 |
mordred | looks gross | 02:13 |
greghaynes | seems similar to the tox 2.0 breakage | 02:13 |
mordred | oh. yeah. very well might be | 02:14 |
*** sigmavirus24 is now known as sigmavirus24_awa | 02:15 | |
*** sigmavirus24_awa is now known as sigmavirus24 | 02:40 | |
*** sigmavirus24 is now known as sigmavirus24_awa | 02:52 | |
*** HenryG has quit IRC | 03:51 | |
*** stevemar has joined #openstack-sdks | 04:01 | |
*** pm90_ has joined #openstack-sdks | 04:43 | |
*** stevemar has quit IRC | 04:46 | |
*** pm90__ has joined #openstack-sdks | 04:47 | |
*** pm90_ has quit IRC | 04:47 | |
*** HenryG has joined #openstack-sdks | 05:38 | |
*** aufi has joined #openstack-sdks | 05:50 | |
*** pm90__ has quit IRC | 05:58 | |
*** aufi has quit IRC | 07:08 | |
*** Guest19895 has quit IRC | 07:21 | |
*** aufi has joined #openstack-sdks | 07:27 | |
*** chlong has quit IRC | 07:32 | |
*** Yanyanhu has quit IRC | 09:46 | |
*** thrash|g0ne is now known as thrash | 10:00 | |
*** Qiming_ has quit IRC | 10:02 | |
*** Qiming has joined #openstack-sdks | 11:04 | |
*** Qiming_ has joined #openstack-sdks | 11:23 | |
*** Qiming has quit IRC | 11:27 | |
*** Yanyanhu has joined #openstack-sdks | 12:13 | |
*** Yanyan has joined #openstack-sdks | 12:17 | |
*** Yanyanhu has quit IRC | 12:20 | |
*** Yanyan has quit IRC | 12:26 | |
*** Daviey has quit IRC | 12:39 | |
*** Daviey has joined #openstack-sdks | 12:39 | |
*** mattfarina has joined #openstack-sdks | 13:03 | |
*** bknudson has joined #openstack-sdks | 13:23 | |
*** stevemar has joined #openstack-sdks | 14:11 | |
*** Qiming__ has joined #openstack-sdks | 14:12 | |
*** Qiming_ has quit IRC | 14:15 | |
*** sigmavirus24_awa is now known as sigmavirus24 | 14:24 | |
stevemar | dtroyer, 7 nested bugs! | 14:25 |
stevemar | dtroyer, someone wants to do a bp for osc on the ML | 14:28 |
stevemar | i feel like saying 'go for it' just one the premise that it's not an osc member | 14:28 |
briancurtin | terrylhowe: unfortunately we overlooked something with some of these BaseProxy implementations, but should be a very small change to make: path_args. as im building a few things up, im realizing we left that path out. in the BaseProxy changes, it's one arg and we just pass it through, so i might try to tackle them all together instead of by verb - how does | 14:37 |
briancurtin | that sound? | 14:37 |
terrylhowe | I was just messing with object probably looking at the same thing | 14:38 |
openstackgerrit | Terry Howe proposed openstack/python-openstackclient: Fix functional test gate https://review.openstack.org/183546 | 14:40 |
briancurtin | terrylhowe: that's where i first hit it, but was just noticing it as well in database for making list work, and realized im going to keep hitting it | 14:41 |
terrylhowe | briancurtin: it kind of seems like if value isinstance of resource, don’t bother making a new one would help a little | 14:42 |
briancurtin | terrylhowe: yeah that could help as well. are you ok with trying to tackle this in one change since its similar or want them split by verb? | 14:43 |
terrylhowe | one would be better at this point | 14:44 |
*** aufi has quit IRC | 15:00 | |
*** gus has quit IRC | 15:08 | |
*** gus has joined #openstack-sdks | 15:09 | |
openstackgerrit | Terry Howe proposed stackforge/python-openstacksdk: Fix proxy object get comment https://review.openstack.org/183560 | 15:09 |
dtroyer | stevemar: I'm not sure what to make of that bp, I don't like the idea of spewing on a user's disk by default like that | 15:16 |
dtroyer | but I'll add it to the session and if those guys are there they can explain | 15:16 |
*** Qiming_ has joined #openstack-sdks | 15:17 | |
stevemar | maybe not by default | 15:18 |
*** Qiming__ has quit IRC | 15:20 | |
openstackgerrit | Robert Carr proposed openstack/python-openstackclient: --insecure ignores --os_cacert and OS_CACERT https://review.openstack.org/183567 | 15:24 |
openstackgerrit | Terry Howe proposed stackforge/python-openstacksdk: Fix proxy object get comment https://review.openstack.org/183560 | 15:30 |
openstackgerrit | greghaynes proposed openstack/os-client-config: Add set_one_cloud method https://review.openstack.org/183323 | 15:33 |
*** pm90_ has joined #openstack-sdks | 15:33 | |
*** Qiming_ has quit IRC | 15:49 | |
dtroyer | jeez, osc still sucks. trying to use Dreamhost and they don't allow Identity version discovery, their image catalog url has /v2 on it, and osc doesn't handle any of that correctly | 15:50 |
sdague | dtroyer: well, \o/ for service catalog standarization conversation at summit | 15:51 |
dtroyer | and version discovery support ('/' routes) needs to be added to defcore if not already there. oh wait, we have not standard for that either... | 15:53 |
morganfainberg | dtroyer: trying to update devstack to use keystone v3 only Apis with osc has been a pain. | 15:53 |
dtroyer | so all this crap is going to be stuffed into o-c-c to hide this nonsense… | 15:53 |
dtroyer | morganfainberg: I bet, besides all the hard-coded '/v2.0' in there | 15:53 |
morganfainberg | dtroyer: question: how *does* osc end up working w/o a catalog? I can't seem to make it do anything but tell me the catalog is broken when trying to bootstrap a catalog in. | 15:54 |
openstackgerrit | Merged stackforge/python-openstacksdk: Fix proxy object get comment https://review.openstack.org/183560 | 15:54 |
morganfainberg | Or ... Wait is this something magical about an admin token instead of a user/password? | 15:54 |
dtroyer | morganfainberg: dunno, I've never done it with v3 only before. | 15:54 |
dtroyer | well, that's different | 15:55 |
morganfainberg | I think I need to tear this apart more. It's been annoying. | 15:55 |
dtroyer | no catalog involved there, you pass the actual endpoint in --os-url | 15:55 |
dtroyer | of that and —os-token are present, all other auth is totally ignored and that is used directly | 15:56 |
morganfainberg | Ah. Ok I need to fix that help string then. It made reference to "auth endpoint" which was confusing. | 15:56 |
dtroyer | that's the bootstrap. and I've never tried it against v3 | 15:56 |
morganfainberg | Thanks. | 15:56 |
dtroyer | yeah, if we have 'auth' in connection to —os-url, that's a bug | 15:56 |
morganfainberg | I've been doing endless combinations of things to make this work :P | 15:56 |
morganfainberg | Well I want to kill admin token. | 15:57 |
morganfainberg | This cycle. | 15:57 |
dtroyer | like me and dreamhost…I finally booted a vm last night \o/ | 15:57 |
dtroyer | so then catalog bootstrap is a keystone-manage thing? | 15:57 |
morganfainberg | Make it a bootstrap a user/project and assignment via keystone-manage | 15:57 |
morganfainberg | Then use hat user to bootstrap everything. | 15:58 |
dtroyer | that is actually better | 15:58 |
morganfainberg | But I'll need a way to ignore catalog errors. | 15:58 |
morganfainberg | You'd still catalog bootstrap via the api. Just less logic in manage. | 15:58 |
morganfainberg | Cause manage gets scary if t gets too sky-net-y | 15:58 |
morganfainberg | But getting a basic user/role/project should be enough to get started. | 15:59 |
morganfainberg | And as a plus, no more insecure token "is_admin" middleware. | 15:59 |
dtroyer | i like this cleanliness trend | 16:00 |
*** shakamunyi has quit IRC | 16:00 | |
morganfainberg | Trying! :) | 16:00 |
*** barra204 has quit IRC | 16:00 | |
*** aayush_ has joined #openstack-sdks | 16:15 | |
aayush_ | Hi All, | 16:15 |
aayush_ | Can anyone please help with Openstack Python SDK | 16:16 |
aayush_ | ? | 16:16 |
briancurtin | aayush_: what about it? | 16:16 |
aayush_ | I am using code base from github page | 16:16 |
aayush_ | and trying to create image using | 16:17 |
aayush_ | conn.image.create_image(name='foo') | 16:17 |
aayush_ | but this is failing with error | 16:17 |
aayush_ | Image is none type | 16:17 |
aayush_ | @briancurtin can you confirm ? | 16:18 |
briancurtin | aayush_: i can't at the moment, need to rebuild my devstack i think. i think you mentioned this before, and i cant imagine it's possible to create an image with only a name. if you just give it a name what would it be creating? | 16:19 |
aayush_ | yes if I give it by command line or by curl then it creates image | 16:20 |
briancurtin | er, well i guess its documented to work that way | 16:20 |
briancurtin | once my devstack is ready i'll try it out. can you send the actual error so i can tell if it's the REST api or something we're returning ourselves? | 16:20 |
aayush_ | Ok | 16:20 |
aayush_ | yes sure | 16:21 |
aayush_ | let me re-create my setup | 16:22 |
dtroyer | briancurtin: image create is odd that way, a name with no file content creates a reservation, then you can do a create again with the file to fill the placeholder. I don't know what the use case for doing it that way was… | 16:28 |
terrylhowe | aayush_: that errror from glance, I know if you specify a disk_format, it still complains. You probably need other parameters and glance is just complaining about the wrong thing. | 16:28 |
*** pm90__ has joined #openstack-sdks | 16:28 | |
*** pm90_ has quit IRC | 16:28 | |
aayush_ | @terrylhowe : what is exact syntax ?? | 16:29 |
terrylhowe | aayush_: I don’t know off the top of my head | 16:29 |
aayush_ | Error goes like this : | 16:38 |
aayush_ | HttpException 400 : Client error Bad request | 16:39 |
*** pm90__ has quit IRC | 16:39 | |
aayush_ | invalid disk format 'None' for image | 16:39 |
*** pm90_ has joined #openstack-sdks | 16:39 | |
terrylhowe | yeh, the glance docs aren’t much help either | 16:39 |
terrylhowe | might help to look at the source to see what it really wants | 16:40 |
aayush_ | i am looking at this http://docs.openstack.org/developer/glance/glanceapi.html#reserve-a-new-image | 16:41 |
aayush_ | @terrylhowe : documentation says it requires 'x-image-meta-name' parameter and other few | 16:45 |
*** subscope has joined #openstack-sdks | 16:47 | |
*** aayush_ has quit IRC | 17:04 | |
*** Guest77307 has joined #openstack-sdks | 17:17 | |
*** Guest77307 is now known as redrobot | 17:20 | |
*** redrobot has left #openstack-sdks | 17:26 | |
briancurtin | wtf? "The resource is not available in a format acceptable to your browser." (trying to get object data) -- terrylhowe you ever seen this? | 17:26 |
*** redrobot has joined #openstack-sdks | 17:26 | |
terrylhowe | content type is application/somethingorother? briancurtin ? | 17:27 |
*** redrobot has quit IRC | 17:29 | |
briancurtin | terrylhowe: hmm. uploaded an image via horizon and it's application/octet-stream in there, but get_object_metadata is telling me text/plain. maybe i'll just forget about that for now, the fact i could even get it to do that is positive for what i'm working on | 17:29 |
briancurtin | (the path args thing) | 17:29 |
terrylhowe | I was only messing with get_object I didn’t try get_object_metadata | 17:30 |
*** redrobot has joined #openstack-sdks | 17:36 | |
*** redrobot is now known as Guest63193 | 17:36 | |
*** Guest63193 is now known as redrobot | 17:39 | |
*** subscope has quit IRC | 17:44 | |
*** subscope has joined #openstack-sdks | 17:49 | |
*** openstackgerrit has quit IRC | 18:22 | |
*** openstackgerrit has joined #openstack-sdks | 18:23 | |
*** shakamunyi has joined #openstack-sdks | 18:36 | |
*** barra204 has joined #openstack-sdks | 18:36 | |
*** stevemar has quit IRC | 18:50 | |
*** stevemar has joined #openstack-sdks | 18:59 | |
*** pm90_ has quit IRC | 19:18 | |
*** pm90_ has joined #openstack-sdks | 19:19 | |
*** pm90_ has quit IRC | 19:23 | |
openstackgerrit | Terry Howe proposed openstack/python-openstackclient: Fix functional test gate https://review.openstack.org/183546 | 19:25 |
openstackgerrit | Brian Curtin proposed stackforge/python-openstacksdk: WIP: accept intermediate path arguments at proxy https://review.openstack.org/183664 | 19:32 |
briancurtin | terrylhowe: can you have a look at https://review.openstack.org/#/c/183664/? no tests yet, have been manually working with it. been testing get_object and get_object_metadata both with string names for obj and cont, and with actual resource objects, and it works fine. i think the API is what it is, but want to make sure the way it's enabled looks reasonable | 19:33 |
briancurtin | before i go too far | 19:33 |
*** sigmavirus24 is now known as sigmavirus24_awa | 19:38 | |
*** barra204 has quit IRC | 19:48 | |
*** shakamunyi has quit IRC | 19:48 | |
*** subscope has quit IRC | 19:54 | |
openstackgerrit | Adam Sheldon proposed stackforge/openstack-sdk-dotnet: Fix: CreateStorageObject stream was getting closed before it was read. https://review.openstack.org/183688 | 20:17 |
openstackgerrit | Merged stackforge/openstack-sdk-dotnet: Fix: CreateStorageObject stream was getting closed before it was read. https://review.openstack.org/183688 | 20:40 |
etoews | is it okay if i rebase and add a patch set to https://review.openstack.org/#/c/181063/ ? | 20:50 |
etoews | it hasn't been touched in a week | 20:51 |
*** mattfarina has quit IRC | 21:04 | |
openstackgerrit | Terry Howe proposed stackforge/python-openstacksdk: No headers in body for create and update https://review.openstack.org/183704 | 21:05 |
terrylhowe | rip it up etoews | 21:08 |
openstackgerrit | Terry Howe proposed stackforge/python-openstacksdk: No headers in body for create and update https://review.openstack.org/183704 | 21:16 |
*** stevemar has quit IRC | 21:17 | |
bknudson | are there docs for sdk? | 21:27 |
bknudson | like for other bindings http://docs.openstack.org/developer/language-bindings.html | 21:27 |
briancurtin | bknudson: eh, i need to do a build of them since we havent in a while, but they live at http://python-openstacksdk.readthedocs.org/en/latest/ | 21:28 |
briancurtin | bknudson: been really cramming to get things done heading into teh summit so user guides and whatnot have been held off until after we get a lot of our current APIs fleshed out, which we're almost through | 21:28 |
bknudson | is there some reason this is stackforge and not openstack? | 21:29 |
bknudson | looks like there's not enough identity to use it in auth_token yet. | 21:30 |
morganfainberg | bknudson: talked with dtroyer and SDK is hopefully going to use keystoneauth | 21:31 |
briancurtin | bknudson: it's in stackforge because that's where it started. Does it need to go somewhere else? | 21:31 |
bknudson | auth_token uses ops like validate_token , fetching revocation lists , fetching certs | 21:31 |
morganfainberg | bknudson: fwiw | 21:31 |
morganfainberg | briancurtin: probably eventually to "openstack" | 21:31 |
morganfainberg | briancurtin: but i mean probably no rush on it. | 21:31 |
briancurtin | I'm in no hurry to change the url of the repo | 21:31 |
bknudson | do contributors to stackforge projects get ATC status? might help. | 21:32 |
bknudson | not that it matters to me personally | 21:32 |
*** breton has joined #openstack-sdks | 21:32 | |
morganfainberg | briancurtin: you might be the one different one in the openstack community then:P seems like everyone wants to be OpenStack namespace ;) | 21:32 |
morganfainberg | briancurtin: [it's not a bad thing] | 21:32 |
bknudson | I'll just mention it would be nice to see the docs in the normal place http://docs.openstack.org/developer/language-bindings.html | 21:33 |
morganfainberg | briancurtin: will come bug you as we get keystoneauth a bit closer to usable. I'd like it to be really generic so SDK can use it too. | 21:33 |
morganfainberg | i know dtroyer is already looking at this stuff too | 21:34 |
briancurtin | bknudson: yeah that'd be nice but havent had time. will try to get things up there in teh next few weeks, with a release probably happening early next week, i hope | 21:35 |
*** thrash is now known as thrash|wknd | 22:01 | |
*** bknudson has quit IRC | 22:10 | |
*** jamielennox|away is now known as jamielennox | 22:48 | |
*** terrylhowe has quit IRC | 23:03 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!