| *** reed_ has quit IRC | 00:00 | |
| *** markwash has joined #openstack-dev | 00:11 | |
| *** otherwiseguy has quit IRC | 00:12 | |
| *** MaxV has joined #openstack-dev | 00:13 | |
| *** MaxV has quit IRC | 00:18 | |
| *** jab416171 has quit IRC | 00:18 | |
| *** mrodden has joined #openstack-dev | 00:20 | |
| *** Mandell has quit IRC | 00:26 | |
| *** Thor has joined #openstack-dev | 00:26 | |
| ayoung | jamielennox, hmmm. not sure | 00:29 |
|---|---|---|
| ayoung | jamielennox, as dolphm pointed out, once we document them, we set them in stone\ | 00:30 |
| ayoung | and the cert stuff is, well, clunky | 00:30 |
| ayoung | jamielennox, but..yeah take them and run with them. | 00:30 |
| *** huats has quit IRC | 00:31 | |
| *** Thor has quit IRC | 00:31 | |
| *** huats has joined #openstack-dev | 00:32 | |
| *** huats has joined #openstack-dev | 00:32 | |
| *** Thor has joined #openstack-dev | 00:33 | |
| *** chenxu has joined #openstack-dev | 00:33 | |
| *** Tross has joined #openstack-dev | 00:34 | |
| *** bdpayne has quit IRC | 00:34 | |
| *** sarob has quit IRC | 00:34 | |
| *** sarob has joined #openstack-dev | 00:35 | |
| *** davidhadas has quit IRC | 00:38 | |
| *** sarob has quit IRC | 00:39 | |
| *** davidhadas has joined #openstack-dev | 00:39 | |
| *** dstanek has joined #openstack-dev | 00:42 | |
| mordred | dolphm: I thnk we should just land the https://review.openstack.org/#/c/48988/ change and worry about the others later | 00:42 |
| *** alop has quit IRC | 00:42 | |
| *** SumitNaiksatam has joined #openstack-dev | 00:44 | |
| *** nosnos has joined #openstack-dev | 00:50 | |
| *** adjohn has quit IRC | 00:50 | |
| *** chenxu has quit IRC | 00:50 | |
| *** freedomhui has joined #openstack-dev | 00:52 | |
| *** otherwiseguy has joined #openstack-dev | 00:52 | |
| *** Javin has joined #openstack-dev | 00:52 | |
| *** shakayumi has quit IRC | 00:56 | |
| *** anteaya has quit IRC | 00:56 | |
| *** colinmcnamara has quit IRC | 00:57 | |
| *** dhubbard_ is now known as dhubbard | 00:59 | |
| *** dhubbard is now known as dhubbard_gone | 00:59 | |
| *** Tross has quit IRC | 01:01 | |
| *** amotoki has joined #openstack-dev | 01:03 | |
| *** roz_ has joined #openstack-dev | 01:04 | |
| *** emagana has quit IRC | 01:04 | |
| *** thingee is now known as thingee_zzz | 01:05 | |
| *** roz has quit IRC | 01:05 | |
| *** sarob has joined #openstack-dev | 01:05 | |
| *** gmurphy has quit IRC | 01:06 | |
| *** chenxu has joined #openstack-dev | 01:06 | |
| *** adjohn has joined #openstack-dev | 01:08 | |
| *** sarob has quit IRC | 01:10 | |
| *** faramir1 has joined #openstack-dev | 01:11 | |
| *** changbl has joined #openstack-dev | 01:12 | |
| *** sarob has joined #openstack-dev | 01:13 | |
| *** FatDarrel has quit IRC | 01:14 | |
| *** dstanek has quit IRC | 01:16 | |
| *** sarob has quit IRC | 01:18 | |
| *** radez_g0n3 is now known as radez | 01:20 | |
| *** nati_ueno has quit IRC | 01:22 | |
| *** gmurphy has joined #openstack-dev | 01:22 | |
| *** bdpayne has joined #openstack-dev | 01:25 | |
| *** RajeshMohan has quit IRC | 01:25 | |
| *** RajeshMohan has joined #openstack-dev | 01:26 | |
| *** edmund has joined #openstack-dev | 01:26 | |
| *** adjohn has quit IRC | 01:26 | |
| *** bdpayne has quit IRC | 01:27 | |
| *** spzala has joined #openstack-dev | 01:27 | |
| *** dguitarbite has joined #openstack-dev | 01:28 | |
| *** melwitt1 has quit IRC | 01:30 | |
| *** erkules_ has joined #openstack-dev | 01:31 | |
| *** erkules has quit IRC | 01:33 | |
| *** boris-42 has quit IRC | 01:33 | |
| *** gmurphy has quit IRC | 01:39 | |
| *** adjohn has joined #openstack-dev | 01:39 | |
| *** danwent has quit IRC | 01:40 | |
| *** adjohn has quit IRC | 01:45 | |
| *** egallen has quit IRC | 01:45 | |
| *** neelashah has joined #openstack-dev | 01:46 | |
| dolphm | ayoung: jamielennox: feature frozen a milestone ahead of the services, and set in stone once the implementation sees a release (bugs and all) | 01:50 |
| dolphm | ayoung: jamielennox: for example, if there was a bug in the oauth1 api or something, we could still fix the api and fix the impl... up until about today | 01:50 |
| *** gmurphy has joined #openstack-dev | 01:52 | |
| *** angdraug has quit IRC | 01:53 | |
| *** adjohn has joined #openstack-dev | 01:56 | |
| jamielennox | dolphm: damn, i was a bit excited then as i was just thinking that the oauth api is just not really right | 01:56 |
| dolphm | jamielennox: how so? | 01:57 |
| jamielennox | dolphm: i've been going back and forwards with stevemar on the client review | 01:57 |
| dolphm | jamielennox: i' | 01:58 |
| jamielennox | i think that an 'entity' whatever that may be should always have a unique id and be a top level namespace | 01:58 |
| jamielennox | at the moment everything about oauth is namespaced behind a user_id | 01:58 |
| jamielennox | so /user/:id/consumer/X | 01:58 |
| jamielennox | which causes problems with how clients are set up | 01:58 |
| dolphm | jamielennox: clientS, or our client? | 01:59 |
| jamielennox | ah, at least ours | 01:59 |
| jamielennox | but i'm pretty sure it's bigger than that | 01:59 |
| dolphm | jamielennox: i'll take the blame for that, then | 01:59 |
| jamielennox | so for future api reviews i'm going to say that everything needs to have a unique id | 01:59 |
| jamielennox | and be at /OS-EXT/entity/:id | 02:00 |
| dolphm | jamielennox: it made sense (at some point) that users own the consumer keys that they create -- do you disagree? | 02:00 |
| jamielennox | and then we just use permissions to check who it actually belongs to | 02:00 |
| jamielennox | so the problem comes in in that managers create entity objects | 02:00 |
| jamielennox | so user = client.users.get(x) | 02:00 |
| jamielennox | that .get() doesn't work now without a user_id | 02:01 |
| jamielennox | there is some stuff i'm hoping we can override - like for user.delete calls client.users.delete(user.id) | 02:01 |
| *** xjiujiu has joined #openstack-dev | 02:01 | |
| jamielennox | i'm hoping if we save user_id in the consumer entity we can still reconstruct the URL - but it means that a lot of the inheritted helper functions just won't work | 02:02 |
| *** sthaha has joined #openstack-dev | 02:03 | |
| jamielennox | and so given that consumer ids and token ids are globally unique anyway it would have just worked better if we'd had them at a top level | 02:03 |
| ayoung | dolphm, I see your point, but I'm starting to think that relationships like "a users entity" should be handled by links, and, where ever possible, we references entities (genereic meaning of the world) by top level URLs | 02:05 |
| *** adjohn has quit IRC | 02:05 | |
| *** novas0x2a|laptop has quit IRC | 02:05 | |
| jamielennox | ayoung: i'm of the opinion that if you define an extension then EVERY url that the extension needs should live under it's own /OS-EXT route | 02:05 |
| *** sarob has joined #openstack-dev | 02:06 | |
| jamielennox | no more intertwining core and extension URIs | 02:06 |
| ayoung | jamielennox, Oh, yeah, under the extensions root is fine | 02:06 |
| ayoung | top level is a relative term | 02:07 |
| jamielennox | the url we currently have could have been expressed as /OS-OAUTH1/users/:user_id/consumer/:consumer_id and be just as expressive | 02:07 |
| jamielennox | exactly the same but with the prefix | 02:07 |
| jamielennox | ayoung: sure - i mean top level as after /v3/ | 02:07 |
| *** briancurtin has quit IRC | 02:08 | |
| ayoung | jamielennox, my point is that it makes more sense if you are going to extend the user object that the things it points to are pointed to by links. It really doesn't matter where the links point to. HAving them in the same subtree is useless without some form of enumeraiton to know that they are there | 02:08 |
| ayoung | jamielennox, I think we are in violent agreement | 02:08 |
| ayoung | What is the best way to read an file designed to set environemt variables from Python? | 02:09 |
| jamielennox | good, cause i'm trying to re-read that last sentence and i'm not sure what you mean exactly | 02:09 |
| ayoung | jamielennox, so, you know that HTML I was tlaking about earlier today? THat kind of thinking. | 02:10 |
| jamielennox | ayoung: line = f.readline() line.split('=') | 02:10 |
| jamielennox | at a guess | 02:10 |
| ayoung | https://review.openstack.org/#/c/29105/ | 02:10 |
| ayoung | jamielennox, yeah, you're probably right...just write the code | 02:10 |
| *** sarob has quit IRC | 02:10 | |
| jamielennox | ayoung: i'm fine with links in user objects, that's a good thing for extensions - i'm still not convinced on html | 02:12 |
| jamielennox | and by fine i mean that sounds like a good idea | 02:12 |
| ayoung | jamielennox, "not conviced on html" what you prefer flash? | 02:13 |
| ayoung | Silverlight | 02:13 |
| jamielennox | ayoung: the other thing i guess you could do is snapshot the environment variables, source your file and then see what changed :) | 02:13 |
| *** Javin has quit IRC | 02:14 | |
| jamielennox | ayoung: heh, no i'm just not sure yet that keystone has any business providing a html front end | 02:15 |
| ayoung | jamielennox, don't think I could do that cleanly from inside python....well, I guess I could in a subshell. Nah, I'll just read them fro now. | 02:15 |
| *** mrda has quit IRC | 02:15 | |
| jamielennox | ayoung: wasn't a serious suggestion | 02:15 |
| ayoung | jamielennox, It should have been there from the start. We'd have a much better web services impl if we had done discovery right from the start | 02:16 |
| jamielennox | ayoung: completely agree about discovery | 02:16 |
| ayoung | jamielennox, I know it wasn't serious...but it had a certain "wouldn't it be cool..." factor | 02:16 |
| ayoung | jamielennox, and...we'll end up with better APIs. cus once you try to do it from the browser and can't, you'll throw in the missing links....like forms | 02:17 |
| *** jab416171 has joined #openstack-dev | 02:17 | |
| jamielennox | compromise on an extension? Anything that goes to /v3/OS-HTML/xxx gets forwarded to /v3/xxx and json pretty printed on the way out? | 02:18 |
| jamielennox | actually i guess it doesn't matter if the html rendering is done in the pipeline like json then it can be optional in production | 02:19 |
| *** anniec has quit IRC | 02:20 | |
| ayoung | jamielennox, nah...that will actually be harder to do. Doing HTML rendering is trivial in a pipeline, and the logic of dealing with contenttypes really should not be sprad out...the XML/JSON thing is already pretty ugly. But that is not where th real work is going to be... | 02:20 |
| ayoung | exactly | 02:20 |
| *** edmund has quit IRC | 02:20 | |
| ayoung | the hard work is going to be figuring out what should be a table (lists) and what should be a form. | 02:20 |
| jamielennox | i keep forgetting to use the pipeline | 02:20 |
| *** colinmcnamara has joined #openstack-dev | 02:21 | |
| ayoung | figuring out how to do "add item" in the current scheme, and Delete for multiples | 02:21 |
| ayoung | Bulk operations in general | 02:21 |
| jamielennox | also determining a RO vs RW backend ahead of time | 02:23 |
| jamielennox | alright, i'll have a look at the html review later as i want discoverability to be a focus of at least mine in Icehouse | 02:24 |
| dolphm | ayoung: (sorry, not really paying attention to IRC) agree, relationships should be handled by entities | 02:24 |
| dolphm | err, links | 02:24 |
| *** spzala has quit IRC | 02:24 | |
| ayoung | dolphm, so, the question then becomes: if we add links to an object (like the user object) from an extension, how do we show that enumeration. I wonder if we should use the links collection, and the rel/rev fields? | 02:25 |
| dolphm | ayoung: yeah, that's what "links" was intended to be | 02:26 |
| *** sarob has joined #openstack-dev | 02:26 | |
| *** RajeshMohan has quit IRC | 02:27 | |
| *** RajeshMohan has joined #openstack-dev | 02:28 | |
| *** sarob has quit IRC | 02:31 | |
| *** AlexF has joined #openstack-dev | 02:31 | |
| *** schwicht has quit IRC | 02:31 | |
| ayoung | dolphm, so, it sounds like things are pretty free form there, as there are no standard meanins forrel and rev. I wonder if we will need to come up with some. For example, under /v3, I would like to enumerate all of the top level packages: user, token, policy. Would it make sense to do 'rel= module' for those, and have a comparable 'rel='extension' for the actual top level URL of the extension? | 02:31 |
| ayoung | and then the rev would be 'container' for both? | 02:32 |
| dolphm | ayoung: "meanins forrel and rev"? very much agree with the basic collections enumerated as links from /v3/ | 02:33 |
| ayoung | dolphm, cool....I can add that to the html review. It is much more useful if we can actually drill down to the collections | 02:35 |
| ayoung | dolphm, once we cut rc1, we are going to branch Havana, and Master becomes Icehouse, right? | 02:36 |
| dolphm | ayoung: yep | 02:36 |
| dolphm | ayoung: https://review.openstack.org/#/c/46193/ https://review.openstack.org/#/c/48988/ then https://review.openstack.org/#/c/48971/ | 02:37 |
| ayoung | dolphm, Keystone work is going to have to be half done by the time the Summit comes around if we do API freeze in I2...and I think we should do it. | 02:37 |
| ayoung | looking | 02:37 |
| *** novas0x2a|laptop has joined #openstack-dev | 02:37 | |
| ayoung | 3? There were only 2 bugs when I looked earlier | 02:37 |
| dolphm | ayoung: well, we already have a lot of api ideas on the table | 02:37 |
| dolphm | ayoung: it's not like everything from the summit will be new | 02:37 |
| dolphm | ayoung: token revocation events, for example | 02:37 |
| ayoung | right...I'd like to get a prototype of that done by summit | 02:38 |
| dolphm | ayoung: ++ | 02:38 |
| ayoung | KDS needs a rebase ,but it should be just about ready to go. | 02:38 |
| ayoung | ah...just saw what the last review was for. Cool. | 02:39 |
| *** dims has quit IRC | 02:40 | |
| *** dguitarbite has quit IRC | 02:40 | |
| dolphm | ayoung: it just took 10 hours for the transifex patch to fail, so this will take awhile this week | 02:40 |
| ayoung | dolphm, let me reinforce how right you were to suggest we freeze early. | 02:42 |
| dolphm | ayoung: ha, thanks | 02:42 |
| dolphm | ayoung: other than transient failures, this release was fairly smooth i thought | 02:42 |
| ayoung | dolphm, yeah. | 02:43 |
| *** sarob has joined #openstack-dev | 02:44 | |
| *** sarob has quit IRC | 02:49 | |
| *** zodiak has joined #openstack-dev | 02:55 | |
| *** sushils has joined #openstack-dev | 02:58 | |
| *** AlexF has quit IRC | 03:00 | |
| *** dguitarbite has joined #openstack-dev | 03:01 | |
| *** dhubbard_gone has quit IRC | 03:08 | |
| *** rcleere has joined #openstack-dev | 03:08 | |
| *** yaguang has joined #openstack-dev | 03:08 | |
| *** bdpayne has joined #openstack-dev | 03:09 | |
| *** chenxu has quit IRC | 03:09 | |
| *** FatDarrel has joined #openstack-dev | 03:10 | |
| *** bdpayne has quit IRC | 03:11 | |
| *** Mandell has joined #openstack-dev | 03:13 | |
| * mordred loves transient failures | 03:13 | |
| *** freedomhui has quit IRC | 03:14 | |
| *** faramir1 has quit IRC | 03:14 | |
| *** adjohn has joined #openstack-dev | 03:15 | |
| *** dhubbard has joined #openstack-dev | 03:16 | |
| *** adjohn_ has joined #openstack-dev | 03:16 | |
| *** paragan has joined #openstack-dev | 03:17 | |
| *** adjohn has quit IRC | 03:19 | |
| *** DennyZhang has joined #openstack-dev | 03:20 | |
| *** prekarat has joined #openstack-dev | 03:20 | |
| *** ArcTanSusan has joined #openstack-dev | 03:22 | |
| *** Tross has joined #openstack-dev | 03:22 | |
| *** Tross has quit IRC | 03:23 | |
| *** sandywalsh has quit IRC | 03:23 | |
| *** senk has joined #openstack-dev | 03:24 | |
| *** yaguang has quit IRC | 03:25 | |
| *** colinmcnamara has quit IRC | 03:26 | |
| *** dstanek has joined #openstack-dev | 03:28 | |
| *** dguitarbite has quit IRC | 03:29 | |
| *** FatDarrel has quit IRC | 03:32 | |
| *** xjiujiu has quit IRC | 03:33 | |
| *** xjiujiu has joined #openstack-dev | 03:34 | |
| *** Mandell has quit IRC | 03:34 | |
| *** adjohn_ has quit IRC | 03:35 | |
| *** radez is now known as radez_g0n3 | 03:37 | |
| *** alunch_ has joined #openstack-dev | 03:38 | |
| *** bnemec has quit IRC | 03:39 | |
| *** alunch has quit IRC | 03:39 | |
| *** alunch_ is now known as alunch | 03:39 | |
| *** zodiak has quit IRC | 03:41 | |
| *** garyk has quit IRC | 03:41 | |
| *** adjohn has joined #openstack-dev | 03:42 | |
| *** jhesketh_ has joined #openstack-dev | 03:42 | |
| *** bnemec has joined #openstack-dev | 03:43 | |
| *** adjohn_ has joined #openstack-dev | 03:43 | |
| *** ArcTanSusan has quit IRC | 03:43 | |
| *** freedomhui has joined #openstack-dev | 03:44 | |
| *** senk has quit IRC | 03:46 | |
| *** adjohn has quit IRC | 03:46 | |
| *** sandywalsh has joined #openstack-dev | 03:49 | |
| *** briancurtin has joined #openstack-dev | 03:51 | |
| *** adjohn_ has quit IRC | 03:53 | |
| *** alexxu has joined #openstack-dev | 03:53 | |
| *** sushils has quit IRC | 03:53 | |
| *** adjohn has joined #openstack-dev | 03:54 | |
| *** AlexF has joined #openstack-dev | 03:55 | |
| *** shinylasers has joined #openstack-dev | 03:55 | |
| *** neelashah has quit IRC | 03:56 | |
| *** ayoung has quit IRC | 03:59 | |
| *** adjohn has quit IRC | 04:00 | |
| *** kushal has joined #openstack-dev | 04:01 | |
| *** dstanek has quit IRC | 04:03 | |
| *** aditirav has joined #openstack-dev | 04:03 | |
| *** radsy has quit IRC | 04:05 | |
| *** adjohn_ has joined #openstack-dev | 04:06 | |
| *** DennyZhang has quit IRC | 04:08 | |
| *** adjohn_ has quit IRC | 04:08 | |
| *** aditirav has quit IRC | 04:10 | |
| *** aditirav has joined #openstack-dev | 04:11 | |
| *** ArcTanSusan has joined #openstack-dev | 04:12 | |
| *** danwent has joined #openstack-dev | 04:13 | |
| *** Mandell has joined #openstack-dev | 04:13 | |
| *** garyk has joined #openstack-dev | 04:14 | |
| *** gimps__ has joined #openstack-dev | 04:15 | |
| *** gimps__ has quit IRC | 04:16 | |
| *** gimps__ has joined #openstack-dev | 04:16 | |
| *** mjb_ has joined #openstack-dev | 04:21 | |
| *** AlexF has quit IRC | 04:22 | |
| *** AlexF has joined #openstack-dev | 04:31 | |
| *** sridevi has joined #openstack-dev | 04:32 | |
| *** xjiujiu has quit IRC | 04:32 | |
| *** erkules_ is now known as erkules | 04:34 | |
| *** chandankumar has joined #openstack-dev | 04:36 | |
| *** colinmcnamara has joined #openstack-dev | 04:37 | |
| *** Ryan_Lane has quit IRC | 04:38 | |
| *** vartom10 has joined #openstack-dev | 04:41 | |
| *** crank has quit IRC | 04:41 | |
| *** colinmcnamara has quit IRC | 04:41 | |
| *** tvb has joined #openstack-dev | 04:46 | |
| *** briancurtin has quit IRC | 04:46 | |
| *** senk has joined #openstack-dev | 04:47 | |
| *** senk has quit IRC | 04:51 | |
| *** vartom11 has joined #openstack-dev | 04:51 | |
| *** vartom12 has joined #openstack-dev | 04:53 | |
| *** vartom10 has quit IRC | 04:54 | |
| *** sumanthns has joined #openstack-dev | 04:54 | |
| *** vartom11 has quit IRC | 04:56 | |
| *** garyk has quit IRC | 04:56 | |
| *** vartom13 has joined #openstack-dev | 04:57 | |
| *** f3nd3 has joined #openstack-dev | 04:58 | |
| *** tvb has quit IRC | 04:58 | |
| *** vartom12 has quit IRC | 04:59 | |
| *** bashok has joined #openstack-dev | 05:00 | |
| *** garyk has joined #openstack-dev | 05:01 | |
| *** freedomhui has quit IRC | 05:01 | |
| *** vartom14 has joined #openstack-dev | 05:05 | |
| *** bashok has quit IRC | 05:05 | |
| *** afazekas has quit IRC | 05:05 | |
| *** bashok has joined #openstack-dev | 05:05 | |
| *** jhesketh_ has quit IRC | 05:06 | |
| *** vartom13 has quit IRC | 05:06 | |
| *** vartom15 has joined #openstack-dev | 05:06 | |
| *** sushils has joined #openstack-dev | 05:07 | |
| *** SergeyLukjanov has joined #openstack-dev | 05:08 | |
| *** Ryan_Lane has joined #openstack-dev | 05:08 | |
| *** vartom14 has quit IRC | 05:09 | |
| *** Ryan_Lane has quit IRC | 05:13 | |
| *** kushal has quit IRC | 05:14 | |
| *** SergeyLukjanov has quit IRC | 05:14 | |
| *** SergeyLukjanov has joined #openstack-dev | 05:15 | |
| *** kushal has joined #openstack-dev | 05:16 | |
| *** otherwiseguy has quit IRC | 05:17 | |
| *** f3nd3 has left #openstack-dev | 05:17 | |
| *** zaitcev has quit IRC | 05:22 | |
| *** AlexF has quit IRC | 05:24 | |
| *** amohn9 has joined #openstack-dev | 05:25 | |
| *** amohn9 has left #openstack-dev | 05:25 | |
| *** thingee_zzz is now known as thingee | 05:32 | |
| *** danwent has quit IRC | 05:33 | |
| *** adjohn has joined #openstack-dev | 05:34 | |
| *** afazekas has joined #openstack-dev | 05:35 | |
| *** jeremyb has quit IRC | 05:36 | |
| *** jeremyb has joined #openstack-dev | 05:36 | |
| *** Ryan_Lane has joined #openstack-dev | 05:39 | |
| *** jasdeepH has joined #openstack-dev | 05:39 | |
| *** adjohn has quit IRC | 05:40 | |
| *** nshaikh has joined #openstack-dev | 05:42 | |
| *** networkstatic has quit IRC | 05:44 | |
| *** networkstatic has joined #openstack-dev | 05:45 | |
| *** jeremyb has quit IRC | 05:46 | |
| *** jeremyb has joined #openstack-dev | 05:46 | |
| *** Javin has joined #openstack-dev | 05:46 | |
| *** senk has joined #openstack-dev | 05:47 | |
| *** Ryan_Lane has quit IRC | 05:47 | |
| *** afazekas is now known as afazekas_no_irq | 05:49 | |
| *** bashok has quit IRC | 05:49 | |
| *** Tross has joined #openstack-dev | 05:50 | |
| *** adjohn has joined #openstack-dev | 05:51 | |
| *** senk has quit IRC | 05:54 | |
| *** tvb|afk has joined #openstack-dev | 05:54 | |
| *** tvb|afk has quit IRC | 05:54 | |
| *** tvb|afk has joined #openstack-dev | 05:54 | |
| *** adjohn has quit IRC | 05:55 | |
| *** adjohn has joined #openstack-dev | 05:56 | |
| *** Javin has quit IRC | 05:57 | |
| *** ctracey has quit IRC | 05:57 | |
| *** adjohn has quit IRC | 06:02 | |
| *** adjohn has joined #openstack-dev | 06:02 | |
| *** ctracey|away has joined #openstack-dev | 06:07 | |
| *** ctracey|away is now known as ctracey | 06:07 | |
| *** Javin has joined #openstack-dev | 06:08 | |
| *** rushiagr has joined #openstack-dev | 06:10 | |
| *** vartom15 has quit IRC | 06:11 | |
| *** doron_afk has joined #openstack-dev | 06:12 | |
| *** yolanda has joined #openstack-dev | 06:13 | |
| *** freedomhui has joined #openstack-dev | 06:14 | |
| *** gmurphy has quit IRC | 06:14 | |
| *** SergeyLukjanov has quit IRC | 06:15 | |
| *** zbitter has joined #openstack-dev | 06:15 | |
| *** vipul is now known as vipul-away | 06:16 | |
| *** stevemar has quit IRC | 06:16 | |
| *** odyssey4me has joined #openstack-dev | 06:18 | |
| *** shinylasers has quit IRC | 06:18 | |
| *** zaneb has quit IRC | 06:18 | |
| *** faramir1 has joined #openstack-dev | 06:18 | |
| *** xqueralt-afk has quit IRC | 06:20 | |
| *** blamar has quit IRC | 06:20 | |
| *** gmurphy has joined #openstack-dev | 06:20 | |
| *** adjohn_ has joined #openstack-dev | 06:22 | |
| *** adjohn has quit IRC | 06:24 | |
| *** doron_afk is now known as doron | 06:25 | |
| *** MaxV has joined #openstack-dev | 06:29 | |
| *** mjb_ has quit IRC | 06:29 | |
| *** AlexF has joined #openstack-dev | 06:29 | |
| *** jbresnah has joined #openstack-dev | 06:30 | |
| *** vipul-away is now known as vipul | 06:30 | |
| *** kushal has quit IRC | 06:31 | |
| *** doron is now known as doron_afk | 06:35 | |
| *** adjohn_ has quit IRC | 06:36 | |
| *** adjohn has joined #openstack-dev | 06:37 | |
| *** aditirav_ has joined #openstack-dev | 06:38 | |
| *** aditirav_ has quit IRC | 06:38 | |
| *** aditirav_ has joined #openstack-dev | 06:39 | |
| *** Ryan_Lane has joined #openstack-dev | 06:39 | |
| *** MaxV has quit IRC | 06:40 | |
| *** shinylasers has joined #openstack-dev | 06:40 | |
| *** aditirav has quit IRC | 06:41 | |
| *** aditirav_ is now known as aditirav | 06:41 | |
| *** o_petit has joined #openstack-dev | 06:42 | |
| *** mrunge has joined #openstack-dev | 06:42 | |
| *** flaper87|afk is now known as flaper87 | 06:43 | |
| *** Ryan_Lane has quit IRC | 06:46 | |
| *** rushiagr has quit IRC | 06:47 | |
| *** kdas__ has joined #openstack-dev | 06:47 | |
| *** adjohn has quit IRC | 06:48 | |
| *** senk has joined #openstack-dev | 06:50 | |
| *** AlexF has quit IRC | 06:54 | |
| *** senk has quit IRC | 06:54 | |
| *** egallen has joined #openstack-dev | 06:57 | |
| *** jprovazn has joined #openstack-dev | 07:00 | |
| *** reidrac has joined #openstack-dev | 07:02 | |
| *** belmoreira has joined #openstack-dev | 07:08 | |
| *** shinylasers has quit IRC | 07:10 | |
| *** basha has joined #openstack-dev | 07:11 | |
| *** jcoufal has joined #openstack-dev | 07:11 | |
| *** thingee is now known as thingee_zzz | 07:11 | |
| *** freedomhui has quit IRC | 07:14 | |
| *** salv-orlando has joined #openstack-dev | 07:15 | |
| *** SergeyLukjanov has joined #openstack-dev | 07:18 | |
| *** gargya has joined #openstack-dev | 07:18 | |
| *** DinaBelova has joined #openstack-dev | 07:19 | |
| *** rushiagr has joined #openstack-dev | 07:20 | |
| *** shinylasers has joined #openstack-dev | 07:23 | |
| *** jbresnah has quit IRC | 07:31 | |
| *** athomas has joined #openstack-dev | 07:31 | |
| *** tsufiev has joined #openstack-dev | 07:32 | |
| *** vsergeyev has left #openstack-dev | 07:33 | |
| *** xga_ has joined #openstack-dev | 07:33 | |
| *** markmc has joined #openstack-dev | 07:35 | |
| *** JordanP has joined #openstack-dev | 07:35 | |
| *** gsilvis has quit IRC | 07:36 | |
| *** tsufiev has quit IRC | 07:37 | |
| *** xga has quit IRC | 07:37 | |
| *** rushiagr has quit IRC | 07:37 | |
| *** jistr has joined #openstack-dev | 07:39 | |
| *** jtomasek has joined #openstack-dev | 07:40 | |
| *** Ryan_Lane has joined #openstack-dev | 07:40 | |
| *** MaxV has joined #openstack-dev | 07:40 | |
| *** rushiagr has joined #openstack-dev | 07:42 | |
| *** fbo_away is now known as fbo | 07:42 | |
| *** Ryan_Lane has quit IRC | 07:44 | |
| *** arezmerita has joined #openstack-dev | 07:46 | |
| *** gmurphy is now known as gmurphy_gone | 07:46 | |
| *** Javin has quit IRC | 07:47 | |
| *** xqueralt has joined #openstack-dev | 07:49 | |
| *** xga has joined #openstack-dev | 07:49 | |
| *** Max__ has joined #openstack-dev | 07:49 | |
| *** MaxV has quit IRC | 07:50 | |
| *** jpich has joined #openstack-dev | 07:50 | |
| *** ygbo has joined #openstack-dev | 07:51 | |
| *** xga_ has quit IRC | 07:53 | |
| *** romcheg has joined #openstack-dev | 07:55 | |
| *** vartom15 has joined #openstack-dev | 07:56 | |
| *** johnthetubaguy has joined #openstack-dev | 07:57 | |
| *** o_petit has quit IRC | 07:57 | |
| *** mkollaro has joined #openstack-dev | 07:57 | |
| *** ndipanov_gone is now known as ndipanov | 07:58 | |
| *** rushiagr has quit IRC | 07:59 | |
| *** DinaBelova has quit IRC | 08:00 | |
| *** vartom15 has quit IRC | 08:00 | |
| *** vartom15 has joined #openstack-dev | 08:01 | |
| *** mmagr has joined #openstack-dev | 08:02 | |
| *** matrohon has joined #openstack-dev | 08:02 | |
| *** AlexF has joined #openstack-dev | 08:04 | |
| *** salv-orlando has quit IRC | 08:05 | |
| *** boden has joined #openstack-dev | 08:05 | |
| *** romcheg has left #openstack-dev | 08:06 | |
| *** o_petit has joined #openstack-dev | 08:06 | |
| *** xga_ has joined #openstack-dev | 08:07 | |
| *** DinaBelova has joined #openstack-dev | 08:08 | |
| *** rushiagr has joined #openstack-dev | 08:08 | |
| *** ArcTanSusan has quit IRC | 08:09 | |
| *** xga has quit IRC | 08:10 | |
| *** mosulica has joined #openstack-dev | 08:11 | |
| *** yassine has joined #openstack-dev | 08:11 | |
| *** gsilvis has joined #openstack-dev | 08:13 | |
| *** tvb|afk has quit IRC | 08:14 | |
| *** JordanP has quit IRC | 08:14 | |
| *** tvb|afk has joined #openstack-dev | 08:14 | |
| *** tvb|afk has quit IRC | 08:14 | |
| *** tvb|afk has joined #openstack-dev | 08:14 | |
| *** kpavel has joined #openstack-dev | 08:16 | |
| *** derekh has joined #openstack-dev | 08:18 | |
| *** vuntz has quit IRC | 08:18 | |
| *** AlexF has quit IRC | 08:19 | |
| *** kpavel has quit IRC | 08:19 | |
| *** vuntz has joined #openstack-dev | 08:19 | |
| *** vuntz has quit IRC | 08:19 | |
| *** vuntz has joined #openstack-dev | 08:19 | |
| *** safchain has joined #openstack-dev | 08:21 | |
| *** gatuus has quit IRC | 08:25 | |
| *** JordanP has joined #openstack-dev | 08:26 | |
| *** rushiagr has quit IRC | 08:27 | |
| *** dnoll has joined #openstack-dev | 08:28 | |
| *** gfidente has joined #openstack-dev | 08:30 | |
| *** gfidente has quit IRC | 08:30 | |
| *** corXi has joined #openstack-dev | 08:30 | |
| *** rushiagr has joined #openstack-dev | 08:30 | |
| *** ygbo has quit IRC | 08:31 | |
| *** pfreund has joined #openstack-dev | 08:35 | |
| *** lucasagomes has joined #openstack-dev | 08:37 | |
| *** eglynn has joined #openstack-dev | 08:39 | |
| *** Ryan_Lane has joined #openstack-dev | 08:40 | |
| *** ygbo has joined #openstack-dev | 08:41 | |
| *** sushils has quit IRC | 08:41 | |
| *** ctracey has quit IRC | 08:42 | |
| *** testingtesting has joined #openstack-dev | 08:43 | |
| *** Ryan_Lane has quit IRC | 08:44 | |
| *** tkammer has joined #openstack-dev | 08:45 | |
| *** danpb has joined #openstack-dev | 08:46 | |
| *** ctracey|away has joined #openstack-dev | 08:46 | |
| *** ctracey|away is now known as ctracey | 08:46 | |
| *** akrogames has joined #openstack-dev | 08:47 | |
| akrogames | Hi all | 08:47 |
| *** AlexF has joined #openstack-dev | 08:47 | |
| *** lsmola has quit IRC | 08:49 | |
| *** rushiagr has quit IRC | 08:49 | |
| *** shinylasers has quit IRC | 08:53 | |
| *** networkstatic has quit IRC | 08:55 | |
| *** flaper87 is now known as flaper87|afk | 08:55 | |
| *** flaper87|afk is now known as flaper87 | 08:56 | |
| *** iartarisi has joined #openstack-dev | 08:59 | |
| *** viktors has joined #openstack-dev | 08:59 | |
| *** feleouet has joined #openstack-dev | 09:00 | |
| *** JordanP has quit IRC | 09:00 | |
| *** xga has joined #openstack-dev | 09:01 | |
| *** JordanP has joined #openstack-dev | 09:02 | |
| *** pixelb has joined #openstack-dev | 09:04 | |
| *** xga_ has quit IRC | 09:04 | |
| *** toabctl has quit IRC | 09:05 | |
| *** toabctl has joined #openstack-dev | 09:06 | |
| *** toabctl has joined #openstack-dev | 09:06 | |
| *** Shaan7 has quit IRC | 09:08 | |
| *** Ryan_Lane has joined #openstack-dev | 09:10 | |
| *** akrogames has quit IRC | 09:10 | |
| *** martyntaylor has joined #openstack-dev | 09:11 | |
| *** AlexF has quit IRC | 09:12 | |
| *** Ryan_Lane has quit IRC | 09:15 | |
| *** senk has joined #openstack-dev | 09:22 | |
| *** dosaboy has quit IRC | 09:22 | |
| *** dosaboy has joined #openstack-dev | 09:22 | |
| *** xga_ has joined #openstack-dev | 09:23 | |
| *** rushiagr has joined #openstack-dev | 09:25 | |
| *** xga has quit IRC | 09:26 | |
| *** senk has quit IRC | 09:26 | |
| *** sridevi has quit IRC | 09:28 | |
| *** faramir1 has quit IRC | 09:29 | |
| *** sridevi has joined #openstack-dev | 09:29 | |
| *** o_petit has quit IRC | 09:33 | |
| *** o_petit has joined #openstack-dev | 09:34 | |
| koolhead17 | garyk: around | 09:37 |
| *** danwent has joined #openstack-dev | 09:38 | |
| *** eglynn has quit IRC | 09:39 | |
| *** eglynn has joined #openstack-dev | 09:40 | |
| *** danwent has quit IRC | 09:43 | |
| *** belmoreira has quit IRC | 09:48 | |
| *** AnilV4 has quit IRC | 09:49 | |
| *** danwent has joined #openstack-dev | 09:51 | |
| *** sushils has joined #openstack-dev | 09:51 | |
| *** prekarat has quit IRC | 09:52 | |
| *** DinaBelova has quit IRC | 09:59 | |
| *** AnilV4 has joined #openstack-dev | 10:07 | |
| *** o_petit has quit IRC | 10:09 | |
| *** sandywalsh has quit IRC | 10:10 | |
| *** Ryan_Lane has joined #openstack-dev | 10:11 | |
| *** Ghe_Holidays is now known as GheRivero | 10:14 | |
| *** Ryan_Lane has quit IRC | 10:15 | |
| *** flaper87 is now known as flaper87|afk | 10:17 | |
| *** freedomhui has joined #openstack-dev | 10:17 | |
| *** pcm_ has joined #openstack-dev | 10:19 | |
| *** jruzicka has joined #openstack-dev | 10:19 | |
| *** pcm_ has quit IRC | 10:21 | |
| *** o_petit has joined #openstack-dev | 10:21 | |
| *** pcm_ has joined #openstack-dev | 10:21 | |
| *** senk has joined #openstack-dev | 10:22 | |
| *** sandywalsh has joined #openstack-dev | 10:23 | |
| *** schwicht has joined #openstack-dev | 10:23 | |
| *** egallen has quit IRC | 10:24 | |
| *** SergeyLukjanov has quit IRC | 10:25 | |
| *** senk has quit IRC | 10:27 | |
| *** AnilV4 has quit IRC | 10:27 | |
| *** SergeyLukjanov has joined #openstack-dev | 10:27 | |
| *** rushiagr has quit IRC | 10:27 | |
| *** novas0x2a|laptop has quit IRC | 10:28 | |
| *** rushiagr has joined #openstack-dev | 10:29 | |
| *** flaper87|afk is now known as flaper87 | 10:29 | |
| *** DinaBelova has joined #openstack-dev | 10:30 | |
| *** adjohn has joined #openstack-dev | 10:31 | |
| *** lsmola has joined #openstack-dev | 10:32 | |
| *** o_petit has quit IRC | 10:33 | |
| *** kaushikc has joined #openstack-dev | 10:33 | |
| *** paragan has quit IRC | 10:34 | |
| *** lbragstad has quit IRC | 10:35 | |
| *** AlexF has joined #openstack-dev | 10:35 | |
| *** danwent has quit IRC | 10:36 | |
| kaushikc | johnthetubaguy: hi | 10:39 |
| johnthetubaguy | hey | 10:39 |
| johnthetubaguy | kaushikc: hows things | 10:39 |
| *** Ryan_Lane has joined #openstack-dev | 10:41 | |
| *** lbragstad has joined #openstack-dev | 10:43 | |
| *** Ryan_Lane has quit IRC | 10:45 | |
| *** zbitter is now known as zaneb | 10:50 | |
| *** gargya has quit IRC | 10:51 | |
| *** gargya has joined #openstack-dev | 10:52 | |
| *** sergmelikyan has joined #openstack-dev | 10:52 | |
| *** rushiagr has quit IRC | 10:53 | |
| *** basha has quit IRC | 10:54 | |
| *** AnilV4 has joined #openstack-dev | 11:00 | |
| *** rushiagr has joined #openstack-dev | 11:03 | |
| *** rkukura has quit IRC | 11:04 | |
| *** AlexF has quit IRC | 11:04 | |
| *** AlexF has joined #openstack-dev | 11:05 | |
| *** boris-42 has joined #openstack-dev | 11:07 | |
| *** dstanek has joined #openstack-dev | 11:09 | |
| *** radez_g0n3 is now known as radez | 11:09 | |
| *** sushils has quit IRC | 11:10 | |
| *** yolanda has quit IRC | 11:10 | |
| *** yolanda has joined #openstack-dev | 11:11 | |
| *** doron_afk is now known as doron | 11:13 | |
| *** flaper87 is now known as flaper87|afk | 11:16 | |
| *** doron is now known as doron_afk | 11:17 | |
| *** dims has joined #openstack-dev | 11:18 | |
| *** doron_afk is now known as doron | 11:19 | |
| *** rushiagr has quit IRC | 11:20 | |
| *** flaper87|afk is now known as flaper87 | 11:21 | |
| *** doron is now known as doron_afk | 11:21 | |
| *** romcheg1 has joined #openstack-dev | 11:22 | |
| *** senk has joined #openstack-dev | 11:23 | |
| *** senk has quit IRC | 11:27 | |
| *** rushiagr has joined #openstack-dev | 11:28 | |
| *** lucasagomes has quit IRC | 11:29 | |
| *** lucasagomes has joined #openstack-dev | 11:29 | |
| *** michchap has quit IRC | 11:29 | |
| *** michchap has joined #openstack-dev | 11:30 | |
| *** doron_afk is now known as doron | 11:30 | |
| *** primemin1sterp has joined #openstack-dev | 11:31 | |
| *** primeministerp has quit IRC | 11:31 | |
| *** nosnos has quit IRC | 11:33 | |
| *** o_petit has joined #openstack-dev | 11:34 | |
| *** drewlander has joined #openstack-dev | 11:36 | |
| *** o_petit_ has joined #openstack-dev | 11:37 | |
| *** matsuhashi has quit IRC | 11:37 | |
| *** o_petit has quit IRC | 11:40 | |
| *** Ryan_Lane has joined #openstack-dev | 11:41 | |
| *** Ryan_Lane has quit IRC | 11:41 | |
| *** Ryan_Lane has joined #openstack-dev | 11:41 | |
| *** markmcclain has quit IRC | 11:42 | |
| *** markmcclain has joined #openstack-dev | 11:43 | |
| *** markmcclain1 has joined #openstack-dev | 11:44 | |
| *** markmcclain has quit IRC | 11:44 | |
| *** AnilV4 has quit IRC | 11:45 | |
| *** alunduil has quit IRC | 11:45 | |
| *** Ryan_Lane has quit IRC | 11:46 | |
| *** sandywalsh has quit IRC | 11:47 | |
| *** markmc has quit IRC | 11:49 | |
| *** testingtesting has quit IRC | 11:50 | |
| *** testingtesting has joined #openstack-dev | 11:51 | |
| *** yassine_ has joined #openstack-dev | 11:52 | |
| *** testingtesting is now known as djoreilly | 11:55 | |
| *** yassine has quit IRC | 11:56 | |
| *** AnilV4 has joined #openstack-dev | 11:56 | |
| *** kdas__ has quit IRC | 11:56 | |
| *** safchain_ has joined #openstack-dev | 11:58 | |
| *** kushal has joined #openstack-dev | 11:58 | |
| *** safchain_ has quit IRC | 11:58 | |
| *** safchain has quit IRC | 11:58 | |
| *** safchain has joined #openstack-dev | 11:58 | |
| *** dolphm has quit IRC | 11:58 | |
| *** nil1511 has joined #openstack-dev | 11:59 | |
| *** venkatesh has joined #openstack-dev | 12:00 | |
| *** sandywalsh has joined #openstack-dev | 12:01 | |
| *** dkranz has joined #openstack-dev | 12:01 | |
| *** mkollaro1 has joined #openstack-dev | 12:01 | |
| *** mkollaro has quit IRC | 12:01 | |
| *** doron is now known as doron_afk | 12:03 | |
| *** tkammer has quit IRC | 12:03 | |
| *** safchain has quit IRC | 12:05 | |
| *** safchain_ has joined #openstack-dev | 12:05 | |
| *** gordc has joined #openstack-dev | 12:05 | |
| *** dnoll has quit IRC | 12:06 | |
| *** dims has quit IRC | 12:07 | |
| *** mohits has quit IRC | 12:08 | |
| *** egallen has joined #openstack-dev | 12:08 | |
| *** dims has joined #openstack-dev | 12:08 | |
| *** rcrit has quit IRC | 12:08 | |
| *** clouded_tux has quit IRC | 12:08 | |
| *** rcrit has joined #openstack-dev | 12:10 | |
| *** larsks is now known as larsks|kids | 12:11 | |
| *** Ryan_Lane has joined #openstack-dev | 12:12 | |
| *** galstrom_zzz is now known as galstrom | 12:12 | |
| *** galstrom is now known as galstrom_zzz | 12:13 | |
| *** mrunge has quit IRC | 12:15 | |
| *** pfreund has quit IRC | 12:16 | |
| *** Ryan_Lane has quit IRC | 12:16 | |
| *** paragan has joined #openstack-dev | 12:18 | |
| *** doron_afk is now known as doron | 12:19 | |
| *** rushiagr has quit IRC | 12:19 | |
| *** flaper87 is now known as flaper87|afk | 12:20 | |
| *** tkammer has joined #openstack-dev | 12:22 | |
| *** romcheg1 has quit IRC | 12:23 | |
| *** romcheg has joined #openstack-dev | 12:23 | |
| *** romcheg1 has joined #openstack-dev | 12:23 | |
| *** senk has joined #openstack-dev | 12:23 | |
| *** vladikr has joined #openstack-dev | 12:24 | |
| *** Ruetobas has quit IRC | 12:25 | |
| *** Ruetobas has joined #openstack-dev | 12:25 | |
| *** romcheg1 has quit IRC | 12:27 | |
| *** romcheg has quit IRC | 12:27 | |
| *** senk has quit IRC | 12:27 | |
| *** romcheg has joined #openstack-dev | 12:28 | |
| ekarlso | annegentle: is there any docs on installing on xenserver atm ? | 12:28 |
| *** jasondot_ has joined #openstack-dev | 12:28 | |
| *** doron is now known as doron_afk | 12:31 | |
| *** dsantos_ has joined #openstack-dev | 12:31 | |
| *** rfolco has joined #openstack-dev | 12:32 | |
| *** jayg|g0n3 is now known as jayg | 12:33 | |
| *** flaper87|afk is now known as flaper87 | 12:33 | |
| *** jayg is now known as jayg|g0n3 | 12:33 | |
| *** jayg|g0n3 is now known as jayg | 12:33 | |
| *** nil1511_ has joined #openstack-dev | 12:34 | |
| *** tellesnobrega has joined #openstack-dev | 12:35 | |
| *** drewlander has quit IRC | 12:35 | |
| *** drewlander has joined #openstack-dev | 12:35 | |
| dsantos_ | hi guys, I'm getting this bug to fix https://bugs.launchpad.net/nova/+bug/1201616 | 12:36 |
| uvirtbot | Launchpad bug 1201616 in nova "misconfigured baremetal deploy environments are hard to debug" [High,Triaged] | 12:36 |
| *** mohits has joined #openstack-dev | 12:36 | |
| dsantos_ | but I don't know what the guy who reported the bug did to get an exception | 12:36 |
| *** Ruetobas has quit IRC | 12:36 | |
| dsantos_ | does someone knows what I can do to reproduce this bug? | 12:37 |
| *** nil1511 has quit IRC | 12:37 | |
| *** blues-man has joined #openstack-dev | 12:39 | |
| *** odyssey4me has quit IRC | 12:39 | |
| *** sgordon_ has joined #openstack-dev | 12:40 | |
| *** thomasm has joined #openstack-dev | 12:40 | |
| *** markmc has joined #openstack-dev | 12:42 | |
| ttx | markmc: around ? Looking into stable/havana cut for oslo-incubator | 12:42 |
| markmc | ttx, hey | 12:42 |
| ttx | markmc: you still have bug 1225099 targeted | 12:43 |
| uvirtbot | Launchpad bug 1225099 in oslo "Encoding hard-coded in gettextutils.Message" [High,In progress] https://launchpad.net/bugs/1225099 | 12:43 |
| *** Ryan_Lane has joined #openstack-dev | 12:43 | |
| markmc | ttx, thanks, looking | 12:43 |
| markmc | ttx, wow, some commit message and discussion in that review | 12:44 |
| markmc | ttx, as bnemec says "this is giving me a headache" :) | 12:44 |
| ttx | markmc: got me worried. Doesn't exactly look like it's around the corner | 12:44 |
| markmc | ttx, right | 12:44 |
| markmc | bnemec, any thoughts on how critical this is for the release? | 12:44 |
| *** jvrbanac has joined #openstack-dev | 12:45 | |
| markmc | ttx, crap, I'm going to be in the air during the project meeting later - just realized | 12:45 |
| ttx | markmc: we can solve most of the questions now if you're around | 12:46 |
| markmc | ttx, yeah - still reading through this stuff on 1225099 | 12:47 |
| ttx | I only had that bug's fate and your signoff for stable/havana cutting | 12:47 |
| *** kincl has joined #openstack-dev | 12:47 | |
| *** Ryan_Lane has quit IRC | 12:47 | |
| *** rushiagr has joined #openstack-dev | 12:47 | |
| *** dvarga has joined #openstack-dev | 12:48 | |
| *** dolphm has joined #openstack-dev | 12:48 | |
| *** sgordon has quit IRC | 12:49 | |
| markmc | ttx, I have to say, this bug makes me very nervous | 12:51 |
| *** kushal has quit IRC | 12:51 | |
| *** gargya_ has joined #openstack-dev | 12:51 | |
| markmc | ttx, I'm not totally clear on the user implications - but it sounds like it could be bad | 12:52 |
| ttx | markmc: how common are those double-byte things ? | 12:52 |
| *** Ruetobas has joined #openstack-dev | 12:52 | |
| *** jvrbanac has quit IRC | 12:52 | |
| markmc | ttx, and I'm far from clear on the code, it's complete voodoo | 12:52 |
| markmc | ttx, I'm assuming they'd have to come from translations | 12:52 |
| ttx | that's my supposition too | 12:52 |
| *** gargya has quit IRC | 12:52 | |
| *** gargya_ is now known as gargya | 12:52 | |
| *** radez is now known as radez_g0n3 | 12:53 | |
| ttx | markmc: can you clarify it (or punt to dhellmann early today) ? | 12:53 |
| markmc | ttx, yeah | 12:53 |
| ttx | markmc: note that we can still cut stable/havana alright, nothing prevents backporting from it | 12:53 |
| *** edmund has joined #openstack-dev | 12:53 | |
| * dhellmann reads backlog | 12:53 | |
| *** kaushikc has quit IRC | 12:53 | |
| ttx | but would like to know if affected projects need to sync it before rc | 12:53 |
| markmc | ttx, yep, agree | 12:53 |
| *** rushiagr has quit IRC | 12:54 | |
| dhellmann | ew, some of this is hairy | 12:54 |
| markmc | ttx, I'm going to target this to nova's rc1 - AFAICT the code is used by nova, and only used by nova | 12:55 |
| *** o_petit_ has quit IRC | 12:55 | |
| markmc | ttx, again, not clear on the user implications for nova yet, but assuming potentially bad | 12:55 |
| *** READ10 has quit IRC | 12:55 | |
| *** xga has joined #openstack-dev | 12:56 | |
| *** Max__ has quit IRC | 12:56 | |
| *** Ruetobas has quit IRC | 12:57 | |
| *** o_petit has joined #openstack-dev | 12:57 | |
| markmc | ttx, dhellmann, oh, ceilometer uses it too apparently | 12:57 |
| *** jcoufal has quit IRC | 12:58 | |
| *** dprince has joined #openstack-dev | 12:58 | |
| *** xga_ has quit IRC | 12:58 | |
| *** safchain_ has quit IRC | 12:58 | |
| *** safchain_ has joined #openstack-dev | 12:59 | |
| *** o_petit has quit IRC | 12:59 | |
| *** safchain_ has quit IRC | 12:59 | |
| *** safchain has joined #openstack-dev | 12:59 | |
| *** o_petit has joined #openstack-dev | 12:59 | |
| dhellmann | markmc: I'm not sure this is the right approach. Isn't the point of lazy evaluation that the translation happens when the message is rendered? This changes when the translation occurs, doesn't it? | 12:59 |
| *** jorisroovers has joined #openstack-dev | 12:59 | |
| *** Ruetobas has joined #openstack-dev | 13:01 | |
| markmc | dhellmann, the new translate_into() is for delayed translation I assume | 13:01 |
| dhellmann | markmc: it seems like the right approach is to set the encoding for Message objects to use instead of 'ascii' | 13:01 |
| dhellmann | afaict, that's the problem described in the (epic) commit message | 13:01 |
| markmc | dhellmann, my instinct on this is that the problem is gnarly enough that we've a very high chance of getting it wrong | 13:02 |
| *** venkatesh has quit IRC | 13:02 | |
| dhellmann | markmc: agreed | 13:02 |
| markmc | dhellmann, so we should consider punting delayed-translation support to icehouse | 13:03 |
| dhellmann | markmc: also agreed | 13:03 |
| markmc | dhellmann, i.e. removing the enable_lazy() calls from projects | 13:03 |
| dhellmann | markmc: also, this solution looks gnarly enough that I suspect it's wrong just on the face of it | 13:03 |
| *** mkollaro has joined #openstack-dev | 13:03 | |
| markmc | dhellmann, yeah | 13:03 |
| dhellmann | markmc: remove them, or make them a no-op? | 13:03 |
| *** sandywalsh has quit IRC | 13:03 | |
| *** anteaya has joined #openstack-dev | 13:03 | |
| markmc | dhellmann, I'd just remove the calls | 13:04 |
| *** xga has quit IRC | 13:04 | |
| markmc | dhellmann, that way work on gettextutils isn't disrupted | 13:04 |
| dhellmann | markmc: fair point | 13:04 |
| markmc | dhellmann, and if people/vendors want to enable it, they can just carry a patch to add back the calls | 13:04 |
| dhellmann | I was worried about something sneaking back in to a project before it was ready | 13:04 |
| dhellmann | true | 13:05 |
| markmc | confused | 13:05 |
| markmc | cinder affected but doesn't call enable_lazy() ? | 13:05 |
| markmc | https://bugs.launchpad.net/cinder/+bug/1229800 | 13:05 |
| uvirtbot | Launchpad bug 1229800 in cinder "UnicodeDecodeError in log when message includes non-ascii characters" [Undecided,New] | 13:05 |
| markmc | luisg, there? | 13:05 |
| *** blues-man has quit IRC | 13:05 | |
| *** davidhadas has quit IRC | 13:06 | |
| *** jasondot_ has quit IRC | 13:06 | |
| *** Ruetobas has quit IRC | 13:06 | |
| *** nshaikh has left #openstack-dev | 13:07 | |
| *** mkollaro1 has quit IRC | 13:07 | |
| *** aditirav has quit IRC | 13:07 | |
| *** nil1511_ has quit IRC | 13:08 | |
| *** sridevi has quit IRC | 13:08 | |
| *** rushiagr has joined #openstack-dev | 13:08 | |
| *** egallen has quit IRC | 13:09 | |
| *** chandankumar has quit IRC | 13:09 | |
| *** o_petit has quit IRC | 13:10 | |
| markmc | dhellmann, ah, it's either enable_lazy() or gettextutils.install('foo', lazy=True) | 13:10 |
| dhellmann | aha | 13:10 |
| *** adalbas has joined #openstack-dev | 13:10 | |
| *** Ruetobas has joined #openstack-dev | 13:12 | |
| *** jasondot_ has joined #openstack-dev | 13:12 | |
| *** o_petit has joined #openstack-dev | 13:12 | |
| dhellmann | I'm trying to figure out why this required a new class method factory, too | 13:12 |
| *** Ryan_Lane has joined #openstack-dev | 13:13 | |
| *** MaxV has joined #openstack-dev | 13:14 | |
| *** egallen has joined #openstack-dev | 13:14 | |
| *** sumanthns has quit IRC | 13:14 | |
| *** yassine has joined #openstack-dev | 13:15 | |
| *** corXi has quit IRC | 13:15 | |
| *** corXi has joined #openstack-dev | 13:15 | |
| *** sumanthns has joined #openstack-dev | 13:16 | |
| markmc | mrodden, there? | 13:16 |
| *** Ryan_Lane has quit IRC | 13:18 | |
| *** zul has quit IRC | 13:18 | |
| *** sandywalsh has joined #openstack-dev | 13:19 | |
| *** yassine_ has quit IRC | 13:19 | |
| *** vartom15 has quit IRC | 13:19 | |
| *** Ruetobas has quit IRC | 13:20 | |
| *** AlexF has quit IRC | 13:20 | |
| *** SumitNaiksatam has left #openstack-dev | 13:21 | |
| *** kbringard has joined #openstack-dev | 13:21 | |
| *** nermina has joined #openstack-dev | 13:22 | |
| *** larsks|kids is now known as larsks | 13:24 | |
| *** jecarey_ has quit IRC | 13:25 | |
| *** jecarey has quit IRC | 13:25 | |
| *** henrynash has joined #openstack-dev | 13:26 | |
| *** jistr has quit IRC | 13:27 | |
| *** Ruetobas has joined #openstack-dev | 13:27 | |
| *** jistr has joined #openstack-dev | 13:28 | |
| *** jasondot_ has quit IRC | 13:30 | |
| markmc | ttx, dropping off again soon | 13:30 |
| *** alexpilotti has joined #openstack-dev | 13:30 | |
| markmc | ttx, posted a summary to https://bugs.launchpad.net/oslo/+bug/1225099 | 13:30 |
| uvirtbot | Launchpad bug 1225099 in oslo "Encoding hard-coded in gettextutils.Message" [High,In progress] | 13:30 |
| *** AlexF has joined #openstack-dev | 13:31 | |
| *** eharney has joined #openstack-dev | 13:32 | |
| ttx | markmc: ok, will read | 13:32 |
| *** zul has joined #openstack-dev | 13:32 | |
| ttx | markmc: is dhellmann replacing you for today's meeting ? | 13:32 |
| markmc | dhellmann, would you mind? | 13:33 |
| *** alunch has quit IRC | 13:33 | |
| *** RajeshMohan has quit IRC | 13:33 | |
| *** blamar has joined #openstack-dev | 13:34 | |
| *** RajeshMohan has joined #openstack-dev | 13:34 | |
| *** alunch has joined #openstack-dev | 13:35 | |
| *** rcrit has quit IRC | 13:35 | |
| *** ifarkas has quit IRC | 13:35 | |
| mrodden | markmc: i'm around but in a call at the moment | 13:36 |
| mrodden | will be kind of slow | 13:36 |
| *** ifarkas has joined #openstack-dev | 13:36 | |
| markmc | mrodden, ok, take a look at https://bugs.launchpad.net/neutron/+bug/1225099/comments/9 when you're ready | 13:36 |
| uvirtbot | Launchpad bug 1225099 in oslo "Encoding hard-coded in gettextutils.Message" [High,In progress] | 13:36 |
| *** rushiagr has quit IRC | 13:37 | |
| *** egallen has quit IRC | 13:38 | |
| *** neelashah has joined #openstack-dev | 13:38 | |
| *** venkatesh has joined #openstack-dev | 13:40 | |
| *** rcrit has joined #openstack-dev | 13:41 | |
| *** davidhadas has joined #openstack-dev | 13:41 | |
| *** jcoufal has joined #openstack-dev | 13:42 | |
| *** alunch_ has joined #openstack-dev | 13:44 | |
| *** Ryan_Lane has joined #openstack-dev | 13:44 | |
| *** blues-man has joined #openstack-dev | 13:44 | |
| *** bknudson has joined #openstack-dev | 13:44 | |
| *** alunch has quit IRC | 13:44 | |
| *** alunch_ is now known as alunch | 13:44 | |
| bnemec | markmc: mrodden: I'm inclined to agree with Mark's comment. The fix is ugly (an immutable object that we want to change on the fly) and intrusive to the other projects using it, which means there's a significant chance of regression if we push it through. | 13:46 |
| bnemec | So I guess I'm +1 on punting for Havana. | 13:46 |
| *** zaitcev has joined #openstack-dev | 13:46 | |
| markmc | bnemec, thanks for the input | 13:46 |
| bnemec | Which is too bad because a lot of work went into this, but I just don't feel that good about it right now. | 13:46 |
| *** rcrit has quit IRC | 13:46 | |
| *** tkammer has quit IRC | 13:47 | |
| *** dkranz has quit IRC | 13:47 | |
| *** colinmcnamara has joined #openstack-dev | 13:47 | |
| *** Ruetobas has quit IRC | 13:48 | |
| *** Ryan_Lane has quit IRC | 13:48 | |
| *** insanidade has joined #openstack-dev | 13:49 | |
| *** Max__ has joined #openstack-dev | 13:49 | |
| *** rcleere has quit IRC | 13:49 | |
| *** MaxV has quit IRC | 13:49 | |
| *** morazi has joined #openstack-dev | 13:51 | |
| *** spzala has joined #openstack-dev | 13:51 | |
| insanidade | anyone: when it comes to neutron plugin development, are we strictly talking about making it able to interact with new switching techonologies? Doesn't it include new routing techonologies ? | 13:52 |
| *** Ruetobas has joined #openstack-dev | 13:52 | |
| insanidade | *technologies | 13:52 |
| *** gargya_ has joined #openstack-dev | 13:52 | |
| *** gargya has quit IRC | 13:53 | |
| *** gargya_ is now known as gargya | 13:53 | |
| *** jecarey has joined #openstack-dev | 13:53 | |
| *** jecarey_ has joined #openstack-dev | 13:53 | |
| *** rcrit has joined #openstack-dev | 13:53 | |
| *** paragan has quit IRC | 13:54 | |
| *** changbl has quit IRC | 13:54 | |
| *** bauzas has quit IRC | 13:56 | |
| *** bauzas has joined #openstack-dev | 13:57 | |
| *** safchain has quit IRC | 13:58 | |
| *** safchain has joined #openstack-dev | 13:58 | |
| *** athomas has quit IRC | 13:58 | |
| *** safchain has quit IRC | 13:58 | |
| *** AlexF has quit IRC | 13:59 | |
| *** safchain has joined #openstack-dev | 14:00 | |
| *** salv-orlando has joined #openstack-dev | 14:00 | |
| *** tkammer has joined #openstack-dev | 14:01 | |
| *** READ10 has joined #openstack-dev | 14:02 | |
| *** sumanthns has quit IRC | 14:02 | |
| *** markmc has quit IRC | 14:03 | |
| *** jtomasek_ has joined #openstack-dev | 14:03 | |
| *** jcoufal has quit IRC | 14:04 | |
| *** jtomasek has quit IRC | 14:04 | |
| *** jvrbanac has joined #openstack-dev | 14:05 | |
| *** markmcclain1 has quit IRC | 14:05 | |
| *** alunch has quit IRC | 14:05 | |
| *** jimfehlig has joined #openstack-dev | 14:05 | |
| *** dstanek has quit IRC | 14:06 | |
| *** freedomhui has quit IRC | 14:06 | |
| mrodden | bnemec: i'm confused... what are we punting on? | 14:07 |
| *** alunch has joined #openstack-dev | 14:07 | |
| *** venkatesh has quit IRC | 14:07 | |
| bnemec | mrodden: Lazy translation | 14:07 |
| *** thedodd has joined #openstack-dev | 14:07 | |
| mrodden | ok yeah i see | 14:07 |
| *** clouded_tux has joined #openstack-dev | 14:08 | |
| mrodden | so mark wants to disabled lazy | 14:08 |
| bnemec | It's broken right now and the proposed fix is...scary. :-) | 14:08 |
| bnemec | Yeah | 14:08 |
| *** rushiagr has joined #openstack-dev | 14:08 | |
| mrodden | yeah the double-byte thing was interesting to discover | 14:08 |
| *** davidhadas_ has joined #openstack-dev | 14:09 | |
| bnemec | Seems like the underlying issues is that the logging module doesn't handle unicode well, but that's still a problem for us. | 14:09 |
| mrodden | yeah | 14:09 |
| mrodden | darn python standard library... | 14:09 |
| mrodden | :P | 14:09 |
| bnemec | :-) | 14:09 |
| *** AlexF has joined #openstack-dev | 14:10 | |
| *** davidhadas has quit IRC | 14:11 | |
| *** tmclaugh[work] has joined #openstack-dev | 14:11 | |
| *** radez_g0n3 is now known as radez | 14:11 | |
| mrodden | i think changing the encoding in __str__ to system default would solve the main issue | 14:12 |
| mrodden | since anyone doing translations to something with double-byte chars would run into the same issue even without lazy=True | 14:13 |
| *** jprovazn is now known as jprovazn_mtg | 14:13 | |
| *** jtomasek_ is now known as jtomasek | 14:13 | |
| *** RajeshMohan has quit IRC | 14:14 | |
| *** egallen has joined #openstack-dev | 14:15 | |
| *** jcoufal has joined #openstack-dev | 14:15 | |
| *** mkoderer has quit IRC | 14:15 | |
| *** ausxxh has quit IRC | 14:15 | |
| *** dstanek has joined #openstack-dev | 14:15 | |
| *** Alex_Gaynor has quit IRC | 14:15 | |
| *** RajeshMohan has joined #openstack-dev | 14:16 | |
| *** clouded_tux has quit IRC | 14:16 | |
| *** mohits has quit IRC | 14:16 | |
| *** s2r2 has quit IRC | 14:17 | |
| *** jab416171 has quit IRC | 14:17 | |
| *** briancurtin has joined #openstack-dev | 14:17 | |
| *** tvb|afk has quit IRC | 14:18 | |
| *** kenperkins has joined #openstack-dev | 14:20 | |
| *** athomas has joined #openstack-dev | 14:21 | |
| *** rkukura has joined #openstack-dev | 14:22 | |
| *** chenxu has joined #openstack-dev | 14:25 | |
| *** FunnyLookinHat has joined #openstack-dev | 14:27 | |
| *** epim has joined #openstack-dev | 14:31 | |
| *** Javin has joined #openstack-dev | 14:32 | |
| *** otherwiseguy has joined #openstack-dev | 14:35 | |
| *** markmcclain has joined #openstack-dev | 14:37 | |
| *** rushiagr has quit IRC | 14:38 | |
| *** davidhadas has joined #openstack-dev | 14:39 | |
| *** mohits has joined #openstack-dev | 14:40 | |
| *** jab416171 has joined #openstack-dev | 14:41 | |
| *** s2r2 has joined #openstack-dev | 14:41 | |
| *** clouded_tux has joined #openstack-dev | 14:41 | |
| *** Alex_Gaynor has joined #openstack-dev | 14:41 | |
| *** imsurit has joined #openstack-dev | 14:42 | |
| *** davidhadas_ has quit IRC | 14:43 | |
| *** kevinconway has quit IRC | 14:43 | |
| *** tkammer has quit IRC | 14:43 | |
| *** tkammer has joined #openstack-dev | 14:44 | |
| *** ayoung has joined #openstack-dev | 14:44 | |
| *** jecarey has quit IRC | 14:44 | |
| *** Ryan_Lane has joined #openstack-dev | 14:45 | |
| *** adjohn has quit IRC | 14:48 | |
| *** Ryan_Lane has quit IRC | 14:49 | |
| *** tellesnobrega has quit IRC | 14:50 | |
| *** imsurit has quit IRC | 14:50 | |
| *** imsurit has joined #openstack-dev | 14:51 | |
| *** imsurit is now known as Guest77088 | 14:51 | |
| *** Guest77088 is now known as imsurit_ | 14:52 | |
| *** changbl has joined #openstack-dev | 14:52 | |
| *** Mandell has quit IRC | 14:53 | |
| *** jruzicka has quit IRC | 14:53 | |
| *** rcleere has joined #openstack-dev | 14:54 | |
| *** imsurit_ has quit IRC | 14:54 | |
| *** jasondot_ has joined #openstack-dev | 14:54 | |
| *** carl_baldwin has joined #openstack-dev | 14:55 | |
| *** epim_ has joined #openstack-dev | 14:55 | |
| *** epim has quit IRC | 14:56 | |
| *** epim_ is now known as epim | 14:57 | |
| *** kevinconway has joined #openstack-dev | 14:58 | |
| *** athomas has quit IRC | 15:00 | |
| *** ndipanov is now known as ndipanov_gone | 15:00 | |
| *** reidrac has quit IRC | 15:00 | |
| *** cdub has quit IRC | 15:00 | |
| *** jprovazn_mtg has quit IRC | 15:00 | |
| *** amuller has quit IRC | 15:01 | |
| *** jcoufal has quit IRC | 15:02 | |
| *** boris-42 has quit IRC | 15:02 | |
| mrodden | jswarren, luisg: looking into the issue a bit more, whats stopping us from attempting to encode into the system default, and if that doesn't work then just punting and returning the original english message in ascii | 15:02 |
| *** jistr has quit IRC | 15:03 | |
| mrodden | stack doesn't fall over | 15:03 |
| *** bpokorny has joined #openstack-dev | 15:03 | |
| *** jtomasek has quit IRC | 15:03 | |
| mrodden | user just doesn't get the translated message since their system encoding is weird | 15:03 |
| *** amuller has joined #openstack-dev | 15:03 | |
| *** DinaBelova has quit IRC | 15:04 | |
| luisg | i think jswarren tried that, but it didn't work b/c the default is likely ascii and so all non-ascii langs won't work | 15:04 |
| jswarren | Yes, I basically came to the conclusion that no matter what you did, there were cases where it broke. | 15:05 |
| *** dkranz has joined #openstack-dev | 15:05 | |
| mrodden | mmk | 15:06 |
| jswarren | It has to do with whether a mod operation involves parameters that are all str or a mixture of str and unicode | 15:06 |
| *** yamahata has joined #openstack-dev | 15:06 | |
| mrodden | yeah... | 15:06 |
| *** pmathews has joined #openstack-dev | 15:08 | |
| *** yamahata has quit IRC | 15:09 | |
| *** dstanek has quit IRC | 15:10 | |
| *** paragan has joined #openstack-dev | 15:10 | |
| *** paragan has quit IRC | 15:10 | |
| *** paragan has joined #openstack-dev | 15:10 | |
| *** aeperezt has joined #openstack-dev | 15:10 | |
| *** xqueralt has quit IRC | 15:10 | |
| *** doron_afk is now known as doron | 15:12 | |
| *** gargya_ has joined #openstack-dev | 15:12 | |
| *** mkollaro has quit IRC | 15:12 | |
| *** basha has joined #openstack-dev | 15:13 | |
| *** dstanek has joined #openstack-dev | 15:13 | |
| *** gargya has quit IRC | 15:13 | |
| *** markwash has quit IRC | 15:14 | |
| *** jtomasek has joined #openstack-dev | 15:15 | |
| *** Ryan_Lane has joined #openstack-dev | 15:16 | |
| *** gargya_ has quit IRC | 15:16 | |
| crazed | henrynash: hey, i haven't touched the keystone ldap code in awhile and noticed there's this new assignment api. is the goal there to move project/role membership to sql and remove ldap from that? i know the code was a bit hairy before | 15:18 |
| *** shinylasers has joined #openstack-dev | 15:19 | |
| *** fandikurnia01 has joined #openstack-dev | 15:19 | |
| fandikurnia01 | hi | 15:19 |
| *** doron is now known as doron_afk | 15:19 | |
| *** Ryan_Lane has quit IRC | 15:20 | |
| *** doron_afk is now known as doron | 15:20 | |
| jswarren | bnemec, the change is ugly in terms of how much code is changed, but what it does in practical terms should not be objectionable: it extends the unicode class by adding one method: translate-into, and this method returns a new unicode object containing the translated text. It no more changes a string on the fly than any other operation, such as mod or slice. | 15:20 |
| *** kbringard has quit IRC | 15:20 | |
| *** qba73 has quit IRC | 15:21 | |
| fandikurnia01 | hi all | 15:21 |
| jswarren | I agree that it's a lot of code changes, but I disagree on the substance. | 15:21 |
| fandikurnia01 | can i ask some thing | 15:21 |
| fandikurnia01 | i am new in openstack | 15:21 |
| fandikurnia01 | error: Installed distribution pyparsing 1.5.7 conflicts with requirement pyparsing>=2.0. | 15:21 |
| bnemec | jswarren: Sure, but it's a fundamental change in how the Message class works. | 15:22 |
| *** adjohn has joined #openstack-dev | 15:22 | |
| bnemec | And the Message class was already an edge case. | 15:22 |
| *** adjohn has quit IRC | 15:22 | |
| jswarren | I would say that the changes make it less of an edge case--they bring it closer to the norm. | 15:22 |
| dhellmann | ttx: I can be in the meeting today, but I'll need to catch up with markmc about status or I won't be much help | 15:23 |
| jswarren | If you used only the extent of changes as a yard stick, no one would ever refactor. | 15:23 |
| *** networkstatic has joined #openstack-dev | 15:23 | |
| bnemec | Right, but at the very end of a release the extent of changes is relevant. | 15:24 |
| jswarren | Agreed. | 15:24 |
| bnemec | I'm not saying this should enver go in, just not right now. | 15:24 |
| *** Ryan_Lane has joined #openstack-dev | 15:24 | |
| bnemec | *never | 15:24 |
| jswarren | Yup. | 15:24 |
| *** senk has joined #openstack-dev | 15:25 | |
| ttx | dhellmann: the only thing up is that new bug he mentioned to you | 15:25 |
| *** basha has quit IRC | 15:25 | |
| mrodden | the unicode extend thing would solve the problem with logger | 15:25 |
| ttx | dhellmann: so I'll just need an up to date status on where that is | 15:25 |
| mrodden | but it has no burn-in yet | 15:26 |
| *** datsun180b has joined #openstack-dev | 15:26 | |
| mrodden | which is the issue there | 15:26 |
| ttx | dhellmann: for reference: https://bugs.launchpad.net/oslo/+bug/1225099 | 15:26 |
| uvirtbot | Launchpad bug 1225099 in oslo "Encoding hard-coded in gettextutils.Message" [High,In progress] | 15:26 |
| *** doron is now known as doron_afk | 15:27 | |
| *** doron_afk is now known as doron | 15:27 | |
| *** arezmerita has quit IRC | 15:27 | |
| dhellmann | ttx: ok, I should be able to provide that | 15:28 |
| *** xga has joined #openstack-dev | 15:28 | |
| luisg | bnemec, we were actually thinking of going with a tactical approach | 15:28 |
| ttx | dhellmann: thanks! looks like the most uncertain thing affecting RC1 at this point | 15:28 |
| dhellmann | bnemec, jswarren : sorry for joining the party late (commute), but it sounds like we're basically in agreement to hold off on merging the change? | 15:29 |
| luisg | so abandoning that change that change s the fundamental way Message works | 15:29 |
| ttx | dhellmann: any clarification on impact, in particular, would be welcome. I.e. should it really be a RC1 blocker | 15:29 |
| jswarren | ...and delaying the strategic approach until Icehouse. | 15:29 |
| *** senk has quit IRC | 15:29 | |
| luisg | dhellmann, yeah i think we want to hold off on that particualr change, because its practically a redesign | 15:29 |
| *** o_petit has quit IRC | 15:29 | |
| luisg | however, there is a tactical solution | 15:29 |
| luisg | the main problem is that the LogHandlers don't think Messages are basestrings | 15:29 |
| dhellmann | luisg: can we return something better from __str__? | 15:30 |
| luisg | so we can extend the Message class from unicode, and the problem will be resolved | 15:30 |
| luisg | dhellmann, i'd like to investigate that approach too | 15:30 |
| luisg | a trial and error on __str__ | 15:30 |
| *** doron has left #openstack-dev | 15:30 | |
| dhellmann | luisg: jswarren's change does subclass from unicode, right? it had other changes, too, so are you saying changing the base class would be good enough? | 15:30 |
| *** jmontemayor has joined #openstack-dev | 15:30 | |
| luisg | bnemec, and dhellmann what do u guys think of that? those more "tactial" solutions, we can delay the strategic to I | 15:31 |
| jswarren | dhellman, a number of different approaches were tried and the conclusion was that the only solution that has worked in all cases so far is when Message extends unicode. | 15:31 |
| *** basha has joined #openstack-dev | 15:31 | |
| jswarren | We tried various encodings, etc. | 15:31 |
| luisg | dhellmann, it did in addition to doing a lot of other stuff to make it look more like unicode in terms of its immutability | 15:31 |
| *** kbringard has joined #openstack-dev | 15:31 | |
| dhellmann | right | 15:31 |
| *** romcheg has quit IRC | 15:31 | |
| bnemec | My major issue here is that unicode support is not well tested in the other projects, so even though lazy translation went in months ago, we're still finding problems now. | 15:31 |
| jswarren | But nothing worked in all possible permutations, especially when it comes to mod operations. | 15:31 |
| dhellmann | jswarren: I'm a bit surprised that using the default system encoding didn't fix the problem | 15:31 |
| bnemec | So if we make a significant change to how Message works now, we have no time to make sure it doesn't trigger some other issue in the other projects. | 15:31 |
| jswarren | Yeah, so was I. | 15:32 |
| dhellmann | bnemec: yep, I'm in favor of markmc's proposal to disable the feature entirely for now | 15:32 |
| bnemec | The system encoding does not necessarily equal the locale of the message though. | 15:32 |
| dhellmann | but I do still want to understand the nature of the issue | 15:32 |
| bnemec | I think that's the issue. | 15:32 |
| mrodden | thats correct | 15:32 |
| mrodden | our messages can be dynamic and independent of system locale | 15:32 |
| mrodden | thats the idea of Message | 15:32 |
| dhellmann | ah, good point, bnemec | 15:32 |
| bnemec | I'm pretty sure I left that comment on a review at some point... | 15:32 |
| *** fungi has quit IRC | 15:33 | |
| mrodden | i did find a library that would help with detecting the encoding of the bytes | 15:33 |
| mrodden | chardet | 15:33 |
| jswarren | dhellman, basically it has to do with what happens during mod operations. When there is a mixture of str and unicode, the str objects are "promoted" to unicode using the default encoding (almost always ascii) | 15:33 |
| dhellmann | does this issue cause problems for other uses of translation, other than logging? | 15:33 |
| *** fungi has joined #openstack-dev | 15:33 | |
| *** o_petit has joined #openstack-dev | 15:33 | |
| *** troytoman-away is now known as troytoman | 15:34 | |
| dhellmann | jswarren: right, that part I do get | 15:34 |
| jswarren | Potentially, but not confirmed. | 15:34 |
| dhellmann | I wonder if we can solve the logging problem independently, with an adapter? | 15:34 |
| jswarren | Well, you also have problems when it's all strings. | 15:34 |
| mrodden | the API translations have been working fine | 15:34 |
| mrodden | since we can control the output at the API level | 15:34 |
| mrodden | logging is all internal to python libs | 15:35 |
| mrodden | unfortunately | 15:35 |
| luisg | yeah so far it appears its all in logging | 15:35 |
| jswarren | dhellman, one approach we tried was to swap out the isinstance method...that worked, but needless to say, not kosher. | 15:35 |
| dhellmann | jswarren: no, that feels pretty ugly :-) | 15:35 |
| *** xga_ has joined #openstack-dev | 15:36 | |
| mrodden | have we investigated a logging wrapper? | 15:36 |
| luisg | so dhellmann basically the fix out there already extends from unicode, and that is why the problem is solved | 15:36 |
| *** basha has quit IRC | 15:36 | |
| jswarren | My thinking was that we didn't want to do anything specific for logging, since logging might change between releases of Python. | 15:36 |
| luisg | but it came in with a bunch of extra stuff that i thinw e can remoce | 15:36 |
| luisg | *remove | 15:36 |
| *** xga has quit IRC | 15:36 | |
| mrodden | a wrapper not a monkey patch to the logging | 15:37 |
| mrodden | wrapper would ideally be portable | 15:37 |
| dhellmann | jswarren: I can see that position. If what we have is working for API messages, though, maybe logging is enough of a special case that we need to treat it differently. | 15:37 |
| *** amuller has quit IRC | 15:37 | |
| dhellmann | jswarren: if only because for logging we can know what the encoding should be | 15:37 |
| mrodden | well we can catch the Messages at a logger adapter level | 15:37 |
| mrodden | translate them to the best known locale at the time | 15:38 |
| dhellmann | mrodden: right, that's why I was wondering if an adapter would let us handle this case | 15:38 |
| *** safchain_ has joined #openstack-dev | 15:38 | |
| mrodden | and let it pass through as a unicode | 15:38 |
| jswarren | I guess if we're talking tactical, an adapter makes sense. | 15:38 |
| *** safchain has quit IRC | 15:38 | |
| *** armax has joined #openstack-dev | 15:38 | |
| jswarren | I don't think it's a good long-term solution, because ultimately _ should behave as much as possible as it did before--other libraries see it as a unicode object and no special treatment is required. | 15:39 |
| *** dkranz has quit IRC | 15:39 | |
| luisg | where would the adapter be placed? | 15:39 |
| mrodden | log.py probably... | 15:39 |
| mrodden | or gettextutils.py and used in log.py | 15:39 |
| bnemec | We could even just coerce it to unicode before logging. | 15:39 |
| dhellmann | jswarren: practicality vs. purity | 15:39 |
| bnemec | It's not like logging needs delayed translation behavior. | 15:40 |
| mrodden | bnemec: yep, that would be the adapter's job i was thinking | 15:40 |
| jswarren | dhellman...right a place and time for both. | 15:40 |
| bnemec | There's even been some debate over whether log messages should be translated. | 15:40 |
| mrodden | yeah | 15:40 |
| dhellmann | bnemec: right, logging translation doesn't need to be delayed -- maybe we don't use _ for logging translations | 15:40 |
| mrodden | i dont like it | 15:40 |
| mrodden | (translating logs) | 15:40 |
| dhellmann | I don't either, but users want it | 15:40 |
| dhellmann | so I'm ok doing it | 15:40 |
| *** AlexF has quit IRC | 15:40 | |
| bnemec | Apparently some users prefer logs in English because it makes googling errors easier. | 15:41 |
| dhellmann | do we need a summit session on this? | 15:41 |
| *** sthaha has quit IRC | 15:41 | |
| dhellmann | bnemec: yep; translation would be optional, right? | 15:41 |
| * bnemec won't be at summit | 15:41 | |
| dhellmann | :-( | 15:41 |
| bnemec | dhellmann: I don't think it is right now. | 15:41 |
| mrodden | yeah i won't make it either | 15:41 |
| dhellmann | bnemec: oh, true, we would need separate domains for the log messages, wouldn't we | 15:42 |
| * dhellmann needs to brush up on gettext | 15:42 | |
| mrodden | yeah that goes way back | 15:42 |
| bnemec | Yeah. | 15:42 |
| clarkb | markmc ^ | 15:42 |
| mrodden | markmc suggested two domains originally | 15:42 |
| mrodden | API_() | 15:42 |
| mrodden | and _() | 15:42 |
| dhellmann | I remember seeing something about that | 15:42 |
| mrodden | or something like that | 15:42 |
| bnemec | I think that predates my involvement in this. :-) | 15:43 |
| mrodden | i think it predated mine too :) | 15:43 |
| * dhellmann checks for a blueprint | 15:43 | |
| luisg | so going back to the adapter idea, if ppl setup the logging through logging.conf how would the adapter be involved | 15:44 |
| briancline | mayhap this is a dumb question; but when would a fix like this make its way down to the precise packages for havana? https://code.launchpad.net/~zulcss/horizon/1210253/+merge/181311 | 15:44 |
| neelashah | luisg: mroddem: would it be possible to see a change proposed for the logging wrapper? | 15:44 |
| *** venkatesh has joined #openstack-dev | 15:44 | |
| luisg | neelashah, yeah i think so, just want to make sure i understand the approach | 15:45 |
| mrodden | actually we already have the wrapper | 15:45 |
| mrodden | its the test handler in the bottom of gettextutils | 15:45 |
| mrodden | its just a Handler right now instead of an Adapter | 15:45 |
| mrodden | its actually a Handler that wraps other Handlers | 15:46 |
| *** mriedem has joined #openstack-dev | 15:46 | |
| mrodden | but it is the same concept | 15:46 |
| luisg | mrodden, u have a good handle on the approach i would say go for it | 15:46 |
| jswarren | How invasive is this change? | 15:47 |
| luisg | :) | 15:47 |
| mrodden | every project that uses Messages would need an additional logging adapter | 15:48 |
| mrodden | in place | 15:48 |
| mrodden | obviously we can do that with code in oslo | 15:48 |
| *** primemin1sterp has quit IRC | 15:49 | |
| dhellmann | I thought we already had some sort of log adapter already, but I don't see one in the incubator code | 15:49 |
| bnemec | briancline: You'd have to ask the Ubuntu people. We don't have anything to do with packaging. | 15:49 |
| *** primeministerp has joined #openstack-dev | 15:50 | |
| mrodden | dhellmann: there is a ContextAdapter in log.py | 15:50 |
| bnemec | briancline: Looks like Chuck is zul here. He might know. | 15:50 |
| mrodden | and LocalizedLogHander (or something) in gettextutils.py | 15:50 |
| dhellmann | mrodden: thanks, I must have typo-ed "adapter" | 15:50 |
| bnemec | Probably spelled it adaptor. I can never remember which it is. :-) | 15:51 |
| *** mlavalle has joined #openstack-dev | 15:51 | |
| *** epim has quit IRC | 15:51 | |
| dhellmann | mrodden: well it looks like the place to set the encoding for log messages is in the LocaleHandler, since that's what it claims to be doing now | 15:52 |
| *** yassine has quit IRC | 15:52 | |
| *** dkranz has joined #openstack-dev | 15:52 | |
| mrodden | dhellmann: correct, but the handler is an example available for users | 15:52 |
| mrodden | it doesn't get utilized yet | 15:53 |
| zul | briancline: it should be getting updated soon | 15:53 |
| mrodden | except by some internal stuff we have | 15:53 |
| bnemec | BTW, this was the comment I made previously on this: http://docs.python.org/2/library/locale.html#locale.nl_langinfo | 15:53 |
| dhellmann | mrodden: do you see an issue with changing that? | 15:53 |
| bnemec | Regarding getting the encoding for the locale. | 15:53 |
| mrodden | a LoggingAdapter would have all messages flow through it | 15:53 |
| dhellmann | I know we need to be careful about tying the logging and gettextutils modules together | 15:53 |
| mrodden | right | 15:54 |
| mrodden | another risk | 15:54 |
| *** SergeyLukjanov has quit IRC | 15:54 | |
| dhellmann | ok, maybe the adapter makes more sense -- I was looking for an existing place to fix it | 15:54 |
| *** networkstatic has quit IRC | 15:54 | |
| *** dprince has quit IRC | 15:54 | |
| bnemec | I'm +1 on the adapter. | 15:54 |
| jswarren | bnemec, aren't locales and encodings independent? Russian, for instance can be encoded in UTF8, UTF16 or KOI8 | 15:54 |
| *** senk has joined #openstack-dev | 15:54 | |
| bnemec | I looked into it before, but I don't know enough about logging to make it work. :-) | 15:55 |
| bnemec | jswarren: Yeah, I was thinking locale.CODESET sounded like what we needed though. | 15:55 |
| bnemec | http://docs.python.org/2/library/locale.html#locale.CODESET | 15:55 |
| bnemec | Whatever that returns should properly encode the locale at least. | 15:56 |
| bnemec | One would hope. | 15:56 |
| bnemec | :-) | 15:56 |
| bnemec | I could be completely wrong too. I haven't looked into it much. | 15:56 |
| *** senk has quit IRC | 15:56 | |
| *** dosaboy has quit IRC | 15:56 | |
| *** dosaboy has joined #openstack-dev | 15:56 | |
| luisg | bnemec, i think that is a good idea | 15:56 |
| luisg | but im kind of worried about the system giving us various encodings | 15:57 |
| luisg | but it looks like it just chooses 1 | 15:57 |
| luisg | maybe u can only use 1 jswarren | 15:57 |
| *** danwent has joined #openstack-dev | 15:57 | |
| *** emagana has joined #openstack-dev | 15:58 | |
| dhellmann | bnemec: locale.getlocale() returns the locale name and an encoding name | 15:58 |
| luisg | so we could in the str method see the locale and use this to get the encoding | 15:58 |
| luisg | and encode to that | 15:58 |
| jswarren | I'm not sure what is being discussed...in any case, changing the default locale is possible, but python must be started with -s option, as I recall. | 15:58 |
| *** jasondot_ has quit IRC | 15:58 | |
| luisg | jswarren, see bnemec proposal above | 15:58 |
| *** senk has joined #openstack-dev | 15:58 | |
| bnemec | Not looking to change the default - just the value used here: https://github.com/openstack/oslo-incubator/blob/master/openstack/common/gettextutils.py#L240 | 15:58 |
| dhellmann | locale.getlocale() honors the posix environment variables to control the locale | 15:58 |
| luisg | we would not change the default locale | 15:58 |
| jswarren | The problem isn't only with the __str__ method that is in Message...it's with encoding/decoding logic in modules we don't control. | 15:59 |
| *** egallen has quit IRC | 15:59 | |
| bnemec | jswarren: The logging adapter would take care of that though. | 15:59 |
| *** xga_ has quit IRC | 15:59 | |
| bnemec | We'd encode the string before it got to that code. | 15:59 |
| bnemec | (as I understand it) | 15:59 |
| dhellmann | jswarren: the proposed fix is to leave Message alone (or make fewer changes) and have a logging adapter that catches Message instances and turns them into encoded strings | 15:59 |
| *** xga_ has joined #openstack-dev | 16:00 | |
| jswarren | The logging formatter perform a mod operation...does that go before or after the adapter? | 16:00 |
| dhellmann | that may not fix the problem for API uses, so we may still need changes in Message | 16:00 |
| dhellmann | after | 16:00 |
| jswarren | Then you still have the same problem. | 16:00 |
| dhellmann | so the logging formatter could even get a unicode string, maybe the adapter doesn't do encoding at all | 16:00 |
| mrodden | but we should be injecting into a unicode | 16:00 |
| *** Ruetobas has quit IRC | 16:01 | |
| jswarren | Yeah, OK. | 16:01 |
| *** gmoro has quit IRC | 16:01 | |
| jswarren | If the adapter makes it a unicode object, then it should be OK. | 16:01 |
| dhellmann | yeah, I got that backwards, but you're right | 16:01 |
| mrodden | the idea is the the adapter will do Message -> Unicode | 16:01 |
| briancline | zul: thanks -- I know that particular one is marked as low urgency, but is that a <7days soon or a <~14 days soon? :) | 16:02 |
| dhellmann | who can summarize this discussion in the bug report so we have a record? | 16:02 |
| *** adalbas has quit IRC | 16:02 | |
| *** chenxu has quit IRC | 16:02 | |
| luisg | k so it looks like we have at least 1 tactical solution that would work | 16:02 |
| luisg | just for completeness did u guys see problems with making Message unicode ? | 16:02 |
| *** paragan has quit IRC | 16:03 | |
| luisg | mrodden, also u mentioned above a library that "detects" encoding? | 16:03 |
| mrodden | chardet | 16:03 |
| *** egallen has joined #openstack-dev | 16:03 | |
| luisg | that woud limit the change to the str() method | 16:03 |
| *** Ruetobas has joined #openstack-dev | 16:04 | |
| luisg | it's along the lines of the fix proposed by bnemec | 16:04 |
| mrodden | right | 16:04 |
| luisg | we could explore that | 16:04 |
| jswarren | I wouldn't call any changes to encoding as doable until tested. | 16:04 |
| bnemec | +1 | 16:04 |
| luisg | jswarren, yeah we would definitely need to test the changes regardless | 16:05 |
| dhellmann | luisg: I don't see a problem with making it a unicode, but is that still needed? | 16:05 |
| dhellmann | luisg: why would chardet be useful? | 16:05 |
| luisg | that woudl fix the problem of the handler trying to str() it | 16:05 |
| *** galstrom_zzz is now known as galstrom | 16:06 | |
| dhellmann | trying to str() the Message object? | 16:06 |
| luisg | yes.. | 16:06 |
| luisg | i think because we can examine the bytes in .msg and we would be able to sort of dynamically adjust the encoding to those bytes | 16:06 |
| luisg | instead of hardcoding it like it currently is | 16:06 |
| *** cdub has joined #openstack-dev | 16:07 | |
| *** FatDarrel has joined #openstack-dev | 16:07 | |
| dhellmann | why is the data inside Message encoded at all? | 16:07 |
| mrodden | i think the worst case there is that the message comes out unreadable because we guess the encoding wrong | 16:07 |
| jswarren | I tried a lot of things involving monkeying with the encoding...giving logging a unicode object (via an adapter or whatever), is the safest bet, IMHO. | 16:07 |
| bnemec | Well, or it fails because str.encode() can't handle characters in it. | 16:07 |
| dhellmann | yeah, we want Message to have unicode objects internally and we want to call unicode() on it to get a unicode string to pass to the logging code | 16:08 |
| dansmith | mordred: nicely put. | 16:08 |
| dhellmann | Message itself doesn't need to be a unicode object | 16:08 |
| luisg | and we can return an encode('replace') | 16:08 |
| mordred | dansmith: thanks! | 16:08 |
| fandikurnia01 | hi there | 16:08 |
| *** Ruetobas has quit IRC | 16:08 | |
| dhellmann | jswarren: right? ^^ | 16:08 |
| fandikurnia01 | has anyone try this | 16:08 |
| fandikurnia01 | 2013-10-01 23:05:26 Installed /opt/stack/nova | 16:08 |
| fandikurnia01 | 2013-10-01 23:05:26 Processing dependencies for nova==2013.1.4.a27.gfaabb91 | 16:08 |
| fandikurnia01 | 2013-10-01 23:05:26 error: Installed distribution pyparsing 1.5.7 conflicts with requirement pyparsing>=2.0.1 | 16:08 |
| jswarren | dhellman, yes. | 16:08 |
| *** crank has joined #openstack-dev | 16:08 | |
| *** markwash has joined #openstack-dev | 16:08 | |
| luisg | i agree on that adapter, was just a bit worried about how to ensure that it is always used | 16:09 |
| dhellmann | so nothing should pass encoded data to Message and expect it to handle that in any useful way (other than repr), and we shouldn't need to guess the encoding | 16:09 |
| luisg | not sure how that was going to work if ppl uses their own loggingn.conf fior instance | 16:09 |
| *** iartarisi has quit IRC | 16:09 | |
| *** blues-man has left #openstack-dev | 16:09 | |
| dhellmann | how does that work with the ContextAdapter now? | 16:10 |
| mrodden | yeah thats a good point... | 16:10 |
| *** bdpayne has joined #openstack-dev | 16:10 | |
| bnemec | Don't all of the projects use getLogger? https://github.com/openstack/oslo-incubator/blob/master/openstack/common/log.py#L453 | 16:10 |
| mrodden | yeah | 16:11 |
| mrodden | i think we are okay | 16:11 |
| mrodden | since its an adapter | 16:11 |
| dhellmann | bnemec: yeah, that's how the ContextAdapter works | 16:11 |
| mrodden | not a handler | 16:11 |
| bnemec | I guess that doesn't guarantee anything, but as long as they're using the standard logging code we can enforce our adapter. | 16:11 |
| dhellmann | right | 16:11 |
| *** sushils has joined #openstack-dev | 16:11 | |
| bnemec | And if they're not using the standard logging code they probably shouldn't be using lazy translation either. :-) | 16:11 |
| luisg | ah ok, so we can even do it there in CA | 16:11 |
| dhellmann | in CA? | 16:11 |
| luisg | ContextAdapter | 16:12 |
| mrodden | context adapter... | 16:12 |
| dhellmann | ok | 16:12 |
| *** boris-42 has joined #openstack-dev | 16:12 | |
| mrodden | but we may want another one to wrap that | 16:12 |
| luisg | err Califronia :P | 16:12 |
| dhellmann | sorry, pre-caffeine :-) | 16:12 |
| bnemec | Blame Canada! | 16:12 |
| bnemec | :-D | 16:12 |
| luisg | ha | 16:12 |
| dhellmann | yes, let's make a new adapter and wrap the ContextAdapter with it (or the other way, around) | 16:12 |
| dhellmann | separation of concerns, etc. | 16:12 |
| luisg | got it, so that looks good to me | 16:12 |
| mrodden | ok... | 16:13 |
| dhellmann | luisg: so you'll write up a quick summary of the agreed approach in the bug for us? | 16:13 |
| mrodden | so who wants to write it :) | 16:13 |
| *** alop has joined #openstack-dev | 16:13 | |
| luisg | for sure | 16:13 |
| dhellmann | cool, thanks | 16:13 |
| luisg | np | 16:13 |
| bnemec | So where does this leave us for rc1? | 16:13 |
| mrodden | and thats the next Q | 16:13 |
| dhellmann | have we come up with anything we think is low enough risk to enable in havana? | 16:13 |
| bnemec | Are we still trying to get this in or waiting for Icehouse? | 16:13 |
| dhellmann | I'm still leaning towards waiting | 16:14 |
| luisg | i think the Adapter approach is low risk enough, because it will only focus on Message objects | 16:14 |
| luisg | while logging | 16:14 |
| dhellmann | OTOH, it might break all of logging :-) | 16:14 |
| bnemec | I like the Adapter change, but we don't have an actual fix to review yet. | 16:14 |
| dhellmann | right | 16:14 |
| *** adalbas has joined #openstack-dev | 16:15 | |
| *** Ruetobas has joined #openstack-dev | 16:15 | |
| dhellmann | if we can get the adapter written, I'm still inclined to have it disabled by default (remove the calls to set up lazy evaluation from the other projects) | 16:15 |
| *** Shaan7 has joined #openstack-dev | 16:15 | |
| luisg | we can take care of that fix | 16:15 |
| luisg | the adapter is actually very low risk, if there is a Message objec,t we unicode it | 16:16 |
| luisg | so it should not really affect anything else | 16:16 |
| bnemec | It's just so late in the cycle to be messing around with fundamental pieces of OpenStack. | 16:17 |
| dhellmann | does Message work correctly, without any changes, for uses other than logging? | 16:17 |
| luisg | it does, i think it has been tested, the issue was just that internal logging | 16:17 |
| dhellmann | agreed | 16:17 |
| luisg | for everythign else it has worked | 16:17 |
| luisg | well | 16:17 |
| luisg | i think it'd be a shame to drop the entire feature :( | 16:17 |
| *** angdraug has joined #openstack-dev | 16:17 | |
| luisg | especially just b/c we have a very low risk solution | 16:18 |
| *** mkoderer_ has joined #openstack-dev | 16:18 | |
| mrodden | yes we test it internally | 16:18 |
| mrodden | it does work | 16:18 |
| bnemec | We have a low-risk plan. ;-) | 16:18 |
| dhellmann | by "internally" you mean at your work on a deployment? | 16:18 |
| mrodden | yes | 16:18 |
| dhellmann | but not with logging? | 16:18 |
| bnemec | I think that's how they found the logging issue, wasn't it? | 16:19 |
| mrodden | we use a custom logging adapter that got dropped from the H release | 16:19 |
| luisg | bnemec, correct | 16:19 |
| mrodden | but yeah | 16:19 |
| dhellmann | ah | 16:19 |
| mrodden | we did find it while debugging internally | 16:19 |
| bnemec | What kind of timeframe are we looking at here? | 16:19 |
| luisg | i think hours tbh, based on that approach | 16:19 |
| bnemec | https://wiki.openstack.org/wiki/Havana_Release_Schedule is kind of vague on specific dates for this part of the cycle. | 16:20 |
| dhellmann | some of that may be decided during the release meeting today, I'm not sure | 16:20 |
| *** SergeyLukjanov has joined #openstack-dev | 16:20 | |
| dhellmann | go ahead and create the adapter | 16:20 |
| bnemec | I mean, even if everything goes perfectly we're looking at a couple of days minimum to get this into Oslo and synced to the other projects. | 16:20 |
| *** DinaBelova has joined #openstack-dev | 16:20 | |
| *** DinaBelova has quit IRC | 16:20 | |
| dhellmann | that way we have something concrete to evaluate | 16:20 |
| bnemec | +1 | 16:21 |
| dhellmann | bnemec: right | 16:21 |
| mrodden | ++ | 16:21 |
| dhellmann | I don't want to rush things, but let's not wait until after the summit to move | 16:21 |
| *** nermina has quit IRC | 16:21 | |
| luisg | makes sense | 16:21 |
| *** flaper87 is now known as flaper87|afk | 16:21 | |
| luisg | so let's get the adapter out there | 16:21 |
| dhellmann | right | 16:21 |
| dhellmann | who's on that? | 16:21 |
| luisg | mrodden, is | 16:22 |
| luisg | i'll document and start testing mrodden change | 16:22 |
| mrodden | sounds good | 16:22 |
| dhellmann | ok, great | 16:23 |
| bnemec | dhellmann: You've got the release meeting covered? | 16:23 |
| luisg | dhellmann, bnemec thank you | 16:23 |
| bnemec | Just want to make sure we don't need to update Mark. | 16:23 |
| *** terryw has joined #openstack-dev | 16:23 | |
| dhellmann | bnemec: yeah, I'll be there | 16:23 |
| bnemec | Cool. | 16:23 |
| dhellmann | I think markmc was in transit at that time | 16:24 |
| bnemec | Sounds like we're good then. | 16:24 |
| dhellmann | yep, thank you all, this is going to be a great team to work with :-) | 16:24 |
| bnemec | +2! | 16:24 |
| bnemec | :-) | 16:24 |
| *** djdeffjamz has joined #openstack-dev | 16:25 | |
| *** gimps_ has joined #openstack-dev | 16:25 | |
| *** otherwiseguy has quit IRC | 16:26 | |
| *** egallen has quit IRC | 16:26 | |
| luisg | :) thx | 16:27 |
| *** Ruetobas has quit IRC | 16:27 | |
| *** DennyZhang has joined #openstack-dev | 16:28 | |
| *** aditirav has joined #openstack-dev | 16:28 | |
| *** DennyZhang has quit IRC | 16:28 | |
| clouded_tux | Hi.. all I just hit this issue with horizon https://lists.launchpad.net/openstack/msg21342.html Can anyone provide me direction on how to debug | 16:30 |
| *** egallen has joined #openstack-dev | 16:30 | |
| *** Ruetobas has joined #openstack-dev | 16:31 | |
| *** DennyZhang has joined #openstack-dev | 16:32 | |
| *** venkatesh has quit IRC | 16:33 | |
| clouded_tux | Hi.. all I just hit this issue with horizon https://lists.launchpad.net/openstack/msg21342.html Can anyone provide me direction on how to debug | 16:33 |
| *** JordanP has quit IRC | 16:35 | |
| *** safchain_ has quit IRC | 16:35 | |
| *** Mandell has joined #openstack-dev | 16:36 | |
| *** shinylasers has quit IRC | 16:36 | |
| *** shinylasers has joined #openstack-dev | 16:37 | |
| *** shinylasers has quit IRC | 16:38 | |
| *** DinaBelova has joined #openstack-dev | 16:38 | |
| *** terryw is now known as otherwiseguy | 16:39 | |
| *** gyee has joined #openstack-dev | 16:40 | |
| *** ygbo has quit IRC | 16:42 | |
| *** garyk has quit IRC | 16:42 | |
| *** hemna has joined #openstack-dev | 16:45 | |
| bigmstone | BobBall: You around? | 16:47 |
| *** devoid has joined #openstack-dev | 16:48 | |
| *** derekh has quit IRC | 16:52 | |
| *** tkammer has quit IRC | 16:52 | |
| crazed | what's the status of keystone using ldap for identity and mysql for assignment? | 16:52 |
| crazed | as far as havana release goes | 16:52 |
| *** mrunge has joined #openstack-dev | 16:53 | |
| dolphm | crazed: supported! | 16:53 |
| crazed | dolphm: is that the new preferred way? | 16:53 |
| *** danpb has quit IRC | 16:53 | |
| dolphm | crazed: i think it will be a popular option, i don't know about preferred | 16:53 |
| crazed | dolphm: what does the config look like to get that going? | 16:53 |
| crazed | i'm trying out a beta release of havana | 16:54 |
| crazed | and noticed all the ldap stuff changed a bit | 16:54 |
| crazed | with the asssignment api | 16:54 |
| dolphm | crazed: just set the [assignment] driver and [identity] drivers appropriately | 16:54 |
| shardy | dolphm, ayoung: I've been looking at this trusts tempest coverage issue | 16:54 |
| crazed | dolphm: i guess but how does the mysql table know what to lookup in ldap | 16:55 |
| shardy | there is no coverage for trusts at all, and IMHO it's not that reasonable to gate the client change on adding it | 16:55 |
| *** emagana has quit IRC | 16:55 | |
| dolphm | crazed: the ldap configuration is otherwise in the same place ([ldap], i believe) | 16:55 |
| *** lucasagomes has quit IRC | 16:55 | |
| shardy | dolphm, ayoung: I'm happy to help adding some during Icehouse however, if we can just raise some bugs and assign them to me? | 16:55 |
| dolphm | crazed: choosing identity for ldap means that users and groups will always be pulled from ldap | 16:55 |
| dolphm | err ldap for identity* | 16:56 |
| *** o_petit has quit IRC | 16:56 | |
| *** rkukura has quit IRC | 16:56 | |
| crazed | dolphm: but for example, how does a role get assigned if it exists in mysql? | 16:56 |
| dolphm | crazed: the assignment is persisted in sql | 16:56 |
| dolphm | crazed: as is the role definition | 16:57 |
| *** eglynn has quit IRC | 16:58 | |
| crazed | dolphm: well for example, does it use the ldap DN for the user_id in the mysql tables or? | 16:58 |
| dolphm | crazed: that's configurable via [ldap] user_id_attribute | 16:59 |
| *** chenxu has joined #openstack-dev | 16:59 | |
| crazed | ah | 17:00 |
| *** Max__ has quit IRC | 17:00 | |
| dolphm | crazed: i don't think dn is the default, but we had a discussion recently that dn would be a more logical, useful default | 17:00 |
| crazed | okay | 17:00 |
| *** emagana has joined #openstack-dev | 17:00 | |
| dolphm | crazed: (i think it defaults to cn) | 17:00 |
| crazed | yeah i have it set to uid | 17:00 |
| dolphm | crazed: which may not play well in multi-domain deploys | 17:00 |
| *** davidhadas has quit IRC | 17:00 | |
| crazed | this is single domain so that's okay | 17:00 |
| *** mlavalle has quit IRC | 17:01 | |
| *** Ryan_Lane has quit IRC | 17:01 | |
| *** davidhadas has joined #openstack-dev | 17:01 | |
| *** JonnyNomad has quit IRC | 17:02 | |
| ayoung | shardy, I've been in meetings...it must be Tuesday | 17:02 |
| *** DennyZhang has quit IRC | 17:02 | |
| *** danwent has quit IRC | 17:02 | |
| crazed | dolphm: so the keystoneclient does not have group commands or am i just running older version of the client | 17:03 |
| ayoung | dolphm, I had a lot of people push back against the DN as the userID attribute. I think that will be a major discussion topic at the Summit | 17:03 |
| *** xga_ has quit IRC | 17:03 | |
| *** mosulica has quit IRC | 17:04 | |
| dolphm | crazed: the keystoneclient library supports groups, but they're exposed on the CLI in python-openstackclient | 17:04 |
| *** alunch_ has joined #openstack-dev | 17:04 | |
| ayoung | crazed, soundsl ike you are using LDAP. I assume that you wnt LDAP in read-only mode? | 17:04 |
| *** alunch has quit IRC | 17:04 | |
| *** alunch_ is now known as alunch | 17:04 | |
| dolphm | ayoung: interesting, what's the big reason why? | 17:04 |
| ayoung | dolphm, if it is read write, he can do the whole thing in LDAP...CERN wants to keep doing that | 17:05 |
| *** afazekas_no_irq has quit IRC | 17:05 | |
| ayoung | dolphm, but the majority wants LDAP read only, and I assume crazed wants that as well | 17:05 |
| Javin | bnemec: I can't reporeduce the issue you had yesterday with https://review.openstack.org/#/c/48386/ | 17:05 |
| dolphm | ayoung: my bad, i was referring to "why not DN?" | 17:06 |
| ayoung | dolphm, Ah...well, one rason they gave was that it was possible that an organization might reorgainzae and the DNs change. that would break everything external | 17:06 |
| crazed | ayoung: dolphm, I want ldap read/write for things like groups | 17:06 |
| *** fbo is now known as fbo_away | 17:06 | |
| Javin | bnemec: Without creating 2 interfaces with the same IP by hand. I'm not sure that is a valid configuration. Maybe just detecting that and producing a valid error is the right approach here? | 17:07 |
| ayoung | but bascially, it is an implementation detail, and not necessarily something that should be tightly linked to keystone | 17:07 |
| ayoung | crazed, that is fine...read/write works well | 17:07 |
| crazed | yeah i think i just figured it out | 17:07 |
| *** danwent has joined #openstack-dev | 17:07 | |
| ayoung | crazed, so you have to chse whether to put the assignemtn s (roles, projects) in LDAP as well. Users and Groups are in the Identity backend, and that is expected to be LDAP | 17:07 |
| crazed | this makes sense | 17:07 |
| bnemec | Javin: That sounds reasonable. I really don't know much about how this networking stuff works, so my setup may just be wrong. :-) | 17:07 |
| bnemec | (although it does seem to work fine) | 17:08 |
| crazed | ayoung: i like roles,projects in mysql personally | 17:08 |
| *** aditirav has quit IRC | 17:08 | |
| *** reed has joined #openstack-dev | 17:08 | |
| crazed | especially when i see things like "LDAP backend is being deprecated" in the code haha | 17:08 |
| crazed | for the assignment backend at least | 17:08 |
| ayoung | crazed, then explicitly set the assignments backend to SQL | 17:08 |
| ayoung | crazed, yeah, but that is no longer true | 17:08 |
| ayoung | crazed, we had a pretty big deployment come back and say "don't deprecate this" | 17:09 |
| crazed | ah | 17:09 |
| Javin | bnemec: Take a look at http://openstack.redhat.com/Neutron_with_existing_external_network I think this is what you likely are trying to do. The real interface (em1) has no bootproto and doesn't get an IP, instead the normal IP is applied to br-ex (but I'm also a bit new at all of this) | 17:09 |
| *** Dr_Who has joined #openstack-dev | 17:09 | |
| *** Dr_Who has joined #openstack-dev | 17:09 | |
| ayoung | crazed, their argument was that LDAP was the only thing that covered their scale, and they were OK with it being single domain | 17:09 |
| crazed | my project/role size will always be fairly small | 17:10 |
| Javin | bnemec: Anyhow, I'll add a simple test there for 2 IPs and re-submit the change | 17:10 |
| bnemec | Javin: Yeah, that's how I configured this, but somehow em1 ended up with an IP too. I have no idea how because I didn't do it intentionally. | 17:10 |
| bnemec | Sounds good. | 17:10 |
| ayoung | shardy, can you talk https://review.openstack.org/#/c/48462/ for a minute? | 17:11 |
| *** cdub_ has quit IRC | 17:11 | |
| shardy | ayoung: definitely :) | 17:11 |
| shardy | ayoung: I had a look at the tempest tests, and the problem is we need the v3 test support (to create the trust) before we can add the v2 consuming trust test | 17:12 |
| crazed | ayoung: dolphm, thanks for talking over the ldap stuff with me, i think ldap for identity and mysql for assignment is what we want | 17:12 |
| ayoung | shardy, here's what I would like to see: in the review, it needs to give a step by step how to test it that someone else can reproduce | 17:12 |
| shardy | ayoung: as mentioned, there's no coverage of trusts at all atm, which obviously would be good to fix :) | 17:12 |
| ayoung | shardy, code would be ideal | 17:13 |
| ayoung | shardy, there is unit test coverage, though | 17:13 |
| ayoung | shardy, trusts are tested inside of the keystone project, but we can't do that with the keystone client...until the keystone client itslef is commited | 17:13 |
| shardy | ayoung: Ok, I can easily create a minimal reproducer and push it to a gist | 17:13 |
| ayoung | shardy, that would be a great first step | 17:13 |
| *** chenxu has quit IRC | 17:13 | |
| dolphm | crazed: glad to help! | 17:14 |
| shardy | ayoung: The code does work, and has been verified as working by those testing the heat change which requires it | 17:14 |
| ayoung | shardy, beyond that, we should push a test to tempest that exercises the client trust functios, same as that script. | 17:14 |
| shardy | ayoung: As I said, I think we need the trusts v3 coverage first | 17:14 |
| *** gyee has quit IRC | 17:14 | |
| shardy | so the auth functions become trust aware and you can generate the trust_id to consume | 17:15 |
| *** djoreilly has quit IRC | 17:15 | |
| ayoung | shardy, I am working on moving some of the other Keystone tests into Tempest as well, especially the client tests. As far as the Trust API goes, those are tested here... | 17:15 |
| *** tellesnobrega has joined #openstack-dev | 17:15 | |
| tellesnobrega | dolphm: hey | 17:15 |
| shardy | ayoung: I am happy to help with all of that, just during Icehouse rather than right now with the RC looming ;) | 17:16 |
| *** AnilV4 has quit IRC | 17:16 | |
| ayoung | https://github.com/openstack/keystone/blob/master/keystone/tests/test_v3_auth.py#L1791 | 17:16 |
| tellesnobrega | dolphm: i would like to ask you some questions about the bug https://bugs.launchpad.net/keystone/+bug/1220913 | 17:16 |
| uvirtbot | Launchpad bug 1220913 in keystone "bp split-identity left unfinished" [Wishlist,Triaged] | 17:16 |
| *** SumitNaiksatam has joined #openstack-dev | 17:16 | |
| *** networkstatic has joined #openstack-dev | 17:16 | |
| ayoung | shardy, understood, but from a Keystone perspective we would rather say no to a feature than have it in untested and broken. | 17:17 |
| ayoung | and there is a real risk of that | 17:17 |
| ayoung | so lets fill in the gaps | 17:17 |
| shardy | ayoung: but you have no test coverage of any of the trusts feature in tempest? | 17:17 |
| *** gyee has joined #openstack-dev | 17:17 | |
| *** dsantos_ is now known as demontiesantos | 17:17 | |
| dolphm | tellesnobrega: o/ | 17:17 |
| shardy | ayoung: this is not a feature, it's just adding an interface to something which is already there, so tbh, I'm confused | 17:17 |
| dolphm | tellesnobrega: feel free to ask the channel, btw -- there's several people here familiar with the work | 17:18 |
| ayoung | shardy, the problem with the client is that it can't be tested in Keystone. Trusts are tested in Keystone itself, against a live server | 17:18 |
| tellesnobrega | dolphm: i started working on it, but there are some stuff that i didnt understand | 17:18 |
| ayoung | shardy, you are adding to the python library that is the primary interface into Keystone. It is so a feature | 17:18 |
| shardy | ayoung: https://github.com/openstack/keystone/blob/master/keystone/tests/test_auth.py#L587 | 17:18 |
| tellesnobrega | i changend the get_project_by_user and all worked nicelly, ran the tests and nothing broke | 17:19 |
| ayoung | shardy, we have keystoneclient tests in Keystone proper, (not test_auth.py) but those are there for a different reason. Those are there to make sure that Keystone doens't change away and break the client. We can't do that, though, for a new function in the client | 17:20 |
| ayoung | its a chicken /egg thing..and Tempest is the solution | 17:20 |
| ayoung | but the test doesn't need to be accepted into tempest to make me happy, just have it submitted | 17:20 |
| shardy | ayoung: so basically you're saying we shouldn't have merged any of the keystoneclient trusts support due to lack of tempest coverage? | 17:21 |
| shardy | nobody ever expressed that requirement to me during the review cycle | 17:21 |
| ayoung | shardy, yiou could, in theory, write an extension to the keystoneclient tests in the Keystone repo. But that would not get accepted into Keystone at this late a date | 17:21 |
| ayoung | shardy, no no no | 17:21 |
| ayoung | I'm saying "Write me a test" | 17:21 |
| ayoung | I'm giving you the tools to do it | 17:21 |
| *** sandywalsh has quit IRC | 17:23 | |
| ayoung | https://review.openstack.org/#/c/48462/1/keystoneclient/tests/v2_0/test_auth.py just says that the code does what you think it should | 17:23 |
| ayoung | and that is what a unit test should be doing | 17:23 |
| ayoung | but we need a functional test. | 17:23 |
| ayoung | It doesn't need to be complete in order to pass that review, but I need to see that it is underway | 17:23 |
| shardy | ayoung: can you clarify what the difference is between the test_token_from_trust* unit tests in test_auth and the test you previously linked? | 17:24 |
| shardy | https://github.com/openstack/keystone/blob/master/keystone/tests/test_auth.py#L758 | 17:24 |
| shardy | AFAICS there is test coverage inside keystone, just not in tempest, which is the same as all of the trusts functionality | 17:24 |
| ayoung | shardy, yeah..one goes against a live server. all of the test_v3 calls go through the whole wsgi stack | 17:25 |
| shardy | ayoung: Ok, thanks, that's the detail I was missing | 17:25 |
| *** mmagr has quit IRC | 17:25 | |
| ayoung | shardy, yeah...understand, that much of the code in Keystone predates Tempest. It wasn't possible to have an external test when test_keystoneclient.py was written | 17:26 |
| ayoung | shardy, so what I want to have is a way to test trusts from client to server...end to end integration. | 17:26 |
| *** dperaza has joined #openstack-dev | 17:27 | |
| *** carl_baldwin has quit IRC | 17:27 | |
| shardy | ayoung: +1, we've invested a lot of time integrating Heat with trusts and I absolutely want us to ensure they're as well tested as possible | 17:28 |
| shardy | ayoung: equally, I'd rather not have Havana Heat trusts functionality totally broken :( | 17:28 |
| shardy | ayoung: I'll go back to looking at tempest, thanks | 17:28 |
| *** prad has joined #openstack-dev | 17:29 | |
| *** jmontemayor has quit IRC | 17:29 | |
| *** jmontemayor has joined #openstack-dev | 17:29 | |
| *** Farooque has joined #openstack-dev | 17:29 | |
| dperaza | Looking for oslo cores to help me figure out if going on right direction here: https://review.openstack.org/#/c/48733 | 17:29 |
| dperaza | is this the right channel for this? | 17:30 |
| *** rkukura has joined #openstack-dev | 17:30 | |
| *** Farooque has left #openstack-dev | 17:31 | |
| *** nati_ueno has joined #openstack-dev | 17:33 | |
| shardy | ayoung: gist example https://gist.github.com/hardys/6435299#file-trust_pw_test_v2-py | 17:34 |
| *** mrunge has quit IRC | 17:35 | |
| *** corXi has quit IRC | 17:37 | |
| bnemec | dperaza: I'm not a database expert, but I don't see any major issues with that (other than the fact that it's failing all the tests ;-). | 17:37 |
| ayoung | shardy, um...getting redirecto login issues...let me try again, but can you fpaste that as well? | 17:38 |
| offenflieg | does anyone know if there is a list of supported s3 api commands for swift3 anywhere? | 17:38 |
| *** nermina has joined #openstack-dev | 17:38 | |
| *** chenxu has joined #openstack-dev | 17:39 | |
| dperaza | bnemec yeah need this guy merged: https://review.openstack.org/#/c/48745 | 17:39 |
| *** sandywalsh has joined #openstack-dev | 17:39 | |
| dperaza | then the theory is they all should pass | 17:39 |
| ayoung | shardy, ah, got it, thanks | 17:39 |
| bnemec | dperaza: Okay, sounds good. | 17:40 |
| ayoung | shardy, that will do, that will do very nicely. dkranz how do we turn this https://gist.github.com/hardys/6435299#file-trust_pw_test_v2-py into a tempest test? | 17:41 |
| *** martines has quit IRC | 17:42 | |
| *** neelashah has quit IRC | 17:42 | |
| bnemec | Lots of angry unicorns at Github today... | 17:42 |
| *** nplanel has joined #openstack-dev | 17:42 | |
| ayoung | bnemec, yeah...where's the Red Bull when you need him? | 17:43 |
| *** sandywalsh has quit IRC | 17:43 | |
| bnemec | ayoung: :-) | 17:43 |
| *** jmontemayor has quit IRC | 17:45 | |
| shardy | ayoung: similar example using curl not keystoneclient http://paste.fedoraproject.org/43444/49537138/ | 17:46 |
| *** Dr_Who has quit IRC | 17:46 | |
| *** amcrn has joined #openstack-dev | 17:47 | |
| dkranz | ayoung: Let me take a look... | 17:47 |
| shardy | dkranz, ayoung: fpaste version of python example since github seems to be having some issues: http://paste.fedoraproject.org/43445/38064971 | 17:49 |
| dkranz | ayoung: I can't load that link. I think there is something wrong with github | 17:49 |
| dkranz | shardy: OK :) | 17:49 |
| dkranz | ayoung: Is this supposed to be an api test or scenario? | 17:50 |
| ayoung | dkranz, keystone client tests cases...I'll fpaste the code | 17:50 |
| *** rkukura has quit IRC | 17:51 | |
| dkranz | ayoung: Oh, ok. | 17:51 |
| *** dstanek_ has joined #openstack-dev | 17:51 | |
| *** chenxu has quit IRC | 17:51 | |
| ayoung | dkranz, shardy http://fpaste.org/43446/80649966/ | 17:53 |
| ayoung | ugly, but you should get the idea. dkranz github is cranky today | 17:53 |
| *** dstanek has quit IRC | 17:53 | |
| *** dstanek_ is now known as dstanek | 17:54 | |
| *** DinaBelova has quit IRC | 17:55 | |
| *** garyk has joined #openstack-dev | 17:56 | |
| shardy | dkranz, ayoung: I can obviously create some cleaner versions of that test code, just need a few pointers on the steps required to create the tempest test (and how to test it locally) | 17:56 |
| *** tvb|afk has joined #openstack-dev | 17:57 | |
| dkranz | shardy: I'm looking at it now and will put some ideas in a paste | 17:57 |
| *** danwent has quit IRC | 17:57 | |
| shardy | dkranz: thanks! | 17:57 |
| dkranz | shardy: Should be straightforward | 17:57 |
| *** chenxu has joined #openstack-dev | 17:58 | |
| *** topol has joined #openstack-dev | 17:58 | |
| ayoung | Thanks, guys.. shardy give me a ping when you have something to show...doesn't have to be perfect, but submit it as a review to Tempest as a WIP | 17:58 |
| shardy | ayoung: Ok, will do, thanks! | 17:59 |
| *** carl_baldwin has joined #openstack-dev | 17:59 | |
| *** Mandell has quit IRC | 17:59 | |
| *** chuckieb has joined #openstack-dev | 17:59 | |
| dkranz | shardy: So this test is a little funky because it is creating lots of clients with different arguments. | 17:59 |
| dkranz | shardy: The place to start is with OfficialClientTest | 18:00 |
| dkranz | shardy: This instantiates many of the python clients and authenticates with the credentials in tempest.conf | 18:01 |
| dkranz | But your subclass can make new clients as well | 18:01 |
| *** jmontemayor has joined #openstack-dev | 18:01 | |
| *** neelashah has joined #openstack-dev | 18:01 | |
| tellesnobrega | can anyone help me understand how the test workflow works, some tests are breaking but i dont understand how it gets to the things i changed | 18:02 |
| *** chuckieb|2 has joined #openstack-dev | 18:02 | |
| *** fandikurnia01 has quit IRC | 18:02 | |
| dkranz | shardy: If this is a prototype for more keystone tests I would put it in a client_lib/keystone dir parallel to scenario and cli | 18:02 |
| *** isd has joined #openstack-dev | 18:03 | |
| shardy | dkranz: Ok, got it, thanks very much for the info! :) | 18:03 |
| *** sarob has joined #openstack-dev | 18:03 | |
| dkranz | shardy: If this moves forward we will want to move that scenario/manager.py file to official_client/manager.py or something | 18:04 |
| *** DinaBelova has joined #openstack-dev | 18:04 | |
| dkranz | shardy: Let me know if you need any other help. | 18:04 |
| shardy | dkranz: are there any docs which describe how to run local tempest tests, prior to submitting them for review? | 18:04 |
| *** chuckieb has quit IRC | 18:05 | |
| dkranz | shardy: There are README files in tempest | 18:05 |
| dkranz | shardy: Are you going to run with local devstack? | 18:05 |
| shardy | dkranz: yes, unless you suggest some other method | 18:06 |
| *** danwent has joined #openstack-dev | 18:06 | |
| dkranz | shardy: No. If you use devstack it installs and configures tempest | 18:06 |
| *** Mandell has joined #openstack-dev | 18:06 | |
| shardy | dkranz: Ok, thanks for all the help, much appreciated | 18:07 |
| dkranz | shardy: THen you can do 'testr run tempest.your.module.here' | 18:07 |
| dkranz | shardy: Or you can use nose | 18:07 |
| shardy | dkranz: awesome, I'll give it a try, thanks! | 18:08 |
| dkranz | shardy: You can also look at tox.ini for more ways to run and with better output if you are using testr | 18:08 |
| dkranz | shardy: Good luck | 18:08 |
| *** networkstatic has quit IRC | 18:08 | |
| isd | Hey everybody, I'm about to try to get my feet wet with the openstack codebase, so I set up devstack and it seems to work - but it isn't clear to me where I'm supposed to find the sources that devstack is using, or more generally how to go about development. Am I missing something obvious? | 18:08 |
| dkranz | shardy: I would like to improve the README if you have any issues. | 18:09 |
| *** shinylasers has joined #openstack-dev | 18:09 | |
| tellesnobrega | dolphm: can you help me understand how the test workflow works, some tests are breaking but i dont understand how it gets to the things i changed | 18:10 |
| *** MaxV has joined #openstack-dev | 18:10 | |
| *** FatDarrel has quit IRC | 18:10 | |
| shardy | dkranz: Ok, np, I'll send some doc patches if I hit anything that's not documented | 18:10 |
| *** gyee has quit IRC | 18:11 | |
| *** ericw has quit IRC | 18:11 | |
| mrodden | bnemec: https://review.openstack.org/#/c/49230/ | 18:12 |
| mrodden | it needs a test case ( think luis is gonna work on that ) and i need to bug out for a bit to grab food since i haven't had a chance to do that yet today... | 18:13 |
| bnemec | mrodden: Bah, eating. :-) | 18:13 |
| bnemec | dhellmann: ^ FYI | 18:13 |
| dhellmann | bnemec: thanks | 18:14 |
| dhellmann | mrodden: I'm taking a look now | 18:14 |
| mrodden | k | 18:14 |
| *** MaxV has quit IRC | 18:14 | |
| *** gyee has joined #openstack-dev | 18:15 | |
| *** chuckieb has joined #openstack-dev | 18:16 | |
| *** alunch_ has joined #openstack-dev | 18:17 | |
| *** alunch has quit IRC | 18:18 | |
| *** alunch_ is now known as alunch | 18:18 | |
| *** davidhadas has quit IRC | 18:20 | |
| *** chuckieb|2 has quit IRC | 18:20 | |
| *** adjohn has joined #openstack-dev | 18:22 | |
| *** chenxu has quit IRC | 18:24 | |
| *** carl_baldwin has quit IRC | 18:26 | |
| *** adjohn has quit IRC | 18:27 | |
| isd | Ah nevermind, found stuff in /opt/stack. | 18:27 |
| *** Lumiere has quit IRC | 18:27 | |
| *** cjellick has quit IRC | 18:28 | |
| *** rnirmal has joined #openstack-dev | 18:31 | |
| *** Lumiere has joined #openstack-dev | 18:31 | |
| *** novas0x2a|laptop has joined #openstack-dev | 18:31 | |
| *** chuckieb has quit IRC | 18:32 | |
| *** carl_baldwin has joined #openstack-dev | 18:34 | |
| *** mjb_ has joined #openstack-dev | 18:38 | |
| *** Ryan_Lane has joined #openstack-dev | 18:40 | |
| *** otherwiseguy has quit IRC | 18:43 | |
| *** DinaBelova has quit IRC | 18:44 | |
| *** jmontemayor has quit IRC | 18:47 | |
| *** neelashah has quit IRC | 18:47 | |
| *** cdub_ has joined #openstack-dev | 18:50 | |
| *** mlavalle has joined #openstack-dev | 18:51 | |
| *** neelashah has joined #openstack-dev | 18:51 | |
| *** avishay has quit IRC | 18:51 | |
| *** jmontemayor has joined #openstack-dev | 18:53 | |
| *** thingee_zzz is now known as thingee | 18:54 | |
| *** egallen has quit IRC | 18:55 | |
| *** Javin has quit IRC | 18:57 | |
| *** derekh has joined #openstack-dev | 18:58 | |
| *** pixelb has quit IRC | 18:59 | |
| ayoung | gyee, jamielennox it just means we use a consistent term. External auth means authentication external to those methods controlled by Keystone itself. REMOTE_USER is the mechanism | 19:00 |
| gyee | jamielennox, ayoung, my patch is backward compatible btw | 19:00 |
| gyee | that admin_token_auth_method is optional | 19:00 |
| jamielennox | gyee: it's not backwards compatibility i'm worried about, it's forwards | 19:01 |
| morganfainberg | henrynash, are we planning on cleaning up the grant tables in Icehouse? | 19:01 |
| jamielennox | gyee: if we can get keystoneclient into auth_token where it belongs i've no idea how to support that | 19:01 |
| gyee | and its forward compatible too | 19:01 |
| jamielennox | gyee: :) | 19:01 |
| henrynash | morganfainberg: yep! | 19:01 |
| morganfainberg | henrynash, if so, i'll be coordinating to get caching enabled after that. | 19:01 |
| morganfainberg | henrynash, awesome. | 19:01 |
| *** nachi1 has joined #openstack-dev | 19:01 | |
| henrynash | have bp arelady assigned to me! | 19:01 |
| *** markwash has quit IRC | 19:01 | |
| ayoung | gyee, INstead of None, make it "external" and I think I can give my blessing, but I am a littel concerned about testing. I'd like to have test cases for it submitted to tempest. We can't put client tests in either keystone or keystone client . | 19:01 |
| morganfainberg | henrynash, nice, more caching! the behind-the-scenes interactions made it hard to cache. | 19:02 |
| thingee | sdague: ping | 19:02 |
| gyee | ayoung, sure, I can change it to external | 19:02 |
| morganfainberg | ok, i'm going to go grab food. back later today. | 19:02 |
| gyee | not sure about the testing part though | 19:02 |
| ayoung | jamielennox, actually, it should probably be "token_request_auth_method" and not "admin..." and then the same option would bne useable from auth_token middleware and from CLI | 19:02 |
| sdague | thingee: pong | 19:02 |
| ayoung | gyee, also drop the admin from it | 19:02 |
| ayoung | it should be the token_request_auth_methdo | 19:03 |
| gyee | ayoung, sure | 19:03 |
| ayoung | jamielennox, does that make sense? | 19:03 |
| gyee | ayoung, service_token_request_auth_method? | 19:03 |
| gyee | its really a service token | 19:03 |
| gyee | technically | 19:03 |
| thingee | sdague: wrt https://review.openstack.org/#/c/43045/2 I was following what nova has done to set a default version from line 81 | 19:03 |
| ayoung | What we are specifying is how to authenticate when a particular client makes a request for a token | 19:03 |
| *** otherwiseguy has joined #openstack-dev | 19:03 | |
| ayoung | gyee, just token_request_auth_method | 19:04 |
| jamielennox | gyee: drop request | 19:04 |
| ayoung | token_auth_method? | 19:04 |
| jamielennox | ayoung: gyee's is better - token_request is confusing | 19:04 |
| gyee | but that's really a service token, used to validate user tokens | 19:04 |
| ayoung | gyee, in this case, yes, but the config option is more general | 19:04 |
| jamielennox | service_token_auth_method | 19:04 |
| *** chenxu has joined #openstack-dev | 19:04 | |
| gyee | jamielennox, we have a winner! | 19:05 |
| ayoung | jamielennox, why not make it general purpose | 19:05 |
| ayoung | gyee, not yet you don't | 19:05 |
| gyee | damn | 19:05 |
| *** melwitt has joined #openstack-dev | 19:05 | |
| jamielennox | mmm, still not a fan of the approach, just giving out a config name | 19:05 |
| jamielennox | ayoung: because you need to distinguish between a validation token and a user token | 19:05 |
| ayoung | jamielennox, we need a way to tell the client which auth method to use when getting a token. | 19:05 |
| jamielennox | and they don't necessarily want to authenticate in the same way | 19:05 |
| sdague | thingee: so if that's not set that way.... how does nova decide what version of cinder client to use? | 19:06 |
| ayoung | jamielennox, no...because auth_token middleware is never going to get any other token | 19:06 |
| gyee | ayoung, you have a point there | 19:06 |
| jamielennox | ayoung: sure, but if you put a config option in auth_token that says token_request_method you are going to confuse people | 19:06 |
| ayoung | jamielennox, but, even if it did, why not use the same method? Unless we do "token for a token" we should specify, generically, "client use auth method X" and we can be consistent with how we specify it | 19:06 |
| ayoung | jamielennox, how about just auth_method? | 19:07 |
| jgriffith | thingee: o/ | 19:07 |
| ayoung | authentication_method, really | 19:07 |
| thingee | sdague: v1 and v2 are available by default in devstack's keystone catalog. nova just needs to import the version they need from the client. | 19:07 |
| thingee | sdague: https://review.openstack.org/#/c/40446/ | 19:07 |
| gyee | ayoung, auth_method then | 19:07 |
| *** chenxu has quit IRC | 19:07 | |
| gyee | short, and sweet | 19:07 |
| jamielennox | ok, auth_method | 19:07 |
| ayoung | gyee, how about authn_method? TO clarify it is for "who are you?" and not "what can you do?" | 19:08 |
| sdague | thingee: ok, I was trying to understand gate implications | 19:08 |
| gyee | ayoung, now you're making me work | 19:08 |
| jamielennox | ayoung: the auth_ prefix is common enough | 19:08 |
| ayoung | gyee, "...make one mistake and support it for the rest of your life." | 19:08 |
| thingee | sdague: gotcha. yeah other projects won't notice unless they specifically use it. | 19:08 |
| jamielennox | authn is technically more correct but i don't think anyone cares | 19:08 |
| ayoung | jamielennox, I know, but auth is genrally understaood to mean both authN and AuthZ. AuthZ is token. We really mean just authN | 19:09 |
| gyee | ayoung, nah, we'll deprecate in one release cycle :) | 19:09 |
| ayoung | I care. | 19:09 |
| sdague | honestly, right now with so many other things in flight I'd rather just let things sit until Havana releases. Can we pick it up again there? | 19:09 |
| mrodden | dhellmann: so i found a bug i think | 19:09 |
| jamielennox | knew i phrased that wrong | 19:09 |
| thingee | sdague: absolutely | 19:09 |
| ayoung | gyee, no, I bet you that if we get this right, we won't have to. | 19:09 |
| mrodden | while re-writing the message adapter to wrap context adapter | 19:09 |
| thingee | sdague: I was just about to recommend that due to me being slow on getting back to ya | 19:09 |
| sdague | thingee: cool, I'll promiss to give it more thought post release | 19:09 |
| sdague | yeh, no worries | 19:09 |
| mrodden | many other functions are looking at ContextAdapters self.logger property and expecting it to be an actual logger | 19:09 |
| gyee | ayoung, jamielennox, I am fine with either auth_method or authn_method | 19:10 |
| mrodden | that doesnt work if i wrap an adapter in another adapter | 19:10 |
| ayoung | you guys cool with authn_method, with an undertanding that we will extend this to be a general purpose config option for the clients? | 19:10 |
| lbragstad | alexpilotti: would you have a minute to discuss https://github.com/openstack/nova/commit/8f28ddd13a91eb25f0f2f3a312068223500ded97#diff-e3f97ae27f34d8632013e768c4bb652cR101 | 19:10 |
| ayoung | extend in Icehouse, that is | 19:10 |
| thingee | sdague: marked wip | 19:10 |
| jamielennox | gyee: ok, from the assumption that I can get keysonteclient into auth_token soon, how do we work with this auth_method? | 19:10 |
| *** davidhadas has joined #openstack-dev | 19:10 | |
| gyee | ayoung, + | 19:10 |
| gyee | jamielennox, load the auth plugin base on method name | 19:11 |
| gyee | same way we does it on the server side | 19:11 |
| alexpilotti | lbragstad: sude | 19:11 |
| jamielennox | i can see if we have it set to 'password' then i need to look for user/pass from config and use that | 19:11 |
| alexpilotti | aka sure :-) | 19:11 |
| *** DinaBelova has joined #openstack-dev | 19:11 | |
| gyee | jamielennox, just password the conf object into the plugin | 19:12 |
| jamielennox | gyee: ok - in which case 'external' is out | 19:12 |
| gyee | plugin will look for the stuff it needs | 19:12 |
| lbragstad | alexpilotti: awesome! I just noticed that a check was added in the case of a compute node not having the image being launch, so it reads the headers of the file to determine the file type and then based on that returns the extention | 19:12 |
| lbragstad | launched* | 19:12 |
| *** martines has joined #openstack-dev | 19:13 | |
| *** galstrom is now known as galstrom_zzz | 19:13 | |
| *** briancurtin has quit IRC | 19:13 | |
| mrodden | bnemec: around? | 19:13 |
| lbragstad | so this now requires hyper-v to only use images with vhdxfile or conectix as the header of the image file | 19:13 |
| bnemec | mrodden: Yes | 19:14 |
| jamielennox | gyee: actually whatever - i can make external work as a plugin, it's just a noop plugin i just need to change my plugin architecture a little | 19:14 |
| mrodden | bnemec: running into issues with having one adapter wrap another | 19:14 |
| mrodden | bnemec: seems that some of the code in log.py is looking at the self.logger attribute of the adapter | 19:15 |
| jamielennox | gyee: fine, auth_method = 'plugin' or 'external' and extend it later, i'll support it | 19:15 |
| mrodden | and expecting it to be an actual logger | 19:15 |
| *** briancurtin has joined #openstack-dev | 19:15 | |
| mrodden | i think i might have accidentally trained my brain to write IRC messages less than 80 chars now | 19:15 |
| mrodden | :( | 19:15 |
| bnemec | Heh | 19:15 |
| *** troytoman is now known as troytoman-away | 19:15 | |
| bnemec | So this is an issue regardless of which adapter is wrapping the other? | 19:16 |
| datsun180b | there are worse vices | 19:16 |
| *** sgordon_ has quit IRC | 19:16 | |
| gyee | jamielennox, cool, will make the changes | 19:16 |
| mrodden | bnemec: yeah | 19:16 |
| *** dvarga has quit IRC | 19:16 | |
| jamielennox | gyee: technically we should have a 'token' option to which is what is used when you provide an auth_token | 19:16 |
| mrodden | bnemec: i *could* change the faulty code, but i don't know if that is the only code expecting the first inner .logger attribute to be a real logger | 19:16 |
| *** eglynn has joined #openstack-dev | 19:16 | |
| mrodden | so back to the inheritance thing | 19:17 |
| jamielennox | gyee: ah, admin_token | 19:17 |
| lbragstad | alexpilotti: I have a .vhd that is returning something different when reading the beginning of the image file and I am hitting the HyperVException(_('Unsupported virtual disk format')). The image is originally transferred to the compute node becuase it is not found in the cache, but it is removed shortly after since the check fails, and Hyper-V cleans up the image. https://github.com/openstack/nova/blob/master/nova/virt/hyperv/vhdutils.py#L182 | 19:17 |
| gyee | right | 19:17 |
| mrodden | i reversed the inheritance, because it seems to make more sense tohave MessageAdapter extend ContextAdapter | 19:17 |
| mrodden | and just wrap ContextAdapter.process in MessageAdapter.process | 19:17 |
| alexpilotti | lbragstad: interesting | 19:17 |
| mrodden | busy channel today | 19:18 |
| alexpilotti | lbragstad: can you send me what you have as a header? | 19:18 |
| bnemec | mrodden: Oh, so there's an alternative? | 19:18 |
| lbragstad | alexpilotti: yep | 19:18 |
| bnemec | I seem to have missed some discussion here. | 19:18 |
| lbragstad | let me grab a pastebin | 19:18 |
| alexpilotti | lbragstad: how did you generate the file? | 19:18 |
| lbragstad | genereate the .vhd? | 19:18 |
| mrodden | bnemec: did you see dhellmann's comment about not doing inheritance | 19:18 |
| mrodden | ? | 19:18 |
| mrodden | that works | 19:18 |
| alexpilotti | lbragstad: yes | 19:18 |
| mrodden | its just not... pretty | 19:18 |
| bnemec | I must not have. | 19:18 |
| mrodden | it makes sense IMO | 19:19 |
| mrodden | https://review.openstack.org/#/c/49230/1/openstack/common/log.py | 19:19 |
| mrodden | bnemec: ^ | 19:19 |
| *** locke1051 has joined #openstack-dev | 19:19 | |
| mrodden | well that way was kind of wonky | 19:19 |
| mrodden | MessageAdapter inherits from ContextAdapter now | 19:19 |
| lbragstad | alexpilotti: it was an existing .vhd that was being used, some rhel vhd I believ e | 19:19 |
| *** rkukura has joined #openstack-dev | 19:19 | |
| *** nermina has quit IRC | 19:19 | |
| ayoung | dripton, question for you. I need to make a call to a third party service upon nova servce.create call...prior to actually creating a VM. What is hte mechanism I need? | 19:19 |
| mrodden | i'll push up what i have | 19:20 |
| ayoung | I don't want this to be user-data | 19:20 |
| dripton | ayoung: I'm not sure offhand. Let me look | 19:20 |
| *** VeggieMeat_ has joined #openstack-dev | 19:20 | |
| alexpilotti | lbragstad: ok, with the first 8 bytes "magic number" I should be able to do some research | 19:20 |
| ayoung | as I need to be able to do it from the Webui without user generating a custom user-data doc | 19:20 |
| lbragstad | alexpilotti: so, compute nodes that have the image cached can launch it if the scheduler assigns to those compute nodes. For other nodes however, it copies and then fails the detection | 19:20 |
| *** rkukura has left #openstack-dev | 19:20 | |
| ayoung | dripton, thanks. dansmith ^^ same question to you in case you know the answer | 19:21 |
| alexpilotti | lbragstad: the check is performed when the image is downloaded from Glance | 19:21 |
| alexpilotti | lbragstad: so if you had it cached before upgrading the Nova code | 19:21 |
| *** annegentle_ has joined #openstack-dev | 19:21 | |
| lbragstad | alexpilotti: yep, right after the image.fetch | 19:21 |
| bnemec | mrodden: Yeah, I suppose the alternative would be to write a third Adapter that basically just calls the other two, but I don't know that there's any benefit to doing it that way over inheritance. | 19:21 |
| *** jyhc_ has joined #openstack-dev | 19:22 | |
| alexpilotti | lbragstad: it'll not check it again | 19:22 |
| mrodden | bnemec: yeah | 19:22 |
| *** jeevan_u1las has joined #openstack-dev | 19:22 | |
| dansmith | ayoung: not sure what you mean.. when we call to glance or cinder we use their client.. more context? | 19:22 |
| *** jeblair_ has joined #openstack-dev | 19:22 | |
| *** iamben_t1 has joined #openstack-dev | 19:22 | |
| *** ivoks_ has joined #openstack-dev | 19:22 | |
| *** jrwren_ has joined #openstack-dev | 19:22 | |
| bnemec | mrodden: Other than maybe code reuse, I suppose. | 19:22 |
| *** sc68cal_ has joined #openstack-dev | 19:22 | |
| bnemec | That way both classes would be usable without the other. | 19:22 |
| *** pmcg_ has joined #openstack-dev | 19:22 | |
| *** fbo has joined #openstack-dev | 19:23 | |
| bnemec | Given that context is pretty specific to OpenStack, I'm not sure that's a major consideration though. | 19:23 |
| lbragstad | alexpilotti: right, because we only grab the image from glance if the vhd_path is none, https://github.com/openstack/nova/blob/master/nova/virt/hyperv/imagecache.py#L119 | 19:23 |
| lbragstad | alexpilotti: http://paste.openstack.org/show/47785/ | 19:23 |
| *** bcwaldon_ has joined #openstack-dev | 19:24 | |
| *** wink_ has joined #openstack-dev | 19:24 | |
| alexpilotti | lbragstad: unfortunately we need a way to be sure that people don't try to boot invalid images | 19:24 |
| *** gyee has quit IRC | 19:24 | |
| *** jeblair has quit IRC | 19:24 | |
| *** DinaBelova has quit IRC | 19:24 | |
| alexpilotti | lbragstad: you can imagine how many people ask for support just to discover that they tray to boot Qcow2 or AMI images ;-) | 19:24 |
| *** rgoodwin_ has joined #openstack-dev | 19:24 | |
| alexpilotti | lbragstad: so we need a way to validate them before | 19:24 |
| *** jeblair_ is now known as jeblair | 19:24 | |
| *** vladan- has joined #openstack-dev | 19:24 | |
| dripton | ayoung: almost all of the service.create calls in nova are for glance. I don't see any general way to hook them for a third-party service. | 19:25 |
| *** GheRiver1 has joined #openstack-dev | 19:25 | |
| alexpilotti | lbragstad: if the magic number is not consistent, we need obviously to resort to another "heuristic" | 19:25 |
| *** sgordon has joined #openstack-dev | 19:25 | |
| *** ttx` has joined #openstack-dev | 19:25 | |
| *** sgordon has quit IRC | 19:25 | |
| *** sgordon has joined #openstack-dev | 19:25 | |
| *** sgran_ has joined #openstack-dev | 19:26 | |
| alexpilotti | lbragstad: so far all the VHDs that we tested were fine, including using qemu-img convert | 19:26 |
| *** CJM has joined #openstack-dev | 19:26 | |
| *** adam_g` has joined #openstack-dev | 19:26 | |
| *** CJM is now known as Guest62309 | 19:26 | |
| *** rosmaita_ has joined #openstack-dev | 19:26 | |
| ayoung | dansmith, we want to register the host with FreeIPA before booting the VM. | 19:26 |
| mrodden | bnemec: yeah its still re-usable, | 19:26 |
| lbragstad | alexpilotti: I tried doing the same thing to the original image before uploading it to glance and the result is the same. So I am going to need new images? | 19:27 |
| *** dabo has joined #openstack-dev | 19:27 | |
| mrodden | and even if it needs refactoring later, thats something for later | 19:27 |
| *** zaitcev has quit IRC | 19:27 | |
| *** sc68cal has quit IRC | 19:27 | |
| *** mmoya has quit IRC | 19:27 | |
| *** locke105 has quit IRC | 19:27 | |
| *** wolfdreamer has quit IRC | 19:27 | |
| *** adam_g has quit IRC | 19:27 | |
| *** annegentle has quit IRC | 19:27 | |
| *** phschwartz has quit IRC | 19:27 | |
| *** edleafe has quit IRC | 19:27 | |
| *** GheRivero has quit IRC | 19:27 | |
| *** oubiwann has quit IRC | 19:27 | |
| *** bcwaldon has quit IRC | 19:27 | |
| *** rosmaita has quit IRC | 19:27 | |
| *** sgran has quit IRC | 19:27 | |
| *** galstrom_zzz has quit IRC | 19:27 | |
| *** iamben_tw has quit IRC | 19:27 | |
| *** troytoman-away has quit IRC | 19:27 | |
| *** pmcg has quit IRC | 19:27 | |
| *** rods` has quit IRC | 19:27 | |
| *** wink has quit IRC | 19:27 | |
| *** jdurgin1 has quit IRC | 19:27 | |
| *** markvoelker has quit IRC | 19:27 | |
| *** xBytez has quit IRC | 19:27 | |
| *** rackerjoe has quit IRC | 19:27 | |
| *** rgoodwin has quit IRC | 19:27 | |
| *** fbo_away has quit IRC | 19:27 | |
| *** dtroyer has quit IRC | 19:27 | |
| *** jkoelker has quit IRC | 19:27 | |
| *** sudorandom has quit IRC | 19:27 | |
| *** CrackerJackMack has quit IRC | 19:27 | |
| *** EmilienM has quit IRC | 19:27 | |
| *** vegarl has quit IRC | 19:27 | |
| *** Anticimex has quit IRC | 19:27 | |
| *** jrwren has quit IRC | 19:27 | |
| *** Guest58099 has quit IRC | 19:27 | |
| *** VeggieMeat has quit IRC | 19:27 | |
| *** jeevan_ullas has quit IRC | 19:27 | |
| *** ttx has quit IRC | 19:27 | |
| *** Adri2000 has quit IRC | 19:27 | |
| *** vladan has quit IRC | 19:27 | |
| *** jyhc has quit IRC | 19:27 | |
| *** ondergetekende has quit IRC | 19:27 | |
| *** ivoks has quit IRC | 19:27 | |
| *** vladan- is now known as vladan | 19:27 | |
| alexpilotti | lbragstad: head -c 8 yourdisk.vhd should do the trick to get the value | 19:27 |
| ayoung | dripton, dansmith I can do it from the command line http://adam.younglogic.com/2013/09/register-vm-freeipa/ | 19:27 |
| *** rgoodwin_ is now known as rgoodwin | 19:27 | |
| alexpilotti | 19:27 | |
| *** phschwartz has joined #openstack-dev | 19:27 | |
| *** davidhadas has quit IRC | 19:27 | |
| dansmith | ayoung: does that really belong in nova? seems like not, to me | 19:27 |
| alexpilotti | lbragstad: a dirty trick is to simply replace the first 8 bytes ;-) | 19:27 |
| *** DinaBelova has joined #openstack-dev | 19:28 | |
| *** davidhadas has joined #openstack-dev | 19:28 | |
| ayoung | dripton, dansmith it is custom integration, but, yes, it need to be in Nova | 19:28 |
| *** ttx` is now known as ttx | 19:28 | |
| *** adam_g` is now known as adam_g | 19:28 | |
| alexpilotti | lbragstad: as soon as I see what you have I'm taking a look | 19:28 |
| bnemec | mrodden: Yeah, works for me. We'll see what dhellmann thinks I guess. | 19:28 |
| *** ttx has quit IRC | 19:28 | |
| *** ttx has joined #openstack-dev | 19:28 | |
| mrodden | k | 19:28 |
| *** rackerjoe has joined #openstack-dev | 19:28 | |
| dhellmann | I'm looking now, but also have a local conversation going on so it may be a few minutes | 19:28 |
| *** oubiwann has joined #openstack-dev | 19:28 | |
| mrodden | k | 19:28 |
| *** galstrom_zzz has joined #openstack-dev | 19:28 | |
| *** galstrom_zzz is now known as galstrom | 19:28 | |
| *** oubiwann is now known as Guest49600 | 19:29 | |
| mrodden | still needs a test case too... which i'm working on | 19:29 |
| *** nachi1 has left #openstack-dev | 19:29 | |
| *** sudorandom has joined #openstack-dev | 19:29 | |
| *** leif has joined #openstack-dev | 19:29 | |
| *** nermina has joined #openstack-dev | 19:29 | |
| ayoung | dansmith, it is an integration point, not something everyone one would need, but for people that are Using IPA, they need to autoregister the hosts. You can do it with user-data, but that requires end user particiapltion, not automated | 19:29 |
| *** leif is now known as Guest44600 | 19:29 | |
| *** rods` has joined #openstack-dev | 19:29 | |
| ayoung | dansmith, it belongs in nova to the same degree that putting an SSH key on the vm belongs in nova | 19:29 |
| ayoung | FreeIPA gets you kerberos and SSO | 19:29 |
| dansmith | ayoung: end user, or the thing automating your deployments? | 19:29 |
| dansmith | ayoung: are you injecting something into the instance that you get from FreeIPA? | 19:30 |
| *** EmilienM has joined #openstack-dev | 19:30 | |
| ayoung | dansmith, well, I would want it kicked off when the user created a new VM either via the command line or Horizon | 19:30 |
| lbragstad | head -c 8 images/RHEL-6.4-X86-64.vhd -> �H��м | 19:30 |
| ayoung | dansmith, freeipa is not creating instnaces | 19:30 |
| *** rkukura has joined #openstack-dev | 19:30 | |
| ayoung | You create the entriies in LDAP in FreeIPA | 19:30 |
| lbragstad | alexpilotti: ^^ | 19:30 |
| ayoung | and get an OTP for registring the VMs | 19:30 |
| ayoung | one time password | 19:30 |
| alexpilotti | lbragstad: hmmm | 19:31 |
| ayoung | so, at that point, yes, you are injecting something into the VM | 19:31 |
| alexpilotti | lbragstad: it definitely does not look right :-) | 19:31 |
| *** Guest62309 is now known as crackerjackmack | 19:31 | |
| lbragstad | alexpilotti: lol yeah I didn't think so either | 19:31 |
| *** dtroyer has joined #openstack-dev | 19:31 | |
| dansmith | ayoung: what specifically gets injected into the instance? | 19:31 |
| alexpilotti | lbragstad: I suppose that you have access to a machine with Hyper-V | 19:31 |
| lbragstad | alexpilotti: yes sir | 19:31 |
| ayoung | dansmith, the One time password...short phrase used to confirm the registration of the instance | 19:32 |
| alexpilotti | lbragstad: can you issue in Powershell: Get-VHD RHEL-6.4-X86-64.vhd | 19:32 |
| alexpilotti | ? | 19:32 |
| lbragstad | alexpilotti: yep, let me hop on quick | 19:32 |
| alexpilotti | lbragstad: this will return details about the image | 19:32 |
| alexpilotti | lbragstad: or an error if it's not valid | 19:32 |
| ayoung | so create a host in FreeIPA with the option that saysd "freeIPA generates the OTP" then take that OTP and pass it to the instance which calls ipa-client-install with the OTP as a parameter, | 19:32 |
| dansmith | ayoung: well, you could propose an integration blueprint if you want, but IMHO, that belongs a layer up | 19:33 |
| *** Javin has joined #openstack-dev | 19:33 | |
| ayoung | dansmith, other Identity systems are going to have to do something comparable. So, while I am solving a specific issue for FreeIPA integration, it will be a general approach for other integration | 19:33 |
| ayoung | dansmith, I thought there might be something already | 19:33 |
| dansmith | ayoung: not that I know of | 19:33 |
| *** FatDarrel has joined #openstack-dev | 19:33 | |
| dansmith | ayoung: other than file injection | 19:33 |
| ayoung | dansmith, yeah, this is something that would have to happen before File injection. You want a trusted user to create the host record and pass off the OTP to the instance. | 19:34 |
| ayoung | only the OTP could be handled via file injection | 19:34 |
| *** ondergetekende has joined #openstack-dev | 19:34 | |
| dansmith | ayoung: before file injection? I'm confused | 19:35 |
| *** vipul is now known as vipul-away | 19:35 | |
| *** davidhadas has quit IRC | 19:35 | |
| *** troytoman-away has joined #openstack-dev | 19:36 | |
| *** jrwren_ has quit IRC | 19:36 | |
| ayoung | dripton, , where would a hook go in Nova to perform an operation comparable to the glance calls? | 19:36 |
| *** dstanek has quit IRC | 19:36 | |
| *** davidhadas has joined #openstack-dev | 19:36 | |
| dhellmann | mrodden: replied on the review | 19:37 |
| *** Ryan_Lane has quit IRC | 19:37 | |
| *** sc68cal_ has quit IRC | 19:37 | |
| *** shinylasers has quit IRC | 19:37 | |
| ayoung | dansmith, yeah, I need to create the file before injecting it into the instance. The file is going to be specific to the instance | 19:37 |
| ayoung | different for every vm, | 19:37 |
| dansmith | ayoung: ...right | 19:37 |
| dripton | ayoung: I don't see an obvious place. We have nova.conductor.service_create but it's basically just a shim around nova.db.service_create, not really a good place to add more code. | 19:37 |
| dripton | ayoung: I think we might want another API call rather than trying to shoehorn it in. | 19:38 |
| dansmith | dripton: service_create has nothing to do with instance creation, which is what he's concerned with, right? | 19:38 |
| mrodden | dhellmann: yeah i suppose that works | 19:38 |
| ayoung | yes, it is instance creation | 19:38 |
| *** FatDarrel_ has joined #openstack-dev | 19:38 | |
| *** FatDarrel has quit IRC | 19:38 | |
| *** FatDarrel_ is now known as FatDarrel | 19:38 | |
| dhellmann | mrodden: you could also make a "multi-adapter" class that gets one of each of the others, but that feels a little over-engineered at this point | 19:39 |
| dripton | ayoung dansmith: you want to register something that will run at later instance creation right? Not attach it to an individual instance creation because then it's "too late"? | 19:39 |
| *** metral has quit IRC | 19:39 | |
| ayoung | dripton, actually, run it before instance creation | 19:39 |
| dripton | ayoung: right s/at/just before/ | 19:40 |
| ayoung | dripton, the order is: ipa.host_create, nova.service.create | 19:40 |
| ayoung | er make that | 19:40 |
| ayoung | nova.servers.create | 19:40 |
| dansmith | (which is why it should be done by something else, IMHO) | 19:40 |
| *** yolanda has quit IRC | 19:40 | |
| dansmith | well, one of the reasons | 19:40 |
| ayoung | what is @hooks.add_hook("create_instance") | 19:41 |
| dansmith | it registers a hook so you can run custom code before or after that | 19:41 |
| ayoung | dansmith, that sounds like what I am looking for | 19:41 |
| dansmith | ayoung: for wedging in some really fragile out-of-tree code? yes :) | 19:42 |
| *** jrwren has joined #openstack-dev | 19:42 | |
| *** sc68cal has joined #openstack-dev | 19:42 | |
| ayoung | dansmith, that is exactly what I am trying to do. | 19:42 |
| dripton | ayoung: what if it fails? | 19:42 |
| ayoung | at least, for Proof-of-concept | 19:43 |
| ayoung | dripton, the VM doesn't get regiwstered, and we can tear it down | 19:43 |
| ayoung | dansmith, out-of-tree, certainly | 19:43 |
| *** metral has joined #openstack-dev | 19:43 | |
| ayoung | hopefully not-too-fragile | 19:43 |
| *** metral has quit IRC | 19:43 | |
| dansmith | ayoung: next time you want help with something horrific, prefix with "out of tree" and I'll be much more helpful :P | 19:44 |
| ayoung | dansmith, why is that horrific? What side effects should I be aware of? | 19:45 |
| *** davidhadas has quit IRC | 19:45 | |
| *** davidhadas has joined #openstack-dev | 19:45 | |
| dansmith | ayoung: I should have s/horrific/proof-of-concept/, but I meant "jamming in an external service integration to the create process", which is what it sounds like you're looking to do, at least at the moment | 19:46 |
| ayoung | dansmith, yeah, but people need this kind of integration in their deployments. I'm trying to figure out how to do it right..ad also how to do it right now if r"right" ain't available. | 19:49 |
| *** metral has joined #openstack-dev | 19:49 | |
| ayoung | dansmith, so, if I called a relatively slow process in that hook, would it tie up the whole API server? | 19:49 |
| dansmith | ayoung: heh | 19:50 |
| jamielennox | ayoung: when you have a moment can you approve: https://review.openstack.org/#/c/44843/ it's basic and its up to 7 +1s | 19:50 |
| *** alunch has quit IRC | 19:51 | |
| dansmith | ayoung: it would hang up one of the workers, yeah | 19:51 |
| *** alunch has joined #openstack-dev | 19:52 | |
| *** Mandell has quit IRC | 19:53 | |
| *** emagana has quit IRC | 19:54 | |
| ayoung | jamielennox, lookinfg now | 19:54 |
| *** Mandell has joined #openstack-dev | 19:54 | |
| ayoung | dansmith, so, how should long running tasks that need to happen in sequence be handled? | 19:55 |
| dansmith | ayoung: in compute/manager | 19:55 |
| ayoung | jamielennox, done | 19:56 |
| ayoung | dansmith, ok let me look | 19:56 |
| *** boris-42 has quit IRC | 19:58 | |
| *** dstanek has joined #openstack-dev | 19:58 | |
| *** boris-42 has joined #openstack-dev | 19:59 | |
| *** drewlander has quit IRC | 19:59 | |
| *** tellesnobrega has quit IRC | 19:59 | |
| ekarlso | dolphm: / ayoung is there any guide for doing AD and keystone ? | 20:00 |
| mrodden | dhellmann: so i see some problems with this approach again... | 20:00 |
| *** stevemar has joined #openstack-dev | 20:00 | |
| mrodden | ContextAdapter is a public API that many methods expect to be available | 20:01 |
| mrodden | so it either has to be the outer Adapter | 20:01 |
| *** markwash has joined #openstack-dev | 20:01 | |
| ayoung | ekarlso, you mean beyond the LDAP guide? | 20:01 |
| *** vkmc has joined #openstack-dev | 20:01 | |
| *** vkmc has quit IRC | 20:01 | |
| *** vkmc has joined #openstack-dev | 20:01 | |
| ekarlso | ayoung: where's that ? :) | 20:01 |
| mrodden | or i can extend it so everything falls through to ContextAdapter when my MessageAdapter doesn't provide all that API | 20:01 |
| dolphm | ekarlso: this is certainly out of date, but worked once upon a time https://gist.github.com/dolph/5489485 | 20:01 |
| *** DinaBelova has quit IRC | 20:01 | |
| *** venkatesh has joined #openstack-dev | 20:02 | |
| *** thingee is now known as thingee_zzz | 20:02 | |
| ayoung | ekarlso, in the docs subdir of keystone there is configuration.rst. | 20:02 |
| ayoung | There is a section in there about LDAP | 20:02 |
| *** gabrielhurley has joined #openstack-dev | 20:02 | |
| ekarlso | what information is stored in LDAP | 20:03 |
| ayoung | https://github.com/openstack/keystone/blob/master/doc/source/configuration.rst | 20:03 |
| ayoung | ekarlso, https://github.com/openstack/keystone/blob/master/doc/source/configuration.rst#configuring-the-ldap-identity-provider | 20:03 |
| *** eglynn has quit IRC | 20:03 | |
| *** otherwiseguy has quit IRC | 20:04 | |
| dhellmann | mrodden: :-| | 20:04 |
| dhellmann | mrodden: what is using ContextAdapter directly? | 20:04 |
| *** MaxV has joined #openstack-dev | 20:04 | |
| *** vegarl has joined #openstack-dev | 20:05 | |
| *** jdurgin1 has joined #openstack-dev | 20:05 | |
| *** Anticimex has joined #openstack-dev | 20:05 | |
| mrodden | dhellmann: everything from what i can tell | 20:05 |
| mrodden | its literally a front end that gets handed out with getLogger | 20:05 |
| *** jvrbanac has quit IRC | 20:05 | |
| dhellmann | but everything should just be treating it like a logger, right? | 20:05 |
| mrodden | theoretically yes | 20:05 |
| *** vipul-away is now known as vipul | 20:05 | |
| dhellmann | mrodden: can you link me to an example use that's causing concern? | 20:06 |
| *** tanisdl has joined #openstack-dev | 20:06 | |
| mrodden | the deprecated method | 20:06 |
| *** enikanorov-w has quit IRC | 20:06 | |
| mrodden | dhellmann: https://github.com/openstack/nova/blob/master/nova/openstack/common/log.py#L244 | 20:07 |
| mrodden | only had the nova one handy | 20:07 |
| dhellmann | ah | 20:07 |
| *** derekh has quit IRC | 20:07 | |
| mrodden | there is also the handlers property | 20:07 |
| dhellmann | so we've added new public methods that we pretend are part of the logger | 20:07 |
| mrodden | yeah | 20:07 |
| dhellmann | ok | 20:08 |
| *** enikanorov-w has joined #openstack-dev | 20:08 | |
| mrodden | so that must be the outer adapter or i need to extend it | 20:08 |
| *** rnirmal has quit IRC | 20:08 | |
| mrodden | outer adapter would require me to change the process function | 20:08 |
| mrodden | which i dont really like that idea | 20:08 |
| mrodden | trying not to touch code that works | 20:08 |
| *** wolfdreamer has joined #openstack-dev | 20:08 | |
| mrodden | is known to work | 20:08 |
| dhellmann | mrodden: agreed | 20:09 |
| * dhellmann thinks | 20:09 | |
| dhellmann | ok, it seems reasonable to me to have ContextAdapter do something like if not isinstance(msg, basestring): msg = unicode(msg) | 20:10 |
| mrodden | ok | 20:10 |
| mrodden | so just add it there? | 20:10 |
| dhellmann | that does 2 things: it doesn't tie ContextAdapter to Message, but it ensures that we're always going to pass unicode down into the logging module | 20:10 |
| dhellmann | yeah | 20:10 |
| * bnemec catches up | 20:10 | |
| dhellmann | with a big comment explaining why :-) | 20:10 |
| mrodden | yeah lol | 20:10 |
| mrodden | extending would probably still be my preferred option | 20:10 |
| mrodden | since we are only replacing the process function with a wrapper essentially | 20:11 |
| dhellmann | mrodden: adding to the class hierarchy just feels like the "hard way" in this case | 20:11 |
| mrodden | yeah | 20:11 |
| mrodden | there's already a couple layers in that file already | 20:11 |
| dhellmann | fwiw, it looks like that deprecated() method should go in BaseLoggerAdapter | 20:12 |
| mrodden | dhellmann: yeah i think a lot of that could be moved up | 20:13 |
| dhellmann | if you did that, it would be safe to use composition instead of inheritance | 20:13 |
| dhellmann | yeah, the handlers() method, too | 20:13 |
| dhellmann | does anything else subclass BaseLoggerAdapter, though? | 20:14 |
| mrodden | lazylogger | 20:14 |
| mrodden | sorry | 20:14 |
| mrodden | lazyadapter | 20:14 |
| *** carl_baldwin has quit IRC | 20:14 | |
| dhellmann | there's no docstring on that class, what is it for? | 20:14 |
| mrodden | its passed back in getLazyLogger | 20:14 |
| dhellmann | and would it be a mistake for it to have those methods that we're talking about moving | 20:15 |
| mrodden | which then i think dynamically swaps out for contextadapter | 20:15 |
| dhellmann | oh, that's the thing where we don't create the loggers until they are being used | 20:16 |
| mrodden | yeah | 20:16 |
| dhellmann | ok, so that shouldn't cause any trouble | 20:16 |
| *** eharney has quit IRC | 20:16 | |
| mrodden | how about i fix the problem at hand first | 20:16 |
| mrodden | and we refactor later | 20:16 |
| mrodden | seems like that needs to happen anyways | 20:17 |
| dhellmann | mrodden: fair enough; let's just do the 2 line change in place, then | 20:17 |
| *** carl_baldwin has joined #openstack-dev | 20:18 | |
| *** nachi has joined #openstack-dev | 20:19 | |
| mrodden | yep | 20:19 |
| mrodden | got tests too... | 20:19 |
| *** stevemar has quit IRC | 20:19 | |
| *** dmakogon_ has joined #openstack-dev | 20:20 | |
| dhellmann | mrodden: cool | 20:20 |
| *** johnthetubaguy has quit IRC | 20:20 | |
| *** otherwiseguy has joined #openstack-dev | 20:22 | |
| *** jkoelker has joined #openstack-dev | 20:24 | |
| *** vipul is now known as vipul-away | 20:24 | |
| *** vipul-away is now known as vipul | 20:25 | |
| *** f13o_ has joined #openstack-dev | 20:25 | |
| *** devoid has left #openstack-dev | 20:26 | |
| *** boden has quit IRC | 20:27 | |
| *** tmclaugh[work] has quit IRC | 20:27 | |
| *** angdraug has quit IRC | 20:28 | |
| *** dolphm has quit IRC | 20:28 | |
| *** angdraug has joined #openstack-dev | 20:29 | |
| *** tmclaugh[work] has joined #openstack-dev | 20:29 | |
| *** tvb|afk has quit IRC | 20:31 | |
| *** sarob has quit IRC | 20:31 | |
| *** sarob has joined #openstack-dev | 20:32 | |
| *** flaper87|afk is now known as flaper87 | 20:32 | |
| *** angdraug has quit IRC | 20:36 | |
| *** gimps_ has quit IRC | 20:36 | |
| *** sarob has quit IRC | 20:36 | |
| *** mjb_ has quit IRC | 20:37 | |
| *** Ryan_Lane has joined #openstack-dev | 20:38 | |
| *** emagana has joined #openstack-dev | 20:40 | |
| *** thomasm has quit IRC | 20:40 | |
| *** venkatesh has quit IRC | 20:44 | |
| *** neoXsys has quit IRC | 20:44 | |
| *** angdraug has joined #openstack-dev | 20:45 | |
| *** thingee_zzz is now known as thingee | 20:47 | |
| *** cjellick has joined #openstack-dev | 20:47 | |
| bigmstone | BobBall: working with Havana now. Same error message as with grizzly. | 20:48 |
| *** gyee has joined #openstack-dev | 20:50 | |
| *** markvoelker has joined #openstack-dev | 20:51 | |
| *** dabo has quit IRC | 20:51 | |
| *** angdraug has quit IRC | 20:51 | |
| *** sarob has joined #openstack-dev | 20:52 | |
| *** eharney has joined #openstack-dev | 20:53 | |
| *** edleafe has joined #openstack-dev | 20:54 | |
| *** demontiesantos has quit IRC | 20:55 | |
| *** termie has joined #openstack-dev | 20:55 | |
| *** termie has quit IRC | 20:55 | |
| *** termie has joined #openstack-dev | 20:55 | |
| mrodden | dhellmann, bnemec: got another patchset ready https://review.openstack.org/#/c/49230/3 | 20:56 |
| *** Ryan_Lane has quit IRC | 20:56 | |
| *** angdraug has joined #openstack-dev | 20:57 | |
| *** donaldh has joined #openstack-dev | 20:59 | |
| *** thomasm has joined #openstack-dev | 21:02 | |
| *** harlowja has quit IRC | 21:02 | |
| *** eglynn has joined #openstack-dev | 21:03 | |
| *** darren0 has left #openstack-dev | 21:05 | |
| *** rfolco has quit IRC | 21:06 | |
| *** dolphm has joined #openstack-dev | 21:06 | |
| *** nermina has quit IRC | 21:08 | |
| *** gyee has quit IRC | 21:08 | |
| *** gyee has joined #openstack-dev | 21:11 | |
| *** sgran_ is now known as sgran | 21:11 | |
| *** eglynn has quit IRC | 21:12 | |
| *** sushils has quit IRC | 21:12 | |
| *** nermina has joined #openstack-dev | 21:13 | |
| *** martyntaylor has quit IRC | 21:16 | |
| *** egallen has joined #openstack-dev | 21:16 | |
| dhellmann | mrodden, bnemec: are you going to be online much longer? | 21:19 |
| mrodden | probably until the issue gets resolved... | 21:19 |
| bnemec | dhellmann: I should be around for a while | 21:19 |
| *** jcoufal has joined #openstack-dev | 21:19 | |
| *** vladikr has quit IRC | 21:19 | |
| dhellmann | ttx would like us to start patching the other projects to disable the lazy translation stuff asap | 21:19 |
| dhellmann | I'm about to get in the car for a rather long commute, so I was hoping you would have time to help out | 21:20 |
| *** jbresnah has joined #openstack-dev | 21:20 | |
| ttx | dhellmann: or find volunteers to do that for you :) | 21:21 |
| dims | dhellmann, mrodden - what needs to be done? i can pitch in | 21:21 |
| bnemec | dhellmann: Yeah, I can help with that. | 21:21 |
| mrodden | its a piece of cake really | 21:21 |
| mrodden | lazy=False | 21:21 |
| mrodden | we made it easy to enable/disable | 21:21 |
| *** jpich has quit IRC | 21:22 | |
| dhellmann | we also need to remove calls to gettestutils.enable_lazy() | 21:22 |
| *** neelashah has quit IRC | 21:22 | |
| dhellmann | markmc found a few of those this morning | 21:22 |
| dims | bnemec, mrodden - you know where to find me if you guys need backup :) | 21:22 |
| mrodden | yeah | 21:22 |
| *** venkatesh has joined #openstack-dev | 21:22 | |
| dhellmann | dims, cool, thanks | 21:22 |
| bnemec | Thanks dims | 21:22 |
| dhellmann | let's start with an etherpad to list all of them, so we don't miss any | 21:22 |
| mrodden | k | 21:22 |
| dhellmann | then we can tag team through the list | 21:22 |
| mrodden | who's working on that? | 21:23 |
| ttx | dhellmann: jd__ willdo the ceilo one with your guidance | 21:23 |
| dhellmann | ttx, ok good | 21:23 |
| dhellmann | mrodden: I'll create the etherpad, just a sec | 21:24 |
| mrodden | k | 21:24 |
| *** zaitcev has joined #openstack-dev | 21:24 | |
| dhellmann | https://etherpad.openstack.org/disable-lazy-translation | 21:24 |
| dhellmann | now to grep... | 21:24 |
| * dhellmann waits for his repos to update | 21:24 | |
| *** henrynash has quit IRC | 21:25 | |
| *** ayoung has quit IRC | 21:25 | |
| bnemec | So I'm looking at heat, and so far I only see changes to the files in bin/. | 21:27 |
| bnemec | Does that sound reasonable? | 21:27 |
| *** Adri2000 has joined #openstack-dev | 21:27 | |
| mrodden | yeah | 21:28 |
| mrodden | nova is in cmd/__init__ | 21:28 |
| mrodden | they are all at the main entry points | 21:28 |
| mrodden | (working on nova) | 21:28 |
| bnemec | Okay, I thought so, but figured I'd make sure. | 21:28 |
| *** nermina has quit IRC | 21:30 | |
| *** vkmc has quit IRC | 21:30 | |
| *** flaper87 is now known as flaper87|afk | 21:30 | |
| *** radez is now known as radez_g0n3 | 21:31 | |
| dhellmann | ok, I think that may cover it | 21:32 |
| *** eglynn has joined #openstack-dev | 21:32 | |
| dhellmann | I used ack to search the openstack and stackforge repos | 21:33 |
| *** dkranz has quit IRC | 21:33 | |
| bnemec | dhellmann: Okay, I've done heat, mrodden was working on nova. | 21:34 |
| dhellmann | I'll look at cinder | 21:34 |
| *** Javin has quit IRC | 21:35 | |
| bnemec | Heh, of course I picked a project I don't have installed on this VM. :-) | 21:35 |
| dhellmann | if someone already has a patch up, let us know what change-id so we can all use the same one | 21:36 |
| bnemec | I haven't pushed anything yet. Running tests just to sanity check. | 21:36 |
| mrodden | yeah unfortunately the tests were enabled as well | 21:36 |
| dhellmann | yeah | 21:36 |
| dhellmann | add skips? | 21:37 |
| mrodden | they shouldn't affect too much but yeah | 21:37 |
| mrodden | some might fail | 21:37 |
| dhellmann | or wait, are they testing the lazy translation itself or just turning them on | 21:37 |
| dolphm | markwash: glance's version bump just failed :-/ | 21:37 |
| mrodden | we can just skip | 21:37 |
| dolphm | markwash: https://jenkins01.openstack.org/job/gate-tempest-devstack-vm-postgres-full/16428/console | 21:37 |
| *** isd has quit IRC | 21:38 | |
| markwash | d'oh! | 21:38 |
| bnemec | dhellmann: mrodden: You know, it might be okay to leave lazy enabled for tests. | 21:38 |
| bnemec | As long as they're passing it won't hurt anything. | 21:38 |
| bnemec | The concern is unexpected unicode, which won't happen in unit tests. | 21:39 |
| dhellmann | bnemec: good point | 21:40 |
| mrodden | yeah | 21:40 |
| mrodden | that is true | 21:40 |
| *** FatDarrel has quit IRC | 21:40 | |
| *** salv-orlando has quit IRC | 21:41 | |
| bnemec | Okay, about to push Ia934a7df9386baf6ae8eb9ff48c24386c47ecd23 | 21:42 |
| *** harlowja has joined #openstack-dev | 21:42 | |
| luisg | bnemec, so we are turning off the t18n :( | 21:42 |
| bnemec | mrodden: dhellmann: Pushed ^ | 21:43 |
| *** changbl has quit IRC | 21:43 | |
| bnemec | luisg: Afraid so. | 21:43 |
| luisg | dhellmann, what did u think of mrodden latest patch for the issue | 21:43 |
| dhellmann | luisg: I need to look again | 21:44 |
| dhellmann | bnemec: thanks | 21:44 |
| dhellmann | I added that changeset to the top of the etherpad for easy reference | 21:44 |
| bnemec | dhellmann: Cool, thanks. | 21:45 |
| *** alunduil has joined #openstack-dev | 21:46 | |
| luisg | dhellmann, i thought since the patch fixes the issue we were going to leave t18n on | 21:47 |
| luisg | and there are not really other issues with non-logging aspects of the class | 21:47 |
| bnemec | Huh, I think ceilometer was enabling this twice. | 21:48 |
| *** Javin has joined #openstack-dev | 21:48 | |
| mrodden | whops | 21:48 |
| bnemec | That's icky. | 21:48 |
| dhellmann | luisg: we either need to disable it, or get the new change to the logging adapter merged into all of the projects | 21:48 |
| bnemec | They did install('ceilometer', True) without the lazy=. | 21:48 |
| luisg | dhellmann, i think we can do that, try to get those changes in across, instead of disabling | 21:48 |
| luisg | or we can disablet the ones that won't take the logger fix only | 21:49 |
| mrodden | i would actually be in favor of that... but i think ttx made up his mind? | 21:49 |
| dhellmann | luisg: let me look at the mrodden's change | 21:49 |
| luisg | dhellmann, k thx | 21:49 |
| mrodden | bnemec: totally borrowing your commit message :) | 21:50 |
| bnemec | mrodden: Then I demand co-author credit. :-P | 21:51 |
| luisg | ha | 21:51 |
| mrodden | does that really count/ | 21:52 |
| mrodden | ? | 21:52 |
| mrodden | :P | 21:52 |
| Javin | bnemec: Can you take a look at the updated pack stack commit when you have some time? https://review.openstack.org/#/c/48386/ | 21:52 |
| dhellmann | mrodden: +2 | 21:52 |
| *** salv-orlando has joined #openstack-dev | 21:52 | |
| *** FatDarrel has joined #openstack-dev | 21:52 | |
| mrodden | dhellmann: thanks | 21:52 |
| luisg | dhellmann, thx! | 21:52 |
| bnemec | Javin: Yeah, I glanced at it. I assume the new if will cause it to fail if there are multiple interfaces? | 21:52 |
| bnemec | Will there be a message that explains why? | 21:52 |
| luisg | we can take care of propagating it | 21:53 |
| luisg | if it gets in | 21:53 |
| bnemec | Crud. ceilometer tox env doesn't want to build. | 21:54 |
| *** venkatesh_ has joined #openstack-dev | 21:54 | |
| *** venkatesh has quit IRC | 21:54 | |
| *** SergeyLukjanov has quit IRC | 21:54 | |
| dhellmann | bnemec: have a link to your change? | 21:54 |
| Javin | bnemec: Yes. It will fail in the same manner as the test right below it. There isn't a custom error yet. | 21:55 |
| bnemec | dhellmann: The heat one? | 21:55 |
| dhellmann | yeah | 21:55 |
| *** johnthetubaguy has joined #openstack-dev | 21:55 | |
| bnemec | dhellmann: https://review.openstack.org/#/c/49266/ | 21:55 |
| *** tmclaugh[work] has quit IRC | 21:55 | |
| *** mlavalle has quit IRC | 21:56 | |
| *** romcheg has joined #openstack-dev | 21:56 | |
| *** neelashah has joined #openstack-dev | 21:56 | |
| *** romcheg1 has joined #openstack-dev | 21:56 | |
| luisg | bnemec, so we got +2 from dhellmann, should we hold on the lazy disabling, until later if we can't actually sync it? | 21:56 |
| *** romcheg has quit IRC | 21:56 | |
| luisg | in reference to the heat patch above | 21:57 |
| *** topol has quit IRC | 21:57 | |
| bnemec | luisg: I'm not sure that's what he was +2ing... | 21:57 |
| *** burt has quit IRC | 21:57 | |
| dhellmann | luisg: I don't want to slam this in at the last minute. I really feel much more comfortable disabling the feature and being careful about re-enabling it. | 21:57 |
| bnemec | Javin: Okay, but the line below will print out a message that the interface wasn't found. This one will fail silently, won't it? | 21:58 |
| luisg | k maybe i missed something, so what is the plan? | 21:58 |
| bnemec | Javin: Wait, maybe not. | 21:58 |
| bnemec | Too many conversations going on at once. :-) | 21:58 |
| luisg | :) | 21:58 |
| *** SergeyLukjanov has joined #openstack-dev | 21:58 | |
| *** READ10 has quit IRC | 21:59 | |
| luisg | dhellmann, could we add a cfg flag that would enable/disable lazy | 21:59 |
| *** dolphm has quit IRC | 21:59 | |
| *** pcm_ has quit IRC | 22:00 | |
| *** eglynn has quit IRC | 22:00 | |
| dhellmann | luisg: that would be an even more complex change than this | 22:00 |
| bnemec | Javin: Okay, I'm just going to +1. If the error handling isn't acceptable I'm sure one of the packstack cores will shoot it down. :-) | 22:00 |
| mrodden | luisg: i already had that discussion with markmc, a config option doesn't really make sense, since that is user configurable | 22:00 |
| Javin | bnemec: I'll test the failure message tomorrow and ensure that it passes a useful message back. | 22:00 |
| dhellmann | the other thing to keep in mind is this adapter change moves the string lookup to a point where it is done for every log message, not just for the ones that will actually be written out | 22:00 |
| *** edmund has quit IRC | 22:01 | |
| *** edmund1 has joined #openstack-dev | 22:01 | |
| *** johnthetubaguy has quit IRC | 22:01 | |
| *** SergeyLukjanov has quit IRC | 22:01 | |
| Javin | bnemec: Thanks | 22:01 |
| bnemec | Javin: Sounds good, thanks. | 22:01 |
| *** thomasm has quit IRC | 22:01 | |
| *** venkatesh_ has quit IRC | 22:02 | |
| *** nati_ueno has quit IRC | 22:02 | |
| mrodden | dhellmann: disabling lazy gettext will have the same affect | 22:03 |
| mrodden | all string lookups will be done immediately | 22:03 |
| bnemec | Tis true | 22:03 |
| *** ifarkas has quit IRC | 22:04 | |
| luisg | it's a shame to disable it really, from what i know customers wanted their responses translated, and tbh i think it was expected we were going to hfnid bugs, we just needed to fix them | 22:04 |
| luisg | :/ | 22:04 |
| mrodden | i dont know how much of a case we have to argue anymore | 22:05 |
| jgriffith | dhellmann: does that apply to the test_wsgi.py or do we care (https://review.openstack.org/#/c/49275/1) | 22:05 |
| mrodden | its probably not fair, as i have seen other individuals/groups shove things into openstack at the last minute that are sure to be problematic | 22:05 |
| dhellmann | mordred: good point | 22:05 |
| dhellmann | er, mrodden, I mean :-) | 22:06 |
| mrodden | but if ttx or the PTLs decides to remove it then we should cooperate | 22:06 |
| dhellmann | jgriffith: we think leaving it enabled in tests is ok | 22:06 |
| jgriffith | dhellmann: works for me, seems like there's some reasoning there | 22:06 |
| bnemec | Well, if we merge mrodden's patch tomorrow and decide it's solid enough to keep in for Havana we can always push another change. | 22:06 |
| jgriffith | dhellmann: thanks | 22:06 |
| bnemec | We'd have to anyway. | 22:06 |
| dhellmann | luisg: the issue is the schedule, unfortunately :-( | 22:07 |
| mrodden | thats the other thing | 22:07 |
| bnemec | So the disable patch would be a bit of extra churn, but othewise we're no worse off. | 22:07 |
| dhellmann | bnemec: right | 22:07 |
| bnemec | *otherwise. | 22:07 |
| luisg | dhellmann, understand | 22:07 |
| bnemec | load average: 8.78, 8.20, 5.74 <-- Outstanding. I knew I bought that 8 core box for a reason. :-) | 22:08 |
| *** rudrarugge_ has joined #openstack-dev | 22:08 | |
| *** jecarey_ has quit IRC | 22:08 | |
| *** bknudson has left #openstack-dev | 22:08 | |
| *** boris-42 has quit IRC | 22:08 | |
| *** kbringard has quit IRC | 22:09 | |
| *** mriedem has left #openstack-dev | 22:09 | |
| *** mriedem has quit IRC | 22:09 | |
| mrodden | nova tests are slow... | 22:09 |
| mrodden | :( | 22:10 |
| bnemec | mrodden: Hence the 8 core box. :-) | 22:10 |
| *** rudrarugge_ is now known as rudrarugge | 22:11 | |
| *** comay has quit IRC | 22:11 | |
| *** comay has joined #openstack-dev | 22:11 | |
| *** rudrarugge has left #openstack-dev | 22:11 | |
| *** insanidade has quit IRC | 22:11 | |
| *** Javin has quit IRC | 22:12 | |
| mrodden | i could probably just push the change to something internal and let my jenkins build slaves do the heavy lifting | 22:12 |
| * mrodden makes a mental note to look into that | 22:13 | |
| bnemec | Hmm, apparently I don't have everything installed to run ceilometer tests. | 22:13 |
| mrodden | you need mongo | 22:13 |
| bnemec | So I see | 22:13 |
| mrodden | don't ask me why that is required for unit tests... | 22:13 |
| bnemec | Maybe they actually test against their production backends. :-) | 22:14 |
| *** nati_ueno has joined #openstack-dev | 22:15 | |
| *** Reapster_ is now known as Reapster | 22:16 | |
| *** romcheg1 has quit IRC | 22:16 | |
| *** rackerjoe has quit IRC | 22:16 | |
| *** angdraug has quit IRC | 22:16 | |
| mrodden | bnemec: well for an integration test that makes sense... | 22:17 |
| *** angdraug has joined #openstack-dev | 22:17 | |
| mrodden | and don't get me started on DB2 as a production backend | 22:17 |
| *** rcleere has quit IRC | 22:17 | |
| *** michchap has quit IRC | 22:17 | |
| *** shinylasers has joined #openstack-dev | 22:17 | |
| *** ondergetekende has quit IRC | 22:17 | |
| *** romcheg has joined #openstack-dev | 22:17 | |
| bnemec | mrodden: People run that in production!? | 22:18 |
| bnemec | :-D | 22:18 |
| mrodden | lol | 22:18 |
| mrodden | apparently... | 22:18 |
| mrodden | do we want to track the review URLs in the etherpad? | 22:18 |
| *** michchap has joined #openstack-dev | 22:18 | |
| bnemec | Should be able to just click on the change id for one of yours, right? | 22:19 |
| *** rackerjoe has joined #openstack-dev | 22:19 | |
| mrodden | oh right | 22:19 |
| *** ondergetekende has joined #openstack-dev | 22:19 | |
| dhellmann | mrodden, bnemec : ceilometer uses mongo for unit tests because the in-memory version we used to use didn't have some of the features of the real server | 22:20 |
| *** datsun180b has quit IRC | 22:20 | |
| * bnemec should have let dhellmann do those changes :-) | 22:20 | |
| mrodden | did we find anything in glance? | 22:20 |
| dhellmann | bnemec: or jd__ :-) | 22:21 |
| dhellmann | mrodden: I didn't, but please feel free to look again in case I botched the search | 22:21 |
| mrodden | yeah i'll check it | 22:21 |
| *** bpokorny has quit IRC | 22:21 | |
| bnemec | In any case the ceilo change is pushed. | 22:22 |
| bnemec | Neutron test cases just blew up, so I need to figure out what happened there. | 22:22 |
| mrodden | that one is always a pita | 22:22 |
| bnemec | Apparently 4 GB of RAM is not enough. | 22:23 |
| mrodden | oh yeah | 22:23 |
| mrodden | you found that too? | 22:23 |
| bnemec | Seems that way. | 22:23 |
| bnemec | I have a bunch of OSError: [Errno 12] Cannot allocate memory | 22:23 |
| mrodden | normally i wouldn't run into it, but our jenkins builders only have 4G without any swap since they are VMs | 22:23 |
| mrodden | so we hit that a lot a few weeks ago | 22:23 |
| * bnemec resizes his neutron VM | 22:24 | |
| *** thedodd has quit IRC | 22:25 | |
| dhellmann | I have to go sit in traffic for a bit, but I'll check the etherpad when I get home | 22:25 |
| dhellmann | 90-120 mins | 22:25 |
| bnemec | dhellmann: Okay. I will likely be somewhere else by then, but I'll try to remember to check in later tonight. | 22:26 |
| *** FunnyLookinHat has quit IRC | 22:26 | |
| *** galstrom is now known as galstrom_zzz | 22:26 | |
| *** flaper87|afk is now known as flaper87 | 22:27 | |
| mrodden | looks like glance is clean | 22:28 |
| *** xqueralt has joined #openstack-dev | 22:29 | |
| bnemec | Yeah, I don't see anything either | 22:29 |
| *** dstanek has quit IRC | 22:29 | |
| bnemec | Heh, there's a change named Do Not Merge in my dash.py check list | 22:30 |
| *** alunduil has quit IRC | 22:30 | |
| mrodden | test_admin_requires_adminness - my favorite keystone test | 22:30 |
| *** emagana has quit IRC | 22:31 | |
| bnemec | A good one, that | 22:31 |
| *** dstanek has joined #openstack-dev | 22:31 | |
| *** anniec has joined #openstack-dev | 22:31 | |
| *** SumitNaiksatam has quit IRC | 22:31 | |
| bnemec | Sounds like a test Joss Whedon would write. :-) | 22:32 |
| *** emagana has joined #openstack-dev | 22:32 | |
| *** armax has quit IRC | 22:32 | |
| bnemec | Heh, I didn't even have to review https://review.openstack.org/#/c/49230/ | 22:33 |
| *** markmcclain has quit IRC | 22:33 | |
| *** xqueralt has left #openstack-dev | 22:33 | |
| *** xqueralt has joined #openstack-dev | 22:34 | |
| mrodden | whoa | 22:34 |
| *** sgordon has quit IRC | 22:34 | |
| mrodden | bnemec: so you have an openstack running ? | 22:35 |
| *** gordc has quit IRC | 22:36 | |
| luisg | i gotta step out, bnemec and dhellmann thx for spending time on this | 22:36 |
| bnemec | mrodden: Yeah, I managed to get Neutron working well enough to be useful. | 22:37 |
| bnemec | luisg: Have a good one. | 22:37 |
| luisg | thx | 22:37 |
| mrodden | luisg: cya, thanks for the help today | 22:38 |
| *** amcrn is now known as amcrn[a] | 22:38 | |
| mrodden | bnemec: is it grizzly or master? | 22:38 |
| luisg | np ty later | 22:38 |
| bnemec | mrodden: Packstack Havana. | 22:38 |
| mrodden | ok | 22:38 |
| mrodden | havent played with that yet | 22:39 |
| bnemec | Figured that's a project I'm technically kind of working on so I oughta learn about it. | 22:39 |
| mrodden | hah, good point | 22:39 |
| *** romcheg has quit IRC | 22:40 | |
| * bnemec should have resized his neutron vm to have more vcpus too :-/ | 22:40 | |
| mrodden | all the cpuz | 22:40 |
| lifeless | all the time | 22:40 |
| bnemec | Heh | 22:41 |
| *** markwash has quit IRC | 22:41 | |
| bnemec | TypeError: <MagicMock name='LinuxBridgeManager().local_ip' id='204068880'> is not JSON serializable | 22:41 |
| bnemec | Hmm. | 22:41 |
| bnemec | That seems unlikely to be related to the lazy translation change. | 22:42 |
| mrodden | odd | 22:42 |
| mrodden | it might be | 22:42 |
| *** gmurphy_gone is now known as gmurphy | 22:42 | |
| bnemec | Pretty sure I got that last time too, but was hoping it was related to the memory issues. | 22:42 |
| mrodden | if something bubbled up in an exception that tried to get serialized | 22:42 |
| mrodden | there was a fix for that in jsonutils | 22:42 |
| bnemec | Yeah, when this run finishes I'll have to take a closer look. | 22:43 |
| mrodden | although usually its a CircularReference issue | 22:44 |
| mrodden | you know what is really amazing | 22:45 |
| bnemec | ??? | 22:45 |
| mrodden | i set my status to "in a meeting" all day and no one has interrupted me | 22:45 |
| *** nati_ueno has quit IRC | 22:45 | |
| bnemec | I just kept dropping off the VPN. Sadly that didn't help with community IRC. :-) | 22:46 |
| mrodden | heh | 22:46 |
| *** vipul is now known as vipul-away | 22:47 | |
| *** vipul-away is now known as vipul | 22:47 | |
| bnemec | Hmm, that may not be a fatal error for the tests. | 22:47 |
| bnemec | It looks like a traceback in the logging code, but there's no test to be found. | 22:48 |
| bnemec | If tox would ever finish I could find out... | 22:48 |
| mrodden | might be some print output from a negative test | 22:49 |
| *** jimfehlig has quit IRC | 22:49 | |
| *** henrynash has joined #openstack-dev | 22:50 | |
| *** dmakogon_ has quit IRC | 22:50 | |
| *** dstanek has quit IRC | 22:52 | |
| *** sarob_ has joined #openstack-dev | 22:54 | |
| *** neelashah has quit IRC | 22:54 | |
| *** cfriesen has joined #openstack-dev | 22:54 | |
| *** emagana has quit IRC | 22:55 | |
| *** shardy is now known as shardy_afk | 22:55 | |
| *** emagana has joined #openstack-dev | 22:55 | |
| *** jmontemayor has quit IRC | 22:57 | |
| *** danwent has quit IRC | 22:58 | |
| *** sarob has quit IRC | 22:58 | |
| *** sarob_ has quit IRC | 22:59 | |
| *** cfriesen has left #openstack-dev | 22:59 | |
| *** SumitNaiksatam has joined #openstack-dev | 22:59 | |
| *** emagana has quit IRC | 23:00 | |
| *** sarob has joined #openstack-dev | 23:01 | |
| mrodden | keystone is up | 23:02 |
| *** danwent has joined #openstack-dev | 23:03 | |
| *** pmathews has quit IRC | 23:04 | |
| *** samuelbercovici has joined #openstack-dev | 23:04 | |
| *** jcoufal has quit IRC | 23:05 | |
| *** sthaha has joined #openstack-dev | 23:06 | |
| *** nati_ueno has joined #openstack-dev | 23:08 | |
| *** samuelbercovici has quit IRC | 23:08 | |
| *** jorisroovers has quit IRC | 23:10 | |
| *** dims has quit IRC | 23:13 | |
| *** reed has quit IRC | 23:13 | |
| *** sarob has quit IRC | 23:15 | |
| bnemec | Hooray! Neutron didn't fail: Ran 18504 (+11156) tests in 1695.958s (-379.421s) | 23:15 |
| bnemec | PASSED (id=1, skips=319) | 23:15 |
| mrodden | impressive | 23:15 |
| *** sarob has joined #openstack-dev | 23:15 | |
| mrodden | down 6 minutes even | 23:15 |
| mrodden | :) | 23:15 |
| bnemec | Yeah, I'm a little confused by that, but whatever. | 23:16 |
| mrodden | that should be all of them then | 23:17 |
| bnemec | Yep, it's up. | 23:17 |
| *** vartom15 has joined #openstack-dev | 23:18 | |
| mrodden | they all look good to me | 23:20 |
| mrodden | heat one looks like its going to merge already | 23:20 |
| *** prad has quit IRC | 23:22 | |
| *** vartom15 has quit IRC | 23:22 | |
| *** colinmcnamara has quit IRC | 23:23 | |
| *** alexpilotti has quit IRC | 23:23 | |
| * bnemec +1s all the things | 23:23 | |
| bnemec | Yeah, only 7 hours and change left before the Heat change gets through the gate. :-) | 23:24 |
| mrodden | lol | 23:24 |
| * bnemec will not be waiting for that one | 23:24 | |
| mrodden | yeah i had two bug fixes for API message localization that were looking promising | 23:24 |
| mrodden | in fact one merged today | 23:24 |
| *** tanisdl has quit IRC | 23:25 | |
| bnemec | Yeah, think I saw that. | 23:26 |
| mrodden | that thing failed gate like 4 or 5 times | 23:26 |
| mrodden | it was done last week sometime... | 23:26 |
| *** dims has joined #openstack-dev | 23:27 | |
| *** otherwiseguy has quit IRC | 23:27 | |
| bnemec | That's always fun. :-) | 23:27 |
| bnemec | Bugger, the Cinder change failed Jenkins | 23:28 |
| mrodden | ouch | 23:28 |
| mrodden | timeout | 23:28 |
| mrodden | probably needs a recheck | 23:29 |
| *** flaper87 is now known as flaper87|afk | 23:30 | |
| *** ctracey has quit IRC | 23:31 | |
| *** medberry is now known as med_ | 23:31 | |
| *** donaldh has quit IRC | 23:32 | |
| *** nachi has left #openstack-dev | 23:32 | |
| *** briancurtin has quit IRC | 23:32 | |
| *** ctracey has joined #openstack-dev | 23:32 | |
| *** carl_baldwin has quit IRC | 23:32 | |
| *** MaxV has quit IRC | 23:34 | |
| mrodden | bnemec: http://logs.openstack.org/75/49275/1/check/check-tempest-devstack-vm-postgres-full/8ae87da/logs/screen-g-api.txt.gz?level=AUDIT | 23:34 |
| mrodden | hmm | 23:34 |
| bnemec | Huh? | 23:35 |
| bnemec | We didn't even touch Glance. :-) | 23:35 |
| mrodden | i know | 23:35 |
| mrodden | thats why its weird | 23:35 |
| mrodden | actually its a problem with swift it seems | 23:35 |
| mrodden | since it was the swift backing store to glance | 23:35 |
| bnemec | I don't see any errors in Swift though. :-/ | 23:37 |
| *** anniec has quit IRC | 23:37 | |
| mrodden | yeah | 23:37 |
| mrodden | that output seems strangely clean | 23:37 |
| mrodden | except for the proxy logs complaining that a request to object containers didn't work | 23:37 |
| *** vipul is now known as vipul-away | 23:38 | |
| mrodden | maybe have someone in infra look? | 23:38 |
| *** sarob has quit IRC | 23:38 | |
| bnemec | Yeah, I was just thinking that. The collie error looks like an issue with the VM. | 23:39 |
| *** sarob has joined #openstack-dev | 23:39 | |
| *** edmund1 has quit IRC | 23:40 | |
| mrodden | i thnk that is always thrown | 23:40 |
| mrodden | its a special backend for glance that i'm not sure is tested or not | 23:41 |
| *** vipul-away is now known as vipul | 23:41 | |
| mrodden | i dont see anything in rechecks that is similar | 23:41 |
| bnemec | Ah, guess I've never had to look at the Glance logs. | 23:41 |
| bnemec | Yeah, I don't either. | 23:41 |
| *** AlexF has joined #openstack-dev | 23:42 | |
| *** sarob has quit IRC | 23:43 | |
| *** rfolco has joined #openstack-dev | 23:47 | |
| *** faramir has joined #openstack-dev | 23:49 | |
| *** shinylasers has quit IRC | 23:50 | |
| bnemec | mrodden: At this point I'm thinking just open a new bug against cinder and recheck against that. | 23:51 |
| bnemec | I don't know which service is at fault here, so we'll start with the one whose tests failed. :-) | 23:51 |
| *** DennyZhang has joined #openstack-dev | 23:52 | |
| *** nati_ueno has quit IRC | 23:54 | |
| mrodden | sure | 23:54 |
| mrodden | launchpad bugs are easy to move later | 23:55 |
| bnemec | Yeah, that's my thought, and they'll probably have a better idea what's going on. | 23:55 |
| bnemec | I'm completely lost. :-) | 23:55 |
| mrodden | yeah it goes from cinder to glance to swift | 23:56 |
| *** rfolco has quit IRC | 23:56 | |
| *** adjohn has joined #openstack-dev | 23:56 | |
| mrodden | i have to step away for awhile to do non-computer related thing | 23:56 |
| mrodden | s | 23:56 |
| *** anniec has joined #openstack-dev | 23:56 | |
| bnemec | Yeah, I'll get this rechecked and then I'm probably done for the night. | 23:56 |
| *** henrynash has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!