| *** internapsjc has joined #openstack-dev | 00:01 | |
| *** mishesks has left #openstack-dev | 00:02 | |
| *** flaviamissi has joined #openstack-dev | 00:05 | |
| *** dtroyer has joined #openstack-dev | 00:06 | |
| *** kbringard has quit IRC | 00:08 | |
| *** anotherjesse is now known as anotherjesse_zz | 00:08 | |
| *** flaviamissi_ has joined #openstack-dev | 00:09 | |
| *** mattstep has quit IRC | 00:10 | |
| *** mattstep has joined #openstack-dev | 00:10 | |
| *** flaviamissi has quit IRC | 00:10 | |
| *** bencherian has joined #openstack-dev | 00:11 | |
| *** misheska has joined #openstack-dev | 00:12 | |
| internapsjc | d | 00:12 |
|---|---|---|
| vishy | kbringard: you need to restart all services | 00:12 |
| vishy | kbringard: everything uses the db | 00:12 |
| *** flaviamissi_ has quit IRC | 00:14 | |
| *** zykes- has quit IRC | 00:14 | |
| *** flaviamissi has joined #openstack-dev | 00:14 | |
| *** internapsjc has quit IRC | 00:15 | |
| *** zykes- has joined #openstack-dev | 00:15 | |
| *** misheska- has joined #openstack-dev | 00:18 | |
| *** flaviamissi has quit IRC | 00:19 | |
| *** lloydde has quit IRC | 00:19 | |
| *** misheska has quit IRC | 00:30 | |
| *** lloydde has joined #openstack-dev | 00:33 | |
| *** flaviamissi has joined #openstack-dev | 00:35 | |
| *** flaviamissi_ has joined #openstack-dev | 00:36 | |
| *** flaviamissi has quit IRC | 00:36 | |
| hub_cap | hey how does nova handle query params in the wsgi layer? Im trying to add it to keystone (it currently ignores queryparams) | 00:36 |
| hub_cap | and i dont want to make it ugly :D | 00:36 |
| *** flaviamissi_ has quit IRC | 00:37 | |
| *** flaviamissi has joined #openstack-dev | 00:37 | |
| *** flaviamissi has quit IRC | 00:38 | |
| *** bencherian has quit IRC | 00:39 | |
| cloudfly | heckj any luck? | 00:39 |
| *** andrewsmedina_ has joined #openstack-dev | 00:40 | |
| heckj | cloudfly: yeah, much farther - I'm stumbling over my own mysql password. Just reset and rant - looks like it's all up | 00:41 |
| heckj | cloudfly - what issue were you seeing? | 00:41 |
| heckj | Just logged into dash - no problem, but I don't recall what issue you had. | 00:42 |
| heckj | creiht: thanks - must have been that older cloudfiles, I'm totally fine now | 00:42 |
| hub_cap | heckj: In regard to the query strings in keystone, should i blindly add _all_ the query params to the list of kwargs that get passed in to a method? as in, if you do a GET on a resource, for instance foo/?p1=blah, you would have to provide kwarg p1 to the method | 00:43 |
| cloudfly | i drop into a stack trace when executing setup.py develop for horizon | 00:43 |
| cloudfly | did it manually same deal | 00:43 |
| heckj | cloudfly: I'm not seeing that issue at all | 00:43 |
| cloudfly | wierd | 00:43 |
| heckj | cloudfly: totally clean ubuntu base image? (mine wasn't!) | 00:44 |
| *** andrewsmedina has quit IRC | 00:44 | |
| *** andrewsmedina_ is now known as andrewsmedina | 00:44 | |
| cloudfly | fairly clean 11.10 | 00:44 |
| heckj | hub_cap: honestly not sure… sec | 00:44 |
| *** vincentricci has joined #openstack-dev | 00:44 | |
| hub_cap | cloudfly: if youre talking about devstack, i was getting that too, and i had to git pull devstack again to get it clean,it was looking for setup.py in /opt/stack/horizon/horizon | 00:44 |
| heckj | hub_cap - what's happening in nova re: this? any idea? | 00:44 |
| hub_cap | ya it passes in the req object | 00:44 |
| hub_cap | and they just inspect it | 00:44 |
| cloudfly | hub_cap yeah that's the bug i think dean merged | 00:44 |
| heckj | cloudfly: I've been also "sudo rm -rf /opt" before re-running to make sure the code is all clean | 00:45 |
| hub_cap | ah well thank you cloudfly ! | 00:45 |
| cloudfly | well i'll try clearing it all again | 00:47 |
| cloudfly | =/ | 00:47 |
| cloudfly | tried 3 times now | 00:47 |
| hub_cap | heckj: i dont mean for you to have to answer it now, think about it. the belongs_to stuff will be waiting for an answer. I talked ot dolphm about it and he said just remove 'fresh' from the params since its a bug w/ keystoneclient. i did that, and then a test failed, which is just a generic wsgi test, but it seems odd that we would rely on _all_ query params as being kwargs :D | 00:47 |
| heckj | hub_cap: reading glance/common/wsgi.py in glance to see what happens there (I'm not awesome as WSGI) | 00:47 |
| hub_cap | ok thats fine, im not either heckj :D | 00:48 |
| hub_cap | ill peep glanc as well | 00:48 |
| heckj | (just seemed easier to tear apart than Nova to me) | 00:48 |
| hub_cap | oh def | 00:48 |
| *** PotHix has quit IRC | 00:48 | |
| *** AlanClark has quit IRC | 00:48 | |
| hub_cap | well nova punts on it w/ method(req) | 00:48 |
| hub_cap | and so does glance :D | 00:49 |
| hub_cap | heckj https://github.com/openstack/glance/blob/master/glance/api/v1/controller.py | 00:49 |
| hub_cap | see the request obj being passed | 00:50 |
| heckj | hub_cap - yeah, looks like they just grabbed context out of the request and passed that into the relevant method there | 00:50 |
| hub_cap | yup | 00:51 |
| hub_cap | that how nova does it too, ks is diff in that it wants kwargs, which i like honestly | 00:51 |
| *** roge has joined #openstack-dev | 00:52 | |
| *** spiffxp has quit IRC | 00:53 | |
| hub_cap | i mean, ive got all but 1 silly example wsgi test fixed by updating all the params w/ the query string and removing fresh as per the bug dolph has assigned to him | 00:53 |
| heckj | hub_cap: sounds like most of a solution then | 00:53 |
| hub_cap | its better than whats thre now ;) | 00:54 |
| *** dtroyer has quit IRC | 00:54 | |
| hub_cap | i can also pass them as query_params kwarg | 00:54 |
| *** lloydde_ has joined #openstack-dev | 00:54 | |
| *** lloydde has quit IRC | 00:57 | |
| *** lloydde_ has quit IRC | 00:58 | |
| hub_cap | heckj: as in http://paste.openstack.org/show/6961/ | 00:59 |
| zaitcev | Someone castrated keystone-manage? It does not have "endpointTemplates list" anymore? | 01:00 |
| *** bencherian has joined #openstack-dev | 01:01 | |
| hub_cap | heckj: shoot me a line (mbasnight@gmail.com) when u make a decision, ill code up what u think is best | 01:03 |
| *** hub_cap has quit IRC | 01:03 | |
| *** bsza has quit IRC | 01:05 | |
| * cloudfly gives it one more shot | 01:06 | |
| *** bencherian has quit IRC | 01:07 | |
| *** vizsla has joined #openstack-dev | 01:07 | |
| *** dolphm has quit IRC | 01:09 | |
| *** jdg has quit IRC | 01:13 | |
| *** shang has quit IRC | 01:16 | |
| cloudfly | nope | 01:17 |
| cloudfly | horizon is still crashing | 01:17 |
| cloudfly | time to set up a vm i guess | 01:17 |
| *** shang has joined #openstack-dev | 01:17 | |
| *** dtroyer has joined #openstack-dev | 01:19 | |
| *** littleidea has joined #openstack-dev | 01:19 | |
| *** bencherian has joined #openstack-dev | 01:20 | |
| cloudfly | looks like it could be cloudfiles oddly enough | 01:21 |
| cloudfly | that is snagging me | 01:21 |
| cloudfly | from cloudfiles import errors as swiftclient <--- is the exception thrower | 01:22 |
| *** anotherjesse_zz is now known as anotherjesse | 01:22 | |
| *** Ryan_Lane has quit IRC | 01:26 | |
| *** vladimir3p has quit IRC | 01:27 | |
| *** jog0 has quit IRC | 01:27 | |
| *** hub_cap has joined #openstack-dev | 01:30 | |
| *** maplebed has quit IRC | 01:34 | |
| *** mnewby has quit IRC | 01:37 | |
| *** martine has joined #openstack-dev | 01:41 | |
| *** sniperd has joined #openstack-dev | 01:46 | |
| *** lloydde has joined #openstack-dev | 01:47 | |
| *** lloydde has quit IRC | 01:48 | |
| *** lloydde has joined #openstack-dev | 01:48 | |
| cloudfly | i may have found the bugger | 01:49 |
| cloudfly | old cloudfiles data in /usr/local | 01:49 |
| cloudfly | that was it | 01:52 |
| *** spiffxp has joined #openstack-dev | 01:52 | |
| *** lloydde has quit IRC | 01:53 | |
| *** littleidea has quit IRC | 01:54 | |
| *** adjohn has quit IRC | 01:58 | |
| *** shang has quit IRC | 01:59 | |
| *** x86brandon has quit IRC | 02:00 | |
| *** adjohn has joined #openstack-dev | 02:01 | |
| *** spiffxp has quit IRC | 02:02 | |
| *** roge has quit IRC | 02:05 | |
| *** adjohn has quit IRC | 02:09 | |
| *** vincentricci has left #openstack-dev | 02:09 | |
| *** bencherian has quit IRC | 02:10 | |
| *** roge has joined #openstack-dev | 02:10 | |
| *** littleidea has joined #openstack-dev | 02:11 | |
| *** dubsquared has joined #openstack-dev | 02:11 | |
| *** ayoung has joined #openstack-dev | 02:15 | |
| *** novas0x2a|laptop has quit IRC | 02:16 | |
| *** ncode has joined #openstack-dev | 02:17 | |
| *** ncode has joined #openstack-dev | 02:17 | |
| *** anotherjesse is now known as anotherjesse_zz | 02:18 | |
| *** mnewby has joined #openstack-dev | 02:18 | |
| *** pixelbeat has quit IRC | 02:19 | |
| *** nati-ueno has quit IRC | 02:20 | |
| *** anotherjesse_zz is now known as anotherjesse | 02:21 | |
| *** heckj has quit IRC | 02:24 | |
| *** Ryan_Lane has joined #openstack-dev | 02:25 | |
| *** shang has joined #openstack-dev | 02:27 | |
| *** ayoung has quit IRC | 02:27 | |
| *** jaypipes has quit IRC | 02:28 | |
| *** gyee has quit IRC | 02:35 | |
| *** hub_cap has quit IRC | 02:42 | |
| *** dalang has quit IRC | 02:44 | |
| *** deshantm has joined #openstack-dev | 02:45 | |
| *** hub_cap has joined #openstack-dev | 02:45 | |
| *** sniperd has quit IRC | 02:45 | |
| *** dubsquared has quit IRC | 02:54 | |
| *** roge has quit IRC | 02:56 | |
| *** x86brandon has joined #openstack-dev | 02:58 | |
| *** sniperd has joined #openstack-dev | 03:00 | |
| *** cp16net has joined #openstack-dev | 03:02 | |
| *** mnewby has quit IRC | 03:09 | |
| *** jdurgin has quit IRC | 03:14 | |
| *** jaypipes has joined #openstack-dev | 03:16 | |
| letterj | LinuxJedi: You around? | 03:17 |
| *** anotherjesse is now known as anotherjesse_zz | 03:26 | |
| *** dubsquared has joined #openstack-dev | 03:29 | |
| *** spiffxp has joined #openstack-dev | 03:40 | |
| *** danwent has quit IRC | 03:42 | |
| notmyname | dweimer: I'm getting ready to submit my versioning patch for review. to make sure everything is sane, I'll be rebasing my branch and using `git push -f` to overwrite what's in my personal repo. if you're following that at all, you'll have to repull | 03:46 |
| *** spiffxp has quit IRC | 03:47 | |
| notmyname | dweimer: also, I'll have to squash all the commits for gerrit. IOW, completely ignore all my code that you've seen so far :-) | 03:48 |
| dweimer | notmyname: Thanks for the heads up. I'm working on a trunk tree at the moment. Seeing if it makes sense to handle posts separately for large file manifests in the proxy server. | 03:57 |
| notmyname | dweimer: I've been working on this for so long, and I'm sure I'll get brutalized in review. I'm nervous about actually submitting it :-) | 03:58 |
| notmyname | I keep thinking I must be forgetting something | 03:58 |
| notmyname | done https://review.openstack.org/#change,5013 | 04:00 |
| *** spiffxp has joined #openstack-dev | 04:02 | |
| dweimer | notmyname: Nice. I'll test it out some more on an all in one server here. | 04:03 |
| dweimer | notmyname: The way to completely delete a file would be delete the older versions first then delete the main file and just deleting the main file would roll-back to the pervious version right? | 04:05 |
| notmyname | dweimer: deleting the main file deletes the current version (like popping the stack) | 04:06 |
| notmyname | it rolls everything back so you can keep sending deletes to the version manifest object | 04:06 |
| *** adjohn has joined #openstack-dev | 04:07 | |
| *** lloydde has joined #openstack-dev | 04:08 | |
| *** andrewsmedina has quit IRC | 04:10 | |
| *** Ryan_Lane has quit IRC | 04:15 | |
| *** vishy has quit IRC | 04:15 | |
| *** vish1 has joined #openstack-dev | 04:16 | |
| *** lloydde has quit IRC | 04:23 | |
| *** martine has quit IRC | 04:24 | |
| *** deshantm has quit IRC | 04:27 | |
| *** spiffxp has quit IRC | 04:27 | |
| *** mdomsch has joined #openstack-dev | 04:31 | |
| hub_cap | ive done some work to python-novaclient to support proxy token/tenant (after fixing keystone to support it) | 04:32 |
| hub_cap | where should i put the tests tho in novaclient? | 04:32 |
| hub_cap | http://paste.openstack.org/show/7006/ | 04:32 |
| hub_cap | ^ ^ work i did to make it a-ok | 04:32 |
| uvirtbot | hub_cap: Error: "^" is not a valid command. | 04:32 |
| hub_cap | d**m you uvirtbot | 04:33 |
| hub_cap | at least someone is talking to me this late :D | 04:33 |
| hub_cap | even tho its a robot | 04:33 |
| *** jaypipes has quit IRC | 04:37 | |
| *** reed has quit IRC | 04:38 | |
| *** adjohn has quit IRC | 04:39 | |
| *** ywu has quit IRC | 04:43 | |
| *** dalang has joined #openstack-dev | 04:44 | |
| *** dubsquared has quit IRC | 04:44 | |
| *** bhall has quit IRC | 04:49 | |
| *** vizsla has quit IRC | 04:52 | |
| *** retr0h has quit IRC | 04:54 | |
| *** reed has joined #openstack-dev | 04:56 | |
| *** anotherjesse_zz is now known as anotherjesse | 04:57 | |
| *** cp16net has quit IRC | 04:57 | |
| *** cp16net has joined #openstack-dev | 04:58 | |
| *** mnewby has joined #openstack-dev | 05:10 | |
| *** x86brandon has quit IRC | 05:11 | |
| anotherjesse | ttx: thoughts on https://bugs.launchpad.net/nova/+bug/851159 - I can have someone on the team look at it tomorrow | 05:13 |
| uvirtbot | Launchpad bug 851159 in nova "ec2 metadata service is very slow" [High,Incomplete] | 05:13 |
| *** maplebed has joined #openstack-dev | 05:18 | |
| *** mdomsch has quit IRC | 05:23 | |
| *** hub_cap has quit IRC | 05:23 | |
| *** mnewby has quit IRC | 05:26 | |
| *** mnewby has joined #openstack-dev | 05:27 | |
| *** cp16net has quit IRC | 05:28 | |
| *** spiffxp has joined #openstack-dev | 05:31 | |
| *** deshantm has joined #openstack-dev | 05:33 | |
| *** sniperd has quit IRC | 05:33 | |
| *** sniperd has joined #openstack-dev | 05:37 | |
| *** dayou has joined #openstack-dev | 05:38 | |
| *** Ryan_Lane has joined #openstack-dev | 05:40 | |
| *** Tesk0 has joined #openstack-dev | 05:45 | |
| *** maplebed has quit IRC | 05:48 | |
| *** mikal has quit IRC | 05:54 | |
| *** hashar has joined #openstack-dev | 05:58 | |
| *** shang has quit IRC | 05:59 | |
| *** ches has quit IRC | 06:00 | |
| *** ches has joined #openstack-dev | 06:01 | |
| *** adjohn has joined #openstack-dev | 06:03 | |
| *** sniperd has quit IRC | 06:12 | |
| *** shang has joined #openstack-dev | 06:16 | |
| *** vincentricci has joined #openstack-dev | 06:23 | |
| *** vincentricci has left #openstack-dev | 06:23 | |
| *** littleidea has quit IRC | 06:23 | |
| *** milner has quit IRC | 06:25 | |
| *** hattwick has quit IRC | 06:31 | |
| *** zaitcev has quit IRC | 06:31 | |
| *** sandywalsh has quit IRC | 06:38 | |
| *** deshantm has quit IRC | 06:42 | |
| *** mnewby has quit IRC | 06:53 | |
| *** gkotton has joined #openstack-dev | 06:53 | |
| *** anotherjesse is now known as anotherjesse_zz | 06:56 | |
| *** milner has joined #openstack-dev | 06:56 | |
| *** ches has quit IRC | 07:04 | |
| *** ches has joined #openstack-dev | 07:05 | |
| *** dalang has quit IRC | 07:11 | |
| *** mikal has joined #openstack-dev | 07:12 | |
| *** hashar has quit IRC | 07:14 | |
| *** sandywalsh has joined #openstack-dev | 07:16 | |
| *** shevek_ has joined #openstack-dev | 07:17 | |
| *** dtroyer has quit IRC | 07:21 | |
| *** mikal has quit IRC | 07:23 | |
| *** hashar has joined #openstack-dev | 07:27 | |
| *** Mkenneth has joined #openstack-dev | 07:28 | |
| *** heckj has joined #openstack-dev | 07:30 | |
| *** mikal has joined #openstack-dev | 07:30 | |
| *** heckj has quit IRC | 07:33 | |
| *** zykes- has quit IRC | 07:36 | |
| *** zykes-_ has joined #openstack-dev | 07:36 | |
| *** Mandell has quit IRC | 07:44 | |
| *** spiffxp has quit IRC | 07:57 | |
| *** reed has quit IRC | 08:11 | |
| *** adjohn has quit IRC | 08:12 | |
| *** reidrac has joined #openstack-dev | 08:14 | |
| *** shang has quit IRC | 08:14 | |
| ttx | anotherjesse: we need to refresh the definition of "slow" on this bug. If it's completely unacceptably-slow and the fix is not too regression-prone, we could definitely have a try. | 08:14 |
| ttx | notmyname: ack, will cut milestone-proposed this morning | 08:16 |
| *** shang has joined #openstack-dev | 08:20 | |
| *** apevec has joined #openstack-dev | 08:21 | |
| *** zigo has joined #openstack-dev | 08:39 | |
| *** zigo is now known as zigo-_- | 08:39 | |
| *** halfss has joined #openstack-dev | 08:44 | |
| *** negronjl has quit IRC | 08:48 | |
| *** hattwick has joined #openstack-dev | 09:00 | |
| *** dneary has joined #openstack-dev | 09:00 | |
| *** zul has quit IRC | 09:06 | |
| *** derekh has joined #openstack-dev | 09:08 | |
| *** shevek_ has quit IRC | 09:08 | |
| *** shevek_ has joined #openstack-dev | 09:14 | |
| *** zul has joined #openstack-dev | 09:20 | |
| *** Ryan_Lane has quit IRC | 09:25 | |
| *** pixelbeat has joined #openstack-dev | 09:29 | |
| *** eglynn__ has quit IRC | 09:32 | |
| *** hashar has left #openstack-dev | 09:35 | |
| *** eglynn has joined #openstack-dev | 09:36 | |
| *** oneiroi has joined #openstack-dev | 09:45 | |
| *** ncode has quit IRC | 09:47 | |
| *** bepernoot has joined #openstack-dev | 09:49 | |
| *** bepernoot has quit IRC | 09:50 | |
| *** bepernoot has joined #openstack-dev | 09:51 | |
| *** zul has quit IRC | 09:51 | |
| *** zul has joined #openstack-dev | 09:58 | |
| *** bepernoot has quit IRC | 10:04 | |
| *** armaan has joined #openstack-dev | 10:19 | |
| *** paulormg has joined #openstack-dev | 10:46 | |
| *** zigo-_- has quit IRC | 11:06 | |
| *** zigo has joined #openstack-dev | 11:10 | |
| *** paulormg is now known as seya_pegasus | 11:19 | |
| *** seya_pegasus is now known as paulormg | 11:19 | |
| *** bepernoot has joined #openstack-dev | 11:20 | |
| *** oneiroi has quit IRC | 11:40 | |
| *** oneiroi has joined #openstack-dev | 11:41 | |
| *** rkukura has quit IRC | 11:45 | |
| *** dbuzz has joined #openstack-dev | 11:54 | |
| *** dbuzz has joined #openstack-dev | 11:54 | |
| *** oneiroi has quit IRC | 11:57 | |
| *** dbuzz is now known as oneiroi | 11:57 | |
| *** rods has joined #openstack-dev | 11:58 | |
| *** bsza has joined #openstack-dev | 12:07 | |
| *** halfss has quit IRC | 12:08 | |
| *** vizsla has joined #openstack-dev | 12:10 | |
| *** zykes-_ has quit IRC | 12:13 | |
| *** zykes- has joined #openstack-dev | 12:15 | |
| *** markvoelker has joined #openstack-dev | 12:18 | |
| *** zykes- has quit IRC | 12:20 | |
| *** rods has quit IRC | 12:22 | |
| *** sandywalsh has quit IRC | 12:23 | |
| zul | Daviey: what if there was something like export NO_GIT=1 ./run_tests.sh -N for keystone tests | 12:31 |
| *** armaan has left #openstack-dev | 12:31 | |
| Daviey | zul: heckj has something up his sleeve.. Personally, i feel a raw -N should not grab external deps. If -N isn't used, then injecting stuff into the virtualenv is free-forall IMO. | 12:33 |
| zul | Daviey: okies | 12:34 |
| *** sandywalsh has joined #openstack-dev | 12:36 | |
| *** zykes- has joined #openstack-dev | 12:39 | |
| *** darraghb has joined #openstack-dev | 12:41 | |
| *** andrewsmedina has joined #openstack-dev | 12:46 | |
| *** sandywalsh has quit IRC | 12:54 | |
| *** andrewsmedina has quit IRC | 12:54 | |
| *** dprince has joined #openstack-dev | 13:07 | |
| *** sandywalsh has joined #openstack-dev | 13:07 | |
| *** littleidea has joined #openstack-dev | 13:10 | |
| *** maploin has joined #openstack-dev | 13:12 | |
| *** maploin has quit IRC | 13:12 | |
| *** maploin has joined #openstack-dev | 13:12 | |
| *** sniperd has joined #openstack-dev | 13:15 | |
| *** crobinso has joined #openstack-dev | 13:18 | |
| *** andrewsmedina has joined #openstack-dev | 13:19 | |
| *** deshantm has joined #openstack-dev | 13:22 | |
| *** littleidea has quit IRC | 13:24 | |
| *** vizsla has quit IRC | 13:25 | |
| *** markmc has joined #openstack-dev | 13:28 | |
| *** roge has joined #openstack-dev | 13:38 | |
| *** sniperd has quit IRC | 13:39 | |
| *** creiht has quit IRC | 13:44 | |
| *** creiht has joined #openstack-dev | 13:44 | |
| *** ChanServ sets mode: +v creiht | 13:44 | |
| *** sniperd has joined #openstack-dev | 13:52 | |
| *** PotHix has joined #openstack-dev | 13:57 | |
| *** littleidea has joined #openstack-dev | 14:10 | |
| *** sniperd has quit IRC | 14:13 | |
| *** rkukura has joined #openstack-dev | 14:16 | |
| *** lts has joined #openstack-dev | 14:19 | |
| *** LinuxJedi has quit IRC | 14:20 | |
| *** mdomsch has joined #openstack-dev | 14:20 | |
| *** LinuxJedi has joined #openstack-dev | 14:24 | |
| *** sniperd has joined #openstack-dev | 14:29 | |
| *** sniperd has quit IRC | 14:30 | |
| *** LinuxJedi_ has joined #openstack-dev | 14:32 | |
| *** LinuxJedi has quit IRC | 14:32 | |
| *** sniperd has joined #openstack-dev | 14:33 | |
| *** vricci has joined #openstack-dev | 14:35 | |
| *** dayou has quit IRC | 14:40 | |
| *** hub_cap has joined #openstack-dev | 14:50 | |
| *** LinuxJedi has joined #openstack-dev | 14:51 | |
| *** LinuxJedi_ has quit IRC | 14:52 | |
| *** ecarlin has joined #openstack-dev | 14:53 | |
| *** ecarlin_ has joined #openstack-dev | 14:54 | |
| *** ecarlin has quit IRC | 14:54 | |
| *** ecarlin_ is now known as ecarlin | 14:54 | |
| *** ecarlin has quit IRC | 14:55 | |
| *** ecarlin has joined #openstack-dev | 14:55 | |
| *** ecarlin has quit IRC | 14:55 | |
| *** dtroyer has joined #openstack-dev | 14:57 | |
| *** LinuxJedi_ has joined #openstack-dev | 14:59 | |
| *** sniperd has quit IRC | 15:01 | |
| *** sniperd has joined #openstack-dev | 15:01 | |
| *** LinuxJedi has quit IRC | 15:02 | |
| *** ecarlin has joined #openstack-dev | 15:03 | |
| *** LinuxJedi_ is now known as LinuxJedi | 15:04 | |
| *** ecarlin has quit IRC | 15:09 | |
| *** vricci has left #openstack-dev | 15:09 | |
| *** ecarlin has joined #openstack-dev | 15:09 | |
| *** lts has quit IRC | 15:09 | |
| *** rnirmal has joined #openstack-dev | 15:10 | |
| *** andrewsmedina has quit IRC | 15:10 | |
| *** sniperd has quit IRC | 15:13 | |
| *** sniperd has joined #openstack-dev | 15:14 | |
| *** rkukura has left #openstack-dev | 15:17 | |
| *** andrewsmedina has joined #openstack-dev | 15:19 | |
| *** rkukura has joined #openstack-dev | 15:29 | |
| *** RobertLaptop has joined #openstack-dev | 15:29 | |
| *** dolphm has joined #openstack-dev | 15:29 | |
| *** RobertLaptop has left #openstack-dev | 15:30 | |
| *** hhoover has joined #openstack-dev | 15:31 | |
| *** negronjl has joined #openstack-dev | 15:33 | |
| *** lts has joined #openstack-dev | 15:36 | |
| *** GheRivero_ has joined #openstack-dev | 15:37 | |
| *** Remco_ has joined #openstack-dev | 15:37 | |
| *** journeeman has joined #openstack-dev | 15:37 | |
| *** zaitcev has joined #openstack-dev | 15:38 | |
| *** bencherian has joined #openstack-dev | 15:39 | |
| *** PotHix has quit IRC | 15:40 | |
| *** hhoover has left #openstack-dev | 15:42 | |
| *** ecarlin has quit IRC | 15:45 | |
| *** Mandell has joined #openstack-dev | 15:50 | |
| *** Mandell_ has joined #openstack-dev | 15:51 | |
| *** Mandell has quit IRC | 15:51 | |
| *** ayoung has joined #openstack-dev | 15:51 | |
| *** PotHix has joined #openstack-dev | 15:53 | |
| *** hub-cap has joined #openstack-dev | 15:56 | |
| *** hub-cap has quit IRC | 15:57 | |
| *** kbringard has joined #openstack-dev | 15:57 | |
| *** hub-cap has joined #openstack-dev | 15:58 | |
| *** hub_cap has quit IRC | 16:00 | |
| *** hub-cap is now known as hub_cap | 16:00 | |
| *** ayoung has quit IRC | 16:06 | |
| *** ayoung has joined #openstack-dev | 16:06 | |
| maploin | can the horizon test suite be run against a deployed server? (instead of the one running from virtualenv) If so how? | 16:07 |
| *** cp16net has joined #openstack-dev | 16:09 | |
| *** hhoover1 has joined #openstack-dev | 16:11 | |
| *** dtroyer has quit IRC | 16:14 | |
| *** oneiroi has quit IRC | 16:15 | |
| *** hhoover1 has left #openstack-dev | 16:15 | |
| *** oneiroi has joined #openstack-dev | 16:15 | |
| *** zzed has joined #openstack-dev | 16:16 | |
| *** hhoover has joined #openstack-dev | 16:16 | |
| GheRivero_ | are the sample flags in nova.conf.sample the defaults is none is set? | 16:16 |
| *** hhoover has left #openstack-dev | 16:17 | |
| *** reidrac has quit IRC | 16:22 | |
| *** ecarlin has joined #openstack-dev | 16:22 | |
| *** jk0 has quit IRC | 16:23 | |
| *** jk0 has joined #openstack-dev | 16:23 | |
| *** jk0 has joined #openstack-dev | 16:23 | |
| *** ChanServ sets mode: +v jk0 | 16:23 | |
| *** dtroyer has joined #openstack-dev | 16:27 | |
| *** hhoover has joined #openstack-dev | 16:28 | |
| *** shang has quit IRC | 16:30 | |
| *** novas0x2a1 has joined #openstack-dev | 16:32 | |
| *** heckj has joined #openstack-dev | 16:34 | |
| *** jdg has joined #openstack-dev | 16:34 | |
| *** vladimir3p has joined #openstack-dev | 16:34 | |
| *** ecarlin has quit IRC | 16:34 | |
| *** novas0x2a has quit IRC | 16:35 | |
| *** oneiroi has quit IRC | 16:35 | |
| *** Remco_ has quit IRC | 16:40 | |
| *** danwent has joined #openstack-dev | 16:46 | |
| *** gyee has joined #openstack-dev | 16:48 | |
| *** dprince has quit IRC | 16:51 | |
| *** vincentricci has joined #openstack-dev | 16:52 | |
| *** Mkenneth has left #openstack-dev | 16:54 | |
| blamar | bcwaldon_away: when you're back, added test to validate id is uuid-like to https://review.openstack.org/#change,4834, logic was already in registry | 16:54 |
| *** hhoover has left #openstack-dev | 16:56 | |
| *** martine has joined #openstack-dev | 16:58 | |
| *** tryggvil has joined #openstack-dev | 17:02 | |
| *** maplebed has joined #openstack-dev | 17:05 | |
| *** maploin has quit IRC | 17:09 | |
| *** ecarlin has joined #openstack-dev | 17:14 | |
| *** dubsquared has joined #openstack-dev | 17:15 | |
| *** armaan has joined #openstack-dev | 17:16 | |
| *** cp16net has quit IRC | 17:20 | |
| *** cp16net has joined #openstack-dev | 17:20 | |
| *** ecarlin has quit IRC | 17:20 | |
| *** x86brandon has joined #openstack-dev | 17:22 | |
| *** armaan has left #openstack-dev | 17:25 | |
| kbringard | anyone know off the top of their head what would be creating this query? | 17:25 |
| kbringard | http://paste.openstack.org/show/7155/ | 17:25 |
| kbringard | it's running constantly | 17:26 |
| *** gkotton has quit IRC | 17:28 | |
| heckj | kbringard: not sure, but it's clearly something that SQLAlchemy is creating. If it's running constantly, I'd expect its some sort of repeated service check. What's running in your openstack world right now? The whole kit, or just some parts? | 17:33 |
| *** derekh has quit IRC | 17:33 | |
| kbringard | whole kit minus keystone and horizon | 17:33 |
| kbringard | it looks like it's the compute nodes doing that check, though | 17:34 |
| kbringard | err, query | 17:34 |
| kbringard | this is diablo vanilla, too, so it could just be the ephed up db con pooling stuff | 17:34 |
| *** spiffxp has joined #openstack-dev | 17:36 | |
| heckj | my suspicion was a repeating status check from horizon, but if it's not running, that's right out. | 17:38 |
| kbringard | yea, it's definitely coming from compute | 17:38 |
| *** ecarlin has joined #openstack-dev | 17:41 | |
| jeblair | pvo: we're getting a bunch of "NucleusClientException: 500 Error"s when launching devstack nodes; https://jenkins.openstack.org/job/devstack-launch-vms/10511/console | 17:43 |
| pvo | jeblair: standby | 17:44 |
| *** dalang has joined #openstack-dev | 17:45 | |
| sleepsonzzz | kbrigand, is there a corresponding update? I believe services generally report their state to the db every so often. That update call is likely preceded by a service_get. | 17:45 |
| *** bencherian has quit IRC | 17:46 | |
| sleepsonzzz | kbringard ^ looks like the default is to report state every 10 seconds, and that does involve service_get | 17:48 |
| kbringard | sleepsonzzz: ah, that would probably be it | 17:48 |
| kbringard | I have 72 compute nodes, so it would make sense that it's scrolling | 17:48 |
| *** dprince has joined #openstack-dev | 17:49 | |
| kbringard | Query_time: 0.000534 Lock_time: 0.000158 Rows_sent: 1 Rows_examined: 53 | 17:49 |
| kbringard | it's not like it's a huge query | 17:49 |
| kbringard | just a crapton of them | 17:49 |
| kbringard | services check in via the db, correct? not via the queue? | 17:50 |
| kbringard | so, if a service is marked as :-) I can assume the node is able to talk to the DB? | 17:51 |
| *** mszilagyi has joined #openstack-dev | 17:51 | |
| *** journeeman has quit IRC | 17:52 | |
| pvo | jeblair: I was able to build an 11.10 image that pinged. are you able to kick that job back off? | 17:54 |
| jeblair | pvo: yeah, still fails: https://jenkins.openstack.org/job/devstack-launch-vms/10513/console | 17:55 |
| *** gkotton has joined #openstack-dev | 17:56 | |
| sleepsonzzz | kbringard - that's right, they talk directly to db when they update | 17:56 |
| *** bencherian has joined #openstack-dev | 17:56 | |
| kbringard | sleepsonzzz: sweet, cool, 's what I thought | 17:57 |
| kbringard | thanks | 17:57 |
| kbringard | :-) | 17:57 |
| *** dneary has quit IRC | 17:57 | |
| *** vladimir3p has quit IRC | 17:58 | |
| *** jog0 has joined #openstack-dev | 18:02 | |
| pvo | jeblair: ops is checking. | 18:02 |
| *** jog0 has left #openstack-dev | 18:02 | |
| pvo | will be back shortly. | 18:02 |
| jeblair | pvo: awesome, thanks! | 18:02 |
| *** ecarlin has quit IRC | 18:03 | |
| *** reed_ has joined #openstack-dev | 18:04 | |
| * zigo just uploaded XCP 1.3.2-2 in Debian. | 18:05 | |
| *** armaan has joined #openstack-dev | 18:07 | |
| *** reed_ is now known as reed | 18:08 | |
| *** cp16net_ has joined #openstack-dev | 18:08 | |
| *** bencherian has quit IRC | 18:09 | |
| *** cp16net has quit IRC | 18:09 | |
| *** cp16net_ is now known as cp16net | 18:09 | |
| *** bencherian has joined #openstack-dev | 18:09 | |
| *** jdurgin has joined #openstack-dev | 18:10 | |
| *** AlanClark has joined #openstack-dev | 18:12 | |
| *** bencherian has quit IRC | 18:13 | |
| *** armaan has left #openstack-dev | 18:16 | |
| *** darraghb has quit IRC | 18:17 | |
| *** rnirmal has quit IRC | 18:17 | |
| *** dolphm has quit IRC | 18:17 | |
| *** oneiroi has joined #openstack-dev | 18:17 | |
| *** adjohn has joined #openstack-dev | 18:18 | |
| *** _adjohn has joined #openstack-dev | 18:20 | |
| *** adjohn has quit IRC | 18:23 | |
| *** _adjohn is now known as adjohn | 18:23 | |
| *** shevek_ has quit IRC | 18:24 | |
| *** zigo has quit IRC | 18:31 | |
| *** dolphm has joined #openstack-dev | 18:32 | |
| *** dolphm has quit IRC | 18:34 | |
| *** dprince has quit IRC | 18:39 | |
| ayoung | heckj, I've come up with the first part of documenting the LDAP setup : http://fpaste.org/jBHY/ Should I submit this as a patch? If so, can I put it under keystone/bin? | 18:40 |
| *** mattstep has quit IRC | 18:40 | |
| *** mattstep has joined #openstack-dev | 18:41 | |
| heckj | ayoung: that would be better under a contrib directory or such... | 18:42 |
| *** camm has quit IRC | 18:43 | |
| heckj | around: looking around in there now to see where might be best | 18:43 |
| ayoung | heckj, do we have one? | 18:43 |
| heckj | ayoung: apparently, not yet... | 18:43 |
| heckj | ayoung: let's follow Nova's pattern and make a top level contrib/ directory - put it in there. | 18:43 |
| *** camm has joined #openstack-dev | 18:44 | |
| ayoung | heckj, it is trivial enough that I think I'd like to hold off doing it just for this one little file | 18:45 |
| ayoung | I'd almost rather put it on the keystone.openstack Wiki | 18:45 |
| ayoung | But I need access, I think? | 18:45 |
| ayoung | heckj, we now have 2 +1s on https://review.openstack.org/#change,4659 and I gave it a pretty thorough review. you or termie can push to master. | 18:46 |
| *** bepernoot has quit IRC | 18:48 | |
| *** novas0x2a|laptop has joined #openstack-dev | 18:52 | |
| heckj | ayoung: wiki.openstack.org or keystone.openstack.org? The second is generated from the ReST text files in keystone/doc. The first should be accessible using your launchpad account. | 18:54 |
| ayoung | heckj, Ah.. so I should be creating patches for Doc...OK, that is what I will do | 18:55 |
| ayoung | I want them keystone specific for now | 18:55 |
| heckj | ayoung: Adding a helper script to contrib/ in the keystone project would also be useful if you want to hit that route. | 18:55 |
| ayoung | heckj, will do | 18:55 |
| termie | ayoung: re: that patch, i've still got a couple things to fix in it that you mentioned | 18:55 |
| termie | ayoung: but i'll do that now | 18:56 |
| ayoung | termie, OK. None of them were show stoppers as I recall | 18:56 |
| heckj | termie: I was just going to sit down and re-read through it, but i'll wait until you've uploaded a new patch. | 18:56 |
| *** hashar has joined #openstack-dev | 18:57 | |
| *** mnewby has joined #openstack-dev | 19:01 | |
| *** bepernoot has joined #openstack-dev | 19:01 | |
| mnewby | sa | 19:01 |
| termie | heckj: running the tests now | 19:02 |
| *** novas0x2a|laptop has quit IRC | 19:02 | |
| *** novas0x2a|laptop has joined #openstack-dev | 19:02 | |
| *** armaan has joined #openstack-dev | 19:02 | |
| *** kbringard has quit IRC | 19:05 | |
| *** kbringard has joined #openstack-dev | 19:06 | |
| termie | heckj: did keystoneclient do a new update? | 19:06 |
| termie | heckj: the bulk of my authz related tests just started failing :/ | 19:08 |
| termie | ... rebasing on master to see if that helps | 19:08 |
| pvo | jeblair: I see you're green again | 19:10 |
| *** eglynn_ has joined #openstack-dev | 19:12 | |
| termie | heckj: looks like maybe a recent update to keystoneclient master now requires some additional stuff in the auth calls? | 19:12 |
| *** Remco_ has joined #openstack-dev | 19:13 | |
| termie | heckj: although it doesn't particularly look like anythign does that | 19:13 |
| *** eglynn has quit IRC | 19:14 | |
| termie | heckj: appears to throw an authorization error if it doesn't get a service catalog? is that new? hmm | 19:16 |
| *** hashar is now known as hasharCall | 19:16 | |
| termie | heckj: aha, it was my old nemesis, bcwaldon | 19:17 |
| termie | https://github.com/openstack/python-keystoneclient/commit/0cc939c9040d9c35865603cc6806deaf6205004e | 19:17 |
| termie | specifically https://github.com/openstack/python-keystoneclient/commit/0cc939c9040d9c35865603cc6806deaf6205004e#L1L114 | 19:18 |
| termie | though it doesn't seeem like there is any reason why we wouldn't be returning that | 19:19 |
| *** vizsla has joined #openstack-dev | 19:21 | |
| termie | ah i see, it is because you don't get a catalog if you don't provide a tenant | 19:22 |
| *** Glacee has joined #openstack-dev | 19:24 | |
| *** vizsla has quit IRC | 19:25 | |
| *** Ryan_Lane has joined #openstack-dev | 19:25 | |
| termie | ... or not, it appears to be pulling it from the response of the auth call which has the keys it is looking for... | 19:29 |
| termie | looks like maybe they changed the exception they throw | 19:29 |
| jeblair | pvo: yeah, looking good now, thanks! | 19:39 |
| *** oneiroi has quit IRC | 19:42 | |
| termie | jeblair: i seem to be getting a difference between what is on github master and what is on gerrit master for python-keystoneclient | 19:42 |
| termie | jeblair: my checkout is from url = https://review.openstack.org/p/openstack/python-keystoneclient.git and a git pull says up-to-date | 19:43 |
| termie | jeblair: but the lines around 116 in client.py are different from github: https://github.com/openstack/python-keystoneclient/blob/master/keystoneclient/v2_0/client.py#L116 | 19:43 |
| jeblair | termie: it may be because of github de-authorizing all the ssh keys after the github compromise | 19:43 |
| termie | ah | 19:43 |
| termie | i reauthorized jenkins | 19:43 |
| jeblair | termie: i re-authorized the keys about 30 mins ago | 19:43 |
| termie | can we get it to re-push or something | 19:44 |
| jeblair | yeah, i'll have it re-push all the projects. just a sec. | 19:44 |
| jeblair | termie: better now? | 19:47 |
| termie | jeblair: yup :p now the broken code shows up so i can fix it :) | 19:48 |
| termie | jeblair: thanks | 19:48 |
| jeblair | yay! np. | 19:48 |
| *** ecarlin has joined #openstack-dev | 19:53 | |
| *** jog0 has joined #openstack-dev | 19:57 | |
| *** gkotton has quit IRC | 19:58 | |
| *** hasharCall is now known as hashar | 20:00 | |
| maplebed | anybody know the history of logging in swift? I'm curious if there's a reason (worth maintaining) that the log format for the container and object servers are different from the proxy server. | 20:00 |
| termie | heckj: keystone is blocked waiting for this: https://review.openstack.org/#change,5037 | 20:01 |
| cloudfly | i think there is definitely some room for log format standardization | 20:01 |
| maplebed | for example, the object server uses the date format [07/Mar/2012:19:59:11 +0000] whereas the proxy server uses the format 07/Mar/2012/19/59/08 | 20:01 |
| * heckj looks | 20:01 | |
| maplebed | another, the object server has the HTTP query in "" marks, wheresa the proxy doesn't. | 20:01 |
| heckj | termie - looks good to me | 20:03 |
| *** dolphm has joined #openstack-dev | 20:03 | |
| *** hashar is now known as hasharCall | 20:04 | |
| cloudfly | i figure at some point a project will pop up to aggregate all the logs and perform heuristic analysis on them | 20:04 |
| cloudfly | heck i may just work on that this week if nothing else gets in my way | 20:04 |
| maplebed | I've written some stuff to get stats into ganglia (for the proxy) | 20:05 |
| maplebed | I was just hoping to use the same thing for the storage nodes. | 20:05 |
| maplebed | (http://ganglia.wikimedia.org/latest/?r=20min&cs=&ce=&m=&tab=v&vn=swift+fronted+proxies) | 20:05 |
| cloudfly | not a huge fan of ganglia to be honest | 20:06 |
| maplebed | it has its benefits. | 20:06 |
| *** vish1 has quit IRC | 20:06 | |
| cloudfly | yes, but it gets ugly over time | 20:06 |
| maplebed | but yeah, there's plenty of other stuff out there. | 20:06 |
| *** ecarlin_ has joined #openstack-dev | 20:06 | |
| *** vishy has joined #openstack-dev | 20:07 | |
| termie | heckj: who else is keystoneclient core? i'd liek somebody else to read my rant :p | 20:08 |
| heckj | gabrielhurley | 20:08 |
| *** vishy has quit IRC | 20:09 | |
| *** ecarlin has quit IRC | 20:09 | |
| *** ecarlin_ is now known as ecarlin | 20:09 | |
| *** vishy has joined #openstack-dev | 20:10 | |
| termie | heckj: don't you work with those guys and make them get online? | 20:10 |
| termie | or are they all off to pycon stuffs? | 20:10 |
| vishy | tr3buchet: so turns out we didn't vif and address in the teardown call | 20:10 |
| heckj | termie - soon, but not yet. I'll get him to look | 20:11 |
| *** adjohn has quit IRC | 20:11 | |
| heckj | termie: sorry, was in meeting | 20:11 |
| tr3buchet | vishy: didn't what? | 20:11 |
| vishy | tr3buchet: I'm thinking to leave it in in case a future driver would need it | 20:11 |
| vishy | tr3buchet: * didn't need | 20:11 |
| tr3buchet | might as well | 20:11 |
| tr3buchet | won't hurt anything now | 20:11 |
| tr3buchet | i don't see how you wouldn't need the address | 20:12 |
| tr3buchet | i didn't look down at the bottom though | 20:12 |
| tr3buchet | vishy: ^ | 20:12 |
| vishy | has anyone seen TypeError: __init__() takes exactly 3 arguments (2 given) | 20:13 |
| vishy | from nose when trying to add a failure? | 20:13 |
| vishy | that is new to me | 20:13 |
| heckj | termie: he's away (lunch?) will grab him as soon as I can to review | 20:13 |
| vishy | tr3buchet: i have a new patch coming up | 20:13 |
| vishy | tr3buchet: i'm just trying to fix a test | 20:13 |
| tr3buchet | kk | 20:13 |
| tr3buchet | just let me know | 20:13 |
| markwash | termie: i'm getting unit test failures in keystone master, is that what your change is trying to fix | 20:13 |
| *** markmc has quit IRC | 20:14 | |
| termie | markwash: yea | 20:15 |
| *** bsza has quit IRC | 20:18 | |
| *** ecarlin has quit IRC | 20:19 | |
| vishy | tr3buchet: see here https://review.openstack.org/5038 | 20:21 |
| vishy | tr3buchet: note we needed to call update_dhcp, not release_dhcp | 20:21 |
| tr3buchet | ah ok | 20:21 |
| vishy | and that is called for the whole network, not the vif/address | 20:21 |
| tr3buchet | i see | 20:21 |
| *** oneiroi has joined #openstack-dev | 20:24 | |
| *** oneiroi has joined #openstack-dev | 20:24 | |
| *** oneiroi has quit IRC | 20:24 | |
| *** oneiroi has joined #openstack-dev | 20:25 | |
| *** bepernoot has quit IRC | 20:26 | |
| *** ecarlin has joined #openstack-dev | 20:28 | |
| *** sandywalsh has quit IRC | 20:31 | |
| *** ches has quit IRC | 20:31 | |
| *** bencherian has joined #openstack-dev | 20:31 | |
| *** ches has joined #openstack-dev | 20:33 | |
| andrewbogott | I think I know the answer to this already, but... are there folsom branches anywhere that people are working on? Or should I just hold back my folsom features until essex is shipped off? | 20:41 |
| *** sandywalsh has joined #openstack-dev | 20:43 | |
| *** cp16net has quit IRC | 20:52 | |
| *** cp16net has joined #openstack-dev | 20:52 | |
| *** ecarlin has quit IRC | 20:56 | |
| *** paulormg has quit IRC | 20:56 | |
| *** ecarlin has joined #openstack-dev | 20:56 | |
| eglynn_ | heckj: quick keystone auth question? | 21:00 |
| devcamcar | termie; https://review.openstack.org/#change,4880 updated | 21:00 |
| devcamcar | termie: and filed this related bug https://bugs.launchpad.net/keystone/+bug/949373 | 21:01 |
| uvirtbot | Launchpad bug 949373 in keystone "Default configuration values are ignored" [Undecided,New] | 21:01 |
| *** adjohn has joined #openstack-dev | 21:01 | |
| *** lloydde has joined #openstack-dev | 21:02 | |
| *** deshantm has quit IRC | 21:02 | |
| eglynn_ | heckj/termie: what's the mediation between the AWS-style auth and keystone? or ne'er the twain shall meet? | 21:02 |
| eglynn_ | say I have nova and glance configured to use the keystone auth strategy, and I call nova image-list ... | 21:02 |
| eglynn_ | the keystone auth token is propagated from nova to glance, happy days! | 21:02 |
| eglynn_ | but if I call into the EC2 API via euca-describe-images, how would the credentials propagation work in that case? | 21:03 |
| eglynn_ | (it doesn't seem to work BTW, so I'm wondering how it /should/ work) | 21:03 |
| *** hasharCall is now known as hashar | 21:04 | |
| termie | eglynn_: i am about to rub to lunch, but ec2 auth in nova verifies teh request with keystone | 21:05 |
| eglynn_ | termie: a-ha, so the keystone auth token should be available for propogation in that case, cool | 21:05 |
| apevec | eglynn_, does ec2 pipeline include keystone ? | 21:05 |
| eglynn_ | apevec: bingo! it doesn't in my config, I'll try that now ... | 21:07 |
| *** mnewby has quit IRC | 21:07 | |
| *** bencherian has quit IRC | 21:07 | |
| termie | eglynn_: vishy could tell you more | 21:07 |
| *** mnewby has joined #openstack-dev | 21:08 | |
| *** mnewby has quit IRC | 21:08 | |
| *** mnewby has joined #openstack-dev | 21:08 | |
| *** dprince has joined #openstack-dev | 21:16 | |
| *** GheRivero_ has quit IRC | 21:17 | |
| *** ecarlin has quit IRC | 21:17 | |
| *** ecarlin has joined #openstack-dev | 21:17 | |
| adam_g | hmm. is nova-compute involved in the keystone auth flow at all when it requests an image from the glance server? for the first time ive run into an issue where both 'nova image-list' / 'euca-describe-images' both complete okay in terms of keystone token auth, but when running an instance, the request from nova-compute is rejected by glance-api with 'Invalid user token - rejecting request' | 21:21 |
| *** ecarlin has quit IRC | 21:28 | |
| *** jaypipes has joined #openstack-dev | 21:28 | |
| *** ecarlin has joined #openstack-dev | 21:28 | |
| *** reed has quit IRC | 21:30 | |
| *** hashar has quit IRC | 21:32 | |
| mattstep | hmm, I need an approver and verifier for a change, anyone up for it? just some updates to --help for nova clis from the docday yesterday | 21:35 |
| annegentle | mattstep: I'm not nova core but I'll take a look! | 21:35 |
| mattstep | https://review.openstack.org/#change,5000 | 21:36 |
| mattstep | annegentle : thanks | 21:36 |
| *** markvoelker has quit IRC | 21:40 | |
| *** hashar has joined #openstack-dev | 21:41 | |
| *** glenc_ has joined #openstack-dev | 21:44 | |
| *** glenc has quit IRC | 21:46 | |
| *** reed has joined #openstack-dev | 21:46 | |
| *** reed_ has joined #openstack-dev | 21:52 | |
| *** ecarlin has quit IRC | 21:52 | |
| cp16net | python-novaclient looks some what incomplete anyone here to answer a question about it? | 21:55 |
| *** reed has quit IRC | 21:55 | |
| cp16net | I have noticed a few cmds like servers.delete() reboot() update() do not return anything | 21:56 |
| cp16net | so any call being made with this client assumes that it completed successfully | 21:56 |
| cp16net | I would expect a return status code or something to see if the command completed successfully or not | 21:57 |
| *** anotherjesse_zz is now known as anotherjesse | 21:57 | |
| *** roge has quit IRC | 21:58 | |
| *** sniperd has quit IRC | 21:58 | |
| *** mszilagyi has quit IRC | 22:00 | |
| *** mszilagyi has joined #openstack-dev | 22:02 | |
| comstud | cp16net: an exception would be raised if they failed | 22:02 |
| comstud | i mean, if the request failed | 22:03 |
| comstud | otherwise, the actions happen in the background in nova... so there's no way for novaclient to know whether or not they truly succeeded | 22:03 |
| cloudfly | yeah at that point it enters the rabbit queue | 22:04 |
| *** hashar has quit IRC | 22:05 | |
| *** dubsquared has quit IRC | 22:06 | |
| *** dubsquared has joined #openstack-dev | 22:07 | |
| cp16net | comstud: ok thanks for that i didnt realize an exception would be raised. | 22:10 |
| *** armaan has left #openstack-dev | 22:11 | |
| comstud | cp16net: np | 22:11 |
| heckj | termie: I flipped the approval bit on https://review.openstack.org/#change,5037 | 22:13 |
| *** martine has quit IRC | 22:13 | |
| *** Remco_ has quit IRC | 22:14 | |
| *** dolphm has quit IRC | 22:15 | |
| *** Remco_ has joined #openstack-dev | 22:15 | |
| *** dolphm has joined #openstack-dev | 22:16 | |
| *** Remco_ has quit IRC | 22:20 | |
| anotherjesse | johan_-_: updated https://review.openstack.org/#change,5042 and provided numbers | 22:21 |
| heckj | anotherjesse: would you give a review on https://review.openstack.org/5046, needed to enable the bind_host patch for keystone. | 22:22 |
| anotherjesse | heckj: why don't we make bind_host optional / have a default of 0.0.0.0? | 22:23 |
| heckj | anotherjesse: it doesn't appear to be wired up in keystone correctly at the moment, but that's what Devin's patch enables. It fails because it's not the config already because the "default" keyword isn't working (logged as a separate bug against Keystone earlier today) | 22:24 |
| anotherjesse | k | 22:24 |
| anotherjesse | heckj: is the fix default keyword going to be a requirement for RC / release? | 22:24 |
| *** pixelbeat has quit IRC | 22:24 | |
| heckj | Are you not going to approve that if it isn't :-) | 22:25 |
| heckj | (honestly haven't thought it about, but it seems very likely) | 22:25 |
| anotherjesse | heckj: if we aren't going to require fixing default then we have required bugs to update docs about the requiredness of bind_host | 22:25 |
| anotherjesse | it seems easier to fix default | 22:25 |
| *** dolphm has quit IRC | 22:25 | |
| heckj | anotherjesse: agreed. I'll set it as such | 22:26 |
| anotherjesse | deal | 22:26 |
| anotherjesse | anyone using images that hit EC2 metadata service might want to help review https://review.openstack.org/#change,5042 | 22:27 |
| anotherjesse | it vastly improves performance for me | 22:27 |
| vishy | heckj, termie: is it possible for some special user in keystone to get a token for a tenant even if they aren't a member of the tenant? | 22:27 |
| cloudfly | i've been asking for service users forever. | 22:28 |
| anotherjesse | cloudfly: the latest devstack uses service users now - but they aren't special | 22:28 |
| anotherjesse | too much to change for essex rc | 22:28 |
| heckj | cloudfly: they're in now | 22:28 |
| *** pixelbeat has joined #openstack-dev | 22:28 | |
| heckj | (what anotherjesse said) | 22:28 |
| vishy | heckj: here is the problem I'm trying to solve | 22:29 |
| cloudfly | so i can see | 22:29 |
| vishy | heckj: As an admin, I want to trigger a resize for a user | 22:29 |
| vishy | heckj: or a block-migration command | 22:29 |
| vishy | heckj: at some point that command has to interact with external services, in this case glance | 22:29 |
| *** andrewsmedina has quit IRC | 22:30 | |
| vishy | heckj: so if i need to access that tenant's images, I need a token that is valid for that tenant. | 22:30 |
| heckj | vishy: totally makes sense - I don't think there's a path for that today (at least not a common one). There might be a way to do it, but I think we're missing that at the moment. | 22:31 |
| rmk | Ugh. My vnc console connections (using wsproxy) always fail the first time and succeed the second. | 22:33 |
| rmk | Diablo Stable. | 22:33 |
| vishy | heckj: this is blocking (ha!) block-migrate right now | 22:33 |
| vishy | and I don't have a good solution for it | 22:33 |
| heckj | vishy: let me huddle up with termie and see what we can come up with | 22:33 |
| anotherjesse | rmk: there were issues with it in diablo so the vnc feature was significantly revamped by sleepsonzzz in essex - including adding support for xen | 22:33 |
| vishy | heckj: so we need some sort of workaround | 22:34 |
| *** zzed has quit IRC | 22:34 | |
| heckj | vishy: I'm going to make a bug about this - also just added it to http://wiki.openstack.org/KeystoneUseCases - to track. Just a sec | 22:34 |
| anotherjesse | heckj: I think vishy is saying we might need it in essex | 22:34 |
| vishy | heckj: it is extra tricky because I need to be able to auth as an admin | 22:34 |
| vishy | heckj: but then somehow switch to user context when communicating with other services | 22:35 |
| heckj | yeah, I definitely got that. | 22:35 |
| rmk | anotherjesse: Definitely works more reliably in essex -- though I am using the wsproxy stuff sleepsonzzz wrote | 22:35 |
| vishy | heckj: so block-migrate is an admin only api | 22:35 |
| rmk | As in, using it on Diablo | 22:35 |
| anotherjesse | rmk: sleepsonzzz is the best person to ask - he is currently afk | 22:35 |
| heckj | anotherjesse: ^^ was meant for you - that I got we need this for Essex. | 22:35 |
| anotherjesse | heckj: :) | 22:35 |
| rmk | Is there a migration path from Diablo to E4? I'm about ready to make the leap right now. | 22:36 |
| rmk | Sorry I don't meant to distract from more important convos. | 22:36 |
| heckj | vishy: I don't have a clear immediate "how we could" plan of attack in my head, but I'll dig into this tonight and see what I can figure out. Termie's a bit clearer on the paths underneath, so I'm going to need to lean on him for some of this. | 22:36 |
| anotherjesse | heckj: I'm going to stay out of this one unless you yell for more hands .. going to focus on the memcache for auth_tokens | 22:37 |
| heckj | anotherjesse: KK - thanks | 22:37 |
| heckj | vishy: did we have a way we were doing this previously? | 22:40 |
| heckj | vishy: https://bugs.launchpad.net/keystone/+bug/949467 | 22:40 |
| uvirtbot | Launchpad bug 949467 in keystone "need a way to get a token for acting on behalf of a user to another service" [High,Confirmed] | 22:40 |
| *** dolphm has joined #openstack-dev | 22:43 | |
| *** cp16net has quit IRC | 22:44 | |
| jdg | vishy: I just noticed something on bug 928819 (Precis Launching instance fails) | 22:45 |
| uvirtbot | Launchpad bug 928819 in nova "Launching instance fails in Precise A2 - network/manager.py too many values to unpack" [High,Confirmed] https://launchpad.net/bugs/928819 | 22:45 |
| jdg | The ec2 cloud tests are actually calling the function that was changed. | 22:46 |
| jdg | In other words there is "some" unit test coverage | 22:46 |
| *** hub_cap has quit IRC | 22:48 | |
| heckj | termie: ping | 22:53 |
| *** dolphm has quit IRC | 22:55 | |
| *** rkukura has quit IRC | 23:00 | |
| *** lts has quit IRC | 23:01 | |
| *** apevec has quit IRC | 23:02 | |
| termie | heckj: heyo | 23:07 |
| heckj | termie: howdy sir - vishy brought up and interesting "how do I" problem - wanted your thoughts on it: https://bugs.launchpad.net/keystone/+bug/949467 | 23:08 |
| uvirtbot | Launchpad bug 949467 in keystone "need a way to get a token for acting on behalf of a user to another service" [High,Confirmed] | 23:08 |
| heckj | (also in scrollback if you want to just read upwards) | 23:08 |
| *** Glacee has quit IRC | 23:08 | |
| termie | heckj, vishy: the token they gave you when you got the original request is valid for that user | 23:09 |
| vishy | termie: this is not for a request from the user | 23:09 |
| vishy | termie: specific example is i need to migrate a users vm | 23:09 |
| heckj | termie: the idea is an "admin" user acting on behalf of an existing user as I understand it. | 23:10 |
| vishy | termie: the destination node has to get an image from glance | 23:10 |
| termie | that's a fine idea, not supported today though | 23:10 |
| vishy | termie: that image is private | 23:10 |
| termie | right now the apis are all designed in such a way that make it hard to do what you awnt | 23:10 |
| vishy | termie: == migration explodes | 23:10 |
| termie | but in general your service account could give you admin rights that glacne accepts and glance would then treat you as an admin | 23:11 |
| vishy | termie: the only workaround i can think of is to give the admin user that is calling migrate some superuser role in glance that lets them do anything | 23:11 |
| termie | for example, your service token could have a role like compute:admin | 23:11 |
| termie | and it could also have image:admin | 23:11 |
| termie | so that when glance validates your token it sees an admin for glance also | 23:12 |
| vishy | termie: I'm kind of hoping to avoid that only because it forces the compute service to also have management control over glance | 23:12 |
| termie | that is exactly what you are asking for | 23:12 |
| vishy | termie: but I can't think of a better short term solution | 23:12 |
| termie | you can make it a specific role that can only do a specific action | 23:12 |
| termie | owner | admin | nova:admin | 23:12 |
| vishy | jaypipes: are you here? | 23:13 |
| termie | so glance will allow a compute:admin to make that one call | 23:13 |
| *** AlanClark has quit IRC | 23:14 | |
| *** dprince has quit IRC | 23:20 | |
| vishy | termie: now i need to find out if such a thing exists in glance | 23:21 |
| termie | vishy: well they hvae the policy code, yea? | 23:22 |
| vishy | termie: yeah, but I don't know if they have the concept of someone who can see all images or not | 23:22 |
| vishy | termie: might need a patch | 23:22 |
| vishy | :) | 23:22 |
| vishy | termie: where is walrus when you need him? | 23:23 |
| *** jog0 has quit IRC | 23:23 | |
| termie | i know right | 23:23 |
| termie | i'm going to have some fun trying to stall keystone for a week while i am at sxsw | 23:23 |
| *** dubsquared has quit IRC | 23:25 | |
| *** andrewsmedina has joined #openstack-dev | 23:27 | |
| *** littleidea has quit IRC | 23:30 | |
| *** PotHix has quit IRC | 23:33 | |
| *** kbringard has quit IRC | 23:33 | |
| jaypipes | vishy: in pycon tutorial.. but yeah | 23:34 |
| vishy | jaypipes: is there anything like a superuser in glance? | 23:36 |
| mtaylor | hey termie: https://jenkins.openstack.org/job/keystone-coverage/386/console | 23:36 |
| jaypipes | vishy: any user with the admin role... | 23:36 |
| vishy | jaypipes: a user who can access all images | 23:36 |
| mtaylor | termie: I get some test failures when I run keystone through code coverage | 23:36 |
| vishy | jaypipes: word! we are good then :) | 23:36 |
| mtaylor | heckj: ^^ | 23:36 |
| jaypipes | vishy: glad to know :) | 23:36 |
| vishy | jaypipes: danger averted :) | 23:37 |
| mtaylor | termie, heckj: running at the moment via "tox -evenv -- /bin/bash run_tests.sh -N -P --cover-package=keystone --cover-erase --with-xcoverage" | 23:37 |
| heckj | mtaylor: yeah, I've seen that before - not sure why. | 23:37 |
| mtaylor | heckj: k. just wanted to point it out | 23:37 |
| mtaylor | it's not putting a huge crimp in my day or anything | 23:37 |
| vishy | heckj, termie : urgency on that bug can be downgraded | 23:37 |
| mtaylor | but you are the last guys with a broken coverage job | 23:37 |
| heckj | vishy: kk | 23:37 |
| vishy | I think we are good to work around for the moment | 23:37 |
| termie | mtaylor: coverage is not supported on keystone | 23:42 |
| mtaylor | termie: great! | 23:42 |
| mtaylor | termie: well, that's certainly easy then | 23:42 |
| termie | mtaylor: :D | 23:42 |
| termie | mtaylor: we do some stuff in there that coverage doesn't like and it has been worth the time to fix | 23:43 |
| mtaylor | termie: should I just disable your coverage job for now then? | 23:43 |
| termie | mtaylor: if somebody wants to dump a lot of time into it then sure, but not a priority for me | 23:43 |
| heckj | something we did in the tests is horking the coverage mechanism horribly, but I don't know where it is | 23:43 |
| termie | mtaylor: as long as it isn't gating i don't mind it | 23:43 |
| mtaylor | k. I'll leave it | 23:43 |
| mtaylor | also - if you guys get bored - there's one thing preventing us from using nose directly, which is that some of the tests are asusming that the cwd is the test dir (for finding config files) | 23:44 |
| mtaylor | I started poking at trying to fix it sensibly and got bored | 23:44 |
| mtaylor | again - low priority :) | 23:44 |
| termie | mtaylor: i'd love to ahve a bug or something noting that, 'cause i remember that also | 23:44 |
| mtaylor | termie: I'll file a bug | 23:44 |
| heckj | mtaylor: I'm trying to track down the test failures in this changeset: https://review.openstack.org/#change,4880 | 23:47 |
| heckj | WHen i run the tests in my own virtualenv (from clean), they're working fine. | 23:47 |
| mtaylor | heckj: are you driving via tox or via run_tests? (asking just in case there's a difference somewhere) | 23:47 |
| heckj | ./run_tests was the first - couldn't repro with that | 23:48 |
| *** littleidea has joined #openstack-dev | 23:48 | |
| heckj | Then I just activated the venv and ran "/bin/bash run_tests.sh -N --no-pep8" - that also didn't reproduce the error. | 23:48 |
| heckj | Trying tox now | 23:48 |
| heckj | mtaylor: anything specific to run tox other than just invoking it outright? | 23:49 |
| mtaylor | heckj: awesome. I'm going to be thrilled if it's a) only in tox or b) only on one of our slaves | 23:49 |
| mtaylor | heckj: nope. it's going to run pep8, py27 and py27 by default | 23:49 |
| mtaylor | so you might want to try "tox -v -epy27" | 23:49 |
| mtaylor | to keep it simple | 23:49 |
| heckj | mtaylor: cool - trying that now | 23:49 |
| mtaylor | heckj: also, in case it helps for your debugging, tox makes its venv in .tox/$ENV_NAME - so tox -epy27 will make a venv in .tox/py27 | 23:50 |
| heckj | mtaylor: cool - waiting now while the deps download. This'll be interesting! | 23:51 |
| mtaylor | heckj: the other difference between tox and install_venv.py is that tox makes an sdist and installs that into the venv | 23:51 |
| heckj | ah, neat. | 23:52 |
| heckj | Kind of gets us a packaging verification step with that, eh? | 23:52 |
| mtaylor | heckj: yup :) | 23:52 |
| mtaylor | heckj: (also, and I should send this to the list now that we've got configs in all of the trees) it respects the PIP_DOWNLOAD_CACHE env var - so if you set that to something, like PIP_DOWNLOAD_CACHE=~/cache/pip (that's mine) it'll skip re-downloading even if you blow away the venv and start fresh | 23:53 |
| mtaylor | afk for a bit | 23:54 |
| *** jaypipes has quit IRC | 23:55 | |
| heckj | mtaylor: I couldn't reproduce it with TOX either. Doing a full tox run to see... | 23:57 |
| *** lloydde has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!