| *** gyee has quit IRC | 00:08 | |
| *** jakedahn_zz is now known as jakedahn | 00:12 | |
| *** reed has quit IRC | 00:15 | |
| *** wiliam has quit IRC | 00:16 | |
| *** harlowja has quit IRC | 00:24 | |
| jgriffith | vishy: ping | 00:25 |
|---|---|---|
| vishy | jgriffith: sup | 00:25 |
| jgriffith | vishy: Yo, wanted your input on all of these tests in nova (related to volumes) | 00:26 |
| jgriffith | vishy: Theyr'e all ported into cinder now | 00:26 |
| vishy | jgriffith: what about them? | 00:26 |
| jgriffith | Wondering how much (if any) to duplicate/leave in nova | 00:26 |
| jgriffith | and what to modify to call cinder or nova volumes depending on a flag | 00:27 |
| jgriffith | The reason I'm thinking about is all of the tests doing direct db calls still | 00:27 |
| jgriffith | I can refactor all of those tests with an api call... but do we need the same tests in both projects? | 00:28 |
| *** thingee has quit IRC | 00:29 | |
| *** wiliam has joined #openstack-dev | 00:30 | |
| vishy | jgriffith: no i don't think so | 00:33 |
| vishy | jgriffith: we probably just need tests for cinder.py and we can toss the other ones when we pull the nova-volumes code | 00:34 |
| jgriffith | Sounds good... I'll just refactor the ones like ec2 etc that aren't "volume" tests per say. | 00:34 |
| vishy | jgriffith: k | 00:35 |
| jgriffith | thanks | 00:35 |
| *** littleidea has joined #openstack-dev | 00:36 | |
| *** jgriffith is now known as jgriffith_away | 00:38 | |
| *** blamar has left #openstack-dev | 00:39 | |
| *** jog0 has joined #openstack-dev | 00:40 | |
| *** crobinso has quit IRC | 00:45 | |
| *** torgomatic has quit IRC | 00:50 | |
| *** s0mik has quit IRC | 00:54 | |
| *** rods has quit IRC | 00:58 | |
| *** bencherian has quit IRC | 01:01 | |
| *** winston-d has joined #openstack-dev | 01:04 | |
| *** adjohn has quit IRC | 01:04 | |
| winston-d | termie, ping | 01:04 |
| *** johnpostlethwait has quit IRC | 01:07 | |
| *** e1mer has quit IRC | 01:08 | |
| *** Aaton is now known as Aaton_off | 01:12 | |
| *** aa has quit IRC | 01:12 | |
| *** Mandell has quit IRC | 01:16 | |
| winston-d | jeblair, hi | 01:17 |
| *** e1mer has joined #openstack-dev | 01:23 | |
| *** e1mer has quit IRC | 01:23 | |
| *** e1mer has joined #openstack-dev | 01:23 | |
| winston-d | jeblair, i saw you revert "blueprint <multi-process-api-service>" change due to 'nova test runner is leaking process'. Could you give me more details so that I can fix it? | 01:23 |
| jeblair | winston-d: hi. I didn't revert it -- I proposed a change to revert it. The details are on the mailing list. there's a thread called "Lossage in nova test suite" | 01:24 |
| winston-d | jeblair, thx. let's read through the thread first. | 01:27 |
| winston-d | jeblair, there's no problem running tests on my laptop. do you mind give me more details so that i can reproduce this issue? | 01:31 |
| jeblair | winston-d: the problem is that after the tests are complete, there are python processes left over that continuously respawn more processes. Kevin, Chris, Gabe, and I have all seen it. | 01:41 |
| jeblair | winston-d: there's no problem running the tests themselves; they succeed. | 01:41 |
| winston-d | jeblair, hmm, i've git clone a clean nova code base to do unittest, all unittest passed, no python process remained. i'm running Fedora 17 btw. | 01:42 |
| jeblair | winston-d: the jenkins slaves run oneiric. | 01:43 |
| winston-d | jeblair, ok, i'll get a oneiric to do some test. thx for the info. | 01:45 |
| *** edygarcia has joined #openstack-dev | 01:45 | |
| *** feiyuliu has joined #openstack-dev | 01:46 | |
| *** johnpostlethwait has joined #openstack-dev | 01:49 | |
| *** torgomatic has joined #openstack-dev | 01:59 | |
| *** edygarcia has quit IRC | 02:01 | |
| *** shang has joined #openstack-dev | 02:03 | |
| *** jemartin has quit IRC | 02:04 | |
| *** novas0x2a|laptop has joined #openstack-dev | 02:05 | |
| *** andrewsben_zz is now known as andrewsben | 02:08 | |
| *** danwent has quit IRC | 02:17 | |
| *** ayoung has quit IRC | 02:18 | |
| *** edygarcia has joined #openstack-dev | 02:21 | |
| *** jgriffith has joined #openstack-dev | 02:30 | |
| westmaas | jeblair: heh, now how to get it through jenkins? | 02:33 |
| *** feiyuliu_ has joined #openstack-dev | 02:44 | |
| *** feiyuliu has quit IRC | 02:45 | |
| *** anderstj has joined #openstack-dev | 02:53 | |
| *** reed has joined #openstack-dev | 02:54 | |
| *** anderstj has quit IRC | 02:56 | |
| *** jemartin has joined #openstack-dev | 03:00 | |
| *** littleidea has quit IRC | 03:09 | |
| *** littleidea has joined #openstack-dev | 03:14 | |
| *** anderstj has joined #openstack-dev | 03:15 | |
| *** relateable has joined #openstack-dev | 03:18 | |
| *** winston-d has quit IRC | 03:18 | |
| *** bencherian has joined #openstack-dev | 03:27 | |
| *** winston-d has joined #openstack-dev | 03:31 | |
| *** edygarcia has quit IRC | 03:32 | |
| *** Mandell has joined #openstack-dev | 03:34 | |
| *** edygarcia has joined #openstack-dev | 03:44 | |
| *** danwent has joined #openstack-dev | 03:47 | |
| *** feiyuliu_ has quit IRC | 03:50 | |
| *** feiyuliu has joined #openstack-dev | 03:54 | |
| *** e1mer has quit IRC | 03:55 | |
| *** johnpostlethwait has quit IRC | 03:55 | |
| *** e1mer has joined #openstack-dev | 04:05 | |
| *** e1mer has joined #openstack-dev | 04:05 | |
| *** nunosantos has quit IRC | 04:12 | |
| *** e1mer has quit IRC | 04:16 | |
| *** e1mer has joined #openstack-dev | 04:17 | |
| *** e1mer has joined #openstack-dev | 04:17 | |
| jeblair | westmaas: i'll see if i can kick enough jenkins slaves to get it through. :) | 04:18 |
| jgriffith | jeblair: Hate to bug you, but... https://jenkins.openstack.org/job/gate-cinder-python26/17/console | 04:20 |
| jgriffith | Is this the same thing westmaas is seeing? | 04:20 |
| jeblair | jgriffith: that should clear up once i restart all the jenkins slave and the fix for nova is merged | 04:20 |
| jgriffith | jeblair: Great, thanks. I'll resubmit it later. | 04:21 |
| *** e1mer has quit IRC | 04:23 | |
| *** hattwick has quit IRC | 04:25 | |
| *** bencherian has quit IRC | 04:27 | |
| *** garyk has quit IRC | 04:29 | |
| *** e1mer has joined #openstack-dev | 04:32 | |
| *** e1mer has joined #openstack-dev | 04:32 | |
| *** mnewby has quit IRC | 04:34 | |
| *** jgriffith has quit IRC | 04:37 | |
| *** jkoelker has left #openstack-dev | 04:39 | |
| *** jkoelker has joined #openstack-dev | 04:39 | |
| *** maoy has quit IRC | 04:43 | |
| *** spiffxp has quit IRC | 04:45 | |
| *** s0mik has joined #openstack-dev | 04:46 | |
| *** s0mik has quit IRC | 04:49 | |
| *** johnpostlethwait has joined #openstack-dev | 04:50 | |
| *** jemartin has quit IRC | 04:54 | |
| *** ywu_ has quit IRC | 04:58 | |
| *** edygarcia has quit IRC | 05:02 | |
| *** b1rkh0ff has quit IRC | 05:07 | |
| *** openstackjenkins has quit IRC | 05:15 | |
| *** openstackjenkins has joined #openstack-dev | 05:16 | |
| *** bencherian has joined #openstack-dev | 05:17 | |
| *** bencherian has left #openstack-dev | 05:17 | |
| *** reed has quit IRC | 05:18 | |
| *** b1rkh0ff has joined #openstack-dev | 05:21 | |
| *** davidha has quit IRC | 05:22 | |
| *** openstackjenkins has quit IRC | 05:23 | |
| *** spiffxp has joined #openstack-dev | 05:25 | |
| *** openstackjenkins has joined #openstack-dev | 05:25 | |
| *** zaitcev has quit IRC | 05:26 | |
| *** littleidea has quit IRC | 05:27 | |
| *** dachary has joined #openstack-dev | 05:28 | |
| *** johnpostlethwait has quit IRC | 05:28 | |
| *** ewindisch has quit IRC | 05:42 | |
| *** ewindisch has joined #openstack-dev | 05:44 | |
| jeblair | westmaas: merged | 05:52 |
| *** Mandell has quit IRC | 05:53 | |
| westmaas | yay | 05:53 |
| *** Slower has quit IRC | 05:55 | |
| *** torgomatic has quit IRC | 05:56 | |
| *** garyk has joined #openstack-dev | 05:57 | |
| *** danwent has quit IRC | 06:03 | |
| *** spiffxp has quit IRC | 06:05 | |
| *** dachary has quit IRC | 06:08 | |
| *** spiffxp has joined #openstack-dev | 06:09 | |
| *** jakedahn is now known as jakedahn_zz | 06:10 | |
| *** sleepsonthefloor is now known as sleepsonzzz | 06:20 | |
| *** mrunge has joined #openstack-dev | 06:25 | |
| *** anderstj has quit IRC | 06:41 | |
| *** wiliam has quit IRC | 06:44 | |
| *** feiyuliu has quit IRC | 06:47 | |
| *** feiyuliu has joined #openstack-dev | 06:53 | |
| *** pmezard has joined #openstack-dev | 06:53 | |
| *** spiffxp has quit IRC | 06:56 | |
| *** corXi has joined #openstack-dev | 07:01 | |
| *** dachary has joined #openstack-dev | 07:17 | |
| * mikal reads the instance disk on LVM patch | 07:25 | |
| *** pixelbeat has joined #openstack-dev | 07:34 | |
| *** roge has quit IRC | 07:39 | |
| *** adjohn has joined #openstack-dev | 07:53 | |
| *** hattwick has joined #openstack-dev | 07:58 | |
| *** derekh has joined #openstack-dev | 08:04 | |
| *** adjohn has quit IRC | 08:18 | |
| *** winston-d has quit IRC | 08:51 | |
| *** davidha has joined #openstack-dev | 09:03 | |
| *** matwood has quit IRC | 09:09 | |
| *** davidha has quit IRC | 09:11 | |
| *** eglynn has joined #openstack-dev | 09:12 | |
| ttx | notmyname: Swift 1.5.0 ready for your release announcement at https://launchpad.net/swift/folsom/1.5.0 | 09:13 |
| *** apevec has joined #openstack-dev | 09:23 | |
| *** koolhead17 has joined #openstack-dev | 09:39 | |
| koolhead17 | apevec: around | 09:39 |
| koolhead17 | https://bugs.launchpad.net/keystone/+bug/999999 | 09:39 |
| uvirtbot | Launchpad bug 999999 in keystone "keystone does not support postgres database" [Low,Invalid] | 09:39 |
| koolhead17 | did you even read what i reported before providing solution and marking it invalid? | 09:39 |
| apevec | koolhead17, hi - ah yes, could you try with connection in keystone conf? | 09:39 |
| koolhead17 | [15:09] <koolhead17> did you even read what i reported before providing solution and marking it invalid? | 09:39 |
| koolhead17 | apevec: what you refered there was for sqlite | 09:40 |
| koolhead17 | in the example | 09:40 |
| apevec | koolhead17, I didn't see reply from you in LP | 09:40 |
| apevec | koolhead17, what do you mean? | 09:40 |
| koolhead17 | i want postgres to work with my keystone | 09:40 |
| apevec | parameter name is simply "connection" | 09:40 |
| apevec | not sql_connection | 09:41 |
| koolhead17 | i dont need sqlite | 09:41 |
| apevec | sqlite doesn't matter | 09:41 |
| apevec | please try with connection = postgresql://keystone:keystone@192.168.1.3/keystone | 09:41 |
| *** sannes has quit IRC | 09:43 | |
| apevec | if you don't trust me, here is the code https://github.com/openstack/keystone/blob/master/keystone/common/sql/core.py#L150 | 09:43 |
| koolhead17 | apevec: thanks. let me try with it | 09:44 |
| *** semyazz has joined #openstack-dev | 09:44 | |
| *** semyazz has left #openstack-dev | 09:44 | |
| *** markmc has joined #openstack-dev | 09:45 | |
| *** sannes has joined #openstack-dev | 09:46 | |
| apevec | I see where the confusion comes from, it's sql_connection in nova.conf | 09:46 |
| apevec | and glance | 09:47 |
| koolhead17 | apevec: and also from Diablo installation steps http://docs.openstack.org/diablo/openstack-compute/install/content/install-keystone.html | 09:47 |
| koolhead17 | apevec: sql_connection = mysql://keystone:yourpassword@192.168.206.130/keystone :) | 09:47 |
| apevec | yeah, keystone light refactoring side-effect - ask termie why he didn't like sql_connection :) | 09:48 |
| * koolhead17 kicks termie | 09:48 | |
| koolhead17 | :P | 09:48 |
| koolhead17 | apevec: this is where all confusion came in, its cool now. i will test it and keep you updated. cheers!! | 09:49 |
| *** Ryan_Lane has joined #openstack-dev | 10:01 | |
| *** winston-d has joined #openstack-dev | 10:07 | |
| *** davidha has joined #openstack-dev | 10:08 | |
| winston-d | i've problem running nova unittest with this error: http://paste.openstack.org/show/18346/ does anybody know how can i fix it? | 10:08 |
| markmc | garyk, hey, I'm just going through the quantum essex fixes and your mail | 10:09 |
| garyk | markmc: hi, thanks | 10:10 |
| garyk | markmc: between a rock and a hard place here | 10:10 |
| markmc | garyk, could you explain the background a bit? | 10:10 |
| garyk | markmc: problem is that quantum essex stable was created at a point in time where some code was mistakenly removed (not sure if that makes sense) | 10:10 |
| garyk | markmc: i'll elaborate | 10:11 |
| garyk | makrmc: quantum client was removed - the common config dir was removed with this, then essex stable was created, then the code was returned. | 10:11 |
| garyk | markmc: so i am in a position where i need to back port fixes to a branch that does not have all of the base code | 10:12 |
| apevec | garyk, what about re-adding quantumclient to stable? | 10:13 |
| markmc | cripes | 10:13 |
| * markmc shakes head | 10:13 | |
| markmc | garyk, I'll poke a bit more | 10:13 |
| garyk | markmc: hence the rock and the hard place. | 10:13 |
| garyk | markmc: at the moment i have a number of problems: | 10:14 |
| koolhead17 | Ryan_Lane: hey there | 10:14 |
| garyk | 1. quantum baseline is missing files (https://review.openstack.org/#/c/8023/) | 10:14 |
| Ryan_Lane | koolhead17: howdy | 10:14 |
| garyk | 2. nova jenkins fails (https://review.openstack.org/#/c/7970/) | 10:15 |
| koolhead17 | Ryan_Lane: all good. thanks :) | 10:15 |
| *** rods has joined #openstack-dev | 10:15 | |
| markmc | LinuxJedi, jeblair, mtaylor, there? see jenkins failures in https://review.openstack.org/#/c/7970/ | 10:16 |
| Ryan_Lane | koolhead17: ready to do some docs? :) | 10:16 |
| koolhead17 | Ryan_Lane: sure. | 10:17 |
| garyk | markmc: can the jedi come and help me with the dark side of the force | 10:17 |
| Ryan_Lane | koolhead17: we'll need to get an account set up. | 10:17 |
| Ryan_Lane | koolhead17: can you add your info here? http://www.mediawiki.org/wiki/Developer_access | 10:17 |
| koolhead17 | ok | 10:18 |
| Ryan_Lane | I really want to open up registration, but we need to work out some issues first :) | 10:19 |
| koolhead17 | :P | 10:19 |
| Ryan_Lane | thinking of issues… we're doing ipv6 for wikipedia right now, and don't have ipv6 in labs, because of openstack... | 10:19 |
| Ryan_Lane | we launched prior to openstack having ipv6 support, and now we don't know how to turn it on | 10:20 |
| Ryan_Lane | since you're expected to create the ipv6 network with the ipv4 network. | 10:20 |
| Ryan_Lane | I have no clue what'll happen if I update the database directly. | 10:20 |
| Ryan_Lane | it's kind of silly to have to create them together. | 10:21 |
| *** sandywalsh has joined #openstack-dev | 10:29 | |
| *** e1mer has quit IRC | 10:29 | |
| *** e1mer has joined #openstack-dev | 10:35 | |
| markmc | garyk, so, it looks to me like it's just that essex quantum isn't designed to work with folsom quantumclient | 10:49 |
| markmc | garyk, we need to get stable/essex quantum unit tests running with 2012.1 quantumclient | 10:49 |
| markmc | garyk, probably best to talk to mtaylor about that, but pushing quantumclient-2012.1 to pypi seems like the right thing | 10:49 |
| *** bsza has joined #openstack-dev | 10:52 | |
| garyk | markmc: thanks. | 10:52 |
| *** rkukura has quit IRC | 10:57 | |
| *** hugokuo1 has joined #openstack-dev | 11:17 | |
| *** hugokuo has quit IRC | 11:20 | |
| *** corXi has quit IRC | 11:22 | |
| garyk | markmc: can you please look at http://fpaste.org/SMp8/ I am not 100% sure that I am on the right tack here. I feel like I am waisting everyones time and efforts | 11:28 |
| markmc | garyk, "Error 500: Sorry, you broke our server. You might have reached the 512KiB limit! " | 11:29 |
| markmc | garyk, if it's the quantum cfg patch, it's no my list for later | 11:29 |
| markmc | garyk, if it's the quantum cfg patch, it's on my list for later | 11:29 |
| garyk | markmc: no - stable essex - http://fpaste.org/veuB/ | 11:30 |
| garyk | markmc: i think i am good with the cfg stuff (you helped a lot here). your inputs on the work done will be greatly appreciated | 11:30 |
| markmc | garyk, right, it's not in pypi yet http://pypi.python.org/pypi/python-quantumclient/ | 11:30 |
| garyk | markmc: ok. what do you suggest? | 11:33 |
| markmc | garyk, talk to mtaylor about getting it added to pypi | 11:33 |
| markmc | garyk, he holds OpenStack's pypi keys AFAIK | 11:33 |
| garyk | markmc: ok, will do. thanks | 11:34 |
| *** markvoelker has joined #openstack-dev | 11:42 | |
| *** sandywalsh has quit IRC | 11:47 | |
| *** salgado has joined #openstack-dev | 11:54 | |
| *** mrunge has quit IRC | 11:54 | |
| *** sandywalsh has joined #openstack-dev | 11:57 | |
| *** sandywalsh has quit IRC | 11:59 | |
| eglynn | markmc: thanks for your review on the security groups de-dupe, got a minute to discuss? | 12:02 |
| markmc | eglynn, sure, fire ahead | 12:03 |
| eglynn | markmc: just this comment on the refresh ... "The idea is that when we're adding or removing a bunch of rules, we only want to trigger an expensive refresh after the last update." | 12:03 |
| eglynn | markmc: so that's exactly the idea of the refresh_rules parameter | 12:03 |
| eglynn | markmc: i.e. here's a bunch of rules in addition to the current one to refresh | 12:04 |
| eglynn | markmc: if the refresh_rules is set to None, then the refresh is deferred | 12:04 |
| markmc | eglynn, right | 12:05 |
| markmc | eglynn, so my comment was that jay didn't want the abstraction to leak details of this "refresh" issue | 12:05 |
| markmc | eglynn, you moved it from being a method to being a method parameter, but it still leaks through | 12:05 |
| *** maoy has joined #openstack-dev | 12:05 | |
| markmc | eglynn, but I don't have a problem with it leaking though :) | 12:06 |
| *** sandywalsh has joined #openstack-dev | 12:06 | |
| markmc | eglynn, the semantics of the refresh_rules param seemed a little strange to me | 12:06 |
| *** eglynn_ has joined #openstack-dev | 12:06 | |
| markmc | eglynn_, what was the last comment you saw? | 12:06 |
| *** alaski has joined #openstack-dev | 12:07 | |
| * markmc shrugs, guess you've got eavesdrop | 12:07 | |
| *** e1mer has quit IRC | 12:07 | |
| markmc | eglynn, what did you think of my batch add/remove suggestion? that way it doesn't need to leak through at all | 12:07 |
| *** e1mer has joined #openstack-dev | 12:07 | |
| *** e1mer has joined #openstack-dev | 12:07 | |
| eglynn_ | markmc: connection dropped there, last I saw was ... the semantics of the refresh_rules | 12:07 |
| markmc | eglynn_, http://eavesdrop.openstack.org/irclogs/%23openstack-dev/%23openstack-dev.2012-06-05.log | 12:08 |
| *** eglynn has quit IRC | 12:08 | |
| eglynn_ | markmc: ok, see what you mean ... so you want the leakage of the refresh action to be more explicit than shoe-horned into the refresh_rules, right? | 12:10 |
| markmc | eglynn_, either that, or not leaked through at all by adding add_rules()/remove_rules() | 12:10 |
| *** e1mer has quit IRC | 12:11 | |
| eglynn_ | markmc: yep, gotcha ... makes sense | 12:11 |
| markmc | eglynn, what did you think of my batch add/remove suggestion? | 12:11 |
| eglynn_ | markmc: I'm thinking that's potentially the neatest approach | 12:11 |
| markmc | eglynn_, cool | 12:12 |
| markmc | eglynn_, wasn't sure whether I'd missed something | 12:12 |
| eglynn_ | markmc: I'll code it up and see how it feels ... | 12:12 |
| *** maploin has joined #openstack-dev | 12:15 | |
| *** maploin has quit IRC | 12:15 | |
| *** maploin has joined #openstack-dev | 12:15 | |
| *** roge has joined #openstack-dev | 12:17 | |
| *** littleidea has joined #openstack-dev | 12:21 | |
| *** semyazz has joined #openstack-dev | 12:23 | |
| *** semyazz has left #openstack-dev | 12:23 | |
| *** littleidea has quit IRC | 12:23 | |
| *** e1mer has joined #openstack-dev | 12:23 | |
| *** e1mer has joined #openstack-dev | 12:23 | |
| zykes- | has anyone run into that if you kill off dnsmasq and delete gw-%s interfaces on a networking node with quantum and restart nova-network dns names for instances are not used anymore ? | 12:24 |
| zykes- | instead you get entries like host-%d | 12:25 |
| *** lts has joined #openstack-dev | 12:38 | |
| *** ywu has joined #openstack-dev | 12:38 | |
| *** apevec has quit IRC | 12:43 | |
| *** koolhead17 has quit IRC | 12:45 | |
| *** stuntmachine has joined #openstack-dev | 12:52 | |
| *** mdomsch has quit IRC | 12:55 | |
| *** maoy has quit IRC | 13:02 | |
| *** ywu has quit IRC | 13:08 | |
| mtaylor | hey garyk | 13:09 |
| mtaylor | markmc, garyk: yes, we do want to push quantumclient to pypi - we may not want to version it at 2012.1 ... there's a longer discussion with ttx about client releases that we're almost finished with | 13:10 |
| mtaylor | markmc, garyk: in the mean time, putting in a depend on a github zipball tied to a sha1 hash is a workaround | 13:11 |
| markmc | mtaylor, you mean "push essex quantumclient to pypi" ? | 13:11 |
| mtaylor | markmc: I mean push quantumclient to pypi | 13:11 |
| markmc | mtaylor, that works in pip-requires? | 13:11 |
| mtaylor | markmc: it does | 13:11 |
| markmc | mtaylor, cool on the latter | 13:11 |
| markmc | mtaylor, if you mean push folsom quantumclient to pypi, that doesn't help our stable/essex issues | 13:12 |
| mtaylor | markmc: then that's a bug in the client lib and should be fixed | 13:12 |
| markmc | mtaylor, retrospectively? | 13:12 |
| mtaylor | markmc: OR - something we'll have to figure out for quantum since it was core and since we didn't have this figured out anyway | 13:12 |
| mtaylor | markmc: yeah, we may have to special-case something here | 13:12 |
| *** littleidea has joined #openstack-dev | 13:13 | |
| markmc | mtaylor, the issue here is we don't want to say "this is the client lib for essex" etc. ? | 13:13 |
| markmc | mtaylor, but rather "this is v1.1, it works with essex and all later versions of the server"? | 13:14 |
| markmc | mtaylor, hmm, got an example of a project using the zipball thing? I vaguely recall seeing one go by | 13:16 |
| mtaylor | markmc: yes to both | 13:20 |
| mtaylor | markmc: one sec... | 13:20 |
| mtaylor | markmc: glance uses it: just plop this in: https://github.com/openstack/python-keystoneclient/zipball/master#egg=python-key | 13:20 |
| mtaylor | markmc: or https://github.com/openstack/python-quantumclient/zipball/master#egg=python-quantumclient | 13:21 |
| mtaylor | if you want to do quantumclient | 13:21 |
| *** timello has joined #openstack-dev | 13:22 | |
| *** ayoung has joined #openstack-dev | 13:23 | |
| markmc | mtaylor, ah, glanceclient uses it ... thanks | 13:24 |
| markmc | garyk, try adding https://github.com/openstack/python-quantumclient/zipball/2012.1#egg=python-quantumclient to pip-requires | 13:25 |
| *** e1mer has quit IRC | 13:28 | |
| *** kbringard has joined #openstack-dev | 13:36 | |
| *** dachary1 has joined #openstack-dev | 13:37 | |
| *** dachary has quit IRC | 13:39 | |
| *** wiliam has joined #openstack-dev | 13:39 | |
| *** JordanRinke has quit IRC | 13:39 | |
| *** JordanRinke has joined #openstack-dev | 13:40 | |
| *** nunosantos has joined #openstack-dev | 13:40 | |
| *** timello has quit IRC | 13:41 | |
| *** timello has joined #openstack-dev | 13:43 | |
| *** edygarcia has joined #openstack-dev | 13:44 | |
| *** e1mer has joined #openstack-dev | 13:45 | |
| *** e1mer has joined #openstack-dev | 13:45 | |
| devananda | morning, all | 13:46 |
| *** nikhil_ has joined #openstack-dev | 13:46 | |
| *** nikhil_ is now known as Guest19011 | 13:46 | |
| *** epim_ has joined #openstack-dev | 13:47 | |
| *** rkukura has joined #openstack-dev | 13:48 | |
| *** GheRivero_ has joined #openstack-dev | 13:49 | |
| mtaylor | markmc: and we _will_ get the larger question sorted | 13:49 |
| mtaylor | morning devananda | 13:49 |
| winston-d | Hi all. i've run into problem running nova unittest with this error: http://paste.openstack.org/show/18346/ does anybody know how can i fix it? | 13:49 |
| *** epim has quit IRC | 13:51 | |
| *** epim_ is now known as epim | 13:51 | |
| *** galstrom has joined #openstack-dev | 13:53 | |
| *** sstent has quit IRC | 13:54 | |
| *** sstent has joined #openstack-dev | 13:55 | |
| *** matiu has quit IRC | 13:57 | |
| *** blamar has joined #openstack-dev | 13:58 | |
| *** e1mer has quit IRC | 13:59 | |
| *** crobinso has joined #openstack-dev | 14:00 | |
| *** maoy has joined #openstack-dev | 14:02 | |
| *** matiu has joined #openstack-dev | 14:03 | |
| *** matiu has quit IRC | 14:03 | |
| *** matiu has joined #openstack-dev | 14:03 | |
| *** andrewsben is now known as andrewsben_zz | 14:03 | |
| *** dachary1 has quit IRC | 14:04 | |
| *** Mandell has joined #openstack-dev | 14:09 | |
| *** rnirmal has joined #openstack-dev | 14:13 | |
| *** epim has quit IRC | 14:22 | |
| *** Mandell_ has joined #openstack-dev | 14:25 | |
| *** Mandell has quit IRC | 14:25 | |
| *** anderstj has joined #openstack-dev | 14:27 | |
| *** feiyuliu has quit IRC | 14:29 | |
| *** danwent has joined #openstack-dev | 14:29 | |
| *** dachary has joined #openstack-dev | 14:30 | |
| *** datsun180b has joined #openstack-dev | 14:30 | |
| *** reed has joined #openstack-dev | 14:31 | |
| *** bsza has quit IRC | 14:31 | |
| *** timello has quit IRC | 14:32 | |
| *** krtaylor has joined #openstack-dev | 14:34 | |
| *** dubsquared has joined #openstack-dev | 14:37 | |
| renier | hey, trying to find documentation on the keystone catalog apis… anyone knows if this exists? | 14:40 |
| kbringard | sleepsonzzz: once you're no awake on the floor let me know :-D | 14:41 |
| kbringard | vishy: quick question if this rings a bell. I'm running diablo-stable, so I was wondering if this is something that had been fixed. Periodically instances fail to spawn giving an iptables "invalid code −1" error | 14:42 |
| * mtaylor is landing a change to turn on our new pypi mirror... everything _should_ still work - and should actually work much quicker... | 14:42 | |
| kbringard | jaypipes suggested you may have a recollection before I go sifting through bugs :-) | 14:42 |
| * jaypipes runs for hills | 14:43 | |
| kbringard | hehe | 14:43 |
| kbringard | go eat lunch :-p | 14:43 |
| *** timello has joined #openstack-dev | 14:45 | |
| *** Mandell_ has quit IRC | 14:50 | |
| garyk | markmc: thanks - testingit now | 14:51 |
| *** pballand has joined #openstack-dev | 14:52 | |
| *** dolphm has joined #openstack-dev | 14:52 | |
| garyk | markmc: looks like it solves the problem | 14:54 |
| *** timello has quit IRC | 14:56 | |
| *** timello has joined #openstack-dev | 14:57 | |
| *** zaitcev has joined #openstack-dev | 14:57 | |
| *** sstent has quit IRC | 15:01 | |
| *** davidha has quit IRC | 15:02 | |
| *** dubsquared has quit IRC | 15:03 | |
| *** pmyers has quit IRC | 15:05 | |
| zykes- | is anyone thinking of doing a broader dns service ? | 15:05 |
| *** pmyers has joined #openstack-dev | 15:06 | |
| *** garyk has quit IRC | 15:06 | |
| *** Gordonz has joined #openstack-dev | 15:08 | |
| *** matwood has joined #openstack-dev | 15:09 | |
| *** Gordonz has quit IRC | 15:09 | |
| *** Gordonz has joined #openstack-dev | 15:10 | |
| *** bencherian has joined #openstack-dev | 15:11 | |
| mtaylor | zykes-: like a dns-as-a-service offering? I, for one, would love it if that were an openstack offering - I use the crap out of rackspaces's | 15:12 |
| *** anderstj has quit IRC | 15:12 | |
| *** spiffxp has joined #openstack-dev | 15:14 | |
| *** jemartin has joined #openstack-dev | 15:16 | |
| Vek | jeblair / mtaylor: Can you confirm the status of https://review.openstack.org/#/c/8161 ? It appears to be approved, but there's no recent movement on it... | 15:19 |
| * Vek just wants to know if he's just being impatient or if the tests ended up hanging somewhere... | 15:20 | |
| jeblair | Vek: I expect the jobs were lost in the badness that happened yesterday; I think you should approvi it again. | 15:21 |
| Vek | OK... | 15:21 |
| *** openstackjenkins has quit IRC | 15:22 | |
| jeblair | I'm restarting jenkins to get it to connect to the last remaining slaves it can't talk to. should be back up in a minute | 15:22 |
| jeblair | It saves its build queue, so it should remember to run the jobs for that change, Vek. | 15:23 |
| *** openstackjenkins has joined #openstack-dev | 15:23 | |
| vishy | kbringard: haven't seen that one | 15:25 |
| zykes- | mtaylor: i suggested it earlier but noone seemed to care ;p | 15:25 |
| zykes- | a service for both tenant dns (internal) and public | 15:25 |
| kbringard | vishy: okie, no worries; I'll investigate further and see if I can figure out what's going on | 15:25 |
| Vek | jeblair: cool, thanks. | 15:26 |
| zykes- | mtaylor: here still ? | 15:29 |
| sdague | jeblair: is this badness of yesterday something which would have scrambled the python2.6 testing? | 15:30 |
| sdague | I'm trying to decypher some build fails for a patch of mine, and they seem completely unrelated | 15:31 |
| jeblair | sdague: absolutely. any weird jenkins test failures yesterday should be suspect. | 15:31 |
| sdague | ok, cool | 15:31 |
| sdague | I had to fix a couple things in the patch anyway. Things good now? | 15:32 |
| jeblair | sdague: yep! | 15:32 |
| *** jgriffith_away has quit IRC | 15:33 | |
| *** jgriffith has joined #openstack-dev | 15:35 | |
| davidkranz | jaypipes: There are four stable/diablo items sitting in the queue that were +1ed by Daryl but no one seems to be able or want to find a diablo to run them against and I don't see that changing. | 15:39 |
| davidkranz | jaypipes: I propose we ask the submitter to make sure they work in diablo and then one of us push them in when he reports success. | 15:39 |
| vishy | Vek: my brain has failed, was it you that did the eventlet stuff or johannes? | 15:40 |
| Vek | define "the eventlet stuff"? | 15:40 |
| Vek | (but probably johannes) | 15:40 |
| vishy | ah yes | 15:41 |
| vishy | it was | 15:41 |
| vishy | I just noticed a new assertion error when i updated i wanted to let him know about | 15:41 |
| Vek | I see. | 15:42 |
| comstud | there are definitely still bugs in eventlet with tpool use mixed with greenthreads | 15:43 |
| Vek | might want to email him; he said earlier that his power had gone out... | 15:43 |
| comstud | as well | 15:43 |
| comstud | eventlet is just kinda screwed in general | 15:43 |
| Vek | *nod* | 15:43 |
| Vek | which reminds me...once the multiprocess stuff is fixed and goes in, it should be noted that turnstile is not (yet) compatible with multiprocess. | 15:44 |
| comstud | ah | 15:44 |
| Vek | (I intend to address that eventually, thanks to input from the HP guys who were also working on multiprocess...) | 15:44 |
| mtaylor | vishy: re: logging during nova test runs ... we have that all working as it has been working from run_tests.sh in the new patch | 15:44 |
| vishy | mtaylor: yay, which was the review? | 15:45 |
| mtaylor | vishy: but ... nose already has a facility for capturing output from runs and logging it appropriately ... is there a reason we couldn't change things a little bit to take advantage of that? | 15:45 |
| vishy | mtaylor: that sounds fine | 15:45 |
| mtaylor | vishy: https://review.openstack.org/#/c/8106/ | 15:46 |
| mtaylor | vishy: ok. I'll work that up as a second patch | 15:46 |
| *** eglynn__ has joined #openstack-dev | 15:46 | |
| mtaylor | vishy: if we do it right, jenkins should even be able to include the appropriate trace logging with the failed test (whee) | 15:46 |
| zykes- | mtaylor: would you be intereseted? | 15:48 |
| *** eglynn_ has quit IRC | 15:48 | |
| *** lorin1 has joined #openstack-dev | 15:52 | |
| *** davidha has joined #openstack-dev | 15:52 | |
| mtaylor | zykes-: I would be interested in its existence... I don't have time to write it :) | 15:52 |
| *** bencherian has quit IRC | 15:53 | |
| *** katell1 has quit IRC | 15:53 | |
| mtaylor | jaypipes, notmyname, vishy, bcwaldon: I just sent a thing to the mailing list about global depends - I have a feeling you might have opinions | 15:54 |
| zykes- | ;p | 15:55 |
| zykes- | mtaylor: tech wise though | 15:55 |
| zykes- | bind or powerdns ? | 15:55 |
| mtaylor | zykes-: well, I always use bind myself - but that doesn't mean it's necessarily the right choice | 15:56 |
| maoy | mtaylor: quick question: does jenkins and smoke tests run against every patch in gerrit or selected ones in some criteria? | 15:57 |
| *** jakedahn_zz is now known as jakedahn | 15:57 | |
| jeblair | mtaylor: actually, you use nsd. :) | 15:57 |
| mtaylor | jeblair: hehe. well, by "use myself" I mean "all of the times I've personally installed it" ... I did not indicate when the last time I did that was :) | 15:59 |
| mtaylor | zykes-: let me clarify - I defer to jeblair | 15:59 |
| mtaylor | maoy: jenkins runs tests against every patch | 15:59 |
| mtaylor | maoy: I do not run smokestack, but I'm pretty sure that dprince also runs tests against every patch to the relevant projects | 16:00 |
| mtaylor | maoy: smokestack does not necessarily report back results on all runs, as dprince verifies that failing results are valid before allowing them to post | 16:00 |
| mtaylor | iirc | 16:00 |
| mtaylor | maoy: for jenkins, we run tests when a patch is uploaded, and then again after approval just before merge | 16:01 |
| maoy | mtaylor: all right. thanks. i was wondering because the reports come back with a random delay. so i thought there was some magic going on. | 16:01 |
| mtaylor | maoy: it just takes a variable amount of time to run them ... | 16:02 |
| *** garyk has joined #openstack-dev | 16:02 | |
| mtaylor | maoy: there is an option to add a "tests started" message, but we removed that long ago because it was very chatty | 16:02 |
| maoy | mtaylor: got it. | 16:03 |
| *** Guest19011 has quit IRC | 16:03 | |
| *** nikhil_ has joined #openstack-dev | 16:03 | |
| *** Adri2000 has quit IRC | 16:03 | |
| *** Adri2000 has joined #openstack-dev | 16:04 | |
| *** nikhil_ is now known as Guest55884 | 16:04 | |
| *** Adri2000 is now known as Guest67276 | 16:04 | |
| *** Guest55884 is now known as help | 16:05 | |
| zykes- | jeblair: ping. | 16:05 |
| *** help is now known as Guest28742 | 16:05 | |
| jeblair | zykes-: pong | 16:06 |
| *** issackelly has joined #openstack-dev | 16:07 | |
| zykes- | jeblair: what would use ? | 16:07 |
| zykes- | Bind or PowerDNS ? | 16:07 |
| *** issackelly has quit IRC | 16:07 | |
| jeblair | i haven't run dns at scale, but the folks i know who do use mostly bind, and some nsd. | 16:08 |
| jeblair | i think the db integration from powerdns is tempting, but... | 16:08 |
| *** sleepsonzzz is now known as sleepsonthefloor | 16:08 | |
| *** sleepsonthefloor is now known as sleepsonzzz | 16:09 | |
| *** Guest28742 is now known as register | 16:09 | |
| *** issackelly has joined #openstack-dev | 16:09 | |
| zykes- | but what jeblair ? ;) | 16:09 |
| *** sleepsonzzz is now known as sleepsonthefloor | 16:09 | |
| jeblair | i honestly think it would be best if the solution could be used with multiple servers | 16:09 |
| zykes- | :p | 16:09 |
| *** register is now known as Guest19777 | 16:09 | |
| zykes- | driver based ? | 16:09 |
| jeblair | having different dns server implementations has proved to be a really good idea in the past for resiliency | 16:10 |
| zykes- | driver / plugin based then i guess ;p0 | 16:10 |
| jeblair | (i believe some of the root servers run nsd for the same reason) | 16:10 |
| jeblair | yeah, that's my 2 cents. hate to sound wishy-washy, but there it is. :) | 16:10 |
| jeblair | (and then i'd probably start by writing the bind driver) | 16:11 |
| * mtaylor agrees with jeblair on pluggability/drivers | 16:11 | |
| mtaylor | that also matches the model of the other openstack projects | 16:11 |
| zykes- | i would go powerdns first :p | 16:11 |
| zykes- | easier to setup for testing i think | 16:11 |
| *** s0mik has joined #openstack-dev | 16:11 | |
| *** eglynn has joined #openstack-dev | 16:11 | |
| jeblair | haven't used powerdns myself. that's probably fine, though i think bind probably has the most installs, so that may increase initial acceptance. | 16:12 |
| garyk | markmc: ping | 16:14 |
| markmc | garyk, hey | 16:14 |
| *** eglynn__ has quit IRC | 16:15 | |
| zykes- | how should it run though jeblair ? | 16:15 |
| zykes- | in a vm inside of nova opr ? | 16:15 |
| garyk | markmc: in bit of a pickle. added the quantum client to pip. when i do ./run_tests all is cool. when i run tox. things are problematic | 16:15 |
| markmc | garyk, pastebin? mtaylor would probably have an idea | 16:16 |
| mtaylor | garyk: yeah - pastebin please | 16:16 |
| jeblair | zykes-: a bit out of my area of expertise. sorry, I don't want to lead you astray. :( | 16:16 |
| garyk | markmc, mtaylor: in a sec | 16:16 |
| zykes- | anyone else ? | 16:17 |
| zykes- | justinsb: ping | 16:17 |
| garyk | markmc, mtaylor: http://fpaste.org/yA2Z/ | 16:19 |
| *** Guest19777 has quit IRC | 16:19 | |
| *** maploin has quit IRC | 16:20 | |
| jaypipes | mtaylor: k, thx' | 16:20 |
| * markmc leaves that one to monty :) | 16:21 | |
| markmc | jeblair, got a sec? | 16:21 |
| jeblair | markmc: sure | 16:21 |
| markmc | jeblair, gerrit permissions for stable branches ... | 16:22 |
| markmc | jeblair, things looked a little different than what I remembered | 16:22 |
| jeblair | markmc: ok (i read your mail, did you see dan's followup?) | 16:23 |
| markmc | jeblair, yeah | 16:23 |
| jaypipes | davidkranz: I will be running them against a stable diablo setup today. | 16:23 |
| markmc | jeblair, actually I'm struggling to identify what I thought was an issue this morning | 16:24 |
| jaypipes | davidkranz: also, we should get with mtaylor and jeblair about setting up alternate jenkins jobs for tempest that pull the stable/X branches of the projects (and tempest) | 16:24 |
| markmc | jeblair, in #924974, you said stable-maint had exclusive perms to +2 on stable/* | 16:24 |
| markmc | jeblair, but it's non-exclusive now | 16:24 |
| markmc | jeblair, which makes sense | 16:24 |
| markmc | jeblair, oh, and I wasn't clear on what gave quantum-core +2 on stable/essex | 16:24 |
| jeblair | markmc: well, it is still exclusive... | 16:25 |
| jeblair | markmc: but occasionally overridden by individual projects | 16:25 |
| *** nikhil__ has joined #openstack-dev | 16:25 | |
| jeblair | (ie, incubated) | 16:25 |
| markmc | jeblair, https://review.openstack.org/#/admin/projects/All-Projects,access | 16:26 |
| markmc | jeblair, Reference: | 16:26 |
| markmc | refs/heads/stable/* | 16:26 |
| markmc | jeblair, no exclusive boxes checked there | 16:26 |
| jeblair | markmc: yes they are? for approved/code review... | 16:26 |
| zykes- | anyone here at rackspace that can get me in contact with the clouddns folks ? | 16:26 |
| markmc | jeblair, hmm, interesting - I don't see that | 16:26 |
| * markmc screenshots | 16:26 | |
| markmc | jeblair, http://markmc.fedorapeople.org/gerrit-perms.png | 16:28 |
| jeblair | markmc: wacky. | 16:29 |
| jeblair | markmc: https://review.openstack.org/gitweb?p=All-Projects.git;a=blob;f=project.config;h=6d98f479e51ef61a42b22e930463fe7ea7f55c32;hb=HEAD#l37 | 16:29 |
| markmc | jeblair, ok, cool :) | 16:30 |
| jeblair | markmc: my screen looks like that, but with the bottom two checkboxes checked. | 16:30 |
| * markmc notes the git access to the raw config | 16:30 | |
| markmc | jeblair, great, false alarm | 16:31 |
| mtaylor | have I mentioned how much I love how much of the config is all in stuff in git repos? | 16:31 |
| * markmc moves along :) | 16:31 | |
| jeblair | markmc: http://ci.openstack.org/gerrit.html#adding-a-new-project-on-the-command-line | 16:31 |
| jeblair | markmc: there's the details on how to pull/checkout the config in your local repo | 16:31 |
| markmc | jeblair, cool | 16:31 |
| jeblair | we got really tired of clicking checkboxes a while back and have pretty much stopped that. :) | 16:32 |
| garyk | mtaylor: anything else i can help with? | 16:32 |
| mtaylor | garyk: this is in essex quantum? | 16:33 |
| jeblair | markmc: so then there's a config in quantum that gives quantum-core exclusive access, and that config for the specific project seems to take precedence over the config for all-projects. | 16:33 |
| mtaylor | garyk: we probably need to update the quantum/openstack/common/setup.py ... you're hitting a bug we fixed a little while ago | 16:33 |
| markmc | jeblair, excellent, you've got it totally covered ... surprise surprise :) | 16:34 |
| jeblair | markmc: whew. :) | 16:34 |
| markmc | mtaylor, was this the fix? https://review.openstack.org/#/c/3801/ | 16:35 |
| garyk | mtaylor: yes, this is stable essex (i am learning the hard way). i can try and dig to find the setup issue - just need to know what the correct entry in the pip should be. | 16:36 |
| mtaylor | markmc: not quite - the fix eventually gets more involved ... | 16:36 |
| mtaylor | markmc: we learned a lot about pip and easy_install and distutils, let me tell you... | 16:37 |
| markmc | mtaylor, not seeing anything likely in log for setup.py | 16:37 |
| markmc | mtaylor, oh, wait - got it | 16:38 |
| markmc | mtaylor, https://review.openstack.org/#/c/7800/ | 16:38 |
| mtaylor | markmc: yup. you need 3801 and 7800 | 16:38 |
| *** s0mik has quit IRC | 16:38 | |
| *** Aaton_off is now known as Aaton | 16:39 | |
| markmc | mtaylor, ok, looks like you just need to update setup.py too - hopefully the update doesn't break other stuff for you | 16:39 |
| markmc | mtaylor, cheers | 16:39 |
| markmc | garyk, ok, looks like you just need to update setup.py too - hopefully the update doesn't break other stuff for you | 16:39 |
| mtaylor | https://review.openstack.org/8188 | 16:39 |
| mtaylor | garyk, markmc ^^ | 16:39 |
| mtaylor | feel free to either use that or I can abandon it | 16:39 |
| garyk | mtaylor: thanks, i'll check it out | 16:40 |
| markmc | mtaylor, yeah garyk will need to squash that into his patch | 16:41 |
| *** epim has joined #openstack-dev | 16:41 | |
| markmc | mtaylor, since tests will fail without garyk's stuff | 16:42 |
| mtaylor | excellent | 16:42 |
| garyk | mtaylor, markmc - so if i understand correctly hen i just need to add this file? | 16:42 |
| *** s0mik has joined #openstack-dev | 16:42 | |
| markmc | garyk, update quantum/openstack/common/setup.py the same way you updated cfg.py | 16:42 |
| garyk | markmc: ok | 16:43 |
| *** lloydde has joined #openstack-dev | 16:43 | |
| *** martine has joined #openstack-dev | 16:48 | |
| *** blamar has quit IRC | 16:49 | |
| *** Guest67276 has quit IRC | 16:49 | |
| *** jakedahn is now known as jakedahn_zz | 16:50 | |
| *** Adri2000_ has joined #openstack-dev | 16:50 | |
| *** Adri2000_ is now known as Guest43626 | 16:50 | |
| *** anderstj has joined #openstack-dev | 16:50 | |
| *** dubsquared has joined #openstack-dev | 16:51 | |
| *** rods has quit IRC | 16:52 | |
| *** danwent has quit IRC | 16:53 | |
| *** epim_ has joined #openstack-dev | 16:55 | |
| *** epim has quit IRC | 16:57 | |
| *** epim_ is now known as epim | 16:57 | |
| *** s0mik has quit IRC | 16:58 | |
| *** Mandell has joined #openstack-dev | 16:58 | |
| *** s0mik has joined #openstack-dev | 16:58 | |
| *** mestery has quit IRC | 17:00 | |
| *** Guest43626 is now known as Adri2000 | 17:00 | |
| *** eglynn has quit IRC | 17:00 | |
| *** Adri2000 has quit IRC | 17:00 | |
| *** Adri2000 has joined #openstack-dev | 17:00 | |
| garyk | mtaylor, markmc: worked locally - hopefully jenkins will be kind. thanks for the help! | 17:01 |
| *** derekh has quit IRC | 17:02 | |
| *** jdurgin has joined #openstack-dev | 17:04 | |
| *** rods has joined #openstack-dev | 17:04 | |
| *** gyee has joined #openstack-dev | 17:07 | |
| *** glenc_ has quit IRC | 17:07 | |
| *** glenc has joined #openstack-dev | 17:07 | |
| *** gyee has quit IRC | 17:11 | |
| *** glenc has quit IRC | 17:11 | |
| *** glenc has joined #openstack-dev | 17:12 | |
| *** semyazz has joined #openstack-dev | 17:13 | |
| *** semyazz has left #openstack-dev | 17:13 | |
| *** jakedahn_zz is now known as jakedahn | 17:13 | |
| *** glenc has quit IRC | 17:13 | |
| *** gyee has joined #openstack-dev | 17:13 | |
| *** glenc has joined #openstack-dev | 17:13 | |
| *** PotHix has joined #openstack-dev | 17:15 | |
| *** dachary has quit IRC | 17:15 | |
| jaypipes | Vek: think I narrowed down the reservation/quota issue... https://bugs.launchpad.net/nova/+bug/1009115 | 17:15 |
| uvirtbot | Launchpad bug 1009115 in nova "Reservations not destroyed when instances are destroyed" [Undecided,New] | 17:15 |
| jaypipes | Vek: turns out cores and instances aren't cleaned up, but the RAM resource *is*... | 17:15 |
| *** dachary has joined #openstack-dev | 17:15 | |
| jaypipes | Vek: perhaps there is a break statement in a for loop somewhere? :) | 17:15 |
| *** danwent has joined #openstack-dev | 17:16 | |
| *** openstackjenkins has quit IRC | 17:19 | |
| *** openstackjenkins has joined #openstack-dev | 17:20 | |
| *** torgomatic has joined #openstack-dev | 17:22 | |
| *** harlowja has joined #openstack-dev | 17:22 | |
| *** glenc has quit IRC | 17:24 | |
| *** glenc has joined #openstack-dev | 17:24 | |
| *** mestery has joined #openstack-dev | 17:29 | |
| *** glenc has quit IRC | 17:32 | |
| *** jakedahn is now known as jakedahn_zz | 17:33 | |
| *** glenc has joined #openstack-dev | 17:33 | |
| *** jakedahn_zz is now known as jakedahn | 17:34 | |
| *** aa has joined #openstack-dev | 17:36 | |
| *** alaski has quit IRC | 17:38 | |
| *** alaski has joined #openstack-dev | 17:39 | |
| *** dubsquared has quit IRC | 17:43 | |
| *** GheRivero_ has quit IRC | 17:43 | |
| *** dubsquared has joined #openstack-dev | 17:43 | |
| *** dubsquared has quit IRC | 17:48 | |
| *** rafaduran has joined #openstack-dev | 17:48 | |
| *** mnewby has joined #openstack-dev | 17:49 | |
| zykes- | jeblair: is it a bad idea to use mongodb as a backend for storing dns zones ? | 17:51 |
| *** blamar has joined #openstack-dev | 17:52 | |
| jeblair | zykes-: i think so, but i'll defer to mtaylor, who is a database expert, on that. | 17:52 |
| jeblair | mtaylor: ^ | 17:52 |
| *** bencherian has joined #openstack-dev | 17:53 | |
| *** rafaduran has quit IRC | 17:55 | |
| mtaylor | zykes-: why would you use mongodb as a backend for storing dns zones? | 17:56 |
| mtaylor | zykes-: as in - what problem are you trying to solve? | 17:56 |
| zykes- | mtaylor: storing dns for example accross datacenters | 17:56 |
| *** Slower has joined #openstack-dev | 17:56 | |
| mtaylor | zykes-: and you're wanting to use database backend as the replication mechaism rather than using the world's most scalable distributed database system already ... dns itself? | 17:57 |
| *** tongli|2 has joined #openstack-dev | 17:57 | |
| zykes- | just wondering mtaylor | 17:57 |
| zykes- | since powerdns has mongodb backends | 17:57 |
| mtaylor | zykes-: I would not depend on that in your design, if it were me, since powerdns is the only one of the dns servers that people use that has that feature | 17:58 |
| mtaylor | zykes-: but at the same time, ALL dns servers know how to delegate to each other | 17:58 |
| mtaylor | zykes-: and as I just (snarkily) mentioned, dns is currently probably the world's largest and most scalable distributed database | 17:58 |
| mtaylor | so I'd use the architecture that it provides as the basis of multi-datacenter work, rather than trying to make something like mongo or couch or mysql work sensibly | 17:59 |
| mtaylor | but that's just my $0.02 :) | 17:59 |
| zykes- | mtaylor: how you mean delegate ? | 18:00 |
| zykes- | i wonder how rs clouddns works :p | 18:00 |
| mtaylor | zykes-: any dns as a service offering is really going to be nothing more than a REST shim in front of DNS ... since DNS is, itself, already a scalable, distributed service | 18:01 |
| *** novas0x2a|laptop has quit IRC | 18:01 | |
| mtaylor | zykes-: or else someone has implemented it very poorly :) | 18:01 |
| zykes- | mtaylor: what about adding zones etc then ? | 18:01 |
| zykes- | i mean powerdns is just couple to a db and that's it basically | 18:01 |
| mtaylor | yeah. that's just backend storage | 18:02 |
| mtaylor | of where you store the information - it doesn't have anything to do with how dns servers share that information with each other | 18:02 |
| jeblair | markmc: hey, the three changes to keystone you just approved were the first gate tests we executed in parallel. :) You may have noticed they merged in rapid succession. | 18:02 |
| dolphm | jeblair: mtaylor: badass! | 18:03 |
| jeblair | markmc: thanks for the excellent well-timed test case. I just sent a message to the mailing list with more details about the system. | 18:03 |
| jeblair | or you can read about it here: http://amo-probos.org/post/14 | 18:03 |
| *** jakedahn is now known as jakedahn_zz | 18:08 | |
| *** danwent_ has joined #openstack-dev | 18:09 | |
| zykes- | so mtaylor you think mongodb is a bad idea ? | 18:09 |
| zykes- | but can't that again be up to the driver beneath the api ? | 18:10 |
| *** lloydde has quit IRC | 18:11 | |
| *** sleepsonthefloor is now known as sleepsonzzz | 18:12 | |
| *** danwent has quit IRC | 18:12 | |
| *** danwent_ is now known as danwent | 18:12 | |
| mtaylor | zykes-: TOTALLY | 18:14 |
| mtaylor | I just think you shouldn't design the dns service to depend on or expect some feature of mongodb under the covers | 18:14 |
| *** s0mik has quit IRC | 18:15 | |
| *** semyazz has joined #openstack-dev | 18:16 | |
| *** semyazz has left #openstack-dev | 18:16 | |
| *** rnirmal has quit IRC | 18:16 | |
| *** datsun180b has quit IRC | 18:17 | |
| *** negronjl has quit IRC | 18:17 | |
| zykes- | ;) | 18:18 |
| zykes- | mtaylor: you work for rs ? | 18:18 |
| *** johnpostlethwait has joined #openstack-dev | 18:19 | |
| davidkranz | jaypipes: Your create_network method does "resp, network = self.client.create(network_name)" but NetworkClient has no "create" method that I can see. Am I missing something obvious? | 18:20 |
| mtaylor | zykes-: nope | 18:21 |
| zykes- | mtaylor: another thing i think would be nice is that mongo allows youto have multiple api endpoints for the same system ? | 18:22 |
| *** negronjl has joined #openstack-dev | 18:25 | |
| jaypipes | davidkranz: doh. sorry, yes, I misunderstood your comment. you're right. | 18:25 |
| jaypipes | davidkranz: thx, fix pushed. | 18:27 |
| mtaylor | zykes-: sure. I'm just opposed to building a system like this to depend on a particular piece of software | 18:27 |
| *** novas0x2a|laptop has joined #openstack-dev | 18:28 | |
| mtaylor | zykes-: also - be careful, mongodb is AGPL, and the rest of openstack is Apache ... so if you were to tie something in dns as a service to mongo specifics, there would be some pretty problematic licensing issues | 18:28 |
| zykes- | mtaylor: I was more wanting to make it driver based | 18:30 |
| zykes- | so people can choose | 18:30 |
| mtaylor | awesome. I'm on board with that then | 18:31 |
| *** rnirmal has joined #openstack-dev | 18:34 | |
| *** issackel_ has joined #openstack-dev | 18:40 | |
| *** issackelly has quit IRC | 18:41 | |
| *** issackel_ is now known as issackelly | 18:41 | |
| *** martine has quit IRC | 18:42 | |
| zykes- | mtaylor: should instance dns be global as well or ? | 18:47 |
| *** dubsquared has joined #openstack-dev | 18:48 | |
| *** reed has quit IRC | 18:49 | |
| *** deva-mobile has joined #openstack-dev | 18:51 | |
| *** mrunge has joined #openstack-dev | 18:52 | |
| *** deva-mobile has quit IRC | 18:53 | |
| *** deva has joined #openstack-dev | 18:53 | |
| *** nunosantos has quit IRC | 18:55 | |
| *** adjohn has joined #openstack-dev | 18:57 | |
| *** martine has joined #openstack-dev | 18:58 | |
| *** dubsquared has quit IRC | 19:01 | |
| *** dubsquared has joined #openstack-dev | 19:01 | |
| *** sleepsonzzz is now known as sleepsonthefloor | 19:16 | |
| *** s0mik has joined #openstack-dev | 19:16 | |
| *** Ruetobas has quit IRC | 19:18 | |
| *** dolphm has quit IRC | 19:19 | |
| zykes- | justinsb: ping | 19:24 |
| *** danwent has quit IRC | 19:25 | |
| *** deva has quit IRC | 19:27 | |
| *** jakedahn_zz is now known as jakedahn | 19:29 | |
| *** stuntmachine has quit IRC | 19:32 | |
| *** stuntmachine has joined #openstack-dev | 19:33 | |
| *** stuntmachine has joined #openstack-dev | 19:33 | |
| *** Ryan_Lane has quit IRC | 19:36 | |
| *** matwood has quit IRC | 19:50 | |
| winston-d | comstud, hi, chris. i'm the author of nova multiprocess patch. sorry for the trouble it caused. | 19:57 |
| comstud | no worries | 19:57 |
| comstud | :) | 19:57 |
| *** danwent has joined #openstack-dev | 20:00 | |
| *** mrunge has quit IRC | 20:01 | |
| winston-d | comstud, i think you are right. the unittest finishes so quickly that api service hasn't even fully started yet. i'll fix it up with new patch set to address your comment and those from johannes. | 20:02 |
| comstud | yep | 20:02 |
| comstud | i'd added some debug to it to figure that out | 20:02 |
| comstud | the service.stop would be called | 20:03 |
| comstud | while it was still in the middle of starting up children | 20:03 |
| comstud | winston-d: Thnx:) | 20:04 |
| *** mattray has joined #openstack-dev | 20:05 | |
| *** glenc has quit IRC | 20:15 | |
| *** matwood has joined #openstack-dev | 20:19 | |
| *** aa has quit IRC | 20:23 | |
| *** wiliam has quit IRC | 20:27 | |
| *** crobinso has quit IRC | 20:27 | |
| *** aa has joined #openstack-dev | 20:30 | |
| *** eglynn has joined #openstack-dev | 20:31 | |
| *** rkukura has quit IRC | 20:35 | |
| *** glenc has joined #openstack-dev | 20:39 | |
| *** reed has joined #openstack-dev | 20:43 | |
| *** tongli|2 has quit IRC | 20:45 | |
| mtaylor | bcwaldon: https://review.openstack.org/#/c/8103/ and https://review.openstack.org/#/c/8143/ ? | 20:46 |
| mtaylor | or any glance-core ? | 20:47 |
| *** stuntmachine has quit IRC | 20:53 | |
| bcwaldon | mtaylor: just got back | 20:58 |
| bcwaldon | mtaylor: will look shortly | 20:58 |
| *** alaski has quit IRC | 20:59 | |
| mtaylor | bcwaldon: thanks! | 20:59 |
| mtaylor | bcwaldon: I'm trying to clear out red dots from jenkins so that we can start paying attention when a job goes red | 21:00 |
| bcwaldon | kk, good goal to have | 21:00 |
| *** lts has quit IRC | 21:03 | |
| *** Ryan_Lane has joined #openstack-dev | 21:03 | |
| *** eglynn has quit IRC | 21:04 | |
| *** maoy has quit IRC | 21:11 | |
| *** maoy has joined #openstack-dev | 21:12 | |
| *** sleepsonthefloor is now known as sleepsonzzz | 21:20 | |
| *** jakedahn is now known as jakedahn_zz | 21:22 | |
| *** martine has quit IRC | 21:23 | |
| *** lorin1 has quit IRC | 21:26 | |
| s1rp | dprince: jk0: if you guys get a sec: https://review.openstack.org/#/c/8007/ | 21:28 |
| jk0 | ok | 21:30 |
| *** Aaton is now known as Aaton_off | 21:32 | |
| *** dubsquared has quit IRC | 21:36 | |
| *** Gordonz has quit IRC | 21:36 | |
| soren | mtaylor: Under what circumstances does git-review not automatically add a Change-Id? | 21:37 |
| *** maoy has quit IRC | 21:38 | |
| *** troytoman-away is now known as troytoman | 21:39 | |
| jeblair | soren: the mechanism is the git "commit-msg" hook, so it runs when that runs, generally on 'git commit'. | 21:41 |
| soren | jeblair: I think I just sorted it out. | 21:42 |
| soren | Not entirely sure why it didn't work before, but meh. | 21:42 |
| jeblair | cool. (as a special case, git-review will ammend your commit so it gets run if git-review has _just_ installed the commit hook, but otherwise doesn't make sure it's there or anything like that) | 21:42 |
| *** rkukura has joined #openstack-dev | 21:42 | |
| *** littleidea has quit IRC | 21:47 | |
| *** dachary has quit IRC | 21:47 | |
| *** s0mik has quit IRC | 21:52 | |
| *** jakedahn_zz is now known as jakedahn | 21:53 | |
| vishy | jgriffith: this is something we definitely need to fix in cinder: https://bugs.launchpad.net/bugs/1008866 | 21:55 |
| uvirtbot | Launchpad bug 1008866 in nova "Creating volume from snapshot on real/production/multicluster installation of OpenStack is broken" [Undecided,New] | 21:55 |
| vishy | jgriffith: I'm not totally sure that the patch is the right solution, but it needs to be addressed | 21:56 |
| jgriffith | vishy: looking at it now | 21:56 |
| *** s0mik has joined #openstack-dev | 21:57 | |
| jgriffith | vishy: Ok, thanks for pointing it out | 22:00 |
| *** rgoodwin is now known as rgoodwin_away | 22:00 | |
| *** rgoodwin_away is now known as rgoodwin | 22:00 | |
| *** ayoung has quit IRC | 22:02 | |
| ttx | vishy: I started working on the new rootwrap stuff. Filters are moving to configuration files. Had a question on config file format for you | 22:03 |
| *** wting has quit IRC | 22:03 | |
| ttx | vishy: I suspect I should try to use INI files like everyone else ? | 22:03 |
| ttx | (I actually want comments, so JSON is not so great) | 22:04 |
| *** wting has joined #openstack-dev | 22:04 | |
| *** torgomatic has quit IRC | 22:05 | |
| *** Ruetobas has joined #openstack-dev | 22:05 | |
| ttx | vishy: with INI files it can look like: http://paste.ubuntu.com/1025836/ | 22:05 |
| *** Aaton_off is now known as Aaton | 22:06 | |
| *** ywu has joined #openstack-dev | 22:06 | |
| *** kbringard has quit IRC | 22:06 | |
| ttx | vishy: the drawback being you need to use a unique name as the key of the key = value list | 22:07 |
| ttx | vishy: YAML would be ideal, but I don't want to dilute our config files formats. | 22:07 |
| vishy | markmc: btw, i noticed we don't support whitespace at the beginning and end of strings in cfg. Thoughts? | 22:08 |
| russellb | should strip() IMO | 22:08 |
| *** edygarcia has quit IRC | 22:08 | |
| ttx | vishy: thoughts ? | 22:08 |
| vishy | russelb: it does, that is the problem | 22:08 |
| ttx | (before I go to bed) | 22:09 |
| vishy | ttx: we have json for policy... | 22:09 |
| russellb | why is that a problem? | 22:09 |
| *** edygarcia has joined #openstack-dev | 22:09 | |
| vishy | russelb, what if you want your config option to end with a space? | 22:09 |
| russellb | i think that would be bizarre :) | 22:09 |
| ttx | vishy: I don't really like JSON, I kinda want to allow comments in there | 22:09 |
| vishy | ttx :) | 22:09 |
| vishy | russellb: for example: cfg.StrOpt('instance_format', default='[instance: %(uuid)s] ', | 22:10 |
| ttx | bah. Will do INI and we'll see how it flies. | 22:10 |
| vishy | ttx: sounds good | 22:10 |
| vishy | russellb: there is actually no way to replicate that option in a config file | 22:10 |
| *** adjohn has quit IRC | 22:10 | |
| russellb | vishy: ah ok. thanks for the use case | 22:11 |
| *** adjohn has joined #openstack-dev | 22:11 | |
| vishy | russellb: my thought is we should support quoted strings | 22:12 |
| *** s0mik has quit IRC | 22:12 | |
| *** somik has joined #openstack-dev | 22:12 | |
| vishy | although I guess then we have to parse \" | 22:12 |
| vishy | russellb: I think it worked at one point, so I"m wondering if it broke when we switched to our own iniparser | 22:13 |
| russellb | quotes would be nice, otherwise it's kind of hard to even see what you've done if whitespace isn't stripped | 22:15 |
| *** sleepsonzzz is now known as sleepsonthefloor | 22:16 | |
| *** e1mer has joined #openstack-dev | 22:18 | |
| *** pmezard has quit IRC | 22:20 | |
| *** milner has quit IRC | 22:20 | |
| *** rnirmal has quit IRC | 22:23 | |
| *** galstrom has quit IRC | 22:26 | |
| *** ncode has joined #openstack-dev | 22:32 | |
| *** e1mer has quit IRC | 22:33 | |
| *** nikhil__ has quit IRC | 22:34 | |
| *** ncode_ has joined #openstack-dev | 22:39 | |
| *** blamar has quit IRC | 22:42 | |
| *** adalbas has quit IRC | 22:42 | |
| *** jgriffith is now known as jgriffith_away | 22:43 | |
| *** pixelbeat has quit IRC | 22:44 | |
| *** relateable has quit IRC | 22:49 | |
| *** edygarcia has quit IRC | 22:50 | |
| *** mattray has quit IRC | 22:50 | |
| *** anderstj has quit IRC | 22:56 | |
| *** roge has quit IRC | 23:00 | |
| *** torgomatic has joined #openstack-dev | 23:02 | |
| *** reed has quit IRC | 23:03 | |
| *** ncode_ has quit IRC | 23:06 | |
| *** winston-d has quit IRC | 23:13 | |
| *** Ryan_Lane has quit IRC | 23:16 | |
| *** anderstj has joined #openstack-dev | 23:17 | |
| *** spiffxp has quit IRC | 23:18 | |
| *** e1mer has joined #openstack-dev | 23:19 | |
| jeblair | jaypipes: looks like there's one failing tempest test: https://jenkins.openstack.org/job/gate-tempest-devstack-vm/1025/testReport/%3Cnose/suite/ContextSuite_context_NetworksTest__setup/ | 23:20 |
| jeblair | jaypipes: thoughts on fixing that and enabling tempest as a gate? | 23:21 |
| *** salgado has quit IRC | 23:26 | |
| jaypipes | jeblair: that's in code review.. | 23:44 |
| jaypipes | jeblair: https://review.openstack.org/#/c/8200/ | 23:44 |
| *** anderstj has quit IRC | 23:48 | |
| smoser | please... https://review.openstack.org/#/c/8126/ someone? "it sucks, -1" is better than nothing. | 23:50 |
| *** jog0 has quit IRC | 23:51 | |
| *** ncode has quit IRC | 23:54 | |
| *** troytoman is now known as troytoman-away | 23:57 | |
| jaypipes | smoser: done. | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
