| *** anotherjesse is now known as anotherjesse_zz | 00:00 | |
| *** reed has quit IRC | 00:33 | |
| *** paulormg has quit IRC | 00:33 | |
| *** mdomsch has quit IRC | 00:49 | |
| *** PotHix has quit IRC | 00:49 | |
| *** littleidea has quit IRC | 00:52 | |
| *** novas0x2a|lapto1 has quit IRC | 00:56 | |
| *** jdurgin has quit IRC | 01:07 | |
| *** littleidea has joined #openstack-dev | 01:11 | |
| *** dolphm has quit IRC | 01:13 | |
| *** dolphm has joined #openstack-dev | 01:15 | |
| *** rackerjoe has joined #openstack-dev | 01:16 | |
| *** rackerjoe has left #openstack-dev | 01:16 | |
| *** flaviamissi_ has joined #openstack-dev | 01:20 | |
| *** flaviamissi has quit IRC | 01:23 | |
| *** vizsla has joined #openstack-dev | 01:24 | |
| *** flaviamissi_ has quit IRC | 01:24 | |
| *** flaviamissi has joined #openstack-dev | 01:24 | |
| *** dolphm has quit IRC | 01:30 | |
| *** Ryan_Lane1 has joined #openstack-dev | 01:31 | |
| *** dolphm has joined #openstack-dev | 01:33 | |
| *** Ryan_Lane has quit IRC | 01:33 | |
| *** andrewsmedina has joined #openstack-dev | 01:34 | |
| *** johnpostlethwait has quit IRC | 01:39 | |
| *** halfss has joined #openstack-dev | 01:40 | |
| *** agonella has joined #openstack-dev | 01:53 | |
| *** dalang has quit IRC | 01:58 | |
| *** westmaas_ has joined #openstack-dev | 01:58 | |
| *** westmaas_ has quit IRC | 01:59 | |
| *** davidkranz has quit IRC | 02:01 | |
| *** breu has joined #openstack-dev | 02:01 | |
| *** vincentricci has left #openstack-dev | 02:03 | |
| *** breu has quit IRC | 02:03 | |
| *** spiffxp has quit IRC | 02:10 | |
| *** dolphm has quit IRC | 02:11 | |
| *** dolphm has joined #openstack-dev | 02:13 | |
| *** ayoung has quit IRC | 02:18 | |
| *** rgoodwin is now known as rgoodwin_away | 02:28 | |
| *** tryggvil_ has quit IRC | 02:30 | |
| *** danwent has quit IRC | 02:35 | |
| *** mdomsch has joined #openstack-dev | 02:42 | |
| *** nati has quit IRC | 02:44 | |
| *** anotherjesse_zz is now known as anotherjesse | 02:46 | |
| *** dwalleck has joined #openstack-dev | 02:55 | |
| *** littleidea has quit IRC | 02:58 | |
| *** anotherjesse is now known as anotherjesse_zz | 03:05 | |
| *** agonella has quit IRC | 03:06 | |
| *** anotherjesse_zz is now known as anotherjesse | 03:06 | |
| *** mnewby has quit IRC | 03:08 | |
| *** dtroyer_zzz is now known as dtroyer | 03:11 | |
| *** pixelbeat has quit IRC | 03:21 | |
| *** mdomsch has quit IRC | 03:23 | |
| *** mdomsch has joined #openstack-dev | 03:25 | |
| *** Ryan_Lane1 has quit IRC | 03:36 | |
| *** danwent has joined #openstack-dev | 03:36 | |
| *** danwent has quit IRC | 03:40 | |
| *** anotherjesse is now known as anotherjesse_zz | 03:47 | |
| *** pixelbeat has joined #openstack-dev | 03:47 | |
| *** anotherjesse_zz is now known as anotherjesse | 03:49 | |
| *** anotherjesse is now known as anotherjesse_zz | 03:51 | |
| *** deshantm has quit IRC | 03:53 | |
| *** dtroyer is now known as dtroyer_zzz | 03:59 | |
| *** jakedahn is now known as jakedahn_zz | 04:00 | |
| *** btorch has quit IRC | 04:17 | |
| *** Ryan_Lane has joined #openstack-dev | 04:22 | |
| *** Ryan_Lane has quit IRC | 04:22 | |
| *** Ryan_Lane has joined #openstack-dev | 04:22 | |
| *** dubsquared has left #openstack-dev | 04:24 | |
| *** thingee has joined #openstack-dev | 04:32 | |
| *** jkoelker has quit IRC | 04:38 | |
| *** anotherjesse_zz is now known as anotherjesse | 04:45 | |
| *** jkoelker has joined #openstack-dev | 04:48 | |
| *** maplebed has quit IRC | 04:48 | |
| *** maplebed has joined #openstack-dev | 04:48 | |
| *** gyee has quit IRC | 04:51 | |
| *** maplebed has quit IRC | 04:53 | |
| *** jakedahn_zz is now known as jakedahn | 04:53 | |
| *** pixelbeat has quit IRC | 05:02 | |
| *** hattwick has quit IRC | 05:02 | |
| *** koolhead17 has joined #openstack-dev | 05:10 | |
| *** jkoelker has quit IRC | 05:14 | |
| *** anotherjesse is now known as anotherjesse_zz | 05:16 | |
| *** mdomsch has quit IRC | 05:17 | |
| *** Mandell has joined #openstack-dev | 05:22 | |
| *** shang has quit IRC | 05:27 | |
| *** vincentricci has joined #openstack-dev | 05:36 | |
| *** jakedahn is now known as jakedahn_zz | 05:37 | |
| *** jakedahn_zz is now known as jakedahn | 05:44 | |
| *** dolphm has quit IRC | 05:45 | |
| *** nati has joined #openstack-dev | 05:46 | |
| *** shang has joined #openstack-dev | 05:47 | |
| *** dwalleck has quit IRC | 06:06 | |
| *** markmc has joined #openstack-dev | 06:25 | |
| *** koolhead17 has quit IRC | 06:36 | |
| *** bepernoot has joined #openstack-dev | 06:36 | |
| *** vincentricci has quit IRC | 06:37 | |
| *** littleidea has joined #openstack-dev | 06:44 | |
| *** Mandell has quit IRC | 06:49 | |
| *** thingee has quit IRC | 06:59 | |
| *** bepernoot has quit IRC | 06:59 | |
| *** bepernoot has joined #openstack-dev | 07:05 | |
| *** zigo has joined #openstack-dev | 07:10 | |
| *** bepernoot has quit IRC | 07:11 | |
| *** winston-d has joined #openstack-dev | 07:12 | |
| *** shang has quit IRC | 07:18 | |
| *** shang has joined #openstack-dev | 07:19 | |
| *** bepernoot has joined #openstack-dev | 07:34 | |
| *** bepernoot has joined #openstack-dev | 07:35 | |
| *** sdake has quit IRC | 07:41 | |
| *** maploin has joined #openstack-dev | 07:42 | |
| *** maploin has quit IRC | 07:42 | |
| *** maploin has joined #openstack-dev | 07:42 | |
| *** bepernoot has quit IRC | 07:47 | |
| *** shang has quit IRC | 07:48 | |
| *** apevec has joined #openstack-dev | 07:49 | |
| *** hattwick has joined #openstack-dev | 07:59 | |
| *** halfss_ has joined #openstack-dev | 08:01 | |
| *** reidrac has joined #openstack-dev | 08:02 | |
| *** bepernoot has joined #openstack-dev | 08:04 | |
| *** halfss has quit IRC | 08:05 | |
| *** vizsla_p has joined #openstack-dev | 08:07 | |
| *** vizsla has quit IRC | 08:10 | |
| *** vizsla_p has quit IRC | 08:11 | |
| *** vincentricci has joined #openstack-dev | 08:13 | |
| *** vincentricci has joined #openstack-dev | 08:14 | |
| *** vincentricci has quit IRC | 08:15 | |
| *** dneary has joined #openstack-dev | 08:23 | |
| *** darraghb has joined #openstack-dev | 08:43 | |
| *** pixelbeat has joined #openstack-dev | 08:55 | |
| *** derekh has joined #openstack-dev | 08:55 | |
| *** winston-d has quit IRC | 08:56 | |
| *** shang has joined #openstack-dev | 09:06 | |
| *** Oneiroi|Zzzz is now known as Oneiroi | 09:40 | |
| *** fc__ has joined #openstack-dev | 10:16 | |
| *** fc__ has joined #openstack-dev | 10:17 | |
| *** tryggvil__ has joined #openstack-dev | 10:21 | |
| *** shang_ has joined #openstack-dev | 10:31 | |
| *** shang_ has quit IRC | 10:31 | |
| *** shang_ has joined #openstack-dev | 10:32 | |
| *** shang_ has quit IRC | 10:32 | |
| *** shang has quit IRC | 10:35 | |
| *** paulormg has joined #openstack-dev | 10:48 | |
| *** shang has joined #openstack-dev | 11:05 | |
| *** shang has quit IRC | 11:06 | |
| *** rohit404 has joined #openstack-dev | 11:51 | |
| *** rohit404 has quit IRC | 11:53 | |
| *** rohit404 has joined #openstack-dev | 11:56 | |
| *** andrewsmedina has quit IRC | 12:04 | |
| *** rods has joined #openstack-dev | 12:08 | |
| *** dneary has quit IRC | 12:12 | |
| *** deshantm has joined #openstack-dev | 12:12 | |
| *** lts has joined #openstack-dev | 12:14 | |
| *** dneary has joined #openstack-dev | 12:21 | |
| *** dneary has quit IRC | 12:21 | |
| *** dneary has joined #openstack-dev | 12:21 | |
| *** sandywalsh has quit IRC | 12:26 | |
| mtaylor | cloudfly: your alternate suggestion doesn't solve jeblair's question at all | 12:27 |
|---|---|---|
| mtaylor | cloudfly: the use case is "I just created a node and now I want to connect to it from somewhere on the internet, what address should I use?" | 12:27 |
| mtaylor | I understand that the theory is that networking can be set up a billion ways | 12:28 |
| mtaylor | but generally speaking, the simple case is that people want to access nodes they create, so that should be handled sensibly | 12:29 |
| mtaylor | and then having some extra-crazy APIs for doing extra-special things is awesome | 12:29 |
| soren | mtaylor, ttx: Is there any particular reason there's no "essex" branch of the various projects yet? | 12:34 |
| soren | I don't care about Folsom yet (and might not for a while). It's awkward that I have to pull something called "milestone-proposed" since a couple of days ago, and then probably switch to something else ("stable/essex" supposedly) later on. | 12:35 |
| soren | It's rather disruptive | 12:36 |
| *** alaski has joined #openstack-dev | 12:36 | |
| *** dolphm has joined #openstack-dev | 12:36 | |
| ttx | soren: stable/essex will be cut from milestone-proposed | 12:36 |
| mtaylor | ttx: I kindof agree with soren | 12:37 |
| mtaylor | ttx: at this point, since master is folsom, I'd argue that milestone-proposed should be named stable/essex even if it hasn't produced an essex release yet - but I might just be crazy | 12:37 |
| ttx | We reuse all the proven toolset that applies to milestone-proposed, i don't really want to experiment with new bugs resulting from a new toolset at this point | 12:37 |
| mtaylor | that's fair | 12:38 |
| ttx | also it's *not* stable/essex. At all. | 12:38 |
| soren | It took me a good while to realise why things were breaking (hadn't changed from master to milestone-proposed). It's worse that I have to do it again (and the exact time depends on people). | 12:38 |
| ttx | You could call it wanabee/essex. But that wouldn't help. | 12:38 |
| soren | Does git support aliases for branches? | 12:38 |
| mtaylor | are we not treating patches to it with a similar scrutiny to how we would look at patches to stable/diablo? | 12:38 |
| soren | If so, can't we set Folsom to point to master, Essex to point to milestone-proposed, and once Essex is out, have essex point to stable/essex? | 12:39 |
| ttx | the rules are different from stable/diablo. | 12:39 |
| ttx | Stable branches are maintained by the stable team. Release branches are maintained by the release team | 12:39 |
| ttx | calling them the same just adds confusion. | 12:39 |
| soren | ttx: I see what you're saying, but I don't like that we're overloading the branch name with designations as to whether something has been released. | 12:39 |
| *** sandywalsh has joined #openstack-dev | 12:39 | |
| mtaylor | ah, ok -sorry | 12:39 |
| soren | ttx: Hence the alias idea. | 12:39 |
| ttx | jenkins/gerrit applies special rules to things named stable/* | 12:40 |
| soren | mtaylor: Do you know? Can we have multiple names for the same branch? | 12:40 |
| mtaylor | soren: I do not believe that we have the technical capability to do the thing you're suggesting | 12:40 |
| ttx | I don't really want to see how many side-effects that would create | 12:40 |
| soren | Ok, let's rewind. | 12:40 |
| soren | Pretend that you're working on Essex. | 12:40 |
| soren | And have "stuff" set up to do things with that. | 12:41 |
| soren | How can I do that with as little disruption as possible? | 12:41 |
| soren | Having to change the branch I pull from at an arbitrary time isn't very elegent. | 12:41 |
| soren | elegant, even. | 12:41 |
| mtaylor | soren: I agree - and I do not believe we have served that particular use case in our branching design - following the soon-to-be-released release in the RC window is a wee-bit different | 12:43 |
| ttx | soren: a branch alias (supposing git supports that) that says "essex" would have helped, *if* set before | 12:43 |
| ttx | we could have redirected it | 12:43 |
| soren | ttx: Why not now? | 12:43 |
| mtaylor | again - branch alias == not possible | 12:43 |
| ttx | changing your tools from master to essex or from master to meilstone-proposed... it's changing anyway | 12:44 |
| soren | mtaylor: 3 minutes ago, it was "I do not believe" :) did you check in the mean time? | 12:44 |
| mtaylor | soren: sorry, that was just me being less direct in my speech | 12:44 |
| ttx | mtaylor: even local aliases ? | 12:44 |
| soren | ttx: The difference is something that changes now vs. someething that changes at an arbitrary later timer. | 12:44 |
| mtaylor | local aliases would be a totally different thing | 12:44 |
| soren | time. | 12:44 |
| soren | mtaylor: Fascinating. Tell me all about it. | 12:45 |
| mtaylor | you can have a local branched called essex that is set up to track remotes/origin/master | 12:45 |
| soren | mtaylor: Except... No, never mind. That's going to suck, too. | 12:45 |
| mtaylor | and then, you can change the upstream tracking branch to be remotes/origin/milestone-proposed | 12:45 |
| mtaylor | etc. | 12:45 |
| ttx | soren: err... what ? arbitrary later time ? | 12:45 |
| soren | ttx: Release time. | 12:45 |
| soren | ttx: The second stable/essex starts existing. | 12:46 |
| mtaylor | I mean - it's not stellar, but _might_ be helpful | 12:46 |
| ttx | ah, so you want something that will make milestone-proposed -> stable/essex transition less painful. Not master -> milestone-proposed. | 12:46 |
| soren | ttx: Right. | 12:46 |
| soren | ttx: For now. | 12:46 |
| soren | ttx: The other problem should be solved for Folsom, but right now I just care about Essex. | 12:46 |
| soren | I guess it's less disruptive than the master -> m-p transition. | 12:47 |
| mtaylor | yes. | 12:47 |
| soren | For m-p -> stable/essex, m-p will just sit there. | 12:47 |
| mtaylor | because milestone-proposed won't become something else | 12:47 |
| soren | No changes for a while. | 12:47 |
| ttx | soren: we can't call one the other, because a lot of the toolchain assumes things based on those names. We could have apointer than points to one and is displaced to point to the other. If git supports it somehow | 12:47 |
| mtaylor | it does not | 12:47 |
| soren | While for the master -> m-p transition, suddenly a bunch of chagnes started landing that I had to back out of. | 12:47 |
| mtaylor | for folsom, I think we should think about the gutter time between the point when master == folsom and stable/folsom exists | 12:48 |
| *** andrewsmedina has joined #openstack-dev | 12:48 | |
| mtaylor | because I think it's a valid use case | 12:48 |
| * soren sucks it up and starts changing everything to point to m-p. | 12:48 | |
| mtaylor | I have no answers for it at the moment though | 12:48 |
| soren | Oh, also very annoying: | 12:48 |
| soren | Keystone doesn't have an m-p branch. | 12:49 |
| *** hugokuo has joined #openstack-dev | 12:49 | |
| soren | So I can't do it across the board. | 12:49 |
| dolphm | is that because we haven't hit RC status? | 12:49 |
| *** mattray has joined #openstack-dev | 12:49 | |
| soren | dolphm: Presumably. | 12:50 |
| ttx | yes | 12:51 |
| mtaylor | soren: oh, there _may_ be a way to do your alias... | 12:51 |
| mtaylor | soren: I need to test to see how it works with gerrit | 12:51 |
| ttx | the toolset generally does "milestone-proposed and if not, master" | 12:51 |
| soren | mtaylor: If it's at all intrusive, don't worry about it. | 12:51 |
| *** dprince has joined #openstack-dev | 12:51 | |
| *** dprince has quit IRC | 12:51 | |
| soren | I'll accept the pain for now. | 12:51 |
| soren | I'll complain and moan, but deal with it | 12:52 |
| mtaylor | soren: I can test it on the gerrit dev server | 12:52 |
| soren | Ah. | 12:52 |
| *** dprince has joined #openstack-dev | 12:53 | |
| markmc | of course, we could just call the stable branch "essex" | 12:55 |
| markmc | and create it a RBP time | 12:55 |
| mtaylor | markmc: we need the stable/ prefix for tooling purposes | 12:56 |
| markmc | meh | 12:56 |
| mtaylor | markmc: otherwise we have no system way to know that the stable maintainers should have access to the branch | 12:56 |
| markmc | we could have rules for the "essex" branch in gerrit | 12:57 |
| markmc | and change those rules after release time | 12:57 |
| mtaylor | yes. we could | 12:57 |
| ttx | markmc: RC12 was RBP, so it's past now | 12:57 |
| mtaylor | but that's actually a bunch of sets of rules | 12:57 |
| ttx | RC1* | 12:57 |
| mtaylor | because we'd have to do that for every project | 12:57 |
| ttx | (for almost everyone) | 12:58 |
| mtaylor | the master vs. stable/* thing is actually working quite well and elegantly | 12:58 |
| markmc | you could have rules for the "essex" branch that applies to All-Projects | 12:58 |
| markmc | ttx, ok, pretend we're talking about folsom | 12:58 |
| markmc | the confusion soren is talking about is a real problem IMHO | 12:58 |
| ttx | frankly I'm not sure the problem is significant enough to warrant changes, but yes, we can discuss Folsom changes :) | 12:58 |
| mtaylor | I do agree with markmc and soren that it's a problem | 12:59 |
| ttx | markmc: so we could have a "folsom" alias that ends up pointing to master, then milestone-proposed | 12:59 |
| ttx | what happens around milestone release times though ? | 12:59 |
| ttx | is the F1 branch "folsom", or is master branch "folsom" ? | 12:59 |
| markmc | well, there's a whole other problem | 13:00 |
| mtaylor | ttx: master | 13:00 |
| markmc | the milestone-proposed branch rebases | 13:00 |
| markmc | that's a no-no for git, really | 13:00 |
| markmc | you'd have folsom-1, folsom-2, folsom-3 branches for cutting those | 13:00 |
| ttx | markmc: I've been discussing release branches separate from master with mtaylor, and apparently they are a no-no too. | 13:00 |
| ttx | and I thought git was flexible :) | 13:00 |
| markmc | mtaylor, why are release branches a no-no? | 13:01 |
| mtaylor | ttx: I do not believe I have said that release branches are a no-no | 13:01 |
| ttx | you said topic branches were. | 13:01 |
| mtaylor | ttx: I believe that I said that the way that you were suggesting using branching would lead to pain | 13:01 |
| mtaylor | topic branches and release branches are completely different things | 13:01 |
| ttx | which are a corollary of release branch. | 13:01 |
| mtaylor | they are not | 13:01 |
| markmc | ttx, git is very flexible - you can rebase branches, but some people go apeshit when they see folks doing that | 13:01 |
| *** deshantm_ has joined #openstack-dev | 13:02 | |
| mtaylor | we may be getting in to the weeds here | 13:02 |
| ttx | yes, it diverges | 13:02 |
| ttx | markmc: we need to define Folsom release cycle before discussing branches and tooling necessary to achieve it | 13:03 |
| ttx | there may just be no such thing as folsom-1. | 13:03 |
| mtaylor | I have to board a plane - I'll be back on in a few minutes... | 13:03 |
| * markmc wanders off too | 13:03 | |
| ttx | mtaylor: I think we are done :) | 13:04 |
| *** deshantm has quit IRC | 13:06 | |
| *** halfss_ has quit IRC | 13:07 | |
| *** kbringard has joined #openstack-dev | 13:30 | |
| *** ayoung has joined #openstack-dev | 13:32 | |
| *** sdake has joined #openstack-dev | 13:34 | |
| *** sdake has quit IRC | 13:42 | |
| sandywalsh | anotherjesse_zz, care to ping me when you un_zz? | 13:42 |
| *** sdake has joined #openstack-dev | 13:43 | |
| *** crobinso has joined #openstack-dev | 13:43 | |
| *** rgoodwin_away is now known as rgoodwin | 13:44 | |
| *** dolphm has quit IRC | 13:49 | |
| zykes- | How is it Nova sends data over AMQP, or more precisely what database does it use or so ? Just data encoded as JSON or ? | 13:49 |
| *** mdomsch has joined #openstack-dev | 13:50 | |
| *** dolphm has joined #openstack-dev | 13:50 | |
| *** maploin` has joined #openstack-dev | 13:52 | |
| *** maploin has quit IRC | 13:54 | |
| *** maploin` is now known as maploin | 13:55 | |
| *** maploin has joined #openstack-dev | 13:55 | |
| *** rohit404 has quit IRC | 13:55 | |
| *** davidkranz has joined #openstack-dev | 13:57 | |
| *** hugokuo has quit IRC | 13:57 | |
| *** dtroyer_zzz is now known as dtroyer | 13:58 | |
| *** martines has quit IRC | 14:03 | |
| dprince | LinuxJedi: The new Gerrit styles are nice! | 14:03 |
| dolphm | LinuxJedi: is there a public repo for that somewhere? | 14:05 |
| russellb | https://github.com/openstack/openstack-ci-puppet/tree/master/modules/gerrit | 14:07 |
| russellb | i think it's in there | 14:08 |
| zykes- | noone got a clue ? | 14:10 |
| russellb | zykes-: the data gets serialized, yes. It's done by the Python library used for AMQP - python-kombu or python-qpid | 14:11 |
| russellb | someone told me python-kombu just uses json, but I don't think python-qpid does. | 14:12 |
| zykes- | ah ok | 14:13 |
| *** martines has joined #openstack-dev | 14:17 | |
| LinuxJedi | dprince: thanks :) | 14:25 |
| LinuxJedi | dolphm: http://www.linuxjedi.co.uk/2012/03/new-gerrit-and-jenkins-styles-are-live.html | 14:25 |
| *** sandywalsh has quit IRC | 14:27 | |
| kbringard | anyone know if the simple scheduler tracks memory usage and allocates, or not, based on that information? | 14:28 |
| dolphm | LinuxJedi: awesome, thanks | 14:28 |
| *** Glacee has joined #openstack-dev | 14:32 | |
| *** danwent has joined #openstack-dev | 14:36 | |
| eglynn__ | russellb: on the subject of AMQP and JSON, you can probably speak to whether this change to set the content-type explicitly to 'application/json' is need for the qpid notifier also ... | 14:37 |
| zykes- | What's the difference of a Routing key and a Queue in AMQP ? | 14:38 |
| eglynn__ | https://review.openstack.org/5594 (see last review comment) | 14:38 |
| russellb | zykes-: a queue is pretty much what it sounds like ... the routing key is used to determine which queue(s) the message ends up in. | 14:39 |
| russellb | how the routing key is processed depends on the message exchange type. | 14:39 |
| russellb | eglynn__: *looks* | 14:39 |
| *** sandywalsh has joined #openstack-dev | 14:41 | |
| *** dtroyer is now known as dtroyer_zzz | 14:41 | |
| *** bepernoot has quit IRC | 14:41 | |
| kbringard | has anyone noticed that dhcp_release doesn't seem to be working anymore? or is that just me? | 14:42 |
| russellb | eglynn__: yes, it looks like a similar change will be needed in qpid, as well. | 14:43 |
| eglynn__ | russellb: cool, thanks for confirming | 14:43 |
| russellb | eglynn__: i'll try to dig up a pointer as to what the change should look like and update the review. | 14:43 |
| eglynn__ | russellb: excellent thanks! | 14:43 |
| *** reed has joined #openstack-dev | 14:43 | |
| *** zzed has joined #openstack-dev | 14:46 | |
| kbringard | does the dhcp_lease_time flag in the nova.conf tell the dhcpbridge how long to wait before allowing an IP to be re-allocated? | 14:49 |
| *** dwalleck has joined #openstack-dev | 14:50 | |
| *** zykes- has quit IRC | 14:57 | |
| *** zykes- has joined #openstack-dev | 14:57 | |
| *** dwalleck has quit IRC | 14:59 | |
| *** vizsla has joined #openstack-dev | 15:00 | |
| *** dneary has quit IRC | 15:06 | |
| *** n0ano has quit IRC | 15:08 | |
| dolphm | LinuxJedi: modules/jenkins_master/files/openstack.css ? | 15:09 |
| LinuxJedi | dolphm: exactly, the heading is rendered in the .js though | 15:09 |
| LinuxJedi | dolphm: which is probably in modules/jenkins_master/templates/ | 15:10 |
| dolphm | LinuxJedi: updated | 15:12 |
| *** dachary has quit IRC | 15:13 | |
| *** rnirmal has joined #openstack-dev | 15:13 | |
| dolphm | LinuxJedi: is this stuff redeployed automatically? | 15:14 |
| LinuxJedi | dolphm: yes | 15:14 |
| LinuxJedi | dolphm: puppet executes every 15 minutes, so it will be within 15 minutes of a merge | 15:15 |
| dolphm | awesome | 15:15 |
| andrewbogott | pvo: Thanks for the merge. Do you know if/how I should merge that change into the hacking guidelines of other projects? | 15:24 |
| pvo | andrewbogott: I don't see why not. | 15:25 |
| andrewbogott | I haven't looked to see if HACKING is identical in each project. I guess it most likely is. | 15:26 |
| andrewbogott | Although with different names :( | 15:26 |
| andrewbogott | pvo: I'm also wondering if encoding is something that people generally already know about, or if it needs to be evangelized. Do you have any instinct about that? (for example, I could schedule a 5-minute encoding rant at the summit.) | 15:29 |
| *** maploin has quit IRC | 15:30 | |
| *** littleidea has quit IRC | 15:30 | |
| soren | dolphm: Got a sec for Keystone question? | 15:31 |
| soren | I'm running into a problem trying to create a service endpoint. I'm using the sql driver, but by the looks of the code, this problem applies globally. | 15:31 |
| *** Susanne-Balle has joined #openstack-dev | 15:32 | |
| pvo | andrewbogott: I think most people do. | 15:32 |
| pvo | I think there are already some tests for this in Nova.. just need to get those added to novacliaent | 15:32 |
| pvo | client | 15:32 |
| andrewbogott | pvo: Great, I'll let it be then. | 15:32 |
| soren | keystone.catalog.core.EndPointController.create_endpoint calls self.catalog_api.service_exists(blah). | 15:33 |
| soren | This fails with an error that service_exists doesn't exist. | 15:33 |
| soren | ...and "grep -r" agrees. | 15:33 |
| soren | I'm a bit confused. | 15:34 |
| soren | My usage of keystone seems *very* basic. I'm creating users and tenants and services and endpoints. | 15:34 |
| soren | Yet it's failing. | 15:34 |
| *** dtroyer_zzz is now known as dtroyer | 15:34 | |
| *** n0ano has joined #openstack-dev | 15:34 | |
| soren | I understand bugs creep in. Happens to the best of us. | 15:34 |
| soren | ..but how does anyone manage to use Keystone without those things? How is anyone else using keystone? | 15:35 |
| *** martines has quit IRC | 15:36 | |
| russellb | soren: https://review.openstack.org/#change,5636 | 15:37 |
| soren | russellb: Ah, lovely. | 15:38 |
| * soren applies | 15:38 | |
| dolphm | soren: just got back to my desk | 15:38 |
| dolphm | soren: there's a critical fix in review for that | 15:39 |
| dolphm | soren: https://review.openstack.org/#change,5636 | 15:39 |
| dolphm | soren: but, no one seems interested in clearing keystone's last RC blocker *shrug* | 15:40 |
| soren | dolphm: What do you mean? It's got 4 +1's? | 15:41 |
| dolphm | waiting on approval | 15:41 |
| dolphm | termie: anotherjesse_zz: heckj: ^^ | 15:42 |
| kbringard | hey soren, quick ?… do you happen to know what the dhcp_lease_time flag in the nova.conf is used for? | 15:42 |
| *** AlanClark has joined #openstack-dev | 15:42 | |
| soren | kbringard: Without actually checking, I'm going to guess that it defined the dhcp lease time. | 15:43 |
| kbringard | it doesn't adjust the lease time dnsmasq hands out, that appears to be hardcoded | 15:43 |
| soren | kbringard: But I'm sure you have a different opinion? | 15:43 |
| kbringard | I knew you did quite a bit with the network code so I thought you may know off the top of your head :-) | 15:43 |
| kbringard | but don't spend any time on it, unless you're curious :-D | 15:44 |
| dolphm | chmouel: i forgot about https://bugs.launchpad.net/keystone/+bug/958135 -- were you able to reproduce? | 15:44 |
| uvirtbot | Launchpad bug 958135 in keystone "UnboundLocalError: local variable 'metadata_ref' referenced before assignment" [Critical,Triaged] | 15:44 |
| chmouel | dolphm: I am working currently with uksysadmin to figure out how to reproduce it | 15:44 |
| soren | kbringard: What makes you say it's hard coded? | 15:45 |
| soren | kbringard: Glancing(!) at the code suggests that it does indeed set the dhcp lease time. | 15:46 |
| dolphm | chmouel: ah cool | 15:46 |
| kbringard | soren: '--dhcp-range=%s,static,120s' % network_ref['dhcp_start'], | 15:46 |
| kbringard | that's the flag dnsmasq uses when it launches | 15:46 |
| kbringard | so long as that is 120s, that's what dnsmasq is going to pass to the clients | 15:46 |
| kbringard | I'm looking at the diablo-stable branch, btw, so maybe it's been changed in essex? | 15:47 |
| soren | Are we looking at the... | 15:47 |
| soren | Ah. | 15:47 |
| kbringard | either way, the crux of my issue is, I want to extend the dhcp lease times | 15:47 |
| soren | Sheesh. Diablo is so last year. | 15:47 |
| kbringard | but, for some reason dhcp_release doesn't seem to be working | 15:48 |
| soren | Essex does it correctly. | 15:48 |
| kbringard | so dnsmasq is keeping leases | 15:48 |
| kbringard | and then nova is trying to reassign them to new VMs | 15:48 |
| kbringard | which needless to say, causes problems | 15:48 |
| soren | I thought I fixed that at some point. | 15:48 |
| *** vizsla has quit IRC | 15:49 | |
| kbringard | well, this seems to be a problem with dhcp_release and dnsmasq… like, running dhcp_release manually doesn't even cause it to release | 15:49 |
| kbringard | I can see nova-network running it | 15:49 |
| kbringard | and the sudoers are setup right | 15:49 |
| kbringard | so I run it as root, and same thing | 15:49 |
| kbringard | I don't see the DHCPRELEASE in the syslog | 15:49 |
| * kbringard bangs his head on the desk | 15:50 | |
| kbringard | which needless to say isn't our problem, here | 15:50 |
| kbringard | but it still makes me crazy :-) | 15:51 |
| kbringard | anyway, I asked because, I saw these entries | 15:53 |
| kbringard | 2012-03-21 20:40:56,277 DEBUG nova.network.manager [baf5ad2f-bad1-4422-8bc2-b9efc3db4525 None None] Leased IP |10.0.4.8| from (pid=31781) lease_fixed_ip /usr/lib/pymodules/python2.7/nova/network/manager.py:645 | 15:53 |
| kbringard | 2012-03-21 20:40:56,327 WARNING nova.network.manager [baf5ad2f-bad1-4422-8bc2-b9efc3db4525 None None] IP |10.0.4.8| leased that isn't allocated | 15:53 |
| *** jdg has joined #openstack-dev | 15:53 | |
| kbringard | and thought that perhaps I could tell the dhcpbridge client to not reallocate those IPs for X time | 15:53 |
| kbringard | which could effectively work around the issu | 15:53 |
| soren | I remember it being a looooong day when I decided to fix it last time. | 15:56 |
| soren | kbringard: Not something I'm willing to take on 5 minutes before EOD. | 15:56 |
| kbringard | was it an issue in dhcp_release or nova? | 15:56 |
| *** jkoelker has joined #openstack-dev | 15:57 | |
| soren | nova | 15:57 |
| kbringard | hmmm, OK, I'll look into it | 15:57 |
| soren | But this might be a completely different problme, of course. | 15:57 |
| kbringard | yea, like I said, I can't even get it to release issuing the dhcp_release command manually | 15:58 |
| kbringard | nova looks like it's doing the right thing | 15:58 |
| * soren wnders off | 16:01 | |
| jaypipes | soren: such a nomad. | 16:03 |
| *** maplebed has joined #openstack-dev | 16:04 | |
| *** GheRivero_ has joined #openstack-dev | 16:06 | |
| kbringard | the wanderlust will get you every time | 16:07 |
| *** dwalleck has joined #openstack-dev | 16:08 | |
| *** dneary has joined #openstack-dev | 16:09 | |
| *** dneary has quit IRC | 16:09 | |
| *** dneary has joined #openstack-dev | 16:09 | |
| *** jdurgin has joined #openstack-dev | 16:09 | |
| *** koolhead17|away has joined #openstack-dev | 16:13 | |
| *** littleidea has joined #openstack-dev | 16:13 | |
| *** spiffxp has joined #openstack-dev | 16:13 | |
| *** nati has joined #openstack-dev | 16:16 | |
| *** reidrac has quit IRC | 16:17 | |
| *** dwalleck has quit IRC | 16:21 | |
| *** hub_cap has joined #openstack-dev | 16:21 | |
| *** martines has joined #openstack-dev | 16:22 | |
| *** dalang has joined #openstack-dev | 16:22 | |
| *** rohit404 has joined #openstack-dev | 16:26 | |
| vishy | Vek: ping | 16:33 |
| *** deshantm_ has quit IRC | 16:34 | |
| *** deshantm has joined #openstack-dev | 16:34 | |
| *** dwalleck has joined #openstack-dev | 16:38 | |
| *** alaski has quit IRC | 16:39 | |
| *** alaski has joined #openstack-dev | 16:39 | |
| *** Mandell has joined #openstack-dev | 16:42 | |
| *** dubsquared has joined #openstack-dev | 16:43 | |
| jaypipes | QA meetign on #openstack-meeting in 15 minutes! | 16:45 |
| chmouel | dolphm: by the way you were right about that /ec2token http://pastie.org/pastes/3648860/text?key=lkofxqjjswqnfudg3rpw0g | 16:46 |
| *** anotherjesse_zz is now known as anotherjesse | 16:46 | |
| chmouel | vishy: do we need something like OS-KSEC2:s3Credentials for s3 endpoint? | 16:46 |
| *** shevek_ has joined #openstack-dev | 16:46 | |
| vishy | Vek: I want to discuss the policy.json on the quotas branch. I don't think it is correct | 16:47 |
| vishy | chmouel: ec2 cres work fine right? | 16:47 |
| vishy | chmouel: it should probably have been called aws creds or something | 16:47 |
| chmouel | vishy: yeah both call works, but on s3 we just have /s3tokens (and agreed on that aws i think this was pointed by jesse sometime ago but we didn't do the change) | 16:48 |
| *** Ryan_Lane has quit IRC | 16:49 | |
| *** Mandell has quit IRC | 16:50 | |
| *** Mandell has joined #openstack-dev | 16:51 | |
| *** Mandell has quit IRC | 16:55 | |
| *** GheRivero_ has quit IRC | 16:56 | |
| *** LinuxJedi has quit IRC | 16:58 | |
| *** LinuxJedi has joined #openstack-dev | 16:58 | |
| *** derekh has quit IRC | 17:00 | |
| *** danwent has quit IRC | 17:01 | |
| *** rohit404 has left #openstack-dev | 17:02 | |
| *** vincentricci has joined #openstack-dev | 17:07 | |
| *** vishy is now known as squish | 17:07 | |
| *** squish is now known as squishy | 17:07 | |
| *** sandywalsh has quit IRC | 17:07 | |
| *** squishy is now known as vishy | 17:08 | |
| *** PotHix has joined #openstack-dev | 17:08 | |
| *** bhall has joined #openstack-dev | 17:10 | |
| *** bhall has quit IRC | 17:10 | |
| *** bhall has joined #openstack-dev | 17:10 | |
| *** zaitcev has joined #openstack-dev | 17:12 | |
| *** stuntmachine has joined #openstack-dev | 17:17 | |
| *** shevek_ has quit IRC | 17:17 | |
| *** sandywalsh has joined #openstack-dev | 17:23 | |
| *** sandywalsh has quit IRC | 17:28 | |
| *** dwalleck has quit IRC | 17:32 | |
| *** jshepher has joined #openstack-dev | 17:34 | |
| jshepher | is a two node quantum/openvswitch setup possible with essex (git clones this morning) | 17:35 |
| jshepher | looks like the gw-<uuid> interface does not get properly created on the second (compute) node.. | 17:35 |
| jshepher | and nova-compute throws errors when trying to attach the tap devices to the br-int bridge | 17:36 |
| *** zigo has quit IRC | 17:37 | |
| *** Oneiroi is now known as Oneiroi^gone | 17:38 | |
| jaypipes | andrewbogott: you just want to stir the hornet's nest, don't you? ;) | 17:38 |
| jaypipes | jshepher: might want to ping danwent when he gets in.. | 17:39 |
| anotherjesse | jaypipes / bcwaldon - any opinions on https://review.openstack.org/#change,5565,patchset=1 (move glance conf dir to /etc/glance) | 17:40 |
| bcwaldon | definitely in support of it, makes it frustrating to check out new code in devstack installs | 17:41 |
| bcwaldon | anotherjesse: ^ | 17:41 |
| *** Ryan_Lane has joined #openstack-dev | 17:41 | |
| *** adjohn has joined #openstack-dev | 17:41 | |
| *** sandywalsh has joined #openstack-dev | 17:41 | |
| jaypipes | anotherjesse: I'm fine with that | 17:41 |
| *** novas0x2a|laptop has joined #openstack-dev | 17:41 | |
| jaypipes | anotherjesse: more consistent with the other services. | 17:41 |
| bcwaldon | +2a'd | 17:42 |
| *** dneary has quit IRC | 17:42 | |
| jeblair | bcwaldon, annegentle: is there a formal api doc for the floating ips extension? | 17:43 |
| jeblair | something like this: http://api-docs.hpcloud.com/hpcloud-compute-supplement/1.0/content/compute-floating-ips.html | 17:43 |
| jeblair | but openstack and not hpcloud? | 17:43 |
| bcwaldon | jeblair: I'll have to let annegentle answer that one, I think there might be, just not sure where its hosted | 17:43 |
| jeblair | annegentle: ^? | 17:44 |
| *** koolhead17|away is now known as koolhead17 | 17:46 | |
| jshepher | jaypipes: thanks i will check with him | 17:50 |
| bcwaldon | vishy: https://review.openstack.org/#change,5696 | 17:50 |
| *** gyee has joined #openstack-dev | 17:52 | |
| vishy | bcwaldon: wow, is it true that instance_type_extra_specs and volume_type_extra_specs, the extra_specs param is underscored ? | 17:53 |
| vishy | :( | 17:53 |
| bcwaldon | vishy: where are you looking? | 17:54 |
| bcwaldon | vishy: oh crap, yeah | 17:55 |
| vishy | bcwaldon: test_volume_types_extra_specs and test_instance_types_extra_specs | 17:55 |
| bcwaldon | vishy: the whole types resource needs to be updated too | 17:55 |
| *** danwent has joined #openstack-dev | 17:55 | |
| vishy | bcwaldon: sigh i guess we should leave it | 17:55 |
| bcwaldon | vishy: leave volume types but fix volume/snap create? | 17:56 |
| *** novas0x2a|laptop has quit IRC | 17:56 | |
| vishy | bcwaldon: the only key is extra_specs | 17:56 |
| vishy | bcwaldon: and it matches flavors | 17:56 |
| vishy | so ... | 17:56 |
| *** zigo has joined #openstack-dev | 17:57 | |
| vishy | bcwaldon: what do you think? | 17:57 |
| *** rohit404 has joined #openstack-dev | 17:57 | |
| vishy | bcwaldon: update instance_types_extra_specs as well? | 17:57 |
| *** novas0x2a|laptop has joined #openstack-dev | 17:57 | |
| bcwaldon | vishy: blargh | 17:57 |
| bcwaldon | leave it | 17:57 |
| *** gyee has quit IRC | 17:59 | |
| vishy | bcwaldon: ok | 17:59 |
| vishy | extra_specs is an extension so we could change it... | 18:01 |
| vishy | but maybe folsom | 18:01 |
| bcwaldon | vishy: its not just an extension though | 18:03 |
| bcwaldon | vishy: since its presented in /types | 18:03 |
| kbringard | quick ?… fixed_ip_disassociate_timeout is how long after a deallocated IP is available for reuse in the dhcp pool, correct? | 18:03 |
| bcwaldon | vishy: and volume types itself is presented as 'volume_types' | 18:03 |
| vishy | bcwaldon: oh rly | 18:04 |
| bcwaldon | vishy: yep | 18:04 |
| vishy | bcwaldon: seems like we should at least change volumeTypes | 18:04 |
| vishy | where is that presented? | 18:04 |
| bcwaldon | vishy: look in api/openstack/volumes/types.py | 18:04 |
| bcwaldon | vishy: returns in the volume types api at /types | 18:04 |
| vishy | oh yes change that as well | 18:05 |
| vishy | but leave extra_specs? | 18:05 |
| vishy | i don't know what to do with extra_specs :( | 18:05 |
| bcwaldon | yeah... its part of the api, not anextension in this context | 18:06 |
| vishy | bcwaldon: it doesn't exist in flavors btw | 18:06 |
| vishy | in the spec | 18:07 |
| vishy | so it is an extension | 18:07 |
| *** btorch has joined #openstack-dev | 18:07 | |
| vishy | 'extension | 18:07 |
| vishy | in compute | 18:07 |
| vishy | so we might as well change it in the volume api since it is part of v1 ... | 18:07 |
| bcwaldon | ok | 18:07 |
| bcwaldon | on it | 18:07 |
| annegentle | jeblair: I have talked to the content creator at HP who wrote that document, but he hasn't given it back yet, waiting on legal approval | 18:11 |
| annegentle | jeblair: sorry for the long delay, was in a meeting :) | 18:11 |
| annegentle | jeblair: I'd like to get something like that supplement for all the Extensions, and had hoped HP would give it to OpenStack | 18:11 |
| sdague | I've noticed quite a large number of items in the bug tracker with the forward-port-needed that have fixes in an NTT branch from Nov of last year? is there still a desire to get all those fixes into the tree? | 18:11 |
| *** deshantm_ has joined #openstack-dev | 18:13 | |
| *** deshantm_ has quit IRC | 18:13 | |
| *** deshantm_ has joined #openstack-dev | 18:14 | |
| *** gyee has joined #openstack-dev | 18:16 | |
| *** deshantm has quit IRC | 18:16 | |
| jeblair | annegentle: who wrote that/who are you dealing with at HP? | 18:16 |
| jeblair | annegentle: I believe HPCS has blanket permission to contribute to openstack. | 18:17 |
| nati | sdague: We wrote many unittests and report bugs https://bugs.launchpad.net/nova/+bugs?field.tag=forward-port-needed | 18:17 |
| nati | sdague: Unfortunately, unittests are outdated now. | 18:17 |
| nati | sdague: for essex | 18:17 |
| nati | sdague: Important bugs looks already fixed | 18:18 |
| annegentle | jeblair: yeah and those are all the "contrib" OS extensions. should be fine. See PM | 18:18 |
| sdague | nati: ok, some of the fixes still seem relevant, at least one that I rebased and pushed for review the other day | 18:18 |
| sdague | nati: does that mean that those should be closed out entirely? | 18:18 |
| nati | sdague: I don't think it should be closed. These bugs are filed and triaged by ttx on launchpad | 18:19 |
| nati | sdague: And we are working on the bug from high priority ones | 18:19 |
| sdague | nati: ok, well if you would like some help I'm happy to lend a hand to work through some of them | 18:20 |
| sandywalsh | anotherjesse, ping | 18:20 |
| anotherjesse | sandywalsh: pong | 18:20 |
| sandywalsh | skype? | 18:20 |
| sandywalsh | or type? | 18:20 |
| anotherjesse | sure - let me get to a room - will ping you on skype | 18:20 |
| sandywalsh | k | 18:21 |
| nati | sdague: Thanks! | 18:21 |
| sdague | nati: just point me at an item you'd like me to work on, and I'll do it. Just want to make sure I'm not duplicating other folks work | 18:22 |
| nati | sdague: Thanks. There are no worries for depilation of work. Because launchpad bug will be assign one guy. | 18:23 |
| *** troytoman-away is now known as troytoman | 18:24 | |
| nati | sdague: I'm really appreciate your help, but we should fix more high priority bugs. | 18:24 |
| nati | sdague: forward-port-needed bugs have some medium priority bug, but some high priority bugs not assigned https://bugs.launchpad.net/nova/+bugs?search=Search&field.importance=High&field.status=New&field.status=Incomplete&field.status=Confirmed&field.status=Triaged&field.status=In+Progress&field.status=Fix+Committed | 18:25 |
| nati | sdague: How about https://bugs.launchpad.net/nova/+bug/938317 ? | 18:26 |
| uvirtbot | Launchpad bug 938317 in nova "rapid firing from command line can go over quota" [High,Confirmed] | 18:26 |
| nati | sdague: And also low-hanging-fruit one are easy to fixed one https://bugs.launchpad.net/nova/+bugs?field.tag=low-hanging-fruit | 18:27 |
| *** dalang is now known as fattarsi | 18:28 | |
| *** littleidea has quit IRC | 18:31 | |
| *** littleidea has joined #openstack-dev | 18:33 | |
| *** fattarsi has quit IRC | 18:36 | |
| *** dalang has joined #openstack-dev | 18:37 | |
| *** adjohn has quit IRC | 18:39 | |
| *** dalang has quit IRC | 18:42 | |
| *** fattarsi has joined #openstack-dev | 18:45 | |
| sdague | nati: I'll look into 938317 as well as low-hanging-fruit. Given that I'm still very much on the beginning of the learning curve, though I'd just run after easy stuff first. :) | 18:45 |
| *** jshepher has quit IRC | 18:45 | |
| *** jshepher_ has joined #openstack-dev | 18:45 | |
| nati | sdague: In case you can try low-hanging-fruit one :) | 18:45 |
| *** dalang has joined #openstack-dev | 18:46 | |
| *** dalang has quit IRC | 18:47 | |
| sdague | nati: sure will do. The reason I original asked as that there were many medium serverity bugs marked with forward-port-needed, so I was wondering if those were more important than the low-hanging-fruit, or what the story was with them. But I'll dive in over on low-hanging-fruit for a bit. Thanks | 18:48 |
| *** anotherjesse is now known as anotherjesse_zz | 18:50 | |
| nati | sdague: Thanks. low-hanging-fruit is easy to fix one, So it is good to learning. After you learn how to fix bugs, you should work on bugs from high priority ones | 18:50 |
| *** Glacee has quit IRC | 18:52 | |
| *** mattray1 has joined #openstack-dev | 18:54 | |
| *** shevek_ has joined #openstack-dev | 18:54 | |
| *** mattray has quit IRC | 18:57 | |
| *** mattray1 is now known as mattray | 18:59 | |
| *** mattray has joined #openstack-dev | 18:59 | |
| *** fattarsi has quit IRC | 19:03 | |
| *** dspano has joined #openstack-dev | 19:04 | |
| *** fc__ has quit IRC | 19:05 | |
| *** GheRivero_ has joined #openstack-dev | 19:10 | |
| *** jdg has quit IRC | 19:10 | |
| *** mnewby has joined #openstack-dev | 19:14 | |
| *** darraghb has quit IRC | 19:15 | |
| *** jakedahn is now known as jakedahn_zz | 19:16 | |
| *** fattarsi has joined #openstack-dev | 19:16 | |
| *** johnpostlethwait has joined #openstack-dev | 19:16 | |
| *** davidkranz has quit IRC | 19:18 | |
| *** anotherjesse_zz is now known as anotherjesse | 19:19 | |
| mnewby | keystone devs? | 19:20 |
| dolphm | mnewby: o/ | 19:20 |
| mnewby | dolphm: I have change: https://review.openstack.org/#change,5595 | 19:21 |
| *** jakedahn_zz is now known as jakedahn | 19:22 | |
| mnewby | dolphm: joe wants to have the test install swift if required. The netifaces library needs to be installed, too. I'm wondering if using checkout_vendor, as he suggested, is the best way to do it, or if putting the requirements in tools/test-requires might be a better solution. | 19:23 |
| dolphm | i'd vote for test-requires | 19:23 |
| dolphm | well hmm | 19:23 |
| mnewby | What are the downsides to using test-requires? | 19:24 |
| mnewby | dolphm: And what are the upsides to checkout_vendor? | 19:24 |
| dolphm | checkout vendor ensures you're always testing against the latest checkout | 19:24 |
| dolphm | test-requires not so much | 19:24 |
| mnewby | I'm not sure that is important for swift... | 19:24 |
| mnewby | The code that needs to be used is pretty stable. | 19:25 |
| mnewby | It's things like acl and referrel parsing. | 19:25 |
| mnewby | referrer, sorry. | 19:25 |
| dolphm | yeah, i could probably go either way | 19:25 |
| dolphm | termie: ^^ might have a stronger opinion | 19:25 |
| *** stuntmachine has quit IRC | 19:25 | |
| rohit404 | dolphm: hi, if you are done with mnewby, please comment on how can i make a request to keystone from nova or any other service for an admin token...any code pointers would be very helpful | 19:27 |
| *** nati has quit IRC | 19:27 | |
| dolphm | rohit404: from python for from the CLI? | 19:27 |
| rohit404 | dolphm : python | 19:27 |
| mnewby | dolphm: danke. I'll try to get termie's opinion. | 19:27 |
| dolphm | rohit404: keystoneclient? | 19:28 |
| chmouel | rohit404: something like that https://github.com/openstack/nova/blob/master/nova/api/ec2/__init__.py#L219 ? | 19:28 |
| dolphm | rohit404: https://github.com/openstack/python-keystoneclient | 19:28 |
| rohit404 | dolph: yeah, that's where i am heading as well...tests/v2_0/test_service_catalog | 19:29 |
| rohit404 | chmouel: great thanks ! | 19:30 |
| rohit404 | thanks dolphm | 19:31 |
| *** alaski has quit IRC | 19:33 | |
| adam_g | mnewby: whatever you guys decide, if you could please keep git checkout conditional on an import error, it would make life much easier for us | 19:33 |
| *** alaski has joined #openstack-dev | 19:33 | |
| mnewby | adam_g: will do, that's definitely better for me and anybody who's actually developing with swift. | 19:34 |
| *** koolhead17 has quit IRC | 19:45 | |
| *** rkukura has quit IRC | 19:46 | |
| *** jdg has joined #openstack-dev | 19:49 | |
| andrewbogott | annegentle: Are .rst files in the openstack source trees used to generate web pages someplace? | 19:51 |
| annegentle | andrewbogott: yes, nova/doc/source publishes to nova.openstack.org | 19:52 |
| annegentle | andrewbogott: as an example, same for other project names | 19:52 |
| *** jshepher_ has quit IRC | 19:52 | |
| andrewbogott | OK... I'm about to move HACKING.rst into the top-level dir of openstack-common. Can we add a line to a script someplace to ensure that that file is also webified? | 19:52 |
| *** jshepher has joined #openstack-dev | 19:53 | |
| *** markmc has quit IRC | 19:53 | |
| anotherjesse | andrewbogott: don't know if hacking.rst is including in sphinx docs (which is what nova.openstack.org / swift.openstack.org / … are) | 19:54 |
| anotherjesse | andrewbogott: grep through docs should answer if it is needed | 19:54 |
| andrewbogott | anotherjesse: Sorry, I don't know what 'grep through docs' means. Which docs? | 19:55 |
| anotherjesse | andrewbogott: the docs directory in the nova repo (that is where the sphinx docs live) | 19:55 |
| andrewbogott | (background: I am trying to address Doug Hellmann's comment that the styleguide should be in 'a wiki'. I don't actually think it should be a wiki, but it should be online someplace.) | 19:55 |
| andrewbogott | ok, looking... | 19:55 |
| *** maploin has joined #openstack-dev | 19:56 | |
| *** maploin has quit IRC | 19:57 | |
| *** maploin has joined #openstack-dev | 19:57 | |
| andrewbogott | The string HACKING appears nowhere in nova/docs. So I need to add something to... find_autodoc_modules.sh? | 19:57 |
| dspano | Has anyone run into this error when trying to run an instance? Bad magic number in super-block while trying to open /dev/mapper/nbd15p1\r\nCouldn't find valid filesystem superblock.\n" | 20:02 |
| *** rackerjoe has joined #openstack-dev | 20:02 | |
| dspano | The image works perfectly fine in kvm | 20:02 |
| *** rackerjoe has left #openstack-dev | 20:02 | |
| dspano | I'm using diablo. | 20:03 |
| *** johnpur has joined #openstack-dev | 20:03 | |
| *** ChanServ sets mode: +v johnpur | 20:03 | |
| dspano | Whoops. Sorry. I thought I was in the openstack channel! | 20:03 |
| dspano | Nevermind. | 20:03 |
| anotherjesse | andrewbogott: we have sphinx interlinking - so we should be able to add a common.openstack.org that renders openstack-common --- then have nova/swift/... sphinx link to hacking guide on common.openstack.org | 20:03 |
| andrewbogott | anotherjesse, annegentle: Hm, the doc build in nova/doc seems pretty broken. Shall I fix, or is someone on top of that already? | 20:04 |
| anotherjesse | andrewbogott: if you have time, plz do help :) there is a jenkins job that does it on each commit - so you can see the output on the server | 20:04 |
| andrewbogott | anotherjesse: Mind if I just file a bug about that rather than trying to understand how to do what you just said? | 20:04 |
| anotherjesse | andrewbogott: sure :) | 20:04 |
| andrewbogott | re: the docs being broken, I think it's a simple fix. At least to get it partially working. | 20:05 |
| anotherjesse | https://jenkins.openstack.org/view/Nova/job/nova-docs/2555/console <- those errors? | 20:05 |
| *** tryggvil__ has quit IRC | 20:06 | |
| andrewbogott | anotherjesse: Hm, no, I wasn't getting that far. Jenkins must be doing the 'make' using a different cwd. | 20:07 |
| andrewbogott | If you do 'make' in the docs dir it fails immediately. | 20:07 |
| andrewbogott | But I guess I should leave that be; don't want to break jenkins. | 20:07 |
| *** mszilagyi has joined #openstack-dev | 20:07 | |
| andrewbogott | I presume those other warnings are about modules that have been removed/renamed and the docs aren't caught up yet. | 20:07 |
| anotherjesse | andrewbogott: we can see exactly what jenkins does here: https://jenkins.openstack.org/view/Nova/job/nova-docs/configure | 20:07 |
| andrewbogott | It doesn't even use the makefile | 20:08 |
| *** bepernoot has joined #openstack-dev | 20:11 | |
| andrewbogott | I'm pretty baffled by setup.py :( | 20:11 |
| anotherjesse | andrewbogott: think you mean openstack-common in your commit message on https://review.openstack.org/#change,5704 | 20:13 |
| * andrewbogott nods | 20:13 | |
| andrewbogott | jaypipes: It's the newline before """ that you care about? Or are there other formatting concerns? | 20:19 |
| *** Glacee has joined #openstack-dev | 20:19 | |
| *** GheRivero_ has quit IRC | 20:20 | |
| jaypipes | andrewbogott: other stuff includes the insistence in the Nova HACKING that a docstring MUST have a short description and MUST have a period at the end of it, along with the notes about not having extra whitespace on the inside of enclosing triple-quotes... | 20:21 |
| *** adjohn has joined #openstack-dev | 20:22 | |
| *** pixelbeat has quit IRC | 20:23 | |
| jaypipes | andrewbogott: those Makefiles in the doc directory are an artificat of Sphinx' quickstart utility... they can be removed from the source trees AFAIAC | 20:24 |
| *** LinuxJedi has quit IRC | 20:24 | |
| *** kindaopsdevy has joined #openstack-dev | 20:25 | |
| andrewbogott | jaypipes: Is the docstring thing something there's already been discussion of? It's already established that no agreement can be made? | 20:25 |
| andrewbogott | jaypipes: OK, I'll ignore the doc thing; I was just concerned that a 'Make' failed. But sounds like no one actually uses that. | 20:25 |
| jaypipes | andrewbogott: heh, yeah, it's been discussed numerous times... :) | 20:26 |
| jaypipes | andrewbogott: I'll respond in depth to your ML post. | 20:26 |
| andrewbogott | great, thanks. I didn't intend to start a fight, really; just wanted to get more eyes on my encoding rule. | 20:26 |
| *** crobinso has quit IRC | 20:27 | |
| *** dolphm has quit IRC | 20:29 | |
| *** LinuxJedi has joined #openstack-dev | 20:29 | |
| *** mattray1 has joined #openstack-dev | 20:32 | |
| *** bepernoot has quit IRC | 20:33 | |
| *** johnpur has quit IRC | 20:34 | |
| *** kindaopsdevy_ has joined #openstack-dev | 20:34 | |
| *** mattray has quit IRC | 20:35 | |
| jaypipes | andrewbogott: lol, no worries mate :) I got a bit of a chuckle out of it. There's certain things that always seem to fire people up. this is one of em. | 20:35 |
| *** kindaopsdevy__ has joined #openstack-dev | 20:35 | |
| * andrewbogott backs away slowly, keeps hands where everyone can see them | 20:36 | |
| jaypipes | andrewbogott: :) | 20:36 |
| *** kindaopsdevy has quit IRC | 20:37 | |
| *** kindaopsdevy__ is now known as kindaopsdevy | 20:37 | |
| *** kindaopsdevy_ has quit IRC | 20:39 | |
| *** pixelbeat has joined #openstack-dev | 20:40 | |
| *** mattray1 is now known as mattray | 20:40 | |
| *** mattray has joined #openstack-dev | 20:40 | |
| anotherjesse | andrewbogott: in the future http://www.amazon.com/NcStar-Crossbow-with-Red-Dot/dp/B000A7PIEG/ is useful when you start talking about coding conventions ;) | 20:44 |
| andrewbogott | well, I am safely 1000 miles away from everyone involved. so I've got that going for me. | 20:45 |
| *** roge has quit IRC | 20:45 | |
| andrewbogott | Yesterday I sucked it up and didn't complain about someone's K&R brace style, so I'm doing pretty well this week on average. | 20:45 |
| *** roge has joined #openstack-dev | 20:46 | |
| *** dprince has quit IRC | 20:47 | |
| *** bepernoot has joined #openstack-dev | 20:49 | |
| *** bepernoot has quit IRC | 20:50 | |
| andrewbogott | Vek: Thanks for the review. I'm looking at http://en.wikipedia.org/wiki/List_of_HTTP_status_codes#2xx_Success and I'm confused about 200 vs. 202 vs. 204. | 20:58 |
| andrewbogott | That wiki page implies that a synchronous delete might be a 204 instead of a 202...? | 20:58 |
| bcwaldon | jeblair: is there a way to force-abandon reviews? | 21:03 |
| bcwaldon | jeblair: ...if you don't own them | 21:03 |
| jeblair | bcwaldon: not at the moment. we're planning on including that when we add the work-in-progress state | 21:03 |
| bcwaldon | jeblair: so that would allow me to abandon glance reviews that I don't own? | 21:04 |
| jeblair | but it'll be a little while before we get to that. | 21:04 |
| jeblair | it would allow you to "work in progress" them at least, which will also get them out of your sight. | 21:04 |
| bcwaldon | jeblair: cool, as long as its on the backlog :) | 21:04 |
| jeblair | right now, either ask the owner or wait 2 weeks. :) | 21:04 |
| bcwaldon | kk | 21:04 |
| *** utlemming has quit IRC | 21:05 | |
| *** mattray has left #openstack-dev | 21:07 | |
| *** mattray has joined #openstack-dev | 21:08 | |
| *** zigo has quit IRC | 21:08 | |
| sdague | would anyone object to me adding "yes" to the set of true booleans for glance's bool_from_string? | 21:13 |
| *** paulormg has quit IRC | 21:16 | |
| apevec | sdague, that's imported from openstack-common so send patch there | 21:16 |
| sdague | apevec: ok, will do, thanks | 21:18 |
| apevec | at least, it should be | 21:18 |
| *** sandywalsh has quit IRC | 21:19 | |
| *** dspano has quit IRC | 21:20 | |
| sdague | apevec: so how does that work? do changes in openstack-common get automatically or manually pushed into the other projects? | 21:20 |
| apevec | here's how it should work http://wiki.openstack.org/CommonLibrary#Incubation | 21:21 |
| sdague | it's definitely in openstack-common as well, just would like to understand the synchronization process so I know what to expect | 21:21 |
| apevec | but now I see utils is not imported | 21:21 |
| apevec | only cfg | 21:21 |
| apevec | oh well, one day... | 21:22 |
| *** lts has quit IRC | 21:22 | |
| *** dachary has joined #openstack-dev | 21:23 | |
| sdague | apevec: so should I submit it to openstack-common? or to glance, or both, or all the projects? | 21:23 |
| apevec | actually, I was looking at keystone checkout :) | 21:23 |
| apevec | sdague, best would be to change in openstack-common | 21:23 |
| sdague | heh | 21:23 |
| apevec | then change other projects to import from common | 21:24 |
| apevec | but looks like now every project has its own utils | 21:24 |
| apevec | it will take some time to consolidate that | 21:24 |
| *** LinuxJedi has quit IRC | 21:24 | |
| *** statik has quit IRC | 21:25 | |
| apevec | eglynn__, jaypipes - glance is not importing openstack-common as described here? http://wiki.openstack.org/CommonLibrary#Incubation | 21:26 |
| sdague | apevec: yeh, it looks like they vary a lot. Ok, I'll just push the review for common, then figure out the rest later. | 21:26 |
| andrewbogott | openstack-common could be managed as a git submodule rather than with auto-sych scripts. I'm not sure how that would play with the current git backend though. | 21:26 |
| apevec | I think markmc looked at using git submods | 21:27 |
| *** heckj has joined #openstack-dev | 21:27 | |
| ttx | heckj: yo | 21:28 |
| *** deshantm_ has quit IRC | 21:30 | |
| *** mattray has quit IRC | 21:30 | |
| *** mattray has joined #openstack-dev | 21:31 | |
| sdague | apevec: ok, pushed for review - https://review.openstack.org/#change,5712 | 21:34 |
| *** sandywalsh has joined #openstack-dev | 21:34 | |
| *** dolphm has joined #openstack-dev | 21:36 | |
| *** dolphm_ has joined #openstack-dev | 21:37 | |
| jaypipes | apevec: we should be, yes... feel free to submit a patch. Glance was (one of) the first to adopt the common library stuff, and I don't think the strategy of using an /openstack directory was established at rthat point | 21:37 |
| *** LinuxJedi has joined #openstack-dev | 21:37 | |
| *** dolphm has quit IRC | 21:40 | |
| *** apevec has quit IRC | 21:42 | |
| *** alaski has quit IRC | 21:42 | |
| *** vizsla has joined #openstack-dev | 21:45 | |
| *** LinuxJedi has quit IRC | 21:45 | |
| chmouel | where should I report a bug against gerrit? | 21:47 |
| *** LinuxJedi has joined #openstack-dev | 21:47 | |
| dolphm_ | chmouel: https://bugs.launchpad.net/openstack-ci | 21:48 |
| *** LinuxJedi has quit IRC | 21:49 | |
| *** vizsla has quit IRC | 21:49 | |
| *** LinuxJedi has joined #openstack-dev | 21:49 | |
| chmouel | tks | 21:51 |
| chmouel | dolphm_: btw: what do you think about this https://review.openstack.org/#change,5713 | 21:55 |
| dolphm_ | chmouel: raise exception.UserNotFound(user_id=user_id) and same for tenant_id | 21:57 |
| dolphm_ | chmouel: the rest of the message is already written in keystone.exception.UserNotFound | 21:57 |
| bcwaldon | vishy: https://review.openstack.org/#change,5714 | 21:58 |
| bcwaldon | vishy: works with novaclient without any extra changes | 21:58 |
| dolphm_ | chmouel: otherwise, +2; i'm putting together a massive list of all the places we're not doing this kind of validation (and should be) | 21:58 |
| chmouel | dolphm_: nice awesome. | 21:59 |
| dolphm_ | chmouel: if you want to fix all the ec2 ones -- all of these should throw friendly errors http://paste.openstack.org/raw/12048/ | 22:01 |
| dolphm_ | not sure what's going on with ec2-credentials-get, but that one looks client-side | 22:02 |
| *** utlemming has joined #openstack-dev | 22:03 | |
| chmouel | yep can do | 22:04 |
| *** hub_cap has quit IRC | 22:04 | |
| *** ayoung has quit IRC | 22:04 | |
| devcamcar | i just opened up rc2 for horizon | 22:07 |
| devcamcar | https://launchpad.net/horizon/+milestone/essex-rc2 | 22:07 |
| devcamcar | we are getting really close | 22:07 |
| *** kindaopsdevy has left #openstack-dev | 22:09 | |
| *** AlanClark has quit IRC | 22:12 | |
| *** Oneiroi^gone is now known as oneiroi | 22:13 | |
| *** uvirtbot has quit IRC | 22:15 | |
| *** uvirtbot has joined #openstack-dev | 22:15 | |
| *** joe3 has joined #openstack-dev | 22:21 | |
| *** joe3 has left #openstack-dev | 22:30 | |
| *** maploin has quit IRC | 22:31 | |
| bcwaldon | vishy: https://bugs.launchpad.net/nova/+bug/962615 | 22:32 |
| uvirtbot | Launchpad bug 962615 in nova "Unable to list volumes after building from snapshot" [Undecided,New] | 22:32 |
| bcwaldon | vishy: urgent | 22:32 |
| *** mycloud has joined #openstack-dev | 22:33 | |
| chmouel | dolphm_: does that make sense? http://pastie.org/private/ebx5wcy5weu5kuc6jrmhhg | 22:37 |
| chmouel | dolphm_: the second one mostly | 22:37 |
| dolphm_ | chmouel: i think so | 22:38 |
| *** dtroyer is now known as dtroyer_zzz | 22:40 | |
| mycloud | Hi, I installed openstack on Ubuntu 11.10 (single node). Installed from github.com/cloudbuilders/devstack.git. After installation, I get, "AttributeError at /syspanel/ 'Client' object has no attribute 'usage'". And in the screen, I see "Address already in use" for both n-api and for n-obj | 22:41 |
| mycloud | The installation completed successfully giving the ip and login at the end | 22:41 |
| anotherjesse | are there multiple screens? | 22:42 |
| mycloud | yes | 22:42 |
| anotherjesse | screen -list | 22:42 |
| anotherjesse | it could be that you are running n-api / n-obj in another screen? | 22:43 |
| mycloud | sorry no | 22:43 |
| mycloud | there is only one screen | 22:43 |
| devcamcar | mycloud: sudo screen -list? | 22:48 |
| devcamcar | maybe one is running as root? | 22:48 |
| *** zzed has quit IRC | 22:48 | |
| jdg | Weird issue: I added a migration for volume_uuid which includes updating iscsi_targets table, this fails migration test (column already exists) | 22:49 |
| anotherjesse | devcamcar: I'm helping mycloud in another channel out of #dev - the issue seems to be that his box has apt packages installed as well as devstack | 22:49 |
| *** dolphm_ has quit IRC | 22:49 | |
| devcamcar | anotherjesse: word | 22:49 |
| jdg | If I leave it out, ec2.test_create_image fails because there's no iscsi_targets.volume_uuid ??? | 22:50 |
| jdg | Anybody have any ideas? | 22:50 |
| *** mattray has quit IRC | 22:52 | |
| *** kbringard has quit IRC | 22:56 | |
| *** troytoman is now known as troytoman-away | 22:56 | |
| *** dolphm has joined #openstack-dev | 23:00 | |
| *** jakedahn is now known as jakedahn_zz | 23:05 | |
| *** dolphm has quit IRC | 23:10 | |
| *** dolphm has joined #openstack-dev | 23:15 | |
| *** dolphm_ has joined #openstack-dev | 23:17 | |
| *** jakedahn_zz is now known as jakedahn | 23:18 | |
| *** dolphm has quit IRC | 23:20 | |
| *** paulstallworth has joined #openstack-dev | 23:27 | |
| *** littleidea has quit IRC | 23:29 | |
| *** sandywalsh has quit IRC | 23:31 | |
| *** dtroyer_zzz is now known as dtroyer | 23:35 | |
| *** oneiroi is now known as Oneiroi^gone | 23:40 | |
| *** mycloud has quit IRC | 23:44 | |
| *** sandywalsh has joined #openstack-dev | 23:44 | |
| *** Glacee has quit IRC | 23:46 | |
| *** rnirmal has quit IRC | 23:48 | |
| *** spiffxp has quit IRC | 23:51 | |
| *** hub_cap has joined #openstack-dev | 23:55 | |
| *** rods has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!