| *** RobertLaptop has quit IRC | 00:42 | |
| *** RobertLaptop has joined #openstack-dev | 00:46 | |
| *** RobertLaptop has quit IRC | 02:15 | |
| *** lts has joined #openstack-dev | 02:28 | |
| *** RobertLaptop has joined #openstack-dev | 02:30 | |
| *** lts has quit IRC | 02:32 | |
| openstackgerrit | Ziad Sawalha proposed a change to openstack/keystone: Added reponse handling for xsd static file rendering https://review.openstack.org/233 | 02:41 |
|---|---|---|
| openstackgerrit | A change was merged to openstack/keystone: Added reponse handling for xsd static file rendering https://review.openstack.org/233 | 02:44 |
| openstackgerrit | Verification of a change to openstack/keystone failed: Extension handling and static file content reorganization https://review.openstack.org/234 | 02:45 |
| openstackgerrit | A change to openstack/keystone has been rejected: Ziads changes and fixes for them. https://review.openstack.org/230 | 02:45 |
| openstackgerrit | Ziad Sawalha proposed a change to openstack/keystone: Extension handling and static file content reorganization https://review.openstack.org/234 | 02:54 |
| openstackgerrit | Verification of a change to openstack/keystone failed: Added reponse handling for xsd static file rendering https://review.openstack.org/234 | 02:54 |
| openstackgerrit | Ziad Sawalha proposed a change to openstack/keystone: Added reponse handling for xsd static file rendering https://review.openstack.org/234 | 03:00 |
| openstackgerrit | Verification of a change to openstack/keystone failed: Schema updates. Split WADLs and extensions and got xsds to compile https://review.openstack.org/234 | 03:01 |
| tr3buchet | vishy: ping | 03:39 |
| tr3buchet | i'm about 1000 lines in | 03:39 |
| tr3buchet | well it was 8 something last i checked | 03:40 |
| tr3buchet | i'm trying to figure out wtf to do with libvirt's refresh security groups | 03:40 |
| tr3buchet | i have the libvirt get_network_info() function completely removed now and all functions that used it before and their tests passing | 03:41 |
| tr3buchet | there are a few functions which previously optionally accepted network_info left to cleanse | 03:41 |
| *** chomping has joined #openstack-dev | 03:57 | |
| openstackgerrit | Monty Taylor proposed a change to openstack/openstack-ci: Added a version of the rfc script from gluster. https://review.openstack.org/235 | 04:55 |
| vishy | tr3buchet: i already did it and proposed | 05:03 |
| vishy | look at merge queue | 05:03 |
| *** nci has quit IRC | 07:03 | |
| *** reidrac has joined #openstack-dev | 07:04 | |
| *** nickon has joined #openstack-dev | 07:17 | |
| *** darraghb has joined #openstack-dev | 07:29 | |
| *** mnour has joined #openstack-dev | 08:06 | |
| *** phil121 has quit IRC | 08:09 | |
| openstackgerrit | Yogeshwar Srikrishnan proposed a change to openstack/keystone: Additional tests and minor changes to support services CRUD. https://review.openstack.org/236 | 08:26 |
| openstackgerrit | Ziad Sawalha proposed a change to openstack/keystone: Schema updates. Split WADLs and extensions and got xsds to compile https://review.openstack.org/234 | 10:07 |
| openstackgerrit | Verification of a change to openstack/keystone failed: Added reponse handling for xsd static file rendering https://review.openstack.org/234 | 10:08 |
| *** rods has joined #openstack-dev | 10:44 | |
| *** Guest77784 has quit IRC | 11:23 | |
| *** lorin1 has joined #openstack-dev | 11:53 | |
| *** mfer has joined #openstack-dev | 11:59 | |
| *** bsza has joined #openstack-dev | 12:35 | |
| *** lts has joined #openstack-dev | 12:56 | |
| *** chuck_ has quit IRC | 13:00 | |
| *** zul has joined #openstack-dev | 13:01 | |
| *** kbringard has joined #openstack-dev | 13:02 | |
| *** Guest77784 has joined #openstack-dev | 13:03 | |
| *** dprince has joined #openstack-dev | 13:13 | |
| *** ameade has joined #openstack-dev | 13:24 | |
| *** amccabe has joined #openstack-dev | 13:44 | |
| *** Binbin has joined #openstack-dev | 14:08 | |
| *** dolphm has joined #openstack-dev | 14:09 | |
| *** dprince has quit IRC | 14:10 | |
| *** troytoman-away is now known as troytoman | 14:15 | |
| openstackgerrit | Dolph Mathews proposed a change to openstack/keystone: Additional tests and minor changes to support services CRUD. https://review.openstack.org/236 | 14:22 |
| openstackgerrit | A change was merged to openstack/keystone: Additional tests and minor changes to support services CRUD. https://review.openstack.org/236 | 14:25 |
| openstackgerrit | Dolph Mathews proposed a change to openstack/keystone: Added reponse handling for xsd static file rendering https://review.openstack.org/234 | 14:33 |
| *** vladimir3p has joined #openstack-dev | 14:34 | |
| openstackgerrit | A change was merged to openstack/keystone: Added reponse handling for xsd static file rendering https://review.openstack.org/234 | 14:36 |
| tr3buchet | vishy: i hate that we did that. :( | 14:40 |
| *** vladimir3p has quit IRC | 14:46 | |
| *** rnirmal has joined #openstack-dev | 14:55 | |
| *** dragondm has joined #openstack-dev | 15:00 | |
| smoser | is it expected that uploading a vmdk formated full disk image should be supported? | 15:01 |
| jaypipes | dolphm: ping | 15:01 |
| dolphm | jaypipes: pong | 15:02 |
| jaypipes | smoser: in Glance, sure. Not sure about the VMWare driver though | 15:02 |
| *** reidrac has quit IRC | 15:03 | |
| jaypipes | dolphm: hey, I want to get z's git and gerrit stuff sorted... mtaylor and jeblair have responded to his emails about his trouble with Gerrit this weekend. It seems that he's not using topic branches, and by making chnages in the master local branch he's messing something up. | 15:03 |
| jaypipes | dolphm: I noticed you merged in his change on the static file stuff. He mentioned there is "stuff on GitHub to pull in". What is that? | 15:03 |
| dolphm | jaypipes: yeah, i got his gerrit change in this morning.. and he just emailed me about github too | 15:03 |
| smoser | jaypipes, i'm trying vmdk upload and then running under kvm | 15:03 |
| dolphm | jaypipes: he said he "didn't have any way to share his code with Jorge" so he pushed it to github | 15:04 |
| dolphm | jaypipes: instead of, i don't know, opening a review | 15:04 |
| jaypipes | smoser: let us know if it works! ;) | 15:04 |
| smoser | well, it doesn't :) | 15:05 |
| smoser | jaypipes, https://bugs.launchpad.net/nova/+bug/826798 | 15:05 |
| uvirtbot | Launchpad bug 826798 in nova "ubuntu vmdk uploaded does not boot" [Undecided,New] | 15:05 |
| jaypipes | dolphm: heh. alright, so he pushed his code by doing a git remote add github.com/rackspace/keystone... | 15:05 |
| jaypipes | smoser: ah, well that was a quick answer :) | 15:05 |
| *** cp16net has joined #openstack-dev | 15:05 | |
| smoser | i'm digging info on whether or not glance was involved at all, this is a public openstack on oneiric that i'm not aware of the details of | 15:05 |
| *** mnour has quit IRC | 15:06 | |
| *** mnour has joined #openstack-dev | 15:06 | |
| jaypipes | dolphm: because I get he tried to git review that code but it kept having merge conflicts. the reason it was having merge conflicts was because he did a git pull in his local master, that pulled in commits, then he made code changes in his master branch then git rebase'd and stuff got messed up.... | 15:07 |
| *** troytoman is now known as troytoman-away | 15:07 | |
| jaypipes | smoser: k, lemme know if it's Glance related. | 15:07 |
| dolphm | jaypipes: should i setup a how-to-git-gerrit session with him? | 15:07 |
| smoser | glance was used, jaypipes but i know nothing else at the moment. | 15:08 |
| jaypipes | smoser: k | 15:08 |
| jaypipes | dolphm: I think we need to. We need to get him to stop pushing his private history to GitHub... | 15:08 |
| smoser | jaypipes, you can easily try to reproduce though | 15:08 |
| dolphm | jaypipes: 110% agree | 15:09 |
| jaypipes | dolphm: well, he is in training much of today on some new release and code management tool that RAX development teams are now going to be using, so perhaps tomorrow? | 15:09 |
| dolphm | jaypipes: he's a little aggressive on +2's, and, umm, passionate about "more commits = better" | 15:09 |
| Daviey | smoser: The cloud implementation you are talking about doesn't use glance, it currently uses nova-objectstore. | 15:10 |
| Daviey | err, i am a plonker, scrub that | 15:10 |
| jaypipes | dolphm: more commits == better is fine for local development. But git's workflow really pushes for rebasing those local commits into a single changeset when you push public history. | 15:10 |
| smoser | Daviey, i was told by agy it uses glance just minutes ago | 15:10 |
| Daviey | smoser: yes, i am a plonker. | 15:11 |
| smoser | you are. | 15:11 |
| jaypipes | dolphm: and ack on the +2 stuff. That is a real problem that is going to be an issue with the incubated -> core process, just to let you know... | 15:11 |
| dolphm | jaypipes: no no, i mean "more commits in openstack/keystone NOW NOW NOW no matter what they are = better" | 15:11 |
| dolphm | jaypipes: i recognize that and i've tried to subtly suggest that to him | 15:11 |
| jaypipes | dolphm: the +2 stuff is emblematic of the GitHub "just close and merge it"... | 15:12 |
| dolphm | jaypipes: yep | 15:12 |
| jaypipes | dolphm: which works fine when you have 2 or 3 people requesting pulls. Not so much once you grow a bit larger... | 15:12 |
| dolphm | jaypipes: yeah, he hasn't recognized how much benefit gerrit has provided keystone (yet), and he's only impeding that benefit | 15:13 |
| jaypipes | dolphm: anyway, I think the big thing is to a) get Ziad using topic branches and b) walking through the difference between public and private history. notmyname posted an excellent link for an article that I think Z should go through: http://sandofsky.com/blog/git-workflow.html | 15:13 |
| dolphm | jaypipes: ha, i love the opening paragraph.. | 15:14 |
| jaypipes | dolphm: yes, indeed. when I read that article, I tried imagining that creiht was reading it to me, sitting with a pipe in front of a fire in a nice big leather chair. | 15:15 |
| jaypipes | dolphm: it helped me consume the content better ;) | 15:16 |
| dolphm | dolphm: i'll go find a cigar | 15:16 |
| dolphm | jaypipes: ^ lol | 15:16 |
| jaypipes | hehe | 15:16 |
| notmyname | dolphm: follow -up twitter conversation to that git post http://paste.openstack.org/show/2170/ | 15:20 |
| *** letterj has joined #openstack-dev | 15:24 | |
| *** ChanServ sets mode: +v letterj | 15:24 | |
| dolphm | notmyname: for future reference =) http://i.imgur.com/paTr6.jpg | 15:25 |
| * creiht worries about what other fantasies jaypipes has about me | 15:25 | |
| jaypipes | creiht: lol | 15:27 |
| notmyname | dolphm: indeed | 15:28 |
| jaypipes | smoser: interesting thought from you on that bug: "disabling a filesystem check has nothing to do with disabling filesystem checks." | 15:28 |
| smoser | :) | 15:29 |
| smoser | yeah... | 15:29 |
| smoser | "injecting a file has nothing to do with disabling filesystem checks" | 15:29 |
| *** heckj has joined #openstack-dev | 15:32 | |
| jaypipes | smoser: :) | 15:32 |
| smoser | i fixed that. | 15:33 |
| jaypipes | smoser: hehe, no worries. | 15:33 |
| smoser | the bug i meant to give to you was https://bugs.launchpad.net/nova/+bug/826798 | 15:33 |
| uvirtbot | Launchpad bug 826798 in nova "ubuntu vmdk uploaded does not boot" [Undecided,New] | 15:33 |
| smoser | thats the vmdk one | 15:33 |
| jaypipes | smoser: I pretty much knew what you meant ;) was a good tautology though. | 15:33 |
| smoser | yeah. | 15:33 |
| smoser | disabling a filesystem check has *something* to do with disabling filesystem checks, assuming tune2fs works properly | 15:34 |
| mtaylor | jaypipes, dolphm ++ on that article | 15:34 |
| mtaylor | jaypipes: and +10 on creiht reading it in an armchair | 15:34 |
| creiht | how about on a horse? | 15:36 |
| mtaylor | creiht: only if you're riding it backwards bare-chested and also holding a bottle of old spice | 15:38 |
| *** mnour has quit IRC | 15:38 | |
| creiht | look at your git workflow, now look at mine | 15:38 |
| jaypipes | ooooh, the imagery... | 15:40 |
| creiht | look now I am on github... | 15:40 |
| mtaylor | now I'm a clam | 15:40 |
| *** bsza has quit IRC | 15:55 | |
| openstackgerrit | Dolph Mathews proposed a change to openstack/keystone: Added reponse handling for xsd static file rendering https://review.openstack.org/237 | 15:57 |
| dolphm | jaypipes: mtaylor: ^^ these are z's changes from github.com/rackspace/keystone | 15:57 |
| *** bsza has joined #openstack-dev | 15:58 | |
| mtaylor | dolphm: great. now all we've got to do is get him moving forward properly | 15:58 |
| openstackgerrit | Kevin L. Mitchell proposed a change to openstack/glance: Add support for shared images https://review.openstack.org/201 | 15:58 |
| *** martine has joined #openstack-dev | 16:04 | |
| openstackgerrit | A change was merged to openstack/openstack-ci: Add a script to close Github pull requests. https://review.openstack.org/224 | 16:04 |
| dolphm | mtaylor: i don't want these changes merged in yet either... there's some terrible slop in there that needs to be fixed... so this is a good chance for him to learn the review process :) | 16:04 |
| mtaylor | dolphm: yay! | 16:04 |
| openstackgerrit | A change was merged to openstack/openstack-ci-puppet: Add cron job for closing pull requests. https://review.openstack.org/225 | 16:05 |
| openstackgerrit | Dolph Mathews proposed a change to openstack/keystone: Added reponse handling for xsd static file rendering https://review.openstack.org/237 | 16:06 |
| jeblair | dolphm: awesome, thanks! | 16:07 |
| dolphm | jeblair: np | 16:07 |
| *** zaitcev has joined #openstack-dev | 16:13 | |
| openstackjenkins | Project nova build #1,248: SUCCESS in 3 min 27 sec: https://jenkins.openstack.org/job/nova/1248/ | 16:20 |
| openstackjenkins | Tarmac: Add durable flag for rabbit queues. | 16:20 |
| *** zul has quit IRC | 16:24 | |
| openstackgerrit | A change to openstack/keystone has been rejected: Ziad's changes from github.com/rackspace/openstack https://review.openstack.org/237 | 16:25 |
| dolphm | if i -2, the z-hammer can't just +2 and merge it anyway, right? | 16:25 |
| mtaylor | dolphm: I believe that is the case | 16:27 |
| mtaylor | jeblair: ^^ | 16:27 |
| *** markvoelker has joined #openstack-dev | 16:28 | |
| *** zul has joined #openstack-dev | 16:28 | |
| jeblair | mtaylor, dolphm: correct | 16:29 |
| *** Binbin has quit IRC | 16:37 | |
| *** Tushar has joined #openstack-dev | 16:56 | |
| openstackgerrit | A change was merged to openstack/glance: Bug fix lp:824706 https://review.openstack.org/229 | 16:59 |
| openstackgerrit | A change was merged to openstack/glance: Bug Fix lp:825493 https://review.openstack.org/226 | 17:00 |
| openstackgerrit | Verification of a change to openstack/glance failed: Add notifications for uploads, updates and deletes https://review.openstack.org/192 | 17:03 |
| *** mgius has joined #openstack-dev | 17:13 | |
| *** jhtran has joined #openstack-dev | 17:14 | |
| *** negronjl has quit IRC | 17:19 | |
| *** negronjl has joined #openstack-dev | 17:19 | |
| *** lorin1 has quit IRC | 17:27 | |
| *** anotherjesse has joined #openstack-dev | 17:42 | |
| vishy | tr3buchet: I saw you push a branch, was trying to ping you to let you know that I was doing it, but you weren't on irc at the time :( | 17:47 |
| vishy | tr3buchet: I assume you made the same set of changes? | 17:47 |
| *** dprince has joined #openstack-dev | 17:56 | |
| dolphm | open question (vishy?): in keystone, does it make more sense for the "default tenant" relationship to be described as "a Tenant is Owned by a User" and therefore be explicitly modeled as an attribute of the tenant referencing it's owner (TenantModel.owner --> UserModel) instead of an attribute of the User (UserModel.default_tenant --> TenantModel)? | 18:16 |
| vishy | dolphm: i think that breaks down if you have a user that owns multiple tenants | 18:17 |
| dolphm | vishy: then multiple tenants point to a single user, the owner | 18:17 |
| dolphm | vishy: the current model breaks down in that scenario (a user can only point to a single tenant) | 18:18 |
| vishy | right but then when the user logs in, how do you pick its default tenant? | 18:18 |
| *** dprince_ has joined #openstack-dev | 18:18 | |
| *** dprince has quit IRC | 18:18 | |
| dolphm | vishy: if they only own a single tenant, it can be looked up, and a scoped token can be immediately provided | 18:19 |
| *** dprince_ has quit IRC | 18:19 | |
| dolphm | vishy: if they own multiple tenants, they must go through the 2.0 process of selecting a tenant to get a token for | 18:19 |
| vishy | dolphm: isn't that the alternate solution we came up with? | 18:20 |
| vishy | dolphm: I thought that there was a strong feeling that default tenant was preferred | 18:20 |
| dolphm | vishy: or their "unscoped" token is naturally "scoped" to their Own tenants, but not to anything they have a Role on | 18:20 |
| vishy | I don't really care personally | 18:20 |
| dolphm | vishy: we didn't talk about modeling it this way | 18:20 |
| vishy | i don't think they should be properties on either object personally | 18:20 |
| *** mszilagyi has joined #openstack-dev | 18:21 | |
| vishy | user, tenant, association seems way simpler to me | 18:21 |
| *** anotherjesse_ has joined #openstack-dev | 18:21 | |
| dolphm | vishy: i personally agree | 18:21 |
| *** dprince has joined #openstack-dev | 18:22 | |
| vishy | I don't understand why this becomes so complicated | 18:23 |
| vishy | have users tenants | 18:23 |
| vishy | and an association table with role_type | 18:23 |
| vishy | user_id, tenant_id, role_type | 18:23 |
| vishy | you do a join and you have all of the tenants and roles | 18:23 |
| vishy | if you have one, return it... | 18:23 |
| *** anotherjesse has quit IRC | 18:25 | |
| *** anotherjesse_ is now known as anotherjesse | 18:25 | |
| annegentle | you're all invited - doc team meeting in #openstack-meeting channel in about an hour and 15 minutes from now | 18:46 |
| tr3buchet | vishy: they were fairly similar yeah | 18:48 |
| tr3buchet | vishy: same goal too, rip out the libvirt get_network_info nonsense | 18:48 |
| vishy | yeah sorry about that :( | 18:49 |
| vishy | i was fixing a couple of bugs, and it made sense to just go ahead and fix that whole part | 18:49 |
| openstackgerrit | Monty Taylor proposed a change to openstack/openstack-ci: Added a version of the rfc script from gluster. https://review.openstack.org/235 | 18:52 |
| openstackgerrit | Monty Taylor proposed a change to openstack/openstack-ci: Added a version of the rfc script from gluster. https://review.openstack.org/235 | 18:55 |
| jaypipes | vishy, dolphm: since when did users "own" tenants? I thought a user "belonged to" a tenant (i.e. a tenant is the account/project) | 18:55 |
| dolphm | jaypipes: it's just a thought... i'm wondering why it won't work | 18:55 |
| vishy | jaypipes: in the current project model there is an 'owner': called the project manager | 18:55 |
| jaypipes | vishy: but that's a role, not a user, right? | 18:56 |
| vishy | but i think it is way easier to have an association table | 18:56 |
| vishy | jaypipes: it is actaully a field in the projects table | 18:56 |
| vishy | but it shouldn't be :) | 18:56 |
| jaypipes | vishy: I see | 18:56 |
| dolphm | vishy: jaypipes: are ya'll referring to nova? | 18:57 |
| vishy | yes in nova that is how it is | 18:57 |
| jaypipes | yes | 18:57 |
| dolphm | so, isn't this "better" for keystone too, if it also satisfies rackspace's business case? | 18:58 |
| dprince | vishy: when you have a minute I have a question on instance states in nova. Looking at implementing pvo blueprint. | 18:58 |
| vishy | dprince: got a minute | 18:58 |
| vishy | dolphm: no, i think it is too limiting, the association table is way simpler and more flexible | 18:59 |
| vishy | otherwise you have all sorts of special cases in the logic | 18:59 |
| dolphm | vishy: like what? | 18:59 |
| vishy | look at the current code checking for default_tenant, it is in way too many places | 19:00 |
| dprince | vishy: So instead of adding a new column what are your thoughts on reworking how the current 'state' and 'state_description' columns work. | 19:01 |
| dolphm | vishy: my thought is to replace with "unscoped" token with a "default-scope" token... the default scope being tenants the user Owns ... so it simply becomes business logic on the token... "can the token operate on this tenant" | 19:01 |
| dprince | vishy: Specifically. We are considering what if we made them 'vm_state' and 'api_state'. | 19:01 |
| dprince | vishy: The existing columns seem to map more directly to an EC2 API use case. Just wondering if we could rework things to work better for the transision states and then have the OS and EC2 API's map them according to those SPECs. | 19:03 |
| *** jhtran_ has joined #openstack-dev | 19:06 | |
| *** jhtran has quit IRC | 19:06 | |
| *** jhtran_ is now known as jhtran | 19:06 | |
| vishy | dprince: there are some comments in the db models file from ewan | 19:06 |
| *** jhtran has quit IRC | 19:06 | |
| vishy | dprince: talking about how xen models this stuff | 19:06 |
| dprince | looking... | 19:06 |
| vishy | basically the idea is that there are three states | 19:07 |
| vishy | the power state, the vm_state and the current task | 19:07 |
| dprince | Comments in nova/db/sqlalchemy/models.py? | 19:07 |
| vishy | you could probably get away without powerstate | 19:07 |
| vishy | yeah | 19:07 |
| vishy | he wrote an email about it to the list long ago | 19:08 |
| vishy | powerstate should be able to be introspected from vmstate | 19:08 |
| dprince | Sure. | 19:08 |
| vishy | dprince: but in general i think you're correct, we should have rich states internally and map them to states that the api knows | 19:09 |
| vishy | dprince: I think two states internally is probably best though, where it is, and where it is trying to get to | 19:09 |
| vishy | since we don't have the concept of an external task | 19:10 |
| dprince | Okay. We weren't thinking about the 'task_state'. I was going to call Ewans power state 'vm_state'. | 19:10 |
| dprince | But I could go either way. | 19:10 |
| vishy | dolphm: I think ownership is too limiting | 19:10 |
| dprince | This would obviously involve changes all over the place (probably not D4 material) but you would be open with a 2 column model? | 19:10 |
| dolphm | vishy: but this is *less-limiting* than a "default_tenant" | 19:11 |
| vishy | dolphm: ownership is inherently 1 - many and tenants - users should be many - many | 19:11 |
| vishy | dolphm: it just switches the 1 side of the 1 - many | 19:11 |
| dolphm | vishy: ... to the side that makes more sense :) | 19:12 |
| vishy | dolphm: but why have that limitation at all. It just means you have to jump through hoops to do things like roles, because you've limited one side of the relationship | 19:12 |
| vishy | dprince: yes definitely | 19:13 |
| dolphm | vishy: you're just arguing against a default_tenant... and i thought we lost that argument on friday... so i'm just trying to fix rackspace's proprietary-ness | 19:14 |
| vishy | dolphm: you can do the same thing in my model that you can do in yours | 19:15 |
| *** yogirackspace has joined #openstack-dev | 19:15 | |
| vishy | dolphm: if there is only one association, return it as the default | 19:15 |
| dolphm | vishy: i totally agree, but we lost that fight | 19:16 |
| vishy | dolphm: I don't really see how the owner thing is any simpler than default tenant | 19:16 |
| dolphm | for starters, it'd be more intuitive to newcomers | 19:16 |
| vishy | the only difference is you have to do a uniqueness test to make sure you can return it | 19:16 |
| vishy | seems more complicated actually | 19:16 |
| dolphm | "tenants can be owned by user" vs "users have the concept of default tenantness" | 19:16 |
| vishy | dolphm: but the only way that is simpler is if you enforce one owned tenant per user | 19:17 |
| vishy | otherwise you are back in my model anyway | 19:17 |
| vishy | dolphm: i really don't care which way it is modeled | 19:18 |
| vishy | dolphm: i think you are making more work for yourself by changing it without any tangible benefit | 19:18 |
| dolphm | the major benefit is the simplicity of an "unscoped token" --> "default-scoped token" or "self-scoped token" | 19:18 |
| dprince | vishy: Still. Using Ewan's nomanclature to keep it simple... I would probably advocate that we have just vm_state and power_state. How does keeping track of the 'task_state' help us? | 19:19 |
| vishy | self-scoped-token? | 19:19 |
| vishy | dprince: it simplifies the number of different states necessary in vm_state | 19:19 |
| vishy | dprince: for example, you need to do a reboot for rescue, unrescue, reboot, etc. | 19:20 |
| dolphm | vishy: a token that is, by default, scoped to only the Tenant's the User Owns ("scoped to my own stuff"... rather than "scoped to something I only have a Role in") | 19:20 |
| vishy | dprince: with only one state you have to add rescuing, unrescuing, rebooting, etc. | 19:20 |
| vishy | dprince: and then have some logic that can understand that all of those states really mean that the vm is rebooting | 19:21 |
| vishy | dprince: and you can't really know where it is in the process. Is it shutoff, has the ACPI command been sent to the chassis? etc. | 19:21 |
| vishy | dprince: if vm_state actually shows the non-transitory state and task_state shows the action that is being performed | 19:22 |
| vishy | dprince: it is easy to always know what state the vm is actually in. | 19:22 |
| vishy | dprince: but I'm not married to the idea. I just thought ewan made a lot of sense when he described how they do it. | 19:23 |
| vishy | dolphm: that doesn't make a whole lot of sense. From the service perspective we want token = 1 tenant and 1 user | 19:24 |
| dprince | vishy: Sure. Looking for that email. But it looks like our 'state' column would become 'power_state' and the existing state_description would become 'vm_state'. | 19:24 |
| vishy | dprince: that is fine | 19:24 |
| vishy | dprince: i think we can do with 2 columns total | 19:24 |
| dolphm | vishy: what's the implication i'm missing there? | 19:24 |
| vishy | dolphm: when we authenticate a token we need to be told the tenant that the token is for | 19:25 |
| dolphm | vishy: ah.... let me check something | 19:25 |
| vishy | dolphm: it sounds like you are returning an amorphous list | 19:25 |
| dolphm | vishy: yeah, you're right... i'm wondering if i can fix that | 19:27 |
| vishy | dolphm: if you have just one owned tenant, it works just like default tenant | 19:27 |
| vishy | dolphm: if you have multiple, than it works just like some limited version of roles | 19:28 |
| vishy | dolphm: so i don't see what you gain | 19:28 |
| vishy | dolphm: if you want multiple just create a role that says owner | 19:28 |
| dolphm | vishy: the only way i see to fix that is to return a *list* of tokens to the user on login, where each token is scoped to one tenant the user owns... and that's not a place i'd want to change the api contract | 19:29 |
| vishy | dolphm: i suppose authorize could return a list of tenants, but then roles get kind of confusing | 19:29 |
| vishy | dolphm: how is what you're suggesting different than roles? | 19:30 |
| *** alekibango has quit IRC | 19:30 | |
| *** alekibango has joined #openstack-dev | 19:30 | |
| dolphm | vishy: "ownership" and "default_tenant-ness" are just a proprietary "core" roles | 19:31 |
| vishy | well, implement them as roles then. Way easier! | 19:31 |
| dolphm | vishy: and by proprietary i mean proprietary to rackspace. | 19:31 |
| vishy | so do the general user tenant associations | 19:31 |
| vishy | and for rackspace create a "default tenant role" | 19:32 |
| vishy | :) | 19:32 |
| vishy | gotta grab some lunch | 19:32 |
| vishy | continue this later | 19:32 |
| dolphm | vishy: "but then the "simplicity" of auth is reduced for rackspace" | 19:32 |
| dolphm | vishy: -ziad | 19:32 |
| dolphm | reduced = lost | 19:32 |
| *** chipc has joined #openstack-dev | 19:33 | |
| openstackjenkins | Project nova build #1,249: SUCCESS in 3 min 36 sec: https://jenkins.openstack.org/job/nova/1249/ | 19:35 |
| openstackjenkins | Tarmac: Libvirt has some autogenerated network info that is breaking ha network. | 19:35 |
| openstackjenkins | * pases network info from manager wherever it is needed | 19:35 |
| openstackjenkins | * fixes libvirt tests | 19:35 |
| openstackjenkins | * renames allow_project_net_traffic to allow_same_net_traffic | 19:35 |
| openstackjenkins | * makes firewall driver use dhcp_server instead of gateway for dhcp exception. | 19:35 |
| openstackgerrit | Kevin L. Mitchell proposed a change to openstack/glance: Add support for shared images https://review.openstack.org/201 | 19:39 |
| *** darraghb has quit IRC | 19:39 | |
| *** anotherjesse_ has joined #openstack-dev | 19:42 | |
| *** anotherjesse has quit IRC | 19:44 | |
| *** anotherjesse_ is now known as anotherjesse | 19:44 | |
| *** nickon has quit IRC | 19:48 | |
| openstackjenkins | Project nova build #1,250: SUCCESS in 3 min 39 sec: https://jenkins.openstack.org/job/nova/1250/ | 19:48 |
| openstackjenkins | Tarmac: This branch allows the standard inclusion of a body param which most http clients will send along with a POST request. | 19:48 |
| *** ameade has quit IRC | 19:54 | |
| vishy | dolphm: jesse just pointed out that your owner thing doesn't really work | 19:57 |
| anotherjesse | vishy: here is the issue: | 19:58 |
| anotherjesse | def remove_user_from_tenant(self, tenant_id, user_id): | 19:58 |
| anotherjesse | params = {"roleRef": {"tenantId": tenant_id, "roleId": "Member"}} | 19:58 |
| anotherjesse | # FIXME(ja): we have to get the roleref? what is 5? | 19:58 |
| anotherjesse | return self._delete("/users/%s/roleRefs/5" % user_id) | 19:58 |
| vishy | dolphm: what if two users should have the same default tenant | 19:58 |
| vishy | dolphm: it isn't possible in your model | 19:58 |
| annegentle | Doc team meeting in a few minutes in #openstack-meeting | 19:58 |
| anotherjesse | that is how you remove "Member" role from a user in a tenant... | 19:58 |
| anotherjesse | maybe | 19:58 |
| anotherjesse | the api for adding/removing role-ness in a tenant needs to be cleaned? | 19:58 |
| anotherjesse | There are three nouns: user, tenant, role … what is the restifarian way of saying - remove this role from the tenant? | 19:59 |
| dolphm | vishy: anotherjesse: yep, proprietary roles are dumb | 19:59 |
| openstackgerrit | Johannes Erdfelt proposed a change to openstack/glance: Add notifications for uploads, updates and deletes https://review.openstack.org/192 | 20:00 |
| openstackgerrit | A change was merged to openstack/glance: Add notifications for uploads, updates and deletes https://review.openstack.org/192 | 20:02 |
| anotherjesse | dolphm: is there a way to say "delete all roles in user/tenant combo" | 20:03 |
| anotherjesse | dolphm: since I removing a user from a tenant needs to remove all roles they have in the tenant | 20:03 |
| jaypipes | Vek: https://review.openstack.org/#patch,sidebyside,201,5,glance/tests/__init__.py | 20:04 |
| jaypipes | Vek: so... that change is from a commit that is in master, but not related to your patchset 5 for shared image groups. | 20:05 |
| jaypipes | Vek: wondering how that got in there... | 20:05 |
| dolphm | anotherjesse: i would think the method you pasted should implement taht | 20:05 |
| dolphm | *should* | 20:05 |
| openstackgerrit | Dolph Mathews proposed a change to openstack/keystone: Ziad's changes from github.com/rackspace/openstack https://review.openstack.org/238 | 20:06 |
| *** dprince has quit IRC | 20:06 | |
| openstackjenkins | Project nova build #1,251: SUCCESS in 3 min 26 sec: https://jenkins.openstack.org/job/nova/1251/ | 20:06 |
| openstackjenkins | Tarmac: Make response structure for list floating ips conform with rest of openstack api | 20:06 |
| openstackgerrit | Dolph Mathews proposed a change to openstack/keystone: Ziad's changes from github.com/rackspace/openstack https://review.openstack.org/237 | 20:08 |
| openstackjenkins | Project nova-tarball-bzr-delta build #478: FAILURE in 12 sec: https://jenkins.openstack.org/job/nova-tarball-bzr-delta/478/ | 20:10 |
| openstackjenkins | Tarmac: We don't have source for open-wrt in the source tree, so we shouldn't use the images. Since the images are only there for uploading smoketests, They are now replaced with random images. | 20:10 |
| zykes- | open-wrt ? | 20:10 |
| zykes- | what the hell :p | 20:11 |
| anotherjesse | I put that in there over a year ago | 20:11 |
| anotherjesse | since it was the smallest thing we had that should launch | 20:11 |
| zykes- | ah | 20:11 |
| anotherjesse | but since then ami-tiny and openstack occured | 20:11 |
| zykes- | ami and what ? | 20:11 |
| anotherjesse | I added it back before it was called nova and before we even heard about rackspace&openstack | 20:12 |
| vishy | ami-tty | 20:12 |
| anotherjesse | the old days of spring 2010 | 20:12 |
| zykes- | ah | 20:13 |
| zykes- | does swift run as a vm at xenserve R? | 20:13 |
| anotherjesse | zykes-: citrix currently runs everything inside a VPX (application domU) | 20:14 |
| anotherjesse | so - yes for swift & nova | 20:15 |
| zykes- | glance as well ? | 20:15 |
| *** dolphm has quit IRC | 20:19 | |
| openstackgerrit | Verification of a change to openstack/glance failed: Add support for shared images https://review.openstack.org/201 | 20:20 |
| jaypipes | Vek: ignore comment above... looks to be a peculiarity with Gerrit -- it's showing changes to files that were changed in trunk master since the last time you pull master I suppose... | 20:20 |
| *** Tushar has quit IRC | 20:21 | |
| openstackgerrit | Dolph Mathews proposed a change to openstack/keystone: Git-ignore python coverage data https://review.openstack.org/239 | 20:21 |
| openstackgerrit | A change was merged to openstack/keystone: Git-ignore python coverage data https://review.openstack.org/239 | 20:24 |
| vishy | soren: ping | 20:25 |
| soren | vishy: 'sup? | 20:29 |
| vishy | soren: theories on cluster installs for migration. The libvirts need to talk to each other | 20:29 |
| vishy | soren: options as I see it are: give one ssh key to the nova user on every compute host in the cluster | 20:30 |
| vishy | soren: or have libvirt listen tcp over a management interface | 20:30 |
| vishy | soren: is there another option that I haven't considered? | 20:30 |
| vishy | soren: I also have some stuff regarding source security group to discuss | 20:31 |
| soren | I'd definitely shoot for ssh. | 20:32 |
| openstackgerrit | Kevin L. Mitchell proposed a change to openstack/glance: Add support for shared images https://review.openstack.org/201 | 20:32 |
| vishy | soren: you think it is acceptable to have the nova user have an ssh key that allows them to get to every machine in the cluster? | 20:32 |
| soren | The tcp+ssl is much more compllicated and the plain SSL is not going to be acceptable for obvious security reasons. | 20:33 |
| openstackgerrit | Verification of a change to openstack/glance failed: Add support for shared images https://review.openstack.org/201 | 20:33 |
| vishy | soren: well if it is over a private interface, that seems like less of an attack surface then giving logins to all the boxes no? | 20:33 |
| soren | vishy: That's assuming noone ever screws up their network config. | 20:34 |
| soren | vishy: :) | 20:34 |
| vishy | soren: true. I'm not actually thinking of putting this in generally | 20:34 |
| soren | vishy: It only takes a minor routing/firewall misconfiguration to accidentally let guests access the hosts. | 20:34 |
| soren | vishy: What would you do in the more general case? | 20:35 |
| vishy | soren: yes good point | 20:35 |
| soren | I'm assuming it's for live migration and such? | 20:35 |
| vishy | soren: yes live migration | 20:35 |
| vishy | soren: We don't really have a general solution for multimachine installs yet | 20:35 |
| openstackjenkins | Project nova build #1,252: SUCCESS in 3 min 33 sec: https://jenkins.openstack.org/job/nova/1252/ | 20:35 |
| openstackjenkins | * Tarmac: Adding kvm-block-migration feature. | 20:35 |
| openstackjenkins | I wrote some description the below URL. I hope it may help for reviewing. | 20:35 |
| openstackjenkins | <http://etherpad.openstack.org/kvm-block-migration> | 20:35 |
| openstackjenkins | * Tarmac: We don't have source for open-wrt in the source tree, so we shouldn't use the images. Since the images are only there for uploading smoketests, They are now replaced with random images. | 20:35 |
| vishy | whoot block migration is in | 20:35 |
| zykes- | that's only for kvm vishy ? | 20:36 |
| vishy | soren: ok I see your point, ssh is probably the "safer" of the options. Maybe it could be configured to ssh into an especially unpriveleged account | 20:36 |
| vishy | that only has a libvirt socket and basically nothing else | 20:37 |
| vishy | zykes-: yes | 20:37 |
| zykes- | vishy: what about xenserver and so on ? | 20:37 |
| vishy | zykes-: block migration doesn't exist in xen yet | 20:37 |
| zykes- | :/ | 20:37 |
| vishy | zykes-: targeted for 2012 i think | 20:37 |
| zykes- | xenserver then ? | 20:38 |
| vishy | same | 20:38 |
| zykes- | so kvm is ahead of it there actually ;p | 20:38 |
| vishy | xenserver supports live migration on shared storage but not block migration afaik | 20:38 |
| zykes- | what's block storage then ? | 20:38 |
| vishy | soren: so security groups, apparently the initial version of ec2 source groups didn't allow for ports to be specified | 20:39 |
| soren | Eh? | 20:39 |
| vishy | soren: so if you use old boto for example you get a security group rule with a source group and NULL in all of the other fields | 20:39 |
| soren | whuh? | 20:39 |
| vishy | soren: you didn't experience this? | 20:40 |
| openstackgerrit | Verification of a change to openstack/glance failed: Add support for shared images https://review.openstack.org/201 | 20:40 |
| soren | vishy: I did not. | 20:40 |
| vishy | soren: did you try with boto 1.9? | 20:40 |
| * soren checks | 20:40 | |
| * soren twiddles thumbs, waiting for rmadison | 20:40 | |
| soren | There we go. | 20:40 |
| soren | 1.9b-1ubuntu5 | 20:40 |
| soren | So yes, 1.9. | 20:40 |
| soren | So what would happen? | 20:41 |
| vishy | http://pastie.org/2377042 | 20:42 |
| vishy | and it worked for you? | 20:42 |
| vishy | hmm | 20:42 |
| vishy | the new version of boto has a version that passes stuff in differently | 20:43 |
| soren | I absolutely tested and verified this. | 20:43 |
| vishy | but as far as i could tell with euca-authorize it doesn't pass in the ip_protocol when you specify a source group | 20:43 |
| * soren tries to remember if there was anything particular about the testing environment that may have affected the test results.. | 20:44 | |
| soren | It's not entirely impossible that I was using a slightly older (by a few revs) api server, I guess. | 20:44 |
| vishy | soren: http://pastie.org/2377052 | 20:45 |
| jaypipes | Vek: https://jenkins.openstack.org/job/glance/89183/console | 20:45 |
| clayg | so... lucid's libvirt is too old? (Version: 0.7.5-5ubuntu27.16) | 20:45 |
| soren | vishy: That's with boto 1.9? | 20:45 |
| soren | clayg: Good grief, yes. | 20:45 |
| vishy | soren: yes | 20:46 |
| soren | vishy: 1.9b? | 20:46 |
| vishy | yup | 20:46 |
| soren | ISTR the letter there being significant. | 20:46 |
| vishy | and euca2ools version 1.3 i think? checking | 20:46 |
| soren | vishy: Ah, good point. /me checks that too | 20:46 |
| soren | 1.3.1-0ubuntu7 | 20:46 |
| vishy | ah i'm using 1.2 | 20:47 |
| vishy | interesting | 20:47 |
| soren | vishy: Tell you what: I'll look at it in the morning to make sure it still works. | 20:47 |
| soren | vishy: If not, I'll fix it. If yes, we can debug together tomorrow? | 20:47 |
| vishy | soren: for comparison http://pastie.org/2377059 | 20:47 |
| vishy | that is what happens with no source group | 20:47 |
| vishy | soren: the comments in boto makes me think that amazon didn't used to support specifying a port for the rule | 20:48 |
| vishy | soren: that you had to allow everything from another group | 20:48 |
| soren | ?!? | 20:48 |
| soren | really? | 20:48 |
| soren | I.. | 20:48 |
| soren | Hm.. | 20:48 |
| *** Tushar has joined #openstack-dev | 20:48 | |
| vishy | yeah see the first pastie | 20:48 |
| soren | Gosh, you're right. | 20:49 |
| soren | I'm trying desperately to remember how I tested this. | 20:49 |
| openstackgerrit | Verification of a change to openstack/glance failed: Add support for shared images https://review.openstack.org/201 | 20:50 |
| vishy | soren: we can deal with it tomorrow. My general thought is to just interpret the null rules as allow everything | 20:50 |
| vishy | soren: I'm also undecided if we should have a default allow all from the same group | 20:51 |
| vishy | i think that is how it works in amazon, but I kind of like the flexibility of having to specify it... | 20:51 |
| soren | vishy: We already have a flag to specify that, don't we? | 20:52 |
| soren | (come to think of it) | 20:52 |
| vishy | soren: it allows all from the same network | 20:52 |
| vishy | soren: which is fine in vlan mode, but is useless in flatdhcp | 20:52 |
| vishy | (and even in vlan mode it allows everything for all groups in the same project | 20:52 |
| soren | Erk. | 20:52 |
| soren | I don't really like the "same network" think. The statically sized subnets are awkward. :-/ | 20:53 |
| soren | s/think/thing/ | 20:53 |
| jaypipes | s1rp: getting these failures again on Vek's shared image groups: https://jenkins.openstack.org/job/glance/89184/console. Can't understand why though... what did we do to resolve those earlier and what could have changed in Vek's commit? | 20:54 |
| vishy | soren: just tested with euca2ools 1.3 and it works | 20:56 |
| vishy | so apparently it was a euca issue? | 20:56 |
| s1rp | jaypipes: good question, no idea why scrubber code would be affected at all | 20:56 |
| Vek | me either. | 20:56 |
| vishy | soren: so the question is do we have an error message or a workaround? Because right now it just happilly accepts the null values and does nothing | 20:57 |
| vishy | soren: i will do a bug report | 20:57 |
| jaypipes | Vek, s1rp: well, this is clearly repeatable, so we'll need to figure it out, or Vek's patches will remain quarantined. :( | 20:57 |
| * jaypipes trying to remember what the fix for this was earlier... | 20:57 | |
| jaypipes | didn't we see this exact failure before? | 20:57 |
| Vek | funny thing is that the tests work fine on my master, just not on the shared images branch | 20:57 |
| jaypipes | Vek: wait, you can repeat the failure locally? | 20:58 |
| Vek | jaypipes: Yeah, actually, I can. | 20:59 |
| jaypipes | Vek: ah, well in that case... | 20:59 |
| *** bengrue has quit IRC | 21:00 | |
| Vek | master is fine, but in my shared-images branch, two test failures. | 21:00 |
| jaypipes | very odd.. | 21:00 |
| Vek | and I didn't touch any of the scrubber code | 21:00 |
| jaypipes | of course | 21:00 |
| Vek | s1rp: could it be some sort of timing-related change? | 21:01 |
| *** bengrue has joined #openstack-dev | 21:01 | |
| s1rp | Vek: i guess that's possible.. | 21:03 |
| soren | vishy: Awesome, thanks. | 21:03 |
| openstackgerrit | James E. Blair proposed a change to openstack/openstack-ci-puppet: Fix crontab typos for gerrit jobs. https://review.openstack.org/240 | 21:03 |
| openstackgerrit | James E. Blair proposed a change to openstack/openstack-ci-puppet: Move puppet cron job to global server config. https://review.openstack.org/241 | 21:03 |
| soren | vishy: *phew*, thought I was losing it there for a few minutes :) | 21:03 |
| vishy | soren: bug 826966 | 21:04 |
| uvirtbot | Launchpad bug 826966 in nova "authorize source security group fails with euca2ools 1.2" [Undecided,New] https://launchpad.net/bugs/826966 | 21:04 |
| * Vek hrms... | 21:04 | |
| *** jakedahn has joined #openstack-dev | 21:04 | |
| anotherjesse | anyone besides sandy able to push novaclient to pypi? | 21:04 |
| Vek | ok, this is weird | 21:04 |
| *** martine has quit IRC | 21:05 | |
| Vek | test_immediate_delete() cleans up, starts servers, adds an image, then does "SELECT * FROM images WHERE status = 'pending_delete'" and expects to get an empty list back | 21:05 |
| Vek | and it's not. | 21:05 |
| s1rp | Vek: does tests.sqlite need to be deleted first | 21:06 |
| Vek | that's what I'm wondering... | 21:07 |
| *** anotherjesse_ has joined #openstack-dev | 21:07 | |
| * Vek does an rm and sees what happens... | 21:07 | |
| Vek | no, rm wasn't sufficient... | 21:08 |
| *** anotherjesse has quit IRC | 21:09 | |
| *** anotherjesse_ is now known as anotherjesse | 21:09 | |
| *** lts has quit IRC | 21:10 | |
| kbringard | upgrade to rm -rf | 21:11 |
| kbringard | :-p | 21:11 |
| Vek | kbringard: you first :P | 21:12 |
| jaypipes | Vek: do a diff between the tests.functional.__init__.py file in master vs. topic branch... | 21:13 |
| Vek | jaypipes: I diff a full-up diff between the branch; I don't recall __init__.py being touched. | 21:13 |
| Guest77784 | hi all, can i ask a quick question about networking in d3? | 21:14 |
| Vek | I take that back, it was... | 21:14 |
| jaypipes | Vek: yeah, just a little change: https://review.openstack.org/#patch,sidebyside,201,6,glance/tests/functional/__init__.py | 21:14 |
| Vek | but I can't see how this change would cause problems. | 21:14 |
| Vek | +context_class = glance.registry.context.RequestContext | 21:14 |
| jaypipes | Vek: me neither... | 21:14 |
| jaypipes | Vek: still digging.. | 21:15 |
| Vek | I added some debugging prints to that test case, and there are several images already in the database at the point of failure, one in the pending_delete state. | 21:16 |
| Vek | best guess is the db is not getting cleaned up properly | 21:17 |
| *** chipc has quit IRC | 21:17 | |
| jaypipes | Vek: got it. | 21:17 |
| jaypipes | Vek: line 88 of glance/store/scrubber.py | 21:17 |
| jaypipes | Vek: passing in the wrong Context... | 21:17 |
| jaypipes | Vek: needs to be the new RegistryContext I believe. | 21:18 |
| *** letterj has left #openstack-dev | 21:18 | |
| Vek | jaypipes: hmmm, could be... | 21:19 |
| * Vek tries... | 21:19 | |
| jaypipes | Vek: the scrubber test failure messages are a bit confusing... they appear that way if something goes wrong with the scrubber daemon in any way. the error doesn't represent what is really wrong... | 21:19 |
| Vek | yay, that seemed to fix it. | 21:20 |
| openstackgerrit | Kevin L. Mitchell proposed a change to openstack/glance: Add support for shared images https://review.openstack.org/201 | 21:20 |
| Vek | submitted... | 21:21 |
| openstackgerrit | Verification of a change to openstack/glance failed: Add support for shared images https://review.openstack.org/201 | 21:33 |
| jaypipes | Vek: hehe, yet another failure. different this time, though :) https://jenkins.openstack.org/job/glance/89185/console | 21:34 |
| Vek | *sigh* now what. | 21:34 |
| Vek | yeah, and it didn't show up when *I* ran the tests. | 21:34 |
| * Vek restarts | 21:35 | |
| openstackgerrit | A change was merged to openstack/glance: Add support for shared images https://review.openstack.org/201 | 21:37 |
| Vek | ok, that's slightly disturbing. I run tests twice and no error; jenkins runs the tests and gets an obscure error the first time, then succeeds the second time... | 21:39 |
| * Vek doesn't know what to make of that. | 21:39 | |
| *** markvoelker has quit IRC | 21:42 | |
| *** mfer has quit IRC | 21:43 | |
| *** kbringard has quit IRC | 21:55 | |
| openstackgerrit | Yogeshwar Srikrishnan proposed a change to openstack/keystone: Additional changes to fix minor service support stuff and increase test coverage. https://review.openstack.org/242 | 22:07 |
| *** cp16net has quit IRC | 22:07 | |
| *** anotherjesse has quit IRC | 22:10 | |
| *** anotherjesse has joined #openstack-dev | 22:23 | |
| jaypipes | Vek: was a cache problem on the jenkins server... | 22:24 |
| jaypipes | Vek: w00t. \o/ merged finally :) | 22:29 |
| openstackgerrit | Yogeshwar Srikrishnan proposed a change to openstack/keystone: Additional changes to fix minor service support stuff and increase test coverage. https://review.openstack.org/242 | 22:30 |
| Guest77784 | can someone support me for a minute? | 22:33 |
| Guest77784 | i need help to setup network in d3 | 22:33 |
| *** bsza has quit IRC | 22:34 | |
| Guest77784 | ttylinux test vm starts but doesn't recognise eth0 | 22:34 |
| Guest77784 | nor meta-data server | 22:34 |
| Guest77784 | i don't know where can i find the solution | 22:35 |
| Guest77784 | i am blocked since this morning | 22:35 |
| Guest77784 | the error is eth0 is not a recognized interface. | 22:38 |
| *** yogirackspace has left #openstack-dev | 22:39 | |
| *** zorzar has quit IRC | 22:54 | |
| *** zorzar has joined #openstack-dev | 22:56 | |
| *** rnirmal has quit IRC | 22:58 | |
| openstackgerrit | A change was merged to openstack/openstack-ci-puppet: Fix crontab typos for gerrit jobs. https://review.openstack.org/240 | 22:58 |
| openstackgerrit | A change was merged to openstack/openstack-ci-puppet: Move puppet cron job to global server config. https://review.openstack.org/241 | 22:59 |
| *** anotherjesse has quit IRC | 23:11 | |
| *** anotherjesse has joined #openstack-dev | 23:11 | |
| *** dragondm has quit IRC | 23:28 | |
| *** bengrue has quit IRC | 23:35 | |
| *** amccabe has quit IRC | 23:40 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
