| openstackgerrit | A change was merged to openstack/nova: xenapi: Fix reboot with hung volumes https://review.openstack.org/23662 | 00:00 |
|---|---|---|
| *** mrodden has quit IRC | 00:02 | |
| *** baba has quit IRC | 00:02 | |
| *** megha has joined #openstack-dev | 00:03 | |
| openstackgerrit | A change was merged to openstack/oslo-incubator: update.py: support one-module-per-line configuration https://review.openstack.org/24220 | 00:03 |
| openstackgerrit | A change was merged to openstack/glance: Characterize image_member_find https://review.openstack.org/23895 | 00:04 |
| openstackgerrit | A change was merged to openstack/glance: Fix calls to image_member_find in tests https://review.openstack.org/23897 | 00:04 |
| openstackgerrit | A change was merged to openstack/horizon: Fixes error message formatting in Load Balancers panel https://review.openstack.org/23953 | 00:04 |
| openstackgerrit | A change was merged to openstack/keystone: Explain LDAP page_size & default value https://review.openstack.org/23445 | 00:04 |
| *** flaper87 has quit IRC | 00:04 | |
| *** lloydde has quit IRC | 00:06 | |
| *** vipul|away is now known as vipul | 00:07 | |
| openstackgerrit | A change was merged to openstack/quantum: Add l3 db migration for plugins which did not support in folsom https://review.openstack.org/24107 | 00:08 |
| *** sthaha has joined #openstack-dev | 00:12 | |
| cyeoh | is anyone else seeing "ImportError: No module named oslo.config" errors when trying to run run_tests.sh on the various projects? (cinder/keystone?) | 00:12 |
| cyeoh | (this is with creating a venv) | 00:13 |
| *** davidhadas has quit IRC | 00:13 | |
| *** davidha has joined #openstack-dev | 00:14 | |
| *** Tross has quit IRC | 00:17 | |
| *** jamielennox has quit IRC | 00:18 | |
| *** jamielennox has joined #openstack-dev | 00:19 | |
| *** kagan has quit IRC | 00:22 | |
| *** pabelanger has joined #openstack-dev | 00:22 | |
| *** Tross has joined #openstack-dev | 00:30 | |
| *** markmcclain has joined #openstack-dev | 00:30 | |
| *** amotoki has quit IRC | 00:37 | |
| *** Ritz-Laptop has joined #openstack-dev | 00:38 | |
| *** jamielennox has quit IRC | 00:38 | |
| *** jamielennox has joined #openstack-dev | 00:39 | |
| *** mohits has joined #openstack-dev | 00:42 | |
| *** feiyuliu has joined #openstack-dev | 00:42 | |
| *** ewindisch has joined #openstack-dev | 00:44 | |
| *** radez_g0n3 is now known as radez | 00:45 | |
| *** stevemar has joined #openstack-dev | 00:46 | |
| *** sacharya has joined #openstack-dev | 00:46 | |
| *** gyee has quit IRC | 00:48 | |
| *** feiyuliu has quit IRC | 00:49 | |
| *** ijw has joined #openstack-dev | 00:51 | |
| *** stevemar has quit IRC | 00:55 | |
| *** jhesketh_ has joined #openstack-dev | 00:55 | |
| *** Nachi has joined #openstack-dev | 00:56 | |
| *** winston-d has joined #openstack-dev | 00:56 | |
| *** crandquist has joined #openstack-dev | 00:58 | |
| *** nati_ueno has quit IRC | 00:59 | |
| *** kaushikc has joined #openstack-dev | 00:59 | |
| *** jamielennox has quit IRC | 01:00 | |
| *** jamielennox has joined #openstack-dev | 01:01 | |
| *** cp16net|away is now known as cp16net | 01:02 | |
| *** adjohn has quit IRC | 01:02 | |
| *** halfss has joined #openstack-dev | 01:04 | |
| *** dkehn has left #openstack-dev | 01:06 | |
| openstackgerrit | A change was merged to openstack/glance: Retain migration 12 indexes for table image_properties with sqlite. https://review.openstack.org/23876 | 01:07 |
| *** Nachi has quit IRC | 01:08 | |
| *** crandquist has quit IRC | 01:08 | |
| *** vkmc has quit IRC | 01:12 | |
| *** dsanders has quit IRC | 01:12 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Raise stevedore requirement to 0.7 https://review.openstack.org/24153 | 01:14 |
| openstackgerrit | A change was merged to openstack/ceilometer: Use a non-standard port for the test server https://review.openstack.org/23955 | 01:16 |
| *** cp16net is now known as cp16net|away | 01:17 | |
| *** alop has quit IRC | 01:18 | |
| *** alexpilotti has quit IRC | 01:19 | |
| *** jergerber has joined #openstack-dev | 01:20 | |
| *** ladquin has quit IRC | 01:20 | |
| termie | ayoung: are people still using tokens taht don't have tenants? | 01:21 |
| *** jamielennox has quit IRC | 01:22 | |
| *** jamielennox has joined #openstack-dev | 01:23 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Add deprecation warnings for V1 API https://review.openstack.org/24112 | 01:23 |
| *** luis_ has quit IRC | 01:23 | |
| *** novas0x2a|laptop has quit IRC | 01:24 | |
| termie | ayoung: it feels stupid to keep writing backends that support that | 01:25 |
| openstackgerrit | A change was merged to openstack/ceilometer: Ensure the statistics are sorted https://review.openstack.org/23956 | 01:26 |
| *** kaushikc has quit IRC | 01:28 | |
| *** cp16net|away is now known as cp16net | 01:28 | |
| *** jasdeepH has quit IRC | 01:28 | |
| ayoung | termie, my thought is that a token without a tenant should be only for Keystone tasks. | 01:29 |
| ayoung | for example, if you want to use one token to get another token, you use an unscoped to get a scoped, but not the other way around | 01:29 |
| termie | ayoung: by "keystone tasks" i assume you mean using the keystone api, not tasks done by keystone | 01:30 |
| ayoung | the list of revoked tenants is for PKI tokens. PKI tokens can be verified in auth_token middleware without contacting the Keystone server | 01:30 |
| ayoung | termie, that is correct | 01:30 |
| ayoung | things like changing your password | 01:30 |
| ayoung | or for admins, adding new users | 01:30 |
| termie | ayoung: admins should be using admin creds | 01:31 |
| ayoung | termie, so, how would you define admin creds? | 01:31 |
| termie | ayoung: having a keystone admin role | 01:31 |
| ayoung | the thing is, we don't have roles without them being a role in something. | 01:32 |
| ayoung | THere are no global roles | 01:32 |
| termie | ayoung: admin tenant | 01:32 |
| termie | ayoung: this was solved long ago | 01:32 |
| ayoung | termie, yes, before we had domains | 01:32 |
| termie | ayoung: i don't know what design for domains eventually dropped, but mine had no issues with that system | 01:33 |
| ayoung | termie, what we have now is that, by default, there is one domain, and users and projects (tenants) all belong to a domain | 01:33 |
| ayoung | A site can create additional domains | 01:33 |
| ayoung | and users from one domain can be assigned roles in projects in another domain | 01:34 |
| ayoung | but domains are the ownership container | 01:34 |
| termie | i don't see how that changes anything about having admin rights in a domain or keystone in total | 01:35 |
| *** danwent has quit IRC | 01:36 | |
| ayoung | so is there one admin tenant per domain? | 01:37 |
| ayoung | Ot just the admin tenant in the default domain gets admin? | 01:37 |
| termie | ayoung: there can be as many as you want, it is a role | 01:37 |
| *** terryg has quit IRC | 01:38 | |
| ayoung | So the code I inherited had something broken in it...I'll paste the piece I am talking about: | 01:38 |
| ayoung | https://github.com/openstack/keystone/blob/master/keystone/common/wsgi.py#L291 | 01:38 |
| ayoung | THis is enforced in Keystone, and it checks that the user has the admin role on "any" tenant. | 01:39 |
| ayoung | If so, the user is considered "admin" | 01:39 |
| ayoung | Haven't done a git blame on it to see where that came from. | 01:39 |
| ayoung | I just want to fix it. | 01:39 |
| termie | that line is misleading? | 01:40 |
| *** jamielennox has quit IRC | 01:40 | |
| termie | is the "creds.roles()" call the one that is getting a list of roles for all tenants? | 01:41 |
| *** ilblackdragon has joined #openstack-dev | 01:41 | |
| *** jamielennox has joined #openstack-dev | 01:41 | |
| termie | i don't see it scannign through tenants | 01:41 |
| ayoung | nope, it never did | 01:41 |
| ayoung | It just checks for the tenant passed in | 01:42 |
| termie | and what is wrong with that? | 01:42 |
| termie | it doesn't have to be an admin tenant, it just has to be a sufficiently unique role name | 01:42 |
| ayoung | If you have the role admin in *any* tenant you are an admin | 01:42 |
| termie | whatever the policy says lets you control your tenant, your domain or the whole system | 01:42 |
| termie | ayoung: no, only int he tenant you are using | 01:42 |
| ayoung | nope | 01:43 |
| termie | ayoung: the tenant doesn't matter, however, the role does | 01:43 |
| ayoung | that is not how it works in practice | 01:43 |
| ayoung | that is wrong | 01:43 |
| ayoung | it should not be admin in any tenant | 01:43 |
| ayoung | that is messing people up | 01:43 |
| ayoung | I think that is a holdover from global roles | 01:43 |
| ayoung | there, it makes sense | 01:43 |
| termie | the policy enforcement should be better described to peopl ethen | 01:43 |
| ayoung | no | 01:43 |
| termie | there are no global roles | 01:43 |
| ayoung | admin should be scoped appropriately | 01:44 |
| ayoung | I would be ok with | 01:44 |
| termie | it can be scoped however anybody wnats | 01:44 |
| ayoung | "admin on the admin tenant" | 01:44 |
| ayoung | or even | 01:44 |
| ayoung | any role in the admin tenant | 01:44 |
| termie | if you want to make the policy say "tenant=admin" and "role=admin" you can do that fine | 01:44 |
| ayoung | or admin in the default domain | 01:44 |
| termie | two lines in policy.json | 01:44 |
| termie | keystone itself should reserve some roles it cares about for its own enforcement | 01:44 |
| ayoung | Yeah, but that rule was enforced in code, not in policy. | 01:44 |
| termie | self.policy_api.enforce(context, creds, 'admin_required', {}) | 01:45 |
| termie | my ideal design would have admin_required require tenant=admin, role=keystone:admin | 01:45 |
| ayoung | so right now I am proposing this: https://blueprints.launchpad.net/keystone/+spec/remove-tenant-admin | 01:46 |
| termie | for something is a keystone admin | 01:46 |
| ayoung | which basically says | 01:46 |
| ayoung | role = admin, domain = default | 01:46 |
| ayoung | and then we say that admins for the default domain have the ability to create other domains. | 01:46 |
| termie | there are multiple levels of admin | 01:46 |
| ayoung | Yep | 01:46 |
| ayoung | and to be able to, say, add a user to a project, you should only have to be admin on that project | 01:47 |
| *** yaguang has joined #openstack-dev | 01:47 | |
| ayoung | Make sense? | 01:47 |
| termie | ayoung: that last statement makes sense but is something that is already fully possible | 01:48 |
| ayoung | termie, anything is possible, but to be able to do it in the current Keystone takes a python change | 01:48 |
| openstackgerrit | A change was merged to openstack/glance: Fix up publicize_image unit test https://review.openstack.org/24087 | 01:48 |
| openstackgerrit | A change was merged to openstack/glance: Mark password options secret https://review.openstack.org/24176 | 01:48 |
| ayoung | when we make that python change, I want the default policy in place | 01:48 |
| termie | just make the requirement for add_user_to_project be: (role=keystone:admin) or (role:tenant_admin, tenant=current_tenant) | 01:48 |
| ayoung | That is the plan, yes | 01:49 |
| termie | that is a policy.json change | 01:49 |
| ayoung | Yep. | 01:49 |
| *** ewindisch has quit IRC | 01:50 | |
| termie | the only python change is to make all the calls check the policy which i don't know if they are doihjng | 01:50 |
| ayoung | they need to do that, and they need to stop doing the enforce_admin call. | 01:50 |
| termie | enforce_admin was specifically described as "right now keystone supports binary permissions" | 01:51 |
| termie | it stopped people arguing about who they thought hsould be allowed ot do things | 01:51 |
| termie | the policy system has supported all the possibilities fromt eh beginnign, though | 01:51 |
| ayoung | Yeah, what I am trying to do is figure out how to move ahead without breaking everyone that is operating under that assumption | 01:51 |
| *** woodspa has joined #openstack-dev | 01:51 | |
| ayoung | Not a big deal, just trying to spec it out before coding. It is going to require tweaking a bunch of unit tests and I don't feel like doing that multiple times. | 01:52 |
| termie | nova already does all this, just make every call check a policy that has its name | 01:52 |
| termie | k | 01:52 |
| termie | anyway, i don't really care about all that, i just came here to complain that unscoped tokens are stupid | 01:52 |
| *** Tross has quit IRC | 01:52 | |
| termie | and needlessly create duplicate code paths | 01:52 |
| ayoung | OK, I have a patch I need to cleanup and resubmit. It is good to have you back engaged. We've been lacking some perspective on why decisions were made. | 01:53 |
| *** jcmartin has quit IRC | 01:53 | |
| termie | i can't say as that i am happy to be engaged | 01:53 |
| *** ilblackd_ has joined #openstack-dev | 01:53 | |
| *** jog0 has quit IRC | 01:54 | |
| *** ilblackd_ has quit IRC | 01:54 | |
| termie | was feeling much better being in r&d land | 01:54 |
| *** ilblackdragon has quit IRC | 01:56 | |
| *** hejie has joined #openstack-dev | 01:57 | |
| *** hejie has quit IRC | 01:57 | |
| *** xchu has joined #openstack-dev | 01:58 | |
| *** zb has joined #openstack-dev | 01:59 | |
| *** ilblackdragon has joined #openstack-dev | 01:59 | |
| *** xchu has quit IRC | 02:00 | |
| cyeoh | ayoung: hi | 02:01 |
| *** markmcclain has quit IRC | 02:01 | |
| ayoung | cyeoh, how do | 02:01 |
| cyeoh | hey I was wondering if I could ask you about https://review.openstack.org/#/c/23660/ | 02:02 |
| cyeoh | I'm just cleaning it up based on feedback. | 02:02 |
| *** zaneb has quit IRC | 02:03 | |
| cyeoh | But given Dolph's comments I was wondering if there is anything fundamental about it which you don't like? | 02:03 |
| *** ilblackd_ has joined #openstack-dev | 02:03 | |
| *** ewindisch has joined #openstack-dev | 02:03 | |
| ayoung | cyeoh, have to admit, I am just now reading the most recent comments. In general, I like the approach | 02:04 |
| *** woodspa has quit IRC | 02:04 | |
| ayoung | cyeoh, and I desperately want to be able to run our SQL based unit tests against both MySQL and postgres (although not on every run) and if your patch gets us closer to that, I am thrilled | 02:05 |
| *** jamielennox has quit IRC | 02:05 | |
| ayoung | cyeoh, in general, I prefer small changes, so incremental is preferable. | 02:05 |
| *** ilblack__ has joined #openstack-dev | 02:06 | |
| *** jamielennox has joined #openstack-dev | 02:06 | |
| ayoung | cyeoh, sound good? | 02:06 |
| *** ilblackdragon has quit IRC | 02:07 | |
| ayoung | cyeoh, and I am fine with dolph abandoning his. So long as we are aware of each others work, this can proceed. | 02:07 |
| cyeoh | ayoung: ok cool. It does allow for both MySQL and Postgres. I'll clean up the current patch and then start submitting incremental ones based on it with probably one changeset per db migration if thats not considered too small. | 02:07 |
| ayoung | cyeoh, the only thing is sometimes we want to group migrations, so keep that in mind | 02:08 |
| ayoung | we have a tendency to make separate migrations for changing schema and for migrating the actual data, but the tests for those are often executed together | 02:08 |
| *** morganfainberg has quit IRC | 02:08 | |
| *** dims has quit IRC | 02:09 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Add a tox target for building documentation https://review.openstack.org/23935 | 02:09 |
| ayoung | But otherwise, it sounds like a planm | 02:09 |
| *** ilblackd_ has quit IRC | 02:09 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Switch to final 1.1.0 oslo.config release https://review.openstack.org/24207 | 02:10 |
| cyeoh | yea I noticed that when looking at the db migrations. The data injected at each level is persistent (for a given test run) so even though there is pre/post check at each level, the end net result can also be checked after any migration. | 02:10 |
| *** HenryG has quit IRC | 02:10 | |
| *** ilblackdragon has joined #openstack-dev | 02:11 | |
| cyeoh | might need just rework where the test data injected is stored to make accessing the test data easier. | 02:11 |
| *** ilblack__ has quit IRC | 02:11 | |
| *** jasdeepH has joined #openstack-dev | 02:12 | |
| ayoung | that is fine. In general, anything that improves our SQL testing is good. | 02:12 |
| *** markmcclain has joined #openstack-dev | 02:12 | |
| *** ijw has quit IRC | 02:12 | |
| *** ilblackd_ has joined #openstack-dev | 02:12 | |
| *** ilblackd_ has quit IRC | 02:12 | |
| *** ijw has joined #openstack-dev | 02:12 | |
| *** ilblackd_ has joined #openstack-dev | 02:15 | |
| *** ilblackdragon has quit IRC | 02:16 | |
| openstackgerrit | A change was merged to openstack/ceilometer: Add sample configuration files for mod_wsgi https://review.openstack.org/23613 | 02:17 |
| *** danwent has joined #openstack-dev | 02:19 | |
| *** Tross has joined #openstack-dev | 02:20 | |
| *** jamielennox has quit IRC | 02:24 | |
| *** ijw has quit IRC | 02:24 | |
| *** jamielennox has joined #openstack-dev | 02:25 | |
| *** stevemar has joined #openstack-dev | 02:25 | |
| *** ijw has joined #openstack-dev | 02:25 | |
| *** Oneiroi has quit IRC | 02:26 | |
| *** Adri2000 has joined #openstack-dev | 02:27 | |
| *** jcmartin has joined #openstack-dev | 02:27 | |
| *** pixelbeat_ has quit IRC | 02:29 | |
| *** eharney has quit IRC | 02:29 | |
| *** matiu has quit IRC | 02:29 | |
| *** Oneiroi has joined #openstack-dev | 02:31 | |
| termie | ayoung: tokens with no expiration are supported? | 02:33 |
| openstackgerrit | A change was merged to openstack/nova: Fix: Managers that incorrectly derive from SchedulerDependentManager https://review.openstack.org/24174 | 02:34 |
| *** Tross has quit IRC | 02:34 | |
| *** Ryan_Lane has quit IRC | 02:34 | |
| *** cp16net is now known as cp16net|away | 02:36 | |
| *** cp16net|away is now known as cp16net | 02:39 | |
| *** kbrierly has quit IRC | 02:39 | |
| *** kbrierly has joined #openstack-dev | 02:39 | |
| *** monst_ has joined #openstack-dev | 02:39 | |
| *** Tross has joined #openstack-dev | 02:41 | |
| *** Hien has quit IRC | 02:42 | |
| *** Hien has joined #openstack-dev | 02:43 | |
| *** matiu has joined #openstack-dev | 02:47 | |
| *** jamielennox has quit IRC | 02:48 | |
| *** jamielennox has joined #openstack-dev | 02:49 | |
| *** jergerber has quit IRC | 02:49 | |
| *** CaptTofu has joined #openstack-dev | 02:57 | |
| *** hanney has quit IRC | 02:57 | |
| *** olaph_ has joined #openstack-dev | 02:59 | |
| *** jcmartin has quit IRC | 02:59 | |
| *** bdpayne has quit IRC | 03:01 | |
| *** olaph has quit IRC | 03:02 | |
| *** sandywalsh has quit IRC | 03:02 | |
| *** aabes_ has joined #openstack-dev | 03:03 | |
| *** fubada has joined #openstack-dev | 03:05 | |
| termie | ayoung: okay, having some issues with revoked tokens, the call to 'delete' token is supposed to mark the thing as invalid but not actually delete it? | 03:06 |
| termie | ayoung: i'm still not sure of what the purpose of list_revoked_tkoens is | 03:06 |
| fubada | hi is it possible to change which physical volume devstack is using right after running stack.sh? | 03:06 |
| fubada | i cant get any help in the main channel | 03:07 |
| *** ijw has quit IRC | 03:07 | |
| *** ijw has joined #openstack-dev | 03:08 | |
| jgriffith | fubada: You can tweak the cinder.conf setting and restart cinder if you want | 03:09 |
| jgriffith | fubada: "volume_group" is the setting | 03:09 |
| fubada | thanks man! | 03:09 |
| jgriffith | fubada: IIRC you can also change this in your localrc file to begin with, but can't remember if devstack looks for that one or not | 03:10 |
| jgriffith | fubada: no problem | 03:10 |
| fubada | thanks, I had an issue running stack.sh, it completed by saying stack is installed, followed by a permissions error on /home/myuser/.bash_profile | 03:11 |
| fubada | didnt know what to make of that | 03:11 |
| fubada | stack.sh ran as user stack, so its no surprise | 03:11 |
| *** jcmartin has joined #openstack-dev | 03:11 | |
| *** sandeepr has joined #openstack-dev | 03:11 | |
| jgriffith | fubada: hmmm... that's odd actually | 03:11 |
| jgriffith | fubada: you should run devstack as a normal user and no sudo | 03:12 |
| fubada | oh | 03:12 |
| fubada | should i reinstall? I ran it as root, and it jailed into stack | 03:12 |
| jgriffith | fubada: ahh | 03:13 |
| jgriffith | fubada: you can try unstack as root | 03:13 |
| jgriffith | fubada: then run as a normal user | 03:13 |
| *** jamielennox has quit IRC | 03:13 | |
| fubada | okay, i think ill pass :) im 99% sure its ok | 03:13 |
| jgriffith | fubada: if it's a vm with a snapshot might be easiest to start over | 03:13 |
| *** bdpayne has joined #openstack-dev | 03:13 | |
| jgriffith | fubada: also for tips checkout devstack.org | 03:13 |
| fubada | thanks for the volume_group tip, can it be a full path? | 03:13 |
| *** jamielennox has joined #openstack-dev | 03:13 | |
| jgriffith | fubada: if it works roll with it :) | 03:13 |
| *** mohits has quit IRC | 03:13 | |
| fubada | currently its just set to "stack-volumes" | 03:14 |
| jgriffith | fubada: No, you just want the ouptut from "sudo vgs" | 03:14 |
| *** SergeyLukjanov has joined #openstack-dev | 03:14 | |
| fubada | ok got it! | 03:14 |
| fubada | interesting, will it insist on formatting this lvm volume group that I have set up? | 03:14 |
| fubada | because I currently have it mounted as ext4 | 03:15 |
| jgriffith | fubada: ummm | 03:15 |
| jgriffith | fubada: I thnk they're might be a misunderstanding on how this works | 03:15 |
| *** kaushikc has joined #openstack-dev | 03:16 | |
| jgriffith | openstack LVM driver is just going to look for the specfied VolumeGroup to use | 03:16 |
| jgriffith | This can be created from a loopback file (which is what devstack sets up) | 03:16 |
| jgriffith | or it can be based on a true PV | 03:16 |
| fubada | right | 03:16 |
| jgriffith | when you do "cinder create xxxx" it's actually just creating an LVM on the specified VG | 03:16 |
| fubada | i get it, I'd like to recreate/move the loopback file to my existing ext4 fs | 03:17 |
| winston-d | jgriffith: hey, just so you know, the log fix has been merged into oslo, i'll submit a patch to pull it to cinder so that the bug 1126378/1131322 can be fixed. | 03:17 |
| uvirtbot | Launchpad bug 1126378 in cinder "SchedulerHostFilterNotFound: Scheduler Host Filter AvailabilityZoneFilter, CapacityFilter, CapabilitiesFilter could not be found." [High,Fix committed] https://launchpad.net/bugs/1126378 | 03:17 |
| fubada | which happens to be dmraid with luks and lvm ontop | 03:17 |
| fubada | formatted as ext4 | 03:17 |
| jgriffith | winston-d: Ohhh, that's excellent!!! | 03:17 |
| jgriffith | winston-d: thanks you VERY much! | 03:17 |
| *** rushiagr has joined #openstack-dev | 03:18 | |
| winston-d | jgriffith: :) | 03:18 |
| *** ewindisch has quit IRC | 03:20 | |
| jgriffith | winston-d: have you or anybody else heard yet whether oslo.1.1 made it to the mirror yet? | 03:20 |
| winston-d | jgriffith: nope, i haven't | 03:21 |
| jgriffith | winston-d: k | 03:21 |
| openstackgerrit | A change was merged to openstack/tempest: Replace try/except/else with self.assertRaises. https://review.openstack.org/24140 | 03:22 |
| *** ijw1 has joined #openstack-dev | 03:25 | |
| *** stevemar has quit IRC | 03:28 | |
| *** ijw has quit IRC | 03:29 | |
| *** martine has quit IRC | 03:29 | |
| *** jamielennox has quit IRC | 03:32 | |
| *** jamielennox has joined #openstack-dev | 03:33 | |
| *** Tross has quit IRC | 03:36 | |
| *** Mandell has joined #openstack-dev | 03:40 | |
| *** sandywalsh has joined #openstack-dev | 03:40 | |
| *** matiu has quit IRC | 03:41 | |
| *** Tross has joined #openstack-dev | 03:42 | |
| *** matiu has joined #openstack-dev | 03:43 | |
| *** rushiagr has quit IRC | 03:43 | |
| *** yaguang has left #openstack-dev | 03:46 | |
| *** susanne-balle has quit IRC | 03:46 | |
| *** yaguang has joined #openstack-dev | 03:46 | |
| *** monst_ has quit IRC | 03:47 | |
| *** CaptTofu has quit IRC | 03:50 | |
| *** CaptTofu has joined #openstack-dev | 03:50 | |
| *** jamielennox has quit IRC | 03:51 | |
| *** jamielennox has joined #openstack-dev | 03:52 | |
| *** Ryan_Lane has joined #openstack-dev | 03:54 | |
| termie | ayoung: https://review.openstack.org/#/c/24284 | 03:59 |
| termie | ayoung: cassandra backend for tokens | 03:59 |
| *** henrynash has joined #openstack-dev | 04:00 | |
| openstackgerrit | A change was merged to openstack/nova: Delete instance metadata when delete VM https://review.openstack.org/23772 | 04:05 |
| openstackgerrit | A change was merged to openstack/quantum: remove references to netstack in setup.py https://review.openstack.org/24274 | 04:05 |
| *** ijw1 has quit IRC | 04:06 | |
| *** nunosantos has quit IRC | 04:06 | |
| *** torandu_ has quit IRC | 04:08 | |
| *** jamielennox has quit IRC | 04:10 | |
| *** jamielennox has joined #openstack-dev | 04:10 | |
| *** kaushikc has left #openstack-dev | 04:19 | |
| *** woodspa has joined #openstack-dev | 04:19 | |
| *** aeperezt has quit IRC | 04:21 | |
| *** Tross has quit IRC | 04:23 | |
| *** ilblackdragon has joined #openstack-dev | 04:23 | |
| *** Tross has joined #openstack-dev | 04:23 | |
| *** jamielennox has quit IRC | 04:23 | |
| *** jamielennox has joined #openstack-dev | 04:24 | |
| *** olaph has joined #openstack-dev | 04:24 | |
| *** henrynash has quit IRC | 04:25 | |
| *** boris-42 has joined #openstack-dev | 04:25 | |
| *** ilblackd_ has quit IRC | 04:26 | |
| *** kagan has joined #openstack-dev | 04:27 | |
| *** olaph_ has quit IRC | 04:27 | |
| *** pabelanger has quit IRC | 04:29 | |
| *** boris-42 has quit IRC | 04:30 | |
| *** ilblackdragon has quit IRC | 04:31 | |
| *** woodspa has left #openstack-dev | 04:38 | |
| *** SergeyLukjanov has quit IRC | 04:38 | |
| *** radez is now known as radez_g0n3 | 04:40 | |
| *** sacharya has quit IRC | 04:42 | |
| *** jamielennox has quit IRC | 04:44 | |
| openstackgerrit | A change was merged to openstack/quantum: port_security migration does not migrate data https://review.openstack.org/24024 | 04:44 |
| *** boris-42 has joined #openstack-dev | 04:44 | |
| *** jamielennox has joined #openstack-dev | 04:45 | |
| *** pnavarro has quit IRC | 04:48 | |
| *** darjeeling has quit IRC | 04:55 | |
| *** glikson has joined #openstack-dev | 05:00 | |
| *** jamielennox has quit IRC | 05:03 | |
| *** jamielennox has joined #openstack-dev | 05:04 | |
| *** bdpayne has quit IRC | 05:11 | |
| *** ilblackdragon has joined #openstack-dev | 05:12 | |
| *** hattwick has quit IRC | 05:13 | |
| *** thimble has joined #openstack-dev | 05:14 | |
| *** bdpayne has joined #openstack-dev | 05:17 | |
| *** jcmartin has quit IRC | 05:21 | |
| openstackgerrit | A change was merged to openstack/quantum: Ensure port get works when NVP mapping not stored in Quantum DB https://review.openstack.org/24067 | 05:22 |
| *** navid_ has joined #openstack-dev | 05:22 | |
| *** rushiagr has joined #openstack-dev | 05:23 | |
| *** bdpayne has quit IRC | 05:26 | |
| *** glikson has quit IRC | 05:28 | |
| *** jamielennox has quit IRC | 05:28 | |
| *** jamielennox has joined #openstack-dev | 05:29 | |
| openstackgerrit | A change was merged to openstack/nova: Pin SQLAlchemy to 0.7.x https://review.openstack.org/24075 | 05:30 |
| *** boris-42 has quit IRC | 05:32 | |
| *** rushiagr has quit IRC | 05:32 | |
| *** winston-d has quit IRC | 05:35 | |
| *** andrewbogott is now known as andrewbogott_afk | 05:39 | |
| *** thimble has quit IRC | 05:41 | |
| *** monst_ has joined #openstack-dev | 05:42 | |
| *** melwitt1 has joined #openstack-dev | 05:44 | |
| *** melwitt1 has quit IRC | 05:44 | |
| *** melwitt1 has joined #openstack-dev | 05:44 | |
| *** darjeeling has joined #openstack-dev | 05:45 | |
| *** melwitt has quit IRC | 05:47 | |
| *** markmcclain has quit IRC | 05:48 | |
| *** monst_ has quit IRC | 05:48 | |
| *** winston-d has joined #openstack-dev | 05:51 | |
| *** jamielennox has quit IRC | 05:51 | |
| *** jamielennox has joined #openstack-dev | 05:52 | |
| *** rkukura has quit IRC | 05:52 | |
| *** networkstatic has quit IRC | 05:56 | |
| *** tomoe_ has joined #openstack-dev | 05:59 | |
| *** almaisan-away is now known as al-maisan | 06:03 | |
| *** koolhead17 has joined #openstack-dev | 06:03 | |
| *** ilblackdragon has quit IRC | 06:05 | |
| *** JonnyNomad has quit IRC | 06:07 | |
| *** JonnyNomad has joined #openstack-dev | 06:08 | |
| openstackgerrit | A change was merged to openstack/glance: Fix visibility on db image_member_find https://review.openstack.org/23898 | 06:16 |
| *** jamielennox has quit IRC | 06:16 | |
| openstackgerrit | A change was merged to openstack/quantum: Imported Translations from Transifex https://review.openstack.org/24270 | 06:16 |
| *** rkukura has joined #openstack-dev | 06:16 | |
| *** jamielennox has joined #openstack-dev | 06:17 | |
| *** al-maisan is now known as almaisan-away | 06:22 | |
| *** alop has joined #openstack-dev | 06:23 | |
| *** alop has quit IRC | 06:25 | |
| *** alop has joined #openstack-dev | 06:25 | |
| *** rushiagr has joined #openstack-dev | 06:26 | |
| *** alexxu has joined #openstack-dev | 06:29 | |
| *** kbrierly has quit IRC | 06:31 | |
| *** darjeeling has quit IRC | 06:34 | |
| *** Ritz-Laptop has quit IRC | 06:35 | |
| *** cp16net is now known as cp16net|away | 06:36 | |
| *** ilblackdragon has joined #openstack-dev | 06:36 | |
| *** koert has quit IRC | 06:39 | |
| *** jamielennox has quit IRC | 06:40 | |
| *** yaguang has quit IRC | 06:40 | |
| *** jamielennox has joined #openstack-dev | 06:41 | |
| *** ilblackdragon has quit IRC | 06:41 | |
| *** jasdeepH has quit IRC | 06:43 | |
| *** olaph has quit IRC | 06:47 | |
| *** stevebaker has quit IRC | 06:48 | |
| *** olaph has joined #openstack-dev | 06:48 | |
| *** garyk has joined #openstack-dev | 06:50 | |
| *** boris-42 has joined #openstack-dev | 06:51 | |
| *** jamielennox has quit IRC | 06:58 | |
| *** jamielennox has joined #openstack-dev | 06:59 | |
| *** kagan has quit IRC | 06:59 | |
| *** pnavarro has joined #openstack-dev | 07:00 | |
| *** xchu has joined #openstack-dev | 07:02 | |
| *** mrunge has joined #openstack-dev | 07:02 | |
| *** adjohn has joined #openstack-dev | 07:02 | |
| openstackgerrit | A change was merged to openstack/quantum: First havana commit. https://review.openstack.org/24257 | 07:03 |
| *** Mandell has quit IRC | 07:04 | |
| *** kbrierly has joined #openstack-dev | 07:05 | |
| *** yaguang has joined #openstack-dev | 07:06 | |
| *** davidhadas has joined #openstack-dev | 07:12 | |
| *** kpavel has joined #openstack-dev | 07:14 | |
| *** davidha has quit IRC | 07:14 | |
| *** matiu has quit IRC | 07:16 | |
| *** flaper87 has joined #openstack-dev | 07:17 | |
| *** jamielennox has quit IRC | 07:19 | |
| *** jamielennox has joined #openstack-dev | 07:20 | |
| *** renner_ has joined #openstack-dev | 07:20 | |
| *** rmohan has quit IRC | 07:21 | |
| *** rmohan has joined #openstack-dev | 07:21 | |
| *** jkyle has quit IRC | 07:21 | |
| *** renner has quit IRC | 07:22 | |
| *** boris-42 has quit IRC | 07:27 | |
| *** matiu has joined #openstack-dev | 07:28 | |
| *** koolhead17 has quit IRC | 07:33 | |
| *** alop_ has joined #openstack-dev | 07:34 | |
| *** zhidong has joined #openstack-dev | 07:35 | |
| *** darjeeling has joined #openstack-dev | 07:35 | |
| *** jkyle has joined #openstack-dev | 07:35 | |
| *** alop has quit IRC | 07:36 | |
| *** alop_ is now known as alop | 07:36 | |
| *** amerine has quit IRC | 07:36 | |
| *** jamielennox has quit IRC | 07:38 | |
| *** jamielennox has joined #openstack-dev | 07:38 | |
| *** pnavarro has quit IRC | 07:39 | |
| *** stevebaker has joined #openstack-dev | 07:40 | |
| *** SergeyLukjanov has joined #openstack-dev | 07:40 | |
| *** thimble has joined #openstack-dev | 07:42 | |
| *** boris-42 has joined #openstack-dev | 07:43 | |
| *** eglynn has quit IRC | 07:44 | |
| *** Ryan_Lane has quit IRC | 07:45 | |
| *** adjohn has quit IRC | 07:45 | |
| *** alop has quit IRC | 07:49 | |
| *** alop has joined #openstack-dev | 07:49 | |
| *** olaph_ has joined #openstack-dev | 07:50 | |
| *** olaph has quit IRC | 07:53 | |
| *** reidrac has joined #openstack-dev | 07:55 | |
| *** doude has joined #openstack-dev | 07:55 | |
| *** rafaduran has joined #openstack-dev | 07:57 | |
| *** dosaboy has quit IRC | 07:57 | |
| *** jprovazn has joined #openstack-dev | 07:58 | |
| *** jamielennox has quit IRC | 07:59 | |
| *** jamielennox has joined #openstack-dev | 08:00 | |
| *** megha has quit IRC | 08:02 | |
| *** security has joined #openstack-dev | 08:02 | |
| *** otherwiseguy has quit IRC | 08:05 | |
| *** matiu has quit IRC | 08:07 | |
| *** boden has joined #openstack-dev | 08:08 | |
| *** zoresvit has joined #openstack-dev | 08:11 | |
| *** mindpixel has joined #openstack-dev | 08:12 | |
| *** flaper87 has quit IRC | 08:13 | |
| *** flaper87 has joined #openstack-dev | 08:13 | |
| soren | alop: No problem at all. It has gotten the best of me a couple of times, too. If you have any ideas on how to make it more obvious, I'd *love* to hear them. | 08:17 |
| *** cp16net|away is now known as cp16net | 08:18 | |
| *** jamielennox has quit IRC | 08:18 | |
| *** jamielennox has joined #openstack-dev | 08:19 | |
| *** sride has joined #openstack-dev | 08:19 | |
| *** mmagr has joined #openstack-dev | 08:19 | |
| *** avishay has joined #openstack-dev | 08:21 | |
| *** matiu has joined #openstack-dev | 08:24 | |
| *** matiu has quit IRC | 08:24 | |
| *** matiu has joined #openstack-dev | 08:24 | |
| soren | jaypipes: Oh, one thing I noticed. In that pastebin of yours you set a PYTHON_PATH environment variable. The environment variable is called PYTHONPATH (without the underscore). Obviously not the issue at play here, but I wanted to point it out for future reference. http://docs.python.org/2.7/using/cmdline.html#envvar-PYTHONPATH | 08:25 |
| *** jamielennox has quit IRC | 08:26 | |
| *** danwent has quit IRC | 08:26 | |
| *** jgallard has joined #openstack-dev | 08:26 | |
| *** jamielennox has joined #openstack-dev | 08:27 | |
| *** zhidong has quit IRC | 08:29 | |
| *** andreaf has quit IRC | 08:31 | |
| *** andreaf has joined #openstack-dev | 08:31 | |
| *** henrynash has joined #openstack-dev | 08:32 | |
| *** almaisan-away is now known as al-maisan | 08:33 | |
| *** sirushti has quit IRC | 08:33 | |
| *** cp16net is now known as cp16net|away | 08:33 | |
| *** koolhead17 has joined #openstack-dev | 08:35 | |
| *** zoresvit has quit IRC | 08:35 | |
| *** dev_sa has joined #openstack-dev | 08:36 | |
| *** zb is now known as zaneb | 08:38 | |
| *** henrynash has quit IRC | 08:39 | |
| *** sulo_ has joined #openstack-dev | 08:42 | |
| *** jbr_ has joined #openstack-dev | 08:44 | |
| *** jamielennox has quit IRC | 08:45 | |
| *** sirushti has joined #openstack-dev | 08:46 | |
| *** jamielennox has joined #openstack-dev | 08:46 | |
| *** alop has quit IRC | 08:47 | |
| *** security has quit IRC | 08:49 | |
| *** navid_ has quit IRC | 08:50 | |
| *** zoresvit has joined #openstack-dev | 08:51 | |
| *** kashyap has quit IRC | 08:51 | |
| *** corXi has joined #openstack-dev | 08:52 | |
| *** pixelbeat has joined #openstack-dev | 08:52 | |
| koolhead17 | hi all | 08:54 |
| rushiagr | koolhead17: hi! | 08:54 |
| *** boris-42 has quit IRC | 08:54 | |
| *** ththande has joined #openstack-dev | 08:54 | |
| *** security has joined #openstack-dev | 08:54 | |
| koolhead17 | hi rushiagr | 08:54 |
| *** doude has quit IRC | 08:55 | |
| rushiagr | koolhead17: watsup? | 08:55 |
| *** stevebaker has quit IRC | 08:55 | |
| koolhead17 | rushiagr: nothing much. | 08:55 |
| *** adjohn has joined #openstack-dev | 08:56 | |
| *** kashyap has joined #openstack-dev | 08:56 | |
| *** ththande has quit IRC | 08:57 | |
| *** dosaboy has joined #openstack-dev | 08:57 | |
| *** jbr_ has left #openstack-dev | 09:00 | |
| *** matiu has quit IRC | 09:00 | |
| *** navid_ has joined #openstack-dev | 09:00 | |
| *** adjohn has quit IRC | 09:00 | |
| sandeepr | hey koolhead17 - looks all set for the sunday meetup? | 09:00 |
| *** sride has quit IRC | 09:00 | |
| *** stevebaker has joined #openstack-dev | 09:02 | |
| *** nimi has left #openstack-dev | 09:02 | |
| *** nimi has quit IRC | 09:02 | |
| koolhead17 | sandeepr: yes sir. | 09:03 |
| *** nimi has joined #openstack-dev | 09:03 | |
| *** afazekas has joined #openstack-dev | 09:03 | |
| *** kpavel_ has joined #openstack-dev | 09:04 | |
| *** jamielennox has quit IRC | 09:04 | |
| *** jamielennox has joined #openstack-dev | 09:05 | |
| *** eglynn has joined #openstack-dev | 09:05 | |
| koolhead17 | GheRivero: hi there | 09:05 |
| *** melwitt1 has quit IRC | 09:07 | |
| *** kpavel has quit IRC | 09:07 | |
| *** winston-d has quit IRC | 09:08 | |
| notmyname | ttx: I see you've sent an email, so I assume it's a decent hour in France. ping | 09:08 |
| ttx | notmyname: it is indeed | 09:08 |
| ttx | 10am. | 09:08 |
| *** trapni has joined #openstack-dev | 09:08 | |
| *** trapni has joined #openstack-dev | 09:08 | |
| notmyname | :-) | 09:08 |
| *** kpavel_ has quit IRC | 09:09 | |
| ttx | notmyname: do you want CONTRIBUTING.md to appear in source code tarballs ? | 09:10 |
| notmyname | ttx: this is fairly preliminary, but FYI there is a regression of sorts in swift that must be resolved before we can release. I'm hopeful that we can get something tomorrow and still be able to do the RC cut, but I won't do it if we dont' have the resolution by COB. when I get in tomorrow and figure out the patch forward, I'll update LP appropriately | 09:10 |
| ttx | notmyname: if yes, should probably be added to MANIFEST.in ... I can propose that for you | 09:11 |
| ttx | notmyname: thanks for the heads-up | 09:11 |
| notmyname | ttx: ya, that'd be great | 09:11 |
| ttx | will do. | 09:11 |
| *** epim has quit IRC | 09:11 | |
| notmyname | ttx: like I said, still preliminary, but in case something happens, this is an early head's up :-) | 09:11 |
| notmyname | and now for a few hours sleep until then... :-) | 09:13 |
| *** xga has joined #openstack-dev | 09:13 | |
| *** xga_ has joined #openstack-dev | 09:13 | |
| *** henrynash has joined #openstack-dev | 09:16 | |
| *** xchu has quit IRC | 09:16 | |
| *** jpich has joined #openstack-dev | 09:18 | |
| *** romcheg1 has joined #openstack-dev | 09:18 | |
| *** henrynash has quit IRC | 09:19 | |
| *** romcheg2 has joined #openstack-dev | 09:19 | |
| *** romcheg2 has left #openstack-dev | 09:19 | |
| *** johnthetubaguy has joined #openstack-dev | 09:22 | |
| *** romcheg1 has quit IRC | 09:22 | |
| *** alexxu has quit IRC | 09:22 | |
| *** kirminas has joined #openstack-dev | 09:25 | |
| *** jamielennox has quit IRC | 09:27 | |
| *** CaptTofu_ has joined #openstack-dev | 09:27 | |
| *** aabes__ has joined #openstack-dev | 09:27 | |
| *** jamielennox has joined #openstack-dev | 09:27 | |
| *** locke1051 has joined #openstack-dev | 09:28 | |
| *** xchu has joined #openstack-dev | 09:28 | |
| *** tomh__ has joined #openstack-dev | 09:29 | |
| *** Mathnerd314_ has joined #openstack-dev | 09:29 | |
| *** thickskin has joined #openstack-dev | 09:30 | |
| thickskin | when does grizzly-rc release ? | 09:31 |
| thickskin | https://launchpad.net/nova/+milestone/grizzly-rc1 | 09:31 |
| *** xarg_ has joined #openstack-dev | 09:31 | |
| *** tomoe_ has quit IRC | 09:31 | |
| *** Slidey_ has joined #openstack-dev | 09:31 | |
| *** mikal_ has joined #openstack-dev | 09:31 | |
| *** tomoe_ has joined #openstack-dev | 09:32 | |
| *** iccha has joined #openstack-dev | 09:32 | |
| *** darraghb has joined #openstack-dev | 09:33 | |
| *** hattwick has joined #openstack-dev | 09:35 | |
| *** djinni_ has joined #openstack-dev | 09:35 | |
| *** nimi has quit IRC | 09:36 | |
| *** sirushti has quit IRC | 09:36 | |
| *** rmohan has quit IRC | 09:36 | |
| *** mrunge has quit IRC | 09:36 | |
| *** CaptTofu has quit IRC | 09:36 | |
| *** aabes_ has quit IRC | 09:36 | |
| *** alunduil has quit IRC | 09:36 | |
| *** yamahata has quit IRC | 09:36 | |
| *** djinni has quit IRC | 09:36 | |
| *** tomh has quit IRC | 09:36 | |
| *** mikal has quit IRC | 09:36 | |
| *** Mathnerd314 has quit IRC | 09:36 | |
| *** Slidey has quit IRC | 09:36 | |
| *** arosen has quit IRC | 09:36 | |
| *** xarg has quit IRC | 09:36 | |
| *** iccha__ has quit IRC | 09:36 | |
| *** locke105 has quit IRC | 09:36 | |
| *** xarg_ is now known as xarg | 09:36 | |
| *** tomoe_ has quit IRC | 09:36 | |
| *** rmohan has joined #openstack-dev | 09:36 | |
| *** stevebaker has quit IRC | 09:37 | |
| *** techlife has quit IRC | 09:40 | |
| *** mrunge has joined #openstack-dev | 09:42 | |
| *** alunduil has joined #openstack-dev | 09:43 | |
| openstackgerrit | A change was merged to openstack/cinder: Fix bugs for Huawei driver https://review.openstack.org/23338 | 09:43 |
| *** thickskin has left #openstack-dev | 09:43 | |
| *** sirushti has joined #openstack-dev | 09:43 | |
| *** yamahata has joined #openstack-dev | 09:43 | |
| *** maxiz has joined #openstack-dev | 09:44 | |
| *** xga_ has quit IRC | 09:44 | |
| *** xga has quit IRC | 09:44 | |
| *** arosen has joined #openstack-dev | 09:44 | |
| *** nimi has joined #openstack-dev | 09:45 | |
| *** techlife has joined #openstack-dev | 09:45 | |
| *** zoresvit has quit IRC | 09:46 | |
| *** jamielennox has quit IRC | 09:46 | |
| *** jgallard has quit IRC | 09:47 | |
| *** jamielennox has joined #openstack-dev | 09:47 | |
| *** jgallard has joined #openstack-dev | 09:47 | |
| *** andreaf has quit IRC | 09:48 | |
| *** andreaf has joined #openstack-dev | 09:49 | |
| *** NobodyCam has quit IRC | 09:49 | |
| *** kpavel has joined #openstack-dev | 09:50 | |
| *** NobodyCam has joined #openstack-dev | 09:52 | |
| *** tserong_ has joined #openstack-dev | 09:56 | |
| *** tserong has quit IRC | 09:59 | |
| *** HenryG has joined #openstack-dev | 10:00 | |
| *** al-maisan is now known as almaisan-away | 10:00 | |
| *** nimi has quit IRC | 10:02 | |
| *** almaisan-away is now known as al-maisan | 10:03 | |
| *** zoresvit has joined #openstack-dev | 10:03 | |
| *** jamielennox has quit IRC | 10:06 | |
| *** jamielennox has joined #openstack-dev | 10:06 | |
| *** zoresvit has quit IRC | 10:10 | |
| *** vkmc has joined #openstack-dev | 10:10 | |
| kpavel | Hi, does anybody have a reference to how setup simple multinode environment with devstack? Latest instructions i was able to find on google relate to Folsom. | 10:12 |
| zykes- | ayoung: ping | 10:12 |
| koolhead17 | kpavel: is the wiki on devstack site not up2date? | 10:14 |
| kpavel | koolhead17: do you mean this one? http://devstack.org/guides/multinode-lab.html | 10:16 |
| koolhead17 | ya | 10:17 |
| *** glikson has joined #openstack-dev | 10:19 | |
| *** henrynash has joined #openstack-dev | 10:19 | |
| kpavel | koolhead17: didn't work for me. did it for you? | 10:20 |
| *** security has quit IRC | 10:20 | |
| *** baba has joined #openstack-dev | 10:21 | |
| *** baba is now known as Guest13870 | 10:21 | |
| koolhead17 | kpavel: feel free to file a bug. i have not tested it for grizzly yet | 10:21 |
| *** Guest13870 has quit IRC | 10:21 | |
| *** gugl has quit IRC | 10:22 | |
| *** mancdaz_ has quit IRC | 10:22 | |
| *** maxiz has quit IRC | 10:23 | |
| *** mancdaz has joined #openstack-dev | 10:23 | |
| *** Yada has joined #openstack-dev | 10:24 | |
| *** kmartin has quit IRC | 10:24 | |
| *** alexpilotti has joined #openstack-dev | 10:24 | |
| *** zoresvit has joined #openstack-dev | 10:25 | |
| *** jamielennox has quit IRC | 10:26 | |
| *** jamielennox has joined #openstack-dev | 10:26 | |
| kpavel | koolhead17: do you know by any chance one that includes quantum? | 10:27 |
| koolhead17 | kpavel: its supports Quantum for sure | 10:29 |
| *** pcm__ has joined #openstack-dev | 10:30 | |
| *** al-maisan is now known as almaisan-away | 10:31 | |
| *** fire has joined #openstack-dev | 10:32 | |
| koolhead17 | kpavel: http://networkstatic.net/openstack-folsom-quantum-devstack-installation-tutorial/ see if it helps :D | 10:33 |
| ttx | notmyname: see https://review.openstack.org/#/c/24300/ | 10:37 |
| *** danpb has joined #openstack-dev | 10:37 | |
| *** koert has joined #openstack-dev | 10:37 | |
| *** zoresvit has quit IRC | 10:38 | |
| *** davidhadas has quit IRC | 10:38 | |
| *** derekh has joined #openstack-dev | 10:38 | |
| *** davidha has joined #openstack-dev | 10:38 | |
| *** rushiagr has quit IRC | 10:42 | |
| *** egallen has joined #openstack-dev | 10:42 | |
| *** alobbs has quit IRC | 10:45 | |
| *** alobbs has joined #openstack-dev | 10:45 | |
| *** boris-42 has joined #openstack-dev | 10:47 | |
| *** nimi has joined #openstack-dev | 10:49 | |
| *** alobbs has quit IRC | 10:49 | |
| *** tomoe_ has joined #openstack-dev | 10:49 | |
| *** alobbs has joined #openstack-dev | 10:49 | |
| *** nimi has quit IRC | 10:49 | |
| *** kpavel_ has joined #openstack-dev | 10:53 | |
| *** jamielennox has quit IRC | 10:53 | |
| *** jamielennox has joined #openstack-dev | 10:54 | |
| *** tomoe_ has quit IRC | 10:55 | |
| *** kpavel has quit IRC | 10:55 | |
| *** kpavel_ is now known as kpavel | 10:55 | |
| *** tomoe_ has joined #openstack-dev | 10:56 | |
| *** zoresvit has joined #openstack-dev | 10:56 | |
| *** johnthetubaguy1 has joined #openstack-dev | 10:58 | |
| *** johnthetubaguy has quit IRC | 10:59 | |
| *** tomoe_ has quit IRC | 11:00 | |
| *** boris-42 has quit IRC | 11:01 | |
| *** boris-42 has joined #openstack-dev | 11:01 | |
| *** yamahata_ has joined #openstack-dev | 11:01 | |
| *** Yada has quit IRC | 11:02 | |
| *** zoresvit has quit IRC | 11:03 | |
| *** xchu has quit IRC | 11:05 | |
| *** techlife has quit IRC | 11:07 | |
| *** kirminas has quit IRC | 11:08 | |
| *** darjeeling has quit IRC | 11:11 | |
| *** alexisT has joined #openstack-dev | 11:11 | |
| *** alexisT has quit IRC | 11:11 | |
| *** alobbs has quit IRC | 11:14 | |
| *** alobbs has joined #openstack-dev | 11:14 | |
| *** techlife has joined #openstack-dev | 11:17 | |
| *** zoresvit has joined #openstack-dev | 11:18 | |
| *** jamielennox has quit IRC | 11:20 | |
| *** jamielennox has joined #openstack-dev | 11:21 | |
| *** kirminas has joined #openstack-dev | 11:23 | |
| *** mindpixel has quit IRC | 11:23 | |
| *** yamahata_ has quit IRC | 11:26 | |
| *** dims has joined #openstack-dev | 11:27 | |
| *** sandeepr has quit IRC | 11:31 | |
| *** tomoe_ has joined #openstack-dev | 11:32 | |
| *** tomoe_ has quit IRC | 11:33 | |
| *** tomoe_ has joined #openstack-dev | 11:33 | |
| *** yamahata_ has joined #openstack-dev | 11:35 | |
| *** glikson has quit IRC | 11:35 | |
| *** Thor has quit IRC | 11:35 | |
| *** egallen has quit IRC | 11:36 | |
| *** tomoe_ has quit IRC | 11:38 | |
| *** Thor has joined #openstack-dev | 11:38 | |
| *** jamielennox has quit IRC | 11:40 | |
| *** jamielennox has joined #openstack-dev | 11:41 | |
| *** johnthetubaguy1 has quit IRC | 11:41 | |
| *** johnthetubaguy has joined #openstack-dev | 11:47 | |
| openstackgerrit | A change was merged to openstack/oslo-incubator: Sanitize input before creating IPC socket. https://review.openstack.org/24229 | 11:48 |
| *** kbringard has joined #openstack-dev | 11:49 | |
| *** markmc has joined #openstack-dev | 11:49 | |
| *** yamahata_ has quit IRC | 11:50 | |
| trapni | how long will OpenStack support nova-network (aside from its new Quantum) ? | 11:53 |
| *** rushiagr has joined #openstack-dev | 11:54 | |
| *** psedlak has joined #openstack-dev | 11:54 | |
| *** psedlak is now known as psedlak|wfh | 11:54 | |
| *** psedlak|wfh is now known as psedlak | 11:54 | |
| *** halfss has quit IRC | 11:55 | |
| jpich | trapni: Until Quantum reaches feature parity, I believe | 11:55 |
| *** mkollaro has joined #openstack-dev | 11:57 | |
| openstackgerrit | A change was merged to openstack/glance: Switch to final 1.1.0 oslo.config release https://review.openstack.org/24201 | 11:58 |
| *** henrynash has quit IRC | 12:00 | |
| *** toanster_ has quit IRC | 12:03 | |
| *** brianr-gone has quit IRC | 12:03 | |
| *** toanster has joined #openstack-dev | 12:03 | |
| *** brianr-gone has joined #openstack-dev | 12:04 | |
| *** digi has quit IRC | 12:04 | |
| *** pmcg has quit IRC | 12:04 | |
| *** nikhil has quit IRC | 12:04 | |
| *** salv-orlando has joined #openstack-dev | 12:05 | |
| salv-orlando | garyk: are you around? | 12:05 |
| *** pmcg has joined #openstack-dev | 12:06 | |
| garyk | salv-orlando: hi | 12:06 |
| *** nikhil has joined #openstack-dev | 12:06 | |
| salv-orlando | garyk: sorry to bother you... | 12:06 |
| garyk | salv-orlando: np. | 12:06 |
| salv-orlando | I am trying to use quantum with connection pool | 12:06 |
| garyk | do you see exceptions when you restart the quantum service? | 12:07 |
| salv-orlando | yes :) I hoped you had a quick answer | 12:07 |
| salv-orlando | get_server_info attribute not found on connection tuple | 12:07 |
| garyk | it is a tricky one. i only encountered the problems when using ubuntu and rabbitmq. it did not happen with qpid | 12:08 |
| garyk | yong saw the same exception as you. i encountered another exception when i was working. | 12:08 |
| garyk | i think a temporary soln is to maybe increase the max pool size to 20. | 12:09 |
| garyk | i also managed to reproduce the problem here too, | 12:09 |
| salv-orlando | may I ask why? | 12:09 |
| garyk | to be honest i am still investigating | 12:09 |
| garyk | i think, and may be wrong here, when the quantum service is down the messages from the rpc are stored on the message broker. | 12:10 |
| *** alobbs has quit IRC | 12:10 | |
| garyk | when the service starts again. it gets a burst of messages. | 12:10 |
| salv-orlando | Ok right. | 12:10 |
| garyk | i tried limiting this but that did not work. | 12:11 |
| salv-orlando | That makes sense. But I have an even more basic problem :/ Quantum fails at startup when I enable pooling | 12:11 |
| garyk | i was thinking that maybe we should purge the queue before starting the service. | 12:11 |
| *** alobbs has joined #openstack-dev | 12:11 | |
| salv-orlando | I've seen that issue too, but it went away when Yong changed the 'cast' for state report into a 'call' | 12:11 |
| *** nimi has joined #openstack-dev | 12:11 | |
| garyk | i have just spun up a setup. give me a sec and i'll paste the exception that i have to you. | 12:12 |
| garyk | which services are running in your env? | 12:12 |
| salv-orlando | garyk: 1$ key 2$ g-reg 3$ g-api 4$ n-api 5$ q-svc* 6$ q-agt 7$ q-dhcp 8$ q-l3 9$ n-cond 10$ n-cpu 11$ n-crt 12$ n-sch | 12:13 |
| garyk | is the agt, dhcp and l3 running? | 12:14 |
| garyk | salv-orlando: this is the exception that i get when i restart the q-service - http://paste.openstack.org/show/33549/ | 12:15 |
| openstackgerrit | A change was merged to openstack-dev/devstack: Don't directly pip install tempest pip-requires https://review.openstack.org/24291 | 12:15 |
| salv-orlando | that's the one which is driving me crazy too | 12:16 |
| salv-orlando | but I have it even without db pooling | 12:16 |
| salv-orlando | I thought I might solve it activating pooling, but it seems you have it with pooling too :) | 12:16 |
| garyk | with me this happens only with the pooling. | 12:17 |
| *** aloga has quit IRC | 12:17 | |
| *** aloga has joined #openstack-dev | 12:18 | |
| garyk | reading up on it i have found the following - http://stackoverflow.com/questions/7957436/error-exception-must-derive-from-baseexception-even-when-it-does-python-2-7 | 12:18 |
| salv-orlando | garyk: that's interesting. Because without pooling, in my NVP env, it happens only when we attach the 'metadata access' network which is an operation which creates quite a lot of nested session transactions | 12:19 |
| salv-orlando | and the effect is the same of bug 1086173 | 12:19 |
| uvirtbot | Launchpad bug 1086173 in quantum "Lock wait of mysql blocks all eventlet thread" [High,Fix released] https://launchpad.net/bugs/1086173 | 12:19 |
| salv-orlando | all the event let threads are blocked (including the ones the nicira plugin uses to communicate with NVP) | 12:19 |
| garyk | the default timeout is 3600. | 12:20 |
| *** martine_ has joined #openstack-dev | 12:20 | |
| *** henrynash has joined #openstack-dev | 12:21 | |
| garyk | any idea how i can reproduce the problem with the " get_server_info attribute not found on connection tuple". maybe tackling it from a different angle can help | 12:22 |
| *** varto has joined #openstack-dev | 12:22 | |
| salv-orlando | garyk: I can give you the traceback, but that looks like a configuration problem to me | 12:24 |
| garyk | i'll be happy to take a look. | 12:24 |
| *** mikal_ is now known as mikal | 12:26 | |
| enikanorov_ | http://stackoverflow.com/questions/9525220/sqlalchemy-raises-none-causes-typeerror | 12:26 |
| *** dprince has joined #openstack-dev | 12:26 | |
| *** jamielennox has quit IRC | 12:27 | |
| *** aabes__ has left #openstack-dev | 12:27 | |
| garyk | salv-orlando: ^^^^ enikanorov's link is also good | 12:27 |
| *** jamielennox has joined #openstack-dev | 12:28 | |
| salv-orlando | garyk: this is my tb: http://paste.openstack.org/show/33556/ | 12:28 |
| *** navid_ has quit IRC | 12:29 | |
| *** navid_ has joined #openstack-dev | 12:29 | |
| *** [1]glikson has joined #openstack-dev | 12:30 | |
| *** alobbs has quit IRC | 12:31 | |
| garyk | salv-orlando: question - what is the difference between get_admin_context and get_admin_context_without_session. When I use the former to get the context the problem does not occur. This could be the source of the problem | 12:32 |
| *** alobbs has joined #openstack-dev | 12:32 | |
| *** yaguang has quit IRC | 12:32 | |
| salv-orlando | the latter is a function which creates a context without a db session, which is meant to used by agent which should not fetch db session | 12:32 |
| nimi | is/has some one currently working with SR-IOV ? | 12:33 |
| salv-orlando | so it solved your problem by doing a change on the dhcp agent side? | 12:33 |
| *** CaptTofu_ has quit IRC | 12:33 | |
| garyk | salv-orlando: http://paste.openstack.org/show/33557/ | 12:33 |
| *** CaptTofu has joined #openstack-dev | 12:33 | |
| *** alexxu has joined #openstack-dev | 12:33 | |
| garyk | yes | 12:33 |
| garyk | it was on the openvswitch agent. | 12:34 |
| *** tomoe_ has joined #openstack-dev | 12:35 | |
| *** jruzicka has joined #openstack-dev | 12:35 | |
| salv-orlando | interesting. I am looking now at the rpc dispatcher on the server side to see what it makes of the session | 12:36 |
| *** johnthetubaguy1 has joined #openstack-dev | 12:37 | |
| *** henrynash has quit IRC | 12:37 | |
| *** rkukura has quit IRC | 12:37 | |
| salv-orlando | I wonder if the facts that picks a new session at each RPC call, and never release it | 12:38 |
| *** rkukura has joined #openstack-dev | 12:38 | |
| salv-orlando | might cause the pool to exhaust because the timeout is 3600 | 12:38 |
| *** tomoe_ has quit IRC | 12:38 | |
| garyk | could be. i need to look at the code to understand it better | 12:38 |
| *** tomoe_ has joined #openstack-dev | 12:39 | |
| *** READ10 has joined #openstack-dev | 12:39 | |
| *** johnthetubaguy has quit IRC | 12:39 | |
| garyk | i would expect it to have a db session. no? | 12:39 |
| *** yamahata_ has joined #openstack-dev | 12:42 | |
| *** susanne-balle has joined #openstack-dev | 12:43 | |
| *** tomoe_ has quit IRC | 12:43 | |
| ttx | markmc: your rc1 buglist is now empty -- let me know if/when you want to proceed | 12:44 |
| salv-orlando | garyk: that solved the problem for me too. and I am running without db pooling now. | 12:44 |
| garyk | ok, cool. | 12:45 |
| garyk | i can push a patch but would like to get some input from yong. | 12:45 |
| openstackgerrit | A change was merged to openstack/quantum: Switch to final 1.1.0 oslo.config release https://review.openstack.org/24204 | 12:45 |
| salv-orlando | I'm trying to understand why picking a session on the agent side avoids the issue. | 12:46 |
| salv-orlando | garyk: don't you find it a little strange? | 12:46 |
| *** tomoe_ has joined #openstack-dev | 12:47 | |
| garyk | salv-orlando: yes, it is very strange | 12:47 |
| garyk | i would expect that when the code is invoked on the server side a session would be created. | 12:48 |
| *** jamielennox has quit IRC | 12:48 | |
| *** jamielennox has joined #openstack-dev | 12:49 | |
| salv-orlando | actually get_admin_context does not even create a session. It just creates a context with a session method. Hoe | 12:51 |
| salv-orlando | However, that context object is not sent over the wire via RPC! | 12:51 |
| salv-orlando | The actual context which will be used by the server is created in the dispatcher | 12:52 |
| markmc | ttx, yep, fire ahead! thanks :) | 12:55 |
| garyk | interesting. i will look how the dispatcher is defined | 12:55 |
| *** avishay has quit IRC | 12:55 | |
| ttx | markmc: so... will cut stable/grizzly from HEAD (44d2a4c02f0) | 12:55 |
| ttx | in ( min | 12:56 |
| ttx | 5 | 12:56 |
| markmc | 4 | 12:56 |
| markmc | 3 | 12:56 |
| markmc | 2 | 12:56 |
| markmc | 1 | 12:56 |
| markmc | sounds good :) | 12:56 |
| *** yamahata_ has quit IRC | 12:56 | |
| *** alobbs has quit IRC | 12:56 | |
| markmc | we need to bump the version number in setup.py too I guess | 12:56 |
| markmc | not that it means anything for oslo-incubator | 12:56 |
| openstackgerrit | A change was merged to openstack/nova: Fix nova-manage --version https://review.openstack.org/24173 | 12:57 |
| *** alunduil has quit IRC | 12:57 | |
| *** alobbs has joined #openstack-dev | 12:57 | |
| markmc | ttx, who should propose this for quantum stable/grizzly? https://review.openstack.org/24204 | 12:58 |
| markmc | (I can do it) | 12:58 |
| ttx | markmc: that would be for milestone-proposed | 12:58 |
| markmc | ttx, sorry, yes - my hands didn't type what my brain told them to :) | 12:58 |
| *** johnthetubaguy1 has quit IRC | 12:58 | |
| *** SergeyLukjanov has quit IRC | 12:58 | |
| ttx | markmc: if you do it i can approve it. | 12:58 |
| markmc | grand so | 12:59 |
| markmc | ttx, https://review.openstack.org/24313 | 13:00 |
| *** CaptTofu has quit IRC | 13:01 | |
| *** glikson has joined #openstack-dev | 13:01 | |
| *** CaptTofu has joined #openstack-dev | 13:02 | |
| *** sandywalsh has quit IRC | 13:02 | |
| *** rushiagr has quit IRC | 13:03 | |
| *** cp16net|away is now known as cp16net | 13:03 | |
| openstackgerrit | A change was merged to openstack/nova: Switch to final 1.1.0 oslo.config release https://review.openstack.org/24200 | 13:03 |
| *** johnthetubaguy has joined #openstack-dev | 13:04 | |
| *** [1]glikson has quit IRC | 13:04 | |
| *** CaptTofu has quit IRC | 13:05 | |
| *** CaptTofu has joined #openstack-dev | 13:05 | |
| *** darjeeling has joined #openstack-dev | 13:05 | |
| *** ewindisch has joined #openstack-dev | 13:06 | |
| *** joesavak has joined #openstack-dev | 13:06 | |
| *** rushiagr1 has joined #openstack-dev | 13:07 | |
| *** bknudson has joined #openstack-dev | 13:08 | |
| *** yamahata_ has joined #openstack-dev | 13:09 | |
| garyk | salv-orlando: in the PluginRpcDispatcher the user, tenant etc is used to create the context. maybe these are invalid in come cases | 13:09 |
| *** rushiagr has joined #openstack-dev | 13:10 | |
| *** rushiagr1 has quit IRC | 13:10 | |
| *** timello has joined #openstack-dev | 13:10 | |
| salv-orlando | garyk: I've been logging the context dict. I did not see anything weird there. | 13:11 |
| garyk | ok | 13:11 |
| *** imsplitbit has joined #openstack-dev | 13:12 | |
| *** ladquin has joined #openstack-dev | 13:12 | |
| *** egallen has joined #openstack-dev | 13:13 | |
| *** anteaya has joined #openstack-dev | 13:14 | |
| ttx | markmc: done @ https://launchpad.net/oslo/+milestone/grizzly-rc1 | 13:15 |
| ttx | markmc: now you can use grizzly-rc-potential for potential backports, until you collect enough to grant a rc2 | 13:15 |
| markmc | ttx, cool | 13:15 |
| *** sandywalsh has joined #openstack-dev | 13:15 | |
| ttx | markmc: are you going to propose the oslo-incubator sync to quantum as well ? Or is it markmcclain's territory ? | 13:17 |
| *** johnthetubaguy1 has joined #openstack-dev | 13:17 | |
| *** johnthetubaguy has quit IRC | 13:19 | |
| *** rnirmal has joined #openstack-dev | 13:20 | |
| *** aditirav has joined #openstack-dev | 13:20 | |
| *** mrunge has quit IRC | 13:20 | |
| *** adalbas has joined #openstack-dev | 13:21 | |
| *** pfreund has joined #openstack-dev | 13:22 | |
| *** gongysh has quit IRC | 13:22 | |
| *** nunosantos has joined #openstack-dev | 13:23 | |
| *** rushiagr has quit IRC | 13:24 | |
| *** sride has joined #openstack-dev | 13:27 | |
| *** rushiagr1 has joined #openstack-dev | 13:27 | |
| *** rushiagr1 has quit IRC | 13:28 | |
| *** crandquist has joined #openstack-dev | 13:28 | |
| *** rushiagr1 has joined #openstack-dev | 13:29 | |
| *** jhesketh_ has quit IRC | 13:31 | |
| *** radez_g0n3 is now known as radez | 13:31 | |
| markmc | ttx, leaving that to each project individually | 13:33 |
| markmc | ttx, I might look, but can't commit to do them all | 13:33 |
| *** adalbas has quit IRC | 13:33 | |
| markmc | ttx, it's not something you do without looking closely | 13:33 |
| *** kirminas has quit IRC | 13:34 | |
| *** pfreund has quit IRC | 13:34 | |
| *** henrynash has joined #openstack-dev | 13:35 | |
| ttx | markmc: shoudl I push a tag for oslo-incubator ? 2013.1.rc1 ? | 13:35 |
| *** salv-orlando_ has joined #openstack-dev | 13:36 | |
| *** salv-orlando has quit IRC | 13:36 | |
| *** salv-orlando_ is now known as salv-orlando | 13:36 | |
| *** stevemar has joined #openstack-dev | 13:37 | |
| *** rushiagr1 has quit IRC | 13:37 | |
| markmc | ttx, sounds good | 13:37 |
| *** stevemar has quit IRC | 13:37 | |
| *** doude has joined #openstack-dev | 13:38 | |
| *** zul has joined #openstack-dev | 13:40 | |
| *** rnirmal_ has joined #openstack-dev | 13:41 | |
| ttx | markmc: done | 13:42 |
| markmc | ttx, thanks | 13:42 |
| *** digitalsanctum has joined #openstack-dev | 13:42 | |
| *** sandeepr has joined #openstack-dev | 13:44 | |
| *** rnirmal has quit IRC | 13:45 | |
| *** rnirmal_ is now known as rnirmal | 13:45 | |
| *** aeperezt has joined #openstack-dev | 13:45 | |
| *** zul has quit IRC | 13:45 | |
| *** adalbas has joined #openstack-dev | 13:45 | |
| *** rcj has joined #openstack-dev | 13:50 | |
| *** eharney has joined #openstack-dev | 13:50 | |
| *** eharney is now known as Guest28157 | 13:51 | |
| *** cp16net is now known as cp16net|away | 13:51 | |
| *** kaushikc has joined #openstack-dev | 13:52 | |
| *** mtreinish has joined #openstack-dev | 13:52 | |
| *** Gordonz has joined #openstack-dev | 13:52 | |
| *** Guest28157 is now known as eharney | 13:53 | |
| *** eharney has joined #openstack-dev | 13:53 | |
| *** cppcabrera has joined #openstack-dev | 13:55 | |
| *** cppcabrera has left #openstack-dev | 13:55 | |
| *** rcj has quit IRC | 13:56 | |
| *** kaushikc has quit IRC | 13:57 | |
| *** jimfehlig has joined #openstack-dev | 13:57 | |
| *** rcj has joined #openstack-dev | 13:58 | |
| *** Gordonz has quit IRC | 13:59 | |
| *** Gordonz has joined #openstack-dev | 13:59 | |
| *** sacharya has joined #openstack-dev | 13:59 | |
| *** digi has joined #openstack-dev | 14:01 | |
| *** johnthetubaguy1 has quit IRC | 14:01 | |
| *** [1]glikson has joined #openstack-dev | 14:01 | |
| *** gongysh has joined #openstack-dev | 14:01 | |
| *** adalbas has quit IRC | 14:01 | |
| *** cp16net|away is now known as cp16net | 14:02 | |
| *** johnthetubaguy has joined #openstack-dev | 14:02 | |
| *** adalbas has joined #openstack-dev | 14:02 | |
| *** johnthetubaguy has quit IRC | 14:02 | |
| *** kaushikc has joined #openstack-dev | 14:03 | |
| *** cloudchimp has joined #openstack-dev | 14:04 | |
| *** glikson has quit IRC | 14:04 | |
| *** [1]glikson is now known as glikson | 14:04 | |
| *** sacharya has quit IRC | 14:05 | |
| *** cloudchimp has quit IRC | 14:05 | |
| *** sride has quit IRC | 14:06 | |
| *** cloudchimp has joined #openstack-dev | 14:06 | |
| *** sride has joined #openstack-dev | 14:08 | |
| ayoung | henrynash, I just reassigend https://bugs.launchpad.net/python-keystoneclient/+bug/1154144 to you, since you are the keystone dev dealing with it. | 14:08 |
| uvirtbot | Launchpad bug 1154144 in python-keystoneclient "auth_token middleware should default to v2.0 if version is not specified" [Medium,In progress] | 14:08 |
| henrynash | ok, thx | 14:09 |
| ayoung | henrynash, I'm still not sure what part of the auth_token middleware authenticate path is failing | 14:09 |
| ayoung | but do the following | 14:09 |
| ayoung | run devstack, and change the auth_urls to top levle insteado f v2.0. | 14:09 |
| ayoung | That will actually trigger the V3 code path | 14:09 |
| ayoung | and you should find out pretty quickly whether they work or not. | 14:09 |
| ayoung | My guess is you will find something broken in fetching the certs or the revocation list. | 14:10 |
| *** utlemming has quit IRC | 14:10 | |
| henrynash | ayoung: …and, just for clarity, where in devstack are you suggesting I change the auth_urls? | 14:11 |
| ayoung | henrynash, once you figure out what is failing, can you update the bug title? | 14:11 |
| *** utlemming has joined #openstack-dev | 14:11 | |
| *** mrodden has joined #openstack-dev | 14:11 | |
| henrynash | ayoung: (yes) | 14:11 |
| *** utlemming has quit IRC | 14:11 | |
| ayoung | henrynash, ah...do a grep through devstack and you will see them all over the place.... | 14:11 |
| henrynash | ayoung: ok | 14:11 |
| ayoung | It needs to be in the past config for the servers | 14:11 |
| ayoung | so for exaple.... | 14:11 |
| *** SergeyLukjanov has joined #openstack-dev | 14:12 | |
| ayoung | henrynash, http://paste.openstack.org/show/33565/ | 14:12 |
| ayoung | it will likely be the fields like | 14:13 |
| ayoung | files/horizon_settings.py:71:# ('http://cluster1.example.com:5000/v2.0', 'cluster1'), | 14:13 |
| ayoung | iniset | 14:13 |
| ayoung | and so forth | 14:13 |
| henrynash | ayoung: ok | 14:13 |
| *** kgriffs has joined #openstack-dev | 14:13 | |
| *** crandquist has quit IRC | 14:13 | |
| ayoung | henrynash, is the CLI V3 ready? My understanding is that it is not 100% yet. Am I wrong? | 14:14 |
| *** thimble has quit IRC | 14:14 | |
| henrynash | ayoung: so the classes are Ok, I think, in kqystoneclient | 14:14 |
| henrynash | ayoung: and a lot has been added in opensatckclient for the cli, but I'm not sure it is complete | 14:15 |
| ayoung | henrynash, ok, my guess is that will be work we need to knock out early in Havana. | 14:15 |
| henrynash | ayoung: the direction was NOT to just add v3 cli to ksyetoneclient | 14:15 |
| *** otherwiseguy has joined #openstack-dev | 14:15 | |
| henrynash | ayoung: agreed…and the openstackclient has its own schedule, right? | 14:15 |
| ayoung | Ah..right. Unified client. Still, we probably need to either modernize or deprecated the Keystone specific CLI. My guess is we'll have to keep it around for at least one more release. | 14:16 |
| henrynash | ayoung: yep….altough do we add v3 to it? Dolph was pretty adamant that we should not | 14:17 |
| ayoung | but first things first. If auto negotiation of the level is broken in the auth-token middleware, it might be broken in the CLI, too. Can you check> | 14:17 |
| henrynash | ayoung: mm, good idea. will do | 14:17 |
| ayoung | henrynash, lets just say we need to plan out the route, and leave it at that. If we deprecate the client with V2 functionality, we need to add V3 admin functionality to the common client ASAP | 14:18 |
| *** ijw has joined #openstack-dev | 14:18 | |
| henrynash | ayoung: agreed | 14:18 |
| *** dolphm has joined #openstack-dev | 14:18 | |
| *** ijw has quit IRC | 14:18 | |
| *** ijw has joined #openstack-dev | 14:19 | |
| *** kaushikc1 has joined #openstack-dev | 14:20 | |
| *** kaushikc has quit IRC | 14:20 | |
| henrynash | ayoung: btw, back to the auth_urls.... | 14:20 |
| henrynash | ayoung: if we are defaulting any version less url to v3.0 this isn't going to work…since we don\t support an exact superset of v2.0 in v3.0 (example - tokens) | 14:21 |
| henrynash | ayoung: i.e. /v3/tokens is not a valid url, but /v2/tokens is | 14:22 |
| trapni | Hey. I've just ran into a very serious issue. None of my VMs can talk to the DHCP server (dnsmasq on nova-gateway) anymore. I was just about to upgrade from Essex to Folsom (on ubuntu), however, not migrating from nova-network to quantum yet. Now the first VM went down because it couldn't acquire a new lease. I quickly statically assigned the expected IP to it and everything works. but I expect lots of more VMs to die within the next hours | 14:23 |
| trapni | because of that. Restarting nova (first killing all dnsmasq procs) didn't fix it though. | 14:23 |
| trapni | I did tcpdump on the gateway and saw that the actual DHCP requests are arriving. but dnsmasq doesn't seem to respond | 14:25 |
| *** andrewbogott_afk is now known as andrewbogott | 14:26 | |
| *** varto has quit IRC | 14:27 | |
| garyk | salv-orlando: ping | 14:27 |
| *** kaushikc1 has quit IRC | 14:28 | |
| *** kaushikc has joined #openstack-dev | 14:28 | |
| *** rpedde_away is now known as rpedde | 14:28 | |
| *** jsavak has joined #openstack-dev | 14:29 | |
| *** yuanz has joined #openstack-dev | 14:29 | |
| salv-orlando | garyk: hey | 14:29 |
| ayoung | henrynash, I wonder if we can add in /v3/tokens as a router entry without changing the controller. If so, it might be less surprising. But we need to be able to fall back to V2.0 for older keystones regardless. | 14:29 |
| ayoung | and we need to be able to deal with V3.0 tokens, too. | 14:29 |
| henrynash | ayoung: trying to think if there are any other v2.0 urls that aren't supported in v3.0 | 14:30 |
| garyk | salv-orlando: i am still investigating the issue and have seen that the _get_agent_by_type_and_host has a database request. this returns MultipleResultsFound. This is fishy | 14:30 |
| ayoung | henrynash, lets automate that test.... | 14:30 |
| *** alunduil has joined #openstack-dev | 14:31 | |
| henrynash | ayoung: using versionless urls? | 14:31 |
| *** thimble has joined #openstack-dev | 14:31 | |
| ayoung | henrynash, yeah. we should look at the keystoneclient tests we have now and figure out an analogue for V3...with a fallback to v2 for each of those calls. | 14:31 |
| garyk | salv-orlando: there is no key. i am wondering if this has something to do with the exceptions that we have seen. | 14:31 |
| *** kpavel has quit IRC | 14:32 | |
| *** joesavak has quit IRC | 14:32 | |
| henrynash | ayoung:ok | 14:32 |
| *** yamahata_ has quit IRC | 14:32 | |
| ayoung | henrynash, this is ongoing work, and not necessarily something we will be able to get in to the Grizzly release. We just need it mapped out. | 14:33 |
| *** rnirmal has quit IRC | 14:33 | |
| ayoung | and we need to make sure we don;t ship a broken CLI | 14:33 |
| salv-orlando | garyk: it might be. Another way I have been enable to remove the error is I remove the update operation in create_or_update | 14:33 |
| *** woodspa has joined #openstack-dev | 14:33 | |
| salv-orlando | but that still does not fully explain the behaviour (at least I don't have a clear picture) | 14:34 |
| salv-orlando | but do you also see a timeout, like the agents keep waiting for a reply, or just the exception? | 14:34 |
| *** yamahata_ has joined #openstack-dev | 14:35 | |
| *** rnirmal has joined #openstack-dev | 14:36 | |
| fubada | hi folks, i have a fresh install using stack.sh, it worked until i rebooted, now I cant login to the dashboard | 14:36 |
| fubada | authentication errors. can someone take a look at my apache error logs: http://pastie.org/private/hqzrqvptakpsvceeqaopaw | 14:36 |
| salv-orlando | garyk: I wish I could enable the same setup as your with the connection pooling, but all my envs (ubuntu based) are refusing too | 14:36 |
| fubada | openstack_auth.backend:Authorization Failed: [Errno 111] Connection refused is what I need help with | 14:36 |
| garyk | salv-orlando: :) ok. i'll continue to debug and hopefully stumble on something soon | 14:37 |
| *** jgallard has quit IRC | 14:37 | |
| *** jgallard has joined #openstack-dev | 14:38 | |
| *** kaushikc1 has joined #openstack-dev | 14:38 | |
| *** kaushikc has quit IRC | 14:38 | |
| *** sacharya has joined #openstack-dev | 14:38 | |
| sandeepr | jpich, had a chance to review the debug log https://bugs.launchpad.net/horizon/+bug/1143897 | 14:42 |
| uvirtbot | Launchpad bug 1143897 in horizon "Login to Horizon dashboard fails with "An unexpected error has occurred. Try refreshing the page."" [Undecided,Incomplete] | 14:42 |
| *** utlemming has joined #openstack-dev | 14:42 | |
| *** markwash has quit IRC | 14:43 | |
| *** cp16net is now known as cp16net|away | 14:43 | |
| *** markmcclain has joined #openstack-dev | 14:44 | |
| *** thimble has quit IRC | 14:44 | |
| *** navid_ has quit IRC | 14:44 | |
| *** cp16net|away is now known as cp16net | 14:45 | |
| *** alobbs has quit IRC | 14:45 | |
| sandeepr | anyone can comment why this is moved invalid - https://bugs.launchpad.net/nova/+bug/1153698 | 14:46 |
| *** alobbs has joined #openstack-dev | 14:46 | |
| uvirtbot | Launchpad bug 1153698 in nova "nova service crash with error "Table 'nova.services' doesn't exist"" [Undecided,Invalid] | 14:46 |
| jpich | sandeepr: I'd wait for someone more familiar with the ubuntu packages and/or offline compression to chime in. You could try deactivating offline compression, it'll be ugly but may help determine if the compression is hiding the real error or is the cause of the error itself | 14:46 |
| *** winston-d has joined #openstack-dev | 14:46 | |
| *** jgallard has quit IRC | 14:47 | |
| sandeepr | jpich, would you let me know please how to deactivate offline compression? | 14:47 |
| *** rnirmal has quit IRC | 14:47 | |
| *** jgallard has joined #openstack-dev | 14:47 | |
| *** garyTh has joined #openstack-dev | 14:47 | |
| sandeepr | and is this a probable bug or configuration issue? | 14:47 |
| *** rnirmal has joined #openstack-dev | 14:47 | |
| jpich | sandeepr: I think there's an option in the local_settings somewhere that sets it to True | 14:47 |
| jpich | sandeepr: in my opinion, it's either related to the configuration or packaging | 14:48 |
| *** dolphm has quit IRC | 14:49 | |
| *** salv-orlando has quit IRC | 14:50 | |
| *** salv-orlando_ has joined #openstack-dev | 14:50 | |
| *** dprince has quit IRC | 14:50 | |
| *** dolphm has joined #openstack-dev | 14:51 | |
| *** bourke has quit IRC | 14:51 | |
| crazed | ayoung: if i can prove that my patch for dereferencing returns the same output with the ldap live tests (same failing/error ones), and add unit tests is that good enough? i'd like to fix all the live tests | 14:52 |
| crazed | ayoung: but am not sure how to proceed since i don't know if the underlying way things work changed or if they are legit failing | 14:52 |
| sandeepr | jpich, i found your comment on https://bugs.launchpad.net/horizon/+bug/1084119 | 14:52 |
| uvirtbot | Launchpad bug 1084119 in horizon "After upgrade cannot manage projects" [Undecided,Incomplete] | 14:52 |
| *** winston-d has quit IRC | 14:52 | |
| ayoung | crazed, let me turn the tables on you. Say you were the tech lead for this, and one of your developers asked you that. What would you say? Do *you* think it is good enough? | 14:52 |
| *** avishay has joined #openstack-dev | 14:53 | |
| *** egallen has quit IRC | 14:53 | |
| *** bourke has joined #openstack-dev | 14:53 | |
| *** edmund has joined #openstack-dev | 14:54 | |
| *** datsun180b has joined #openstack-dev | 14:54 | |
| jpich | sandeepr: Yeah, that was for fedora/red hat packages though. Usually with those, in that case you see the real error when turning debug to True but according to your latest comments, that didn't help with your problem unfortunately | 14:54 |
| crazed | ayoung: well that's true, it's not good enough.. | 14:54 |
| ayoung | heh. | 14:54 |
| crazed | just a lot of failing tests that seem to be failing with/without the patch | 14:55 |
| *** yolanda has quit IRC | 14:55 | |
| *** anteaya_ has joined #openstack-dev | 14:56 | |
| *** jsavak has quit IRC | 14:56 | |
| *** egallen has joined #openstack-dev | 14:56 | |
| crazed | hopefully i can find some time to fix them if thats what it takes | 14:56 |
| salv-orlando_ | garyk: I managed to enable the db pool at least on my setup. The cause was eventlet and I was being hit by a bug which has already been fixed for nova | 14:56 |
| salv-orlando_ | but 1087054 | 14:57 |
| salv-orlando_ | gut 1087054 | 14:57 |
| *** kmartin has joined #openstack-dev | 14:57 | |
| salv-orlando_ | bug 1087054 | 14:57 |
| uvirtbot | Launchpad bug 1087054 in openstack-manuals "grizzly: Add eventlet db_pool use for mysql" [Medium,Confirmed] https://launchpad.net/bugs/1087054 | 14:57 |
| *** joesavak has joined #openstack-dev | 14:57 | |
| *** egallen has quit IRC | 14:57 | |
| *** egallen has joined #openstack-dev | 14:57 | |
| *** john5223 has joined #openstack-dev | 14:57 | |
| *** egallen has quit IRC | 14:57 | |
| *** anteaya has quit IRC | 14:58 | |
| *** egallen has joined #openstack-dev | 14:58 | |
| garyk | salv-orlando_: please see https://review.openstack.org/#/c/22162/ (i think that yong also tried this and it solved the problems) | 14:59 |
| *** aeperezt has quit IRC | 14:59 | |
| *** kpavel has joined #openstack-dev | 15:00 | |
| *** networkstatic has joined #openstack-dev | 15:00 | |
| salv-orlando_ | uhm.. that fix is for a deadlock problem, is it? | 15:00 |
| *** kaushikc1 has quit IRC | 15:01 | |
| *** kaushikc has joined #openstack-dev | 15:01 | |
| *** yolanda has joined #openstack-dev | 15:01 | |
| *** mtreinish has quit IRC | 15:01 | |
| *** xga has joined #openstack-dev | 15:02 | |
| *** aeperezt has joined #openstack-dev | 15:03 | |
| *** monst_ has joined #openstack-dev | 15:03 | |
| *** ProfFalken has joined #openstack-dev | 15:04 | |
| *** thimble has joined #openstack-dev | 15:04 | |
| ProfFalken | hi all, I just got bitten by the fix for https://bugs.launchpad.net/nova/+bug/1151012 as I have already upgraded from Folsem to Grizzly. If I update the services table to use the hostname instead of the FQDN, will I also need to update the instances table? | 15:04 |
| uvirtbot | Launchpad bug 1151012 in nova "grizzly uses FQDN for services" [Critical,Fix committed] | 15:04 |
| ProfFalken | FWIW, I checked the DB Schema at https://wiki.openstack.org/wiki/Nova_Database_Schema but I can't see a link between the two | 15:04 |
| *** mtreinish has joined #openstack-dev | 15:05 | |
| *** xga has quit IRC | 15:07 | |
| *** koolhead17 has quit IRC | 15:08 | |
| *** terryh has joined #openstack-dev | 15:10 | |
| *** salgado is now known as salgado-lunch | 15:11 | |
| *** sthaha has quit IRC | 15:12 | |
| *** egallen has quit IRC | 15:12 | |
| ProfFalken | heh, looks like they're not related and that nova.instances uses the hostname not the FQDN anyway... | 15:13 |
| *** doude has quit IRC | 15:13 | |
| *** colinmcnamara has joined #openstack-dev | 15:13 | |
| *** bdpayne has joined #openstack-dev | 15:13 | |
| *** mtreinish has quit IRC | 15:13 | |
| *** Oneiroi has quit IRC | 15:13 | |
| *** mtreinish has joined #openstack-dev | 15:15 | |
| *** aeperezt has quit IRC | 15:16 | |
| *** jasdeepH has joined #openstack-dev | 15:16 | |
| ProfFalken | scratch that... I was on the wrong environment | 15:17 |
| trapni | can someone please send me the ps output of dnsmasq if he's running in VlanManager mode ? | 15:18 |
| *** winston-d has joined #openstack-dev | 15:18 | |
| *** cloudchimp has quit IRC | 15:19 | |
| *** aeperezt has joined #openstack-dev | 15:20 | |
| *** jgallard has quit IRC | 15:20 | |
| *** jgallard has joined #openstack-dev | 15:20 | |
| jaypipes | trapni: multihost + VLAN? | 15:21 |
| trapni | yes | 15:21 |
| jaypipes | trapni: sure, one sec | 15:21 |
| trapni | ty | 15:21 |
| *** alexxu has quit IRC | 15:21 | |
| jaypipes | trapni: you just want to see what the ps output looks like, right? | 15:21 |
| jaypipes | to validate your own env? | 15:21 |
| trapni | well, it's more like a fuckup right now | 15:22 |
| *** zoresvit has quit IRC | 15:22 | |
| *** colinmcnamara has quit IRC | 15:23 | |
| jaypipes | trapni: http://paste.openstack.org/show/33572/ | 15:24 |
| trapni | I am currently upgrading from essex to folsom, (network: multi-node vlan mode) | 15:24 |
| jaypipes | trapni: in both? | 15:24 |
| trapni | and for some reason the VMs start loosing their IPs because dnsmasq is not responding anymore | 15:24 |
| trapni | ty for the link | 15:24 |
| jaypipes | trapni: meaning you are going from essex with multihost+VLAN to folsom with the same? | 15:24 |
| jaypipes | trapni: make sure that you have created the networks with nova-manage network create --multi_host='T' | 15:25 |
| jaypipes | trapni:I've been bitten before on that. nova-manage hardcodes its lookup to uppercase 'T' | 15:25 |
| jaypipes | trapni: and if you screw it up, you get odd results. | 15:26 |
| trapni | yes - i am not going into quantum yet | 15:26 |
| jaypipes | trapni: this isn't quantum either. | 15:26 |
| trapni | ah no, multi_host is not set for us | 15:27 |
| trapni | checked configs | 15:27 |
| trapni | so --multi_host=F in our case | 15:27 |
| trapni | we have a central nova-network node (our gateway) though | 15:27 |
| jaypipes | trapni: you will need to change that. | 15:28 |
| jaypipes | trapni: if you are going to use multihost, of course.. | 15:28 |
| trapni | sec | 15:28 |
| *** afazekas has quit IRC | 15:29 | |
| trapni | we have been always in multi_host=F mode, meaning, having a central gateway to pass the traffic through | 15:30 |
| trapni | I hardly can change that in the production env now :) | 15:30 |
| *** sride has quit IRC | 15:30 | |
| trapni | (without downtimes) | 15:30 |
| trapni | my question is, why is dnsmasq not responding to the dhcp requests anymore | 15:30 |
| *** dosaboy has quit IRC | 15:31 | |
| jaypipes | trapni: ah, no worries, sorry, I thought you were in multihost mode. no worries, then, ignore my comment about that then! | 15:33 |
| *** giulivo has quit IRC | 15:33 | |
| jaypipes | trapni: not sure why dnsmasq would not be responding to a particular instance's DHCP requests. Have you tried restarting the nova-network service? | 15:33 |
| *** gongysh has quit IRC | 15:34 | |
| *** boris-42 has quit IRC | 15:34 | |
| SpamapS | OSError: [Errno 24] Too many open files | 15:35 |
| SpamapS | anybody seen this on keystone under devstack? | 15:35 |
| trapni | my question is, why is dnsmasq not responding to the dhcp requests anymore. | 15:35 |
| jaypipes | trapni: not sure why dnsmasq would not be responding to a particular instance's DHCP requests. Have you tried restarting the nova-network service? | 15:35 |
| *** danwent has joined #openstack-dev | 15:36 | |
| trapni | jaypipes, btw, we quick-fixed it by removing "--bind-interfaces" from the command line (we started dnsmasq by hand) - and then dnsmasq is able to respond. so this is some kind of bug, and I wonder how to avoid this the clean way :) | 15:36 |
| *** dosaboy has joined #openstack-dev | 15:36 | |
| dolphm | SpamapS: raise the file descriptor limit (check it with "ulimit -n", raise it with something like "ulimit -n 2048") | 15:36 |
| trapni | jaypipes, dnsmasq wasn't replying et al since the upgrade to folsom | 15:36 |
| dolphm | SpamapS: if it's already high, something could be leaking descriptors | 15:36 |
| jaypipes | trapni: and you restarted the nova-network service or host? | 15:36 |
| *** rkukura has quit IRC | 15:37 | |
| *** colinmcnamara has joined #openstack-dev | 15:37 | |
| SpamapS | dolphm: given that it has exactly 1 user (me) and I left it over night doing nothing.. I'm going to guess that something is indeed leaking descriptors.. likely keystone given the pipe stuff I see there | 15:37 |
| SpamapS | ahhh | 15:39 |
| *** danwent has quit IRC | 15:39 | |
| SpamapS | the VM I have devstack in ran out of memory | 15:39 |
| dolphm | SpamapS: how much memory did it have? | 15:39 |
| SpamapS | 1G | 15:39 |
| SpamapS | I noticed yesterday it was not enough | 15:40 |
| dolphm | ah, that's a bit too low in my experience | 15:40 |
| SpamapS | was using 800M | 15:40 |
| dolphm | it used to be enough :P | 15:40 |
| *** zoresvit has joined #openstack-dev | 15:40 | |
| *** markmcclain has quit IRC | 15:40 | |
| SpamapS | total used free shared buffers cached | 15:40 |
| SpamapS | Mem: 1002 884 118 0 20 67 | 15:40 |
| SpamapS | with quantum, and heat.. not enough :) | 15:40 |
| dolphm | SpamapS: if you are running into a real leak, i'd look first at keystoneclient.middleware.auth_token | 15:41 |
| SpamapS | once mallocs start failing I don't trust python to do the right thing anymore. :) | 15:42 |
| SpamapS | my guess, whatever close() code that should be called on pipes was not reached.. | 15:42 |
| *** torandu has joined #openstack-dev | 15:42 | |
| SpamapS | there's probably a deep bug somewhere "handle malloc/fork fail better" . but... at some point you just have to say "don't run at 90% RAM w/ no swap, kthxbai" | 15:43 |
| *** rushiagr has joined #openstack-dev | 15:47 | |
| *** networkstatic has quit IRC | 15:47 | |
| *** jprovazn has quit IRC | 15:48 | |
| openstackgerrit | A change was merged to openstack/keystone: Catch and log server exceptions https://review.openstack.org/23015 | 15:48 |
| *** markmcclain has joined #openstack-dev | 15:48 | |
| *** salgado-lunch is now known as salgado | 15:49 | |
| *** davidhadas has joined #openstack-dev | 15:50 | |
| ttx | dolphm: ping about python-keystoneclient | 15:51 |
| dolphm | ttx: o/ | 15:51 |
| *** johnthetubaguy has joined #openstack-dev | 15:51 | |
| ttx | dolphm: there was a recent "When will new version python-keystoneclient available on pypi?" thread on the dev ml | 15:51 |
| *** yidclare has joined #openstack-dev | 15:51 | |
| ttx | Looks like some people are waiting for a tag to be pushed | 15:52 |
| ttx | dolphm: do you know what the plan is ? | 15:52 |
| *** rkukura has joined #openstack-dev | 15:52 | |
| *** henrynash has quit IRC | 15:52 | |
| *** henrynash has joined #openstack-dev | 15:54 | |
| dolphm | ttx: we have just a couple in progress reviews to complete v3 support -- perhaps next week / around rc1? | 15:54 |
| *** koolhead17 has joined #openstack-dev | 15:54 | |
| *** [1]glikson has joined #openstack-dev | 15:54 | |
| ttx | dolphm: ok, could someone from keystone answer to that thread ? | 15:54 |
| *** kpavel has quit IRC | 15:55 | |
| dolphm | ttx: will do | 15:55 |
| *** kpavel has joined #openstack-dev | 15:55 | |
| ttx | dolphm: many thanks ! | 15:56 |
| *** glikson has quit IRC | 15:56 | |
| *** [1]glikson is now known as glikson | 15:56 | |
| *** gyee has joined #openstack-dev | 15:56 | |
| *** rpedde has left #openstack-dev | 15:58 | |
| *** danwent has joined #openstack-dev | 15:58 | |
| *** danwent has quit IRC | 16:00 | |
| *** jsavak has joined #openstack-dev | 16:00 | |
| *** thimble has quit IRC | 16:00 | |
| *** varto has joined #openstack-dev | 16:00 | |
| *** mrodden has quit IRC | 16:00 | |
| *** yamahata_ has quit IRC | 16:00 | |
| *** henrynash has quit IRC | 16:01 | |
| *** adjohn has joined #openstack-dev | 16:02 | |
| openstackgerrit | A change was merged to openstack/nova: Sync rpc from oslo-incubator. https://review.openstack.org/24226 | 16:03 |
| *** joesavak has quit IRC | 16:03 | |
| *** joesavak has joined #openstack-dev | 16:03 | |
| openstackgerrit | A change was merged to openstack/glance: Honor metadata_encryption_key in glance domain https://review.openstack.org/23725 | 16:04 |
| *** kaushikc has quit IRC | 16:04 | |
| *** yolanda_ has joined #openstack-dev | 16:04 | |
| *** varto has quit IRC | 16:04 | |
| *** salv-orlando_ has quit IRC | 16:04 | |
| *** yolanda has quit IRC | 16:04 | |
| *** jsavak has quit IRC | 16:04 | |
| *** salv-orlando has joined #openstack-dev | 16:05 | |
| openstackgerrit | A change was merged to openstack/nova: Fix reconnecting to libvirt. https://review.openstack.org/24323 | 16:06 |
| *** jcmartin has joined #openstack-dev | 16:06 | |
| *** reidrac has quit IRC | 16:08 | |
| *** hoodow has quit IRC | 16:09 | |
| *** hoodow has joined #openstack-dev | 16:09 | |
| *** hoodow has joined #openstack-dev | 16:09 | |
| *** rushiagr has quit IRC | 16:10 | |
| *** rushiagr has joined #openstack-dev | 16:10 | |
| *** alop has joined #openstack-dev | 16:11 | |
| *** dontalton has joined #openstack-dev | 16:11 | |
| *** giulivo has joined #openstack-dev | 16:12 | |
| *** dosaboy has quit IRC | 16:12 | |
| *** aditirav has quit IRC | 16:13 | |
| *** rnirmal_ has joined #openstack-dev | 16:14 | |
| *** rnirmal_ has quit IRC | 16:14 | |
| *** mrodden has joined #openstack-dev | 16:14 | |
| *** rnirmal has quit IRC | 16:15 | |
| *** xga has joined #openstack-dev | 16:16 | |
| *** mnewby has joined #openstack-dev | 16:16 | |
| *** dosaboy has joined #openstack-dev | 16:16 | |
| *** mnewby has quit IRC | 16:16 | |
| *** garyk has quit IRC | 16:17 | |
| *** nati_ueno has joined #openstack-dev | 16:18 | |
| *** hemnafk is now known as hemna | 16:18 | |
| *** marun has joined #openstack-dev | 16:19 | |
| trapni | jaypipes, i'm back, sorry, I'd to quickly go afk - well yes and no, re repeately restarted nova-network today, and always ensured to have no dnsmasq process running before restarting | 16:19 |
| trapni | because nova-network does indeed start dnsmasq processes on start() but doesn't stop them on stop() (a bug imho) | 16:19 |
| *** sulo__ has joined #openstack-dev | 16:22 | |
| *** sulo_ has quit IRC | 16:22 | |
| *** sulo__ is now known as sulo_ | 16:22 | |
| *** doude has joined #openstack-dev | 16:23 | |
| *** adjohn has quit IRC | 16:25 | |
| *** giulivo has quit IRC | 16:25 | |
| *** amerine has joined #openstack-dev | 16:26 | |
| *** dprince has joined #openstack-dev | 16:27 | |
| *** otherwiseguy has quit IRC | 16:27 | |
| *** anteaya_ is now known as anteaya | 16:28 | |
| *** networkstatic has joined #openstack-dev | 16:28 | |
| *** giulivo has joined #openstack-dev | 16:29 | |
| *** anniec has joined #openstack-dev | 16:30 | |
| *** yidclare has quit IRC | 16:30 | |
| *** rnirmal has joined #openstack-dev | 16:30 | |
| openstackgerrit | A change was merged to openstack/nova: Lazy load CONF.quota_driver. https://review.openstack.org/24269 | 16:32 |
| *** yidclare has joined #openstack-dev | 16:32 | |
| sandeepr | ttx, !ping | 16:32 |
| *** xga has quit IRC | 16:33 | |
| ttx | sandeepr, !pong | 16:33 |
| sandeepr | lol | 16:33 |
| sandeepr | i am playing a bit with pki | 16:33 |
| *** Gordonz has quit IRC | 16:33 | |
| sandeepr | needed an input | 16:34 |
| sandeepr | reading through launchpad | 16:34 |
| *** thouveng has quit IRC | 16:34 | |
| sandeepr | is there a bug still with pki and keystone? | 16:34 |
| *** jgallard has quit IRC | 16:35 | |
| sandeepr | keystone-manage pki_setup ok and keystone user-list error Authorization Failed: Unable to communicate with identity service: | 16:35 |
| *** jgallard has joined #openstack-dev | 16:35 | |
| ttx | sandeepr: there are probably still bugs, but nothing that I particularly know of | 16:35 |
| sandeepr | i ran the manage pkisetup as root | 16:35 |
| *** koert has quit IRC | 16:36 | |
| sandeepr | could you take a look at http://paste.openstack.org/show/33575/ | 16:36 |
| *** lloydde has joined #openstack-dev | 16:36 | |
| *** colinmcnamara has quit IRC | 16:37 | |
| *** nati_ueno has quit IRC | 16:37 | |
| openstackgerrit | A change was merged to openstack/quantum: sync Oslo Grizzly stable branch with Quantum https://review.openstack.org/24324 | 16:38 |
| *** esp1 has joined #openstack-dev | 16:38 | |
| *** xga has joined #openstack-dev | 16:39 | |
| *** sirushti has quit IRC | 16:39 | |
| * ttx looks | 16:40 | |
| *** dev_sa has quit IRC | 16:40 | |
| *** sulo_ has quit IRC | 16:40 | |
| ttx | sandeepr: that doesn ring a bell, but then I'm not a keystone specialist. | 16:40 |
| sandeepr | oh ok | 16:41 |
| *** kaushikc has joined #openstack-dev | 16:41 | |
| sandeepr | i saw your comment on https://bugs.launchpad.net/keystone/+bug/1031372 | 16:41 |
| uvirtbot | Launchpad bug 1031372 in keystone "keystone-manage pki_setup creates certs owned by root" [High,Fix released] | 16:41 |
| sandeepr | so i thought you lookin keystone | 16:41 |
| ttx | Ha! I comment on everything. But I don't know everything. | 16:42 |
| dolphm | ayoung: Unable to load certificate / Unable to sign token ? | 16:42 |
| dolphm | ayoung: ^ | 16:42 |
| ayoung | sounds right | 16:42 |
| ayoung | dolphm, certs should be owned by root. The correct title is "certs not readable by keystone" | 16:43 |
| sandeepr | ah ok, ayoung, what is the correct way to use pki | 16:43 |
| *** Ritz-Laptop has joined #openstack-dev | 16:43 | |
| ayoung | sandeepr, run keystone-manage pki_setup as root and chown, probably for your set up. What are you using to install? | 16:44 |
| *** reed has joined #openstack-dev | 16:44 | |
| *** koolhead17 has quit IRC | 16:45 | |
| ayoung | dolphm, let me finish cleaning up the revoke token patch and I'll take that one on...it has been bugging me for a while. I think the right thing is to do a chown afterwards, but I am not sure I can get all the variations correct that people are going to use. correc | 16:45 |
| *** Ritz-Mobile has joined #openstack-dev | 16:46 | |
| *** sulo_ has joined #openstack-dev | 16:46 | |
| *** jog0 has joined #openstack-dev | 16:46 | |
| sandeepr | chpwn to keystone/keystone? | 16:46 |
| dolphm | ayoung: easy review for a release blocker https://review.openstack.org/#/c/23934/ | 16:47 |
| ayoung | sandeepr, look at this https://github.com/openstack/keystone/commit/ca2b2cb4903aa1d8e6469b22e2022925ca413a9c | 16:47 |
| ayoung | dolphm, looked like two different reviews dealt with that | 16:47 |
| ayoung | I thought one merged already that did that....let me post | 16:47 |
| *** lcheng has quit IRC | 16:47 | |
| ayoung | dolph what does this do differently : 31df378f2d95374d6662be11fc6487d01e373f6a | 16:47 |
| *** pabelanger has joined #openstack-dev | 16:48 | |
| ayoung | ah..OK just looked at the body. I can ack that one. If it is ready.... | 16:48 |
| dolphm | ayoung: the patch in review filters out legacy_endpoint_id from being exposed on v3 (it should be an implementation detail) | 16:48 |
| ayoung | dolphm, what calls wrap_member? | 16:48 |
| dolphm | ayoung: the one that has merged improves v2<-->v3 compatibility for endpoints | 16:49 |
| *** SergeyLukjanov has quit IRC | 16:49 | |
| dolphm | ayoung: every EndpointV3 method, actually | 16:49 |
| ayoung | yeah, I had just read the summary recently, which is why I got them confused. Looking at the actual code makes it clear | 16:49 |
| openstackgerrit | A change was merged to openstack/nova: Skip deleted fixed ip address for os-fixed-ips extension https://review.openstack.org/23993 | 16:49 |
| dolphm | ayoung: it's called directly by controller methods and by wrap_collection | 16:49 |
| *** arbrandes has joined #openstack-dev | 16:50 | |
| *** giulivo has quit IRC | 16:50 | |
| ayoung | dolphm, looks good. Any reason to hold off pulling the trigger on it? | 16:50 |
| *** AlanClark has joined #openstack-dev | 16:54 | |
| ayoung | dolphm, going once.... | 16:54 |
| *** fire has quit IRC | 16:55 | |
| *** baba has joined #openstack-dev | 16:55 | |
| *** markwash has joined #openstack-dev | 16:55 | |
| ayoung | dolphm, +2...I'll let you approve. https://review.openstack.org/#/c/23934/ | 16:55 |
| *** markmc has quit IRC | 16:56 | |
| *** metral has quit IRC | 16:56 | |
| *** rnirmal_ has joined #openstack-dev | 17:00 | |
| *** baba has quit IRC | 17:00 | |
| *** giulivo has joined #openstack-dev | 17:01 | |
| *** rnirmal_ has quit IRC | 17:01 | |
| *** avishay is now known as avishay-afk | 17:01 | |
| *** rnirmal_ has joined #openstack-dev | 17:02 | |
| *** winston-d has quit IRC | 17:02 | |
| *** rnirmal has quit IRC | 17:02 | |
| *** rnirmal_ is now known as rnirmal | 17:02 | |
| *** arbrandes has quit IRC | 17:04 | |
| *** alobbs1 has joined #openstack-dev | 17:05 | |
| sandeepr | thanks ayoung, chown -R for under ssl worked | 17:05 |
| ayoung | sandeepr, did you look at the change I linked? | 17:05 |
| ayoung | you should be passing the params on the command line | 17:05 |
| ayoung | --keystone-user and --keystone-group | 17:06 |
| sandeepr | btw, i uncommented the section signing in keystone.conf | 17:06 |
| *** alobbs has quit IRC | 17:06 | |
| *** arbrandes has joined #openstack-dev | 17:06 | |
| *** terryg has joined #openstack-dev | 17:07 | |
| *** Mandell has joined #openstack-dev | 17:08 | |
| sandeepr | ayoung, i looked at the change | 17:08 |
| sandeepr | are you saying it should be like this - keystone-manage pki_setup --keystone-user keystone --keystone-group keystone | 17:08 |
| ayoung | sandeepr, that question is rhetorical, right? | 17:09 |
| sandeepr | ayoung, :-) | 17:10 |
| ayoung | I don't know how you are setting up. If you are doing this for production, you should be doing it either in a puppet script or in a package, and it depends on who you run as, permissions, etc. So....probably? | 17:10 |
| *** AlanClark has quit IRC | 17:11 | |
| ayoung | If OTOH, you are just testing things out, please confirm that the above works for you. | 17:11 |
| ayoung | :) | 17:11 |
| sandeepr | the above meaning the chown thing? | 17:11 |
| sandeepr | and how do i know i'm using the PKI feature? | 17:12 |
| *** kagan has joined #openstack-dev | 17:12 | |
| *** metral has joined #openstack-dev | 17:12 | |
| *** terryh has quit IRC | 17:13 | |
| *** xga has quit IRC | 17:13 | |
| *** xga has joined #openstack-dev | 17:14 | |
| sandeepr | oops i missed to read your prior comment | 17:14 |
| sandeepr | its not production yet, i'm testing | 17:14 |
| *** glikson has quit IRC | 17:14 | |
| *** alop has quit IRC | 17:14 | |
| *** mmagr has quit IRC | 17:15 | |
| sandeepr | got it, default is PKI unless you change the token_format to UUID | 17:16 |
| *** morganfainberg has joined #openstack-dev | 17:16 | |
| *** alop has joined #openstack-dev | 17:17 | |
| *** morganfainberg has quit IRC | 17:17 | |
| *** terryh has joined #openstack-dev | 17:17 | |
| enikanorov-w | folks, how do i define list of dict on quantum client? say, i want to update fixed_ips | 17:19 |
| ayoung | dolphm, I have three outstanding reviews that are for backport to Folsom stable. https://review.openstack.org/#/c/24079/ https://review.openstack.org/#/c/23996/ and https://review.openstack.org/#/c/23842/ | 17:19 |
| ayoung | I'd appreciate getting your stamp of approval on the ones that still need more core, and then I'll bug the stable gatekeepers. | 17:20 |
| *** kpavel has quit IRC | 17:20 | |
| *** kpavel has joined #openstack-dev | 17:20 | |
| eglynn | bcwaldon, markwash: care to take another look at https://review.openstack.org/19547 ? | 17:20 |
| markwash | eglynn: sure, and thanks! | 17:21 |
| *** jsavak has joined #openstack-dev | 17:25 | |
| *** roampune has joined #openstack-dev | 17:25 | |
| *** zoresvit has quit IRC | 17:26 | |
| *** jergerber has joined #openstack-dev | 17:27 | |
| *** Ryan_Lane has joined #openstack-dev | 17:27 | |
| *** esp1 has left #openstack-dev | 17:27 | |
| *** terryg has quit IRC | 17:28 | |
| *** nati_ueno has joined #openstack-dev | 17:28 | |
| *** joesavak has quit IRC | 17:29 | |
| *** rafaduran has quit IRC | 17:30 | |
| *** topol has joined #openstack-dev | 17:31 | |
| *** adjohn has joined #openstack-dev | 17:31 | |
| *** xga has quit IRC | 17:32 | |
| *** SergeyLukjanov has joined #openstack-dev | 17:33 | |
| *** corXi has quit IRC | 17:34 | |
| *** pholland has joined #openstack-dev | 17:35 | |
| jog0 | ayoung: ping | 17:37 |
| jog0 | wanted to follow up on https://review.openstack.org/#/c/23970/ | 17:37 |
| *** glikson has joined #openstack-dev | 17:38 | |
| *** rnirmal_ has joined #openstack-dev | 17:38 | |
| *** rnirmal_ has quit IRC | 17:38 | |
| *** dosaboy has quit IRC | 17:38 | |
| *** rdo_ is now known as rdo | 17:38 | |
| *** esp1 has joined #openstack-dev | 17:38 | |
| *** rnirmal_ has joined #openstack-dev | 17:39 | |
| openstackgerrit | A change was merged to openstack/tempest: Convert try/expect/else as per new Tempest style. https://review.openstack.org/24281 | 17:39 |
| *** networkstatic has quit IRC | 17:40 | |
| *** koolhead17 has joined #openstack-dev | 17:41 | |
| ayoung | jog0, oooh pretty | 17:41 |
| ayoung | "service_or_admin": [["rule:admin_required"], ["rule:service_project"]], | 17:42 |
| ayoung | jog0, that looks good. Lets drop the old rule, though | 17:42 |
| *** rnirmal has quit IRC | 17:42 | |
| *** rnirmal_ is now known as rnirmal | 17:42 | |
| jog0 | ayoung: drop the admin_required part? | 17:43 |
| ayoung | nope | 17:43 |
| ayoung | the one I put in there | 17:43 |
| ayoung | "service_project": [["project_id:service"], ["tenant_id:service"]], | 17:43 |
| ayoung | BTW, what is the reason for | 17:43 |
| ayoung | *args in def wrapper(self, context, *args, **kwargs): | 17:43 |
| *** torandu has quit IRC | 17:43 | |
| ayoung | keystone/common/controller.py | 17:43 |
| jog0 | service_project is referenced in the service_or_admin rule | 17:43 |
| ayoung | jog0, right....if we replace it with a role check, we can drop that.... | 17:44 |
| ayoung | I just mean lets nopt put both rules in there.... | 17:45 |
| jog0 | ayoung: added the *args because validate_token uses *args not kwargs | 17:45 |
| ayoung | ok...I am liking this | 17:45 |
| jog0 | ayoung: so "service_or_admin": [["rule:admin_required"], ["project_id:service"], ["tenant_id:service"]] ? | 17:46 |
| ayoung | jog0, what happens if we drop keystone/common/wsgi.py assert_admin now? | 17:46 |
| *** networkstatic has joined #openstack-dev | 17:46 | |
| dolphm | vkmc: trusts should be added to keystoneclient, yes -- i'm not aware of anything in progress though | 17:47 |
| ayoung | that says "any role in the service tenant" and you think it should be a specific role. Either is probably sufficient. | 17:47 |
| jog0 | ayoung: if we remove that then everything goes to RBAC from policy.json | 17:47 |
| koolhead17 | notmyname, ping | 17:47 |
| ayoung | muahahahah | 17:47 |
| ayoung | yes , yes it does | 17:47 |
| notmyname | koolhead17: ack | 17:48 |
| *** yidclare has quit IRC | 17:48 | |
| koolhead17 | notmyname, https://bugs.launchpad.net/swift/+bug/1024477 | 17:48 |
| uvirtbot | Launchpad bug 1024477 in swift "SAIO: logs filling up with "exception dumping recon cache"" [Undecided,Fix released] | 17:48 |
| koolhead17 | it says fixed | 17:48 |
| vkmc | dolphm, Cool, is there a bp or something I could subscribe to get notified on its progress? | 17:48 |
| ayoung | jog0, I think that is the right thing to do...doesn't have to be in this patch, though | 17:48 |
| koolhead17 | notmyname, but on ubuntu 12.04 1.7.4-0ubuntu2~cloud0 i see this again | 17:49 |
| jog0 | ayoung: *nod*, yeah I like the idea of keeping the scope of this patch to just token auth | 17:49 |
| openstackgerrit | A change was merged to openstack/tempest: xml_to_json should not convert xmlns into attribute https://review.openstack.org/23926 | 17:49 |
| jog0 | ayoung: I can start working on a devstack patch to accompany this one. to enable non admin auth_token users | 17:50 |
| koolhead17 | notmyname, http://paste.ubuntu.com/5611372/ this is how the log looks like | 17:51 |
| *** epim has joined #openstack-dev | 17:51 | |
| *** yidclare has joined #openstack-dev | 17:51 | |
| jog0 | ayoung: also can you comment on the patch with the changes you want to see, just to keep the history clean | 17:51 |
| *** AlanClark has joined #openstack-dev | 17:52 | |
| notmyname | koolhead17: what's the user line in your config? ie what user are you running as? | 17:52 |
| koolhead17 | root iteself | 17:52 |
| notmyname | koolhead17: running as root and getting a permission error? that seems odd | 17:53 |
| ayoung | jog0, I am thinking just keep it as is...I think | 17:53 |
| jog0 | kk | 17:53 |
| koolhead17 | notmyname, indeed | 17:53 |
| koolhead17 | :D | 17:53 |
| koolhead17 | notmyname, -rw------- 1 root root 93 Mar 13 23:19 /var/cache/swift/object.recon <--- it requires execute permission too? | 17:54 |
| notmyname | no, it's just a data file | 17:55 |
| notmyname | koolhead17: none of what I'm saying is condoning running swift as root, but it should work | 17:55 |
| *** dsanders has joined #openstack-dev | 17:56 | |
| koolhead17 | notmyname, i see this as a reported bug and fixed too in 1.6 | 17:56 |
| *** jpich has quit IRC | 17:57 | |
| *** rafaduran has joined #openstack-dev | 17:57 | |
| *** torandu has joined #openstack-dev | 17:57 | |
| *** renier has quit IRC | 17:58 | |
| *** morganfainberg has joined #openstack-dev | 17:58 | |
| *** AlanClark has quit IRC | 17:59 | |
| koolhead17 | networkstatic, hello there | 17:59 |
| networkstatic | koolhead17 hey bro, meeting meetings :-/ | 18:00 |
| networkstatic | brb | 18:00 |
| koolhead17 | networkstatic, you need to give me appointment by this week itself :) | 18:00 |
| *** kagan has quit IRC | 18:01 | |
| vkmc | ayoung, If you have a second... Is it possible to add trust support in Keystone v2 client? I saw you developed that feature so you probably could give me some pointers with it :) | 18:02 |
| ayoung | vkmc, well, I am not sure I would recommend that. | 18:04 |
| *** derekh has quit IRC | 18:04 | |
| ayoung | vkmc, the APIs to create and modify trusts are V3 only, but you can create a token from a trust with V2 | 18:04 |
| ayoung | is that sufficient? | 18:05 |
| *** jbresnah has quit IRC | 18:05 | |
| termie | ayoung: i need to run downstairs to catch a trian in a sec, but i assume these are based on getting a library installed: https://review.openstack.org/#/c/24284/ | 18:05 |
| termie | ayoung: (those jenkins failures) | 18:05 |
| *** glikson has quit IRC | 18:05 | |
| *** danpb has quit IRC | 18:05 | |
| termie | ayoung: oh, nevermind, they need a cassandra running :p | 18:06 |
| *** retr0h has joined #openstack-dev | 18:06 | |
| *** retr0h has joined #openstack-dev | 18:06 | |
| ayoung | termie, I'll take a look, though. | 18:06 |
| vkmc | ayoung, How could I create a token from a trust? Is there a guide or a code snippet to look somewhere? | 18:06 |
| termie | ayoung: what is the process nowadays for getting jenkins configured more? i think i've lost my admin bits | 18:06 |
| termie | TRAINTIME | 18:07 |
| ayoung | termie, probably needs devstack support. Let me look. Ping me when you've caught your train etc and I'll have more for you | 18:07 |
| trapni | for how long will nova-network (along with quantum) be supported within openstack ? | 18:07 |
| ayoung | vkmc, there are unit tests that do it, but in general, you pass the trust_id along with the token request....for example | 18:08 |
| *** sulo_ has quit IRC | 18:08 | |
| ayoung | vkmc take a look at the tests here https://github.com/openstack/keystone/blob/master/tests/test_auth.py#L501 | 18:08 |
| ayoung | those do direct calls to the controlers, but it should be clear how to modify the web calls to do the same thing. | 18:09 |
| vkmc | ayoung, Thanks, I will check that test and try to understand how it works :) | 18:09 |
| *** AlanClark has joined #openstack-dev | 18:10 | |
| *** kagan has joined #openstack-dev | 18:10 | |
| notmyname | koolhead17: sorry, had to reboot to fix some local net issues. lost my irc buffers (but hey! no more crazy lag!) | 18:14 |
| koolhead17 | notmyname, :D | 18:15 |
| *** stevebaker has joined #openstack-dev | 18:15 | |
| *** lloydde has quit IRC | 18:15 | |
| *** doude has left #openstack-dev | 18:16 | |
| *** jkordish_ has joined #openstack-dev | 18:16 | |
| *** jkordish has quit IRC | 18:16 | |
| *** darraghb has quit IRC | 18:16 | |
| *** AlanClark has quit IRC | 18:17 | |
| *** rafaduran has quit IRC | 18:18 | |
| openstackgerrit | A change was merged to openstack/tempest: Add base classes for image tests. https://review.openstack.org/24098 | 18:19 |
| *** rnirmal_ has joined #openstack-dev | 18:19 | |
| *** rnirmal_ has quit IRC | 18:20 | |
| *** rnirmal has quit IRC | 18:23 | |
| *** jgallard has quit IRC | 18:24 | |
| koolhead17 | notmyname, what would be your suggestion? reopening the bug? | 18:24 |
| *** stevebaker has quit IRC | 18:25 | |
| *** jkordish_ is now known as jkordish | 18:26 | |
| trapni | ayoung, is openstack hosting their git repos officially there? or is it "just a mirror" ? | 18:26 |
| *** rushiagr has quit IRC | 18:29 | |
| *** colinmcnamara has joined #openstack-dev | 18:29 | |
| *** rnirmal has joined #openstack-dev | 18:30 | |
| *** adalbas has quit IRC | 18:31 | |
| *** tomoe_ has quit IRC | 18:31 | |
| *** tomoe_ has joined #openstack-dev | 18:32 | |
| *** melwitt has joined #openstack-dev | 18:33 | |
| ayoung | trapni, "there" meaning github? Yes that is the source. But we merge in via gerrit. | 18:33 |
| *** salv-orlando has quit IRC | 18:35 | |
| *** salv-orlando has joined #openstack-dev | 18:35 | |
| *** networkstatic has quit IRC | 18:36 | |
| *** tomoe_ has quit IRC | 18:37 | |
| trapni | ayoung, ah, thanks :) | 18:38 |
| *** CaptTofu has quit IRC | 18:38 | |
| *** metral has quit IRC | 18:39 | |
| notmyname | koolhead17: can you paste your error again please? | 18:39 |
| koolhead17 | notmyname, http://paste.ubuntu.com/5611506/ | 18:40 |
| *** zaitcev has joined #openstack-dev | 18:40 | |
| *** johnthetubaguy has quit IRC | 18:41 | |
| *** marun has quit IRC | 18:42 | |
| *** maoy has joined #openstack-dev | 18:44 | |
| *** rnirmal has quit IRC | 18:44 | |
| *** yjiang5 has joined #openstack-dev | 18:44 | |
| *** david-lyle has joined #openstack-dev | 18:46 | |
| *** rnirmal has joined #openstack-dev | 18:47 | |
| *** adalbas has joined #openstack-dev | 18:47 | |
| *** varto has joined #openstack-dev | 18:49 | |
| dolphm | ayoung: critical release blocker looks ready to merge to me https://review.openstack.org/#/c/24078/ | 18:50 |
| dolphm | gyee: ^ | 18:50 |
| *** otherwiseguy has joined #openstack-dev | 18:50 | |
| ayoung | dolphm, I addressed that | 18:51 |
| dolphm | ayoung: ? | 18:51 |
| ayoung | hold off on that one and instead look at this | 18:51 |
| ayoung | https://review.openstack.org/#/c/24010/ | 18:51 |
| *** vipul is now known as vipul|away | 18:51 | |
| *** vipul|away is now known as vipul | 18:51 | |
| *** adalbas has quit IRC | 18:52 | |
| ayoung | dolphm, hmmm | 18:52 |
| ayoung | dolphm, OK, I am going to need to rebase on his changes | 18:52 |
| ayoung | dolphm, OK...I can ACK that. I'll self nack mine. But consider mine a release blocker as well | 18:54 |
| dolphm | ayoung: it is -- but are you addressing more than is described in the commit message & bug description? | 18:54 |
| dolphm | ayoung: i've only reviewed a bit of that patch | 18:54 |
| ayoung | dolphm, it makes all of the revoke calls go through the same code. | 18:55 |
| ayoung | dolphm, tell you what, I'll self nack, but still look at it. | 18:55 |
| *** thimble has joined #openstack-dev | 18:55 | |
| *** trapni has quit IRC | 18:55 | |
| *** bknudson has quit IRC | 18:56 | |
| ayoung | dolphm, one thing I did in the last version was to move delete_tokens_for users into the common/controller. YOu implied in a previous review that I should do as much, and it has been bothering me. I'll do the same with the group function here for consistency. | 18:57 |
| jog0 | dripton: ping | 18:57 |
| dripton | jog0: pong | 18:58 |
| *** stevebaker has joined #openstack-dev | 18:58 | |
| jog0 | dripton: https://bugs.launchpad.net/nova/+bug/1137977 | 18:58 |
| uvirtbot | Launchpad bug 1137977 in nova ""nova-manage db archive_deleted_rows" returns error re foreign key constraints" [High,Triaged] | 18:58 |
| jog0 | any chance that will be fixed for G? | 18:58 |
| *** varto has quit IRC | 18:58 | |
| dripton | jog0: I hope to have time to look at that bug tomorrow. There's a chance it'll be fixed -- I need to reorder or group archiving to satisfy FK constraints. | 18:59 |
| jog0 | dripton: cool, it would be really nice if that works in G | 18:59 |
| jog0 | as its a great feature | 18:59 |
| *** avishay-afk has quit IRC | 18:59 | |
| dripton | jog0: I'll let you know when I have a patch. | 19:00 |
| jog0 | dripton: cool, and if there is anything I can do to help just let me know | 19:00 |
| *** novas0x2a|laptop has joined #openstack-dev | 19:03 | |
| *** stevebaker has quit IRC | 19:04 | |
| openstackgerrit | A change was merged to openstack/glance: Publish API version 2.1 https://review.openstack.org/24267 | 19:05 |
| openstackgerrit | A change was merged to openstack/glance: Insure that migration 6 retains deleted image property index. https://review.openstack.org/23873 | 19:05 |
| *** vipul is now known as vipul|away | 19:06 | |
| *** metral has joined #openstack-dev | 19:07 | |
| jog0 | do tenant/project ids need to be uuids in keystone? https://bugs.launchpad.net/python-novaclient/+bug/1088835 | 19:07 |
| uvirtbot | Launchpad bug 1088835 in python-novaclient ""nova quota-update" command is executed without checking the project_id parameter" [Undecided,Fix committed] | 19:07 |
| jog0 | ayoung: ^? | 19:07 |
| ayoung | jog0, no, but they are autogenerated as UUIDs | 19:08 |
| jog0 | ayoung: thanks | 19:08 |
| openstackgerrit | A change was merged to openstack/glance: Log when image_size_cap has been exceeded https://review.openstack.org/24317 | 19:08 |
| *** sirushti has joined #openstack-dev | 19:08 | |
| *** pfreund has joined #openstack-dev | 19:09 | |
| *** adalbas has joined #openstack-dev | 19:11 | |
| jog0 | ayoung: I ask because no novacient we are looking for an elegant way to validate if a tenant_id is valid or not | 19:11 |
| ayoung | jog0, there is none | 19:12 |
| ayoung | a uuid could refer to a user. or a vm | 19:13 |
| openstackgerrit | A change was merged to openstack/tempest: Disable test_rescued_vm_attach_volume https://review.openstack.org/24276 | 19:14 |
| openstackgerrit | A change was merged to openstack/nova: Prepare services index method for use with cells https://review.openstack.org/23761 | 19:14 |
| openstackgerrit | A change was merged to openstack/nova: Fix behaviour of split_cell_and_item https://review.openstack.org/24135 | 19:14 |
| openstackgerrit | A change was merged to openstack/tempest: Add api version detection support to glance tests. https://review.openstack.org/23848 | 19:14 |
| jog0 | ayoung: so no simple way to confirm if a user typed in a valid tenant_id? any keystone call | 19:15 |
| openstackgerrit | A change was merged to openstack/tempest: Remove unnecessary asserts from test_images.py https://review.openstack.org/24342 | 19:15 |
| ayoung | jog0, Keystone makes no guarantees about IDs other than that they are strings. | 19:16 |
| *** cp16net is now known as cp16net|away | 19:16 | |
| jog0 | ayoung: and what about having novaclient send out a REST call to keystone to check? | 19:17 |
| jog0 | ayoung: b/c right now the quota code takes in a tenant_id as an (unvalidated) parameter | 19:18 |
| ayoung | jog0, yes, you can do that. Assuming you have access to the project. | 19:18 |
| *** mestery has joined #openstack-dev | 19:18 | |
| *** marun has joined #openstack-dev | 19:18 | |
| jog0 | ayoung: thanks | 19:18 |
| *** smulcahy has quit IRC | 19:19 | |
| *** arbrandes has quit IRC | 19:20 | |
| *** markmcclain has quit IRC | 19:23 | |
| *** thimble has quit IRC | 19:24 | |
| *** rnirmal has quit IRC | 19:27 | |
| termie | ayoung: now at the office, the tests use devstack, eh, so need to get cassandra into that? | 19:29 |
| ayoung | termie, is a mock impl possible first? | 19:29 |
| *** stevebaker has joined #openstack-dev | 19:29 | |
| *** digitalsanctum has quit IRC | 19:30 | |
| ayoung | termie, this is cassandra, right: http://wiki.apache.org/cassandra/ | 19:30 |
| termie | ayoung: yeah that's the one | 19:30 |
| termie | ayoung: not aware of any in-process python version | 19:30 |
| ayoung | termie, we have the mildly unpleasant FakeLDAP approach, which is what we used for LDAP | 19:31 |
| *** jbresnah has joined #openstack-dev | 19:32 | |
| ayoung | The thing is , CAssandra is only one of a multitude of NoSQL dbs, and I am not sure that those are really appropriate for Tokens. THe issue is that token revocation is typically desired to be immediate, and "right eventually" semantics kindof go against that.... | 19:32 |
| *** networkstatic has joined #openstack-dev | 19:32 | |
| ayoung | but even if we do take in the cassandra backend, does that mean we end up with Mongo et all as well? | 19:33 |
| termie | ayoung: a variety of dbs support strong write consistency | 19:33 |
| ayoung | that is good | 19:33 |
| termie | ayoung: but even not having that really isn't that much of an issue when you have periodic updates to revoked lists | 19:33 |
| ayoung | termie, that is true, too, although those are not checked for direct Keystone calls. | 19:34 |
| termie | ayoung: but yes, it does mean we will probably have mongo and all the rest | 19:34 |
| ayoung | Keystone goes against the trust backend directly at token validation time | 19:34 |
| ayoung | termie, ok...so I think that this is a good approach... | 19:34 |
| ayoung | assuming that you will be around to consult when we need to abstract the nosql backend to deal with MOngo and variants? | 19:35 |
| termie | it isn't a nosql backend | 19:35 |
| termie | it is a cassandra backend | 19:35 |
| *** networkstatic has quit IRC | 19:35 | |
| termie | the whole point of implementing a new backend is so that you can optimize it for the features of your db | 19:35 |
| termie | the "kvs" backend supports any arbitrary key-value-store | 19:36 |
| termie | it could havebeen used for cassandra but would have been inefficient | 19:36 |
| ayoung | I guess that is true, the APIs are going to be fairly DB specific. | 19:36 |
| ayoung | cassandra seems to have better query capabilities, too | 19:36 |
| termie | and after some changes made to the api it doesn't even actually do what it was intended to any more either | 19:36 |
| sandeepr | ayoung, seems like the PKI does not work | 19:36 |
| termie | sandeepr: LIES | 19:36 |
| termie | sandeepr: FILTHY LIAR | 19:36 |
| ayoung | hah! | 19:36 |
| sandeepr | lol | 19:36 |
| termie | ayoung: it seems people have "extended" the kvs backend to support things that KVSs don't support | 19:37 |
| sandeepr | keystone user-list errors as - Authorization Failed: HTTPConnectionPool(host='10.1.0.16', port=5000): Request timed out. (timeout=600.0) | 19:37 |
| ayoung | termie, you mean like listing tokens and the such? Guilty. guilty as all hell. | 19:38 |
| termie | yes | 19:38 |
| termie | you have to make your own indexes | 19:38 |
| *** stevebaker has quit IRC | 19:38 | |
| termie | otherwise you aren't playing by the rules | 19:38 |
| termie | it is supposed to teach people | 19:38 |
| termie | how to write better db access code | 19:38 |
| *** torandu has quit IRC | 19:38 | |
| ayoung | termie, yeah...I think, maybe, my understanding of the KVS was different from the original intention. I didn't see it being used with persistance anywhere, which lead me to believe it was only being used for testing purposes....what you said before makes me question that | 19:39 |
| ayoung | I was actually proposing we remove the KVS backend and do all the unit tests against in memory sqlite | 19:39 |
| *** stevebaker has joined #openstack-dev | 19:40 | |
| termie | ayoung: the philosophy there is that we shouldn't implement any features that we can't figure out how to implement on a kvs | 19:40 |
| ayoung | termie, it did seem to me that if KVS supported persistance, it would have been used for the memcached backend | 19:40 |
| termie | ayoung: because a kvs will scale forever | 19:40 |
| *** lloydde has joined #openstack-dev | 19:40 | |
| ayoung | termie, well, not once you run out of memory. | 19:40 |
| termie | ayoung: memcache was easy to optimize support for | 19:41 |
| termie | ayoung: no, there are plenty of kvs datastores that are not memory dependent | 19:41 |
| termie | ayoung: the fake implementation in keystone obviously is just in process | 19:41 |
| termie | ayoung: but something like riak, redis, tokyocabinet, etc | 19:41 |
| termie | ayoung: are all clusterable | 19:41 |
| termie | ayoung: and scale extremely well | 19:41 |
| termie | ayoung: in general any time you can do only primary key lookups you are giving yourself extremely good scaling prospects | 19:42 |
| termie | ayoung: you can scale sql databases but it requires much more knowledge | 19:42 |
| *** adalbas has quit IRC | 19:42 | |
| ayoung | termie, interesting thoughts...If that is the case, I would like to make it so that we don't have to reimplement all of the manually indexing logic for each KVS backend we end up supporting.... | 19:43 |
| *** henrynash has joined #openstack-dev | 19:43 | |
| *** cp16net|away is now known as cp16net | 19:43 | |
| termie | ayoung: you are going to anyway once you write an optimized version | 19:43 |
| termie | ayoung: you will index differently on redis than on riak than on cassandra | 19:43 |
| termie | ayoung: the point of kvs is to know what your db has to do | 19:44 |
| *** pfreund has quit IRC | 19:44 | |
| termie | ayoung: _somebody_ has to make those indexes | 19:44 |
| *** jcmartin has quit IRC | 19:44 | |
| termie | ayoung: and if you aren't thinking about them then you aren't thinking about your data model | 19:44 |
| *** jcmartin has joined #openstack-dev | 19:44 | |
| termie | ayoung: which is, apparently, the default thing to not think about every time somebody has an idea | 19:44 |
| termie | ayoung: (not keystone specifically, though obviously it has been a culprit too) | 19:45 |
| *** markmcclain has joined #openstack-dev | 19:45 | |
| termie | ayoung: Learn Databases The Hard Way | 19:45 |
| ayoung | termie, OK, point taken...with the caveat that all the pure KVS like memcached (and I don't include cassandra in there) are going to require the same logic to build the indexes. | 19:45 |
| termie | ayoung: again, no, cassandra has its own secondary indexes, the backend implementation is already using them | 19:46 |
| ayoung | right | 19:46 |
| ayoung | but memcached does not. | 19:46 |
| termie | ayoung: i could have built my own, if i did it would be in a very specific cassandra way that uses wide columns | 19:46 |
| *** mscott has quit IRC | 19:46 | |
| ayoung | I'm assumine that the nosqls in general will have their own way to index. | 19:46 |
| termie | ayoung: for memcache "listing" is sort of a silly prospect altogether | 19:46 |
| *** vipul|away is now known as vipul | 19:46 | |
| ayoung | but that is different from straight KVS. | 19:47 |
| termie | ayoung: because the data is non-persistent | 19:47 |
| termie | ayoung: so you should throw notimplementeds | 19:47 |
| termie | ayoung: and not support pki and whatnot | 19:47 |
| termie | ayoung: a certain amount of stuff can be put into kvs to make it easier to write the backends | 19:47 |
| termie | ayoung: just saying the actual indexing steps are likely going to be implemented differently each time | 19:48 |
| ayoung | termie, even if we use memcached in non-persistant mode, we need a way to be able to revoke tokens based on rules. To do that, we need to find the tokens. EIther we build special index records, or we slice up the keys | 19:48 |
| termie | ayoung: if you use memcache your revokation list can go away at any point, it's not worth it | 19:48 |
| termie | ayoung: you would have to use a real database to store all your stuff | 19:49 |
| ayoung | termie, actually, it is trusts that will make revocation lists go away, but for a different reason | 19:49 |
| termie | ayoung: it is an ephemeral backend, it is a fine cache | 19:49 |
| termie | ayoung: that is unrelated, don't muddy the convo | 19:49 |
| ayoung | trusts will mean we can have short lived tokens | 19:49 |
| ayoung | right | 19:49 |
| termie | ayoung: point is, don't use memcache for things it isn't designed for and you will be happy | 19:50 |
| *** marun has quit IRC | 19:50 | |
| termie | ayoung: and don't shove list operations into KVSs ;) | 19:50 |
| ayoung | termie but that kindof goes for storing tokens in KVS stores across the board | 19:50 |
| termie | ayoung: that goes for everything, actually, because it is a pretty general statement | 19:50 |
| ayoung | termie, which is why I was thinking more in terms of jettisoning kvs than in limiting what we can do with tokens | 19:50 |
| termie | ayoung: what i am trying to get at is that it is better to ditch sql than to ditch kvss ;) | 19:51 |
| ayoung | termie, if tokens live, say, a minute, and we don't have to deal with revocation events, then, yet, memcached it appropriate | 19:51 |
| termie | ayoung: because people writing sql can't be trusted to know what they are doing | 19:51 |
| ayoung | Ha! | 19:51 |
| *** otherwiseguy has quit IRC | 19:51 | |
| ayoung | termie, that goes double for people that choose nosql, non-transactional databases for integrity dependent complex data.... | 19:52 |
| ayoung | bascially, people can't be trusted anyway | 19:52 |
| termie | ayoung: truth, though the definition of "integrity dependent" is highly variable | 19:52 |
| ayoung | fair enough...and since we have multiple back ends, we can't really do foreign keys. | 19:53 |
| termie | ayoung: designing your system to not need "integrity" is usually a better goal than trying to have it | 19:53 |
| *** stevebaker has quit IRC | 19:53 | |
| *** zodiak has joined #openstack-dev | 19:53 | |
| termie | anyway, shame on you | 19:54 |
| termie | but it's fixable | 19:54 |
| termie | and i am not going to fix it until after cassandra goes in | 19:54 |
| ayoung | termie, thing is, it is revocations that make the backend for tokens so painful. If we could do away with revocations, the backend should get simple again | 19:54 |
| termie | ayoung: i would have preferred them to be in largely separate APIs, honestly | 19:54 |
| *** garyk has joined #openstack-dev | 19:54 | |
| termie | ayoung: putting two shapes of pegs into one round hole | 19:55 |
| termie | ayoung: is not good for code consistency | 19:55 |
| ayoung | http://saudadebrothers.files.wordpress.com/2011/12/yin-yang.jpg | 19:55 |
| termie | ayoung: *(can work fine if you planned for multiple shapes to be going in there) | 19:55 |
| termie | ayoung: i would have been happy with an extension that supported it with its own backend | 19:56 |
| ayoung | termie, for revocations? | 19:56 |
| termie | ayoung: basically intercept the request if it looks like pki | 19:56 |
| ayoung | termie, has nothing to do with PKI. Predates it | 19:56 |
| openstackgerrit | A change was merged to openstack/nova: Update to Quantum Client 2.2.0 https://review.openstack.org/23943 | 19:56 |
| ayoung | if a user changes their password, revoke their tokens started it. | 19:57 |
| ayoung | THere you list tokens by user | 19:57 |
| *** pfreund has joined #openstack-dev | 19:57 | |
| termie | ayoung: easier to check the token against a hash of the password | 19:57 |
| ayoung | which was no-opped for the kvs and memcached backends, which basically rendered them unusable | 19:57 |
| ayoung | termie, are you suggesing that the key in the backend should have been a hash of the password? | 19:58 |
| openstackgerrit | A change was merged to openstack/nova: Add missing tests for db.block_device_mapping_* methods https://review.openstack.org/24146 | 19:58 |
| termie | ayoung: changing large amounts of data at write time < ignoring things at read time | 19:58 |
| *** marun has joined #openstack-dev | 19:58 | |
| termie | ayoung: that'd be one implementation, not making any overt suggestions at this point isnce this is the first i heard of it | 19:58 |
| *** stevebaker has joined #openstack-dev | 19:59 | |
| termie | ayoung: but that is one easy technique for solving the "user changed some setting affecting possibly hundreds of rows" | 19:59 |
| *** jbr_ has joined #openstack-dev | 19:59 | |
| termie | denormalize the check | 20:00 |
| termie | so you change one piece of data atomically | 20:00 |
| *** stevebaker_ has joined #openstack-dev | 20:00 | |
| ayoung | termie, so the solution for memcached and for kvs is basically, have a special index record. And that logic can be generalized, I think. | 20:00 |
| ayoung | Which is, I think, really what you are suggesting. | 20:00 |
| ayoung | So you pay the cost at write time, but not at read time. | 20:00 |
| *** stevebaker has quit IRC | 20:01 | |
| ayoung | But it does mean you have to get the locking correct when you do a revocation or you might end up with a race condition, but that is true regardless | 20:01 |
| termie | ayoung: sorry walked away, but no not really | 20:02 |
| *** stevebaker_ is now known as stevebaker | 20:02 | |
| termie | ayoung: this isn't a kvs specific solution, it is just a data model solution | 20:03 |
| ayoung | termie, here's the example I was thinking of | 20:03 |
| *** joesavak has joined #openstack-dev | 20:03 | |
| termie | ayoung: i think list_tokens is possibly useful from a user experience perspective so people can see which tokens they have in existence, though i would rather not bother | 20:03 |
| openstackgerrit | A change was merged to openstack/python-novaclient: Removes tenant IDs checking for nova quota operations. https://review.openstack.org/24314 | 20:04 |
| crazed | topol: hey, i got fairly far to the point where the live ldap tests run, but haven't had much time to fix the failing/erroring ones, i have some output here: http://paste.openstack.org/show/33493/, hoping to pick it up again after i finish some work | 20:04 |
| termie | ayoung: but the main point is doign a list to find a bunch fo tokens and revoke them is slow | 20:04 |
| openstackgerrit | A change was merged to openstack/nova: List ComputeHostNotFound as a client exception https://review.openstack.org/24322 | 20:04 |
| ayoung | termie, agreed, but as you point out, that is something that can be optimized. I just have not seen sufficient reason to optimize it in the KVS backend,. as I would not put that backend into production | 20:04 |
| termie | ayoung: where if you had a hash of the password stored near the token you would only have to check that they match to decide on validity | 20:04 |
| ayoung | memcached, on the other hand, probably should be optimized | 20:04 |
| topol | crazed, OK sounds good Ill take a look | 20:05 |
| termie | ayoung: memcache doens't need to be optimized, just used correctly | 20:05 |
| termie | ayoung: not everything supports every use case | 20:05 |
| termie | ayoung: for the most part you probably shouldn't use the memcache backend | 20:05 |
| termie | ayoung: but it is fine for a small simple system | 20:05 |
| *** esp1 has quit IRC | 20:06 | |
| ayoung | termie, so the reason people want the memcached backedn is because it peridoically cleans up the records, where as the sql one does not. THe sql one could be cleaned up by an async process, but that is not part of the api right now | 20:06 |
| termie | ayoung: as far as "optimizing" the cehck for the kvs backend, it isn't that it is non-optimal right now it is that it is literally not possible for any real kvs the way it is implemented | 20:06 |
| *** otherwiseguy has joined #openstack-dev | 20:06 | |
| termie | ayoung: cassandra happens to do that ;) | 20:06 |
| ayoung | the idea being that it is better to leave the records there for an audit trail | 20:06 |
| termie | ayoung: they are tokens, fuck audit trails | 20:06 |
| ayoung | termie, no no no no no no non nein nyet | 20:06 |
| termie | ayoung: you keep user names with the actions on the services | 20:06 |
| *** jsavak has quit IRC | 20:07 | |
| *** kagan has quit IRC | 20:07 | |
| ayoung | audit trails are not necessary for all deployements | 20:07 |
| ayoung | but for some they are essential | 20:07 |
| termie | ayoung: and the services can make all teh audit trails they want, keeping expired tokens is not part of that | 20:07 |
| ayoung | termie, we'd still need an automated task to clean up the expired tokens in SQL. | 20:08 |
| termie | ayoung: regardless, of auditing, which is beside the point of there not being a list operation on KVSs | 20:08 |
| *** esp1 has joined #openstack-dev | 20:08 | |
| termie | ayoung: and of sql cleanup processes | 20:08 |
| *** mrodden1 has joined #openstack-dev | 20:08 | |
| *** mrodden has quit IRC | 20:08 | |
| ayoung | termie, that was the reason for people wanting to use memcached | 20:08 |
| termie | ayoung: doing the check at access rather than attempting to mutate a bunch of data is usually a good idea | 20:08 |
| ayoung | I think it makes sense as a non-persisted token cached, and it is part of the major distros | 20:08 |
| *** glikson has joined #openstack-dev | 20:09 | |
| ayoung | termie, storing the password hash in a token wouldn't solve all the revocation use cases. We also revoke when role assignements change. | 20:10 |
| termie | hash the roles too? | 20:10 |
| termie | shorter lived tokens would alleviate a decent amount of friction here, why are they so long again? | 20:11 |
| ayoung | perhaps. You could design it other ways, but I'm not convinced that working around the limitations of kvs is should be driving the design decisions | 20:12 |
| *** matiu has joined #openstack-dev | 20:12 | |
| *** matiu has quit IRC | 20:12 | |
| *** matiu has joined #openstack-dev | 20:12 | |
| termie | ayoung: well be convinced | 20:12 |
| ayoung | termie, ah , good question | 20:12 |
| ayoung | they are long because of long running processes....something that can be alleviated | 20:12 |
| *** SergeyLukjanov has quit IRC | 20:12 | |
| termie | ayoung: How I Learned to Stop Worrying and Love Doing Things The Hard Way | 20:12 |
| ayoung | If a token is need for, say migrating a virtual machine from one compute node to another | 20:13 |
| ayoung | and it was doing something like | 20:13 |
| ayoung | shutting down the first, waitign until it was all the way off, bringing up the second, making sure it was all the way up.... | 20:13 |
| ayoung | and you wanted to do this with one token | 20:13 |
| *** sgordon has quit IRC | 20:13 | |
| termie | ayoung: i see | 20:14 |
| ayoung | you need to make sure that token is still valid at the end of the last action, as the user is not going to be involved to get another token an hour later... | 20:14 |
| ayoung | but...that is where trusts come in | 20:14 |
| ayoung | trusts are an agreement that another user can get a token as me at some time in the future | 20:14 |
| termie | ayoung: fine, you can talk about that now if you'd like | 20:14 |
| ayoung | so say this happens a month from now, as part of an High availability agreement | 20:14 |
| ayoung | I create a trust with the the user that the HA process is running as the trustee. I am the trustor. | 20:15 |
| ayoung | A month from now, user HA gets a token as ayoung. BUt that token is limited to a very specific role in a specific project. | 20:15 |
| termie | this is a third-party HA process, i assume | 20:15 |
| ayoung | termie, correct | 20:15 |
| *** lcheng has joined #openstack-dev | 20:16 | |
| ayoung | termie, so say we make a role which is something like "turn of vm" and another which is "turn on vm" | 20:16 |
| *** boden has quit IRC | 20:16 | |
| vishy | nati_ueno: https://bugs.launchpad.net/nova/+bug/1112912 should I move the nova side to havana also? | 20:16 |
| uvirtbot | Launchpad bug 1112912 in quantum "get_firewall_required should use VIF parameter from quantum" [High,In progress] | 20:16 |
| termie | i assume one of those is "off" ? | 20:16 |
| ayoung | we set up the trust so that the HA user only gets those two roles. Limits exposure. | 20:16 |
| ayoung | yeah | 20:17 |
| nati_ueno | vishy: Yes. Thanks | 20:17 |
| * ayoung needs typing lessons | 20:17 | |
| *** rcj has quit IRC | 20:17 | |
| ayoung | termie, so the idea is that anything that is going to require long term access to credential can be migrated to using a trust instead. | 20:17 |
| ayoung | A trust can have a limited time span, so you can say "make this trust only last an hour" and that should handle most of the use cases we currently have where horizon talks to nova which talks to glance...or something like that. | 20:18 |
| ayoung | Then the tokens themselves can last a very short time, just long enough to kick off the processes. And revocation is no longer an issue. | 20:19 |
| *** dprince has quit IRC | 20:19 | |
| termie | ayoung: well, i'm not keen on your terminology (big surprise) but i would probably implement this as delegated auth with a limited scope | 20:19 |
| termie | ayoung: in current implementation that is just a token with a limited role set | 20:20 |
| ayoung | termie, my day job needs my attention. Back in a bit | 20:20 |
| termie | ayoung: in another implementation i'd love to have the ha service provide me with a key that i add a user 'HA' for to my project and give appropraite roles | 20:20 |
| *** dprince has joined #openstack-dev | 20:21 | |
| *** digitalsanctum has joined #openstack-dev | 20:22 | |
| *** stevebaker_ has joined #openstack-dev | 20:25 | |
| *** stevebaker_ has quit IRC | 20:25 | |
| *** torandu has joined #openstack-dev | 20:25 | |
| *** stevebaker_ has joined #openstack-dev | 20:26 | |
| *** stevebaker has quit IRC | 20:27 | |
| *** colinmcnamara has quit IRC | 20:28 | |
| sandeepr | ayoung, !ping | 20:30 |
| *** adjohn has quit IRC | 20:30 | |
| sandeepr | termie, !ping | 20:31 |
| *** adjohn has joined #openstack-dev | 20:31 | |
| termie | sandeepr: why are you talking to me | 20:33 |
| termie | sandeepr: i'm just a heckler | 20:33 |
| termie | sandeepr: this will only lead to sadness | 20:34 |
| sandeepr | termie, i am talking to you since you were talking to ayoung | 20:34 |
| termie | sandeepr: do you have something i can use against him? | 20:34 |
| sandeepr | :-) i needed to get the PKI stuff working for me to continue testing | 20:35 |
| sandeepr | and it is not | 20:35 |
| sandeepr | with UUID it works | 20:35 |
| termie | ayoung: YOU BROKE EVERYTHING | 20:35 |
| termie | sandeepr: there, i helped | 20:35 |
| sandeepr | so i'm little lost how to get this sorted | 20:35 |
| *** adjohn has quit IRC | 20:36 | |
| *** adjohn has joined #openstack-dev | 20:37 | |
| *** stevebaker has joined #openstack-dev | 20:37 | |
| dolphm | ayoung: you have a -2 on a RC1-blocker that i think is ready to go https://review.openstack.org/#/c/23979/ | 20:37 |
| dolphm | ayoung: you -2'd on a lack of test coverage; new test is pretty extensive | 20:38 |
| *** stevebaker_ has quit IRC | 20:38 | |
| ayoung | dolphm, yes, yes it is | 20:38 |
| *** sandywalsh has quit IRC | 20:39 | |
| ayoung | dolphm, is the logic right? | 20:39 |
| ayoung | # roles should be the union of the two role lists | 20:39 |
| dolphm | ayoung: yes, plus the member_role_id added in migration 17 i think | 20:39 |
| ayoung | yes | 20:40 |
| ayoung | dolphm, removed my -2....taking some time to grok this.... | 20:40 |
| *** jsavak has joined #openstack-dev | 20:40 | |
| dolphm | ayoung: cool, take your time | 20:41 |
| dolphm | ayoung: definitely want your eyes on that one | 20:41 |
| markmcclain | mikal: ping | 20:42 |
| ayoung | dolphm, OK, that looks good. | 20:43 |
| ayoung | dolphm, is he around? | 20:43 |
| *** bknudson has joined #openstack-dev | 20:43 | |
| mikal | markmcclain: what's up? | 20:44 |
| *** joesavak has quit IRC | 20:44 | |
| markmcclain | can https://review.openstack.org/#/c/24350/ be approved? | 20:45 |
| mikal | markmcclain: sure, I was just waiting a little longer to let others comment, but that one is pretty safe. | 20:46 |
| *** mrodden has joined #openstack-dev | 20:46 | |
| markmcclain | no problem… wasn't sure if it was an oversight or you want more feedback | 20:46 |
| dolphm | ayoung: he? | 20:47 |
| ayoung | dolphm, the patch author | 20:48 |
| ayoung | jjcannava I guess | 20:48 |
| ayoung | not | 20:48 |
| *** jcannava has joined #openstack-dev | 20:49 | |
| markmcclain | mikal: thanks | 20:49 |
| dolphm | ayoung: jcannava: ^ | 20:49 |
| mikal | markmcclain: NP, I was just being conservative it being so close to RC1 and all | 20:50 |
| ayoung | jcannava, care to talk about https://review.openstack.org/#/c/23979/ | 20:50 |
| markmcclain | understandable | 20:50 |
| jcannava | ayoung sure | 20:50 |
| *** mrodden1 has quit IRC | 20:50 | |
| ayoung | OK...so, would you be willing to be on removing the metadata table now? | 20:51 |
| ayoung | to bet on | 20:51 |
| ayoung | jcannava, I can see what you are doing, and I think it is the right thing.... | 20:52 |
| openstackgerrit | A change was merged to openstack/nova: Virtual Power Driver list running vms quoting error https://review.openstack.org/24236 | 20:52 |
| *** radez is now known as radez_g0n3 | 20:52 | |
| ayoung | and if it is, we should not need the metadata table anymore | 20:52 |
| ayoung | but..are you 100% sure of that? | 20:52 |
| ayoung | I mean, you wrote it, so I am guessing that you are. Can you justify that conclusion? | 20:53 |
| *** CaptTofu has joined #openstack-dev | 20:53 | |
| * ayoung has to leave, though, to pick up my son...just noticed the time. Dagnabit | 20:53 | |
| jcannava | ayoung ok so i'm confident that the migration works and the roles will be in the right place | 20:54 |
| dolphm | ayoung: ping me later about it | 20:54 |
| *** ayoung is now known as ayoung-afk | 20:54 | |
| dolphm | gyee: easy RC1 blocker https://review.openstack.org/#/c/24202/ | 20:54 |
| dolphm | henrynash: ^ | 20:54 |
| *** fire has joined #openstack-dev | 20:54 | |
| *** sandeepr has quit IRC | 20:54 | |
| jcannava | do I need to keep the metadata table around for safety I think is the question, i'm not opposed to leaving it to be sure | 20:54 |
| openstackgerrit | A change was merged to openstack/oslo-incubator: update OpenStack, LLC to OpenStack Foundation https://review.openstack.org/24350 | 20:56 |
| dolphm | jcannava: the table is no longer referenced by the codebase, it's expected to be gone | 20:56 |
| wkelly | and the information in the new table would be what you want to be in the metadata table should you roll the database back to a previous state | 20:56 |
| wkelly | to make sure that you keep role changes on rollback | 20:56 |
| *** stevebaker has quit IRC | 20:57 | |
| wkelly | which seems to be in place in downgrade | 20:57 |
| *** stevebaker has joined #openstack-dev | 20:57 | |
| openstackgerrit | A change was merged to openstack/keystone: Filter out legacy_endpoint_id (bug 1152635) https://review.openstack.org/23934 | 20:59 |
| uvirtbot | Launchpad bug 1152635 in keystone "legacy_endpoint_id returned on v3" [Medium,In progress] https://launchpad.net/bugs/1152635 | 20:59 |
| openstackgerrit | A change was merged to openstack/keystone: Ensure tokens are revoked for relevant v3 api calls https://review.openstack.org/24078 | 20:59 |
| *** henrynash has quit IRC | 21:00 | |
| *** giulivo has quit IRC | 21:04 | |
| *** henrynash has joined #openstack-dev | 21:04 | |
| *** henrynash has quit IRC | 21:06 | |
| *** stevebaker has quit IRC | 21:07 | |
| *** stevebaker has joined #openstack-dev | 21:07 | |
| *** digitals_ has joined #openstack-dev | 21:08 | |
| gyee | dolphm, question on the links | 21:08 |
| gyee | so the links inside an element have no previous or next attributes right? | 21:08 |
| *** jsavak has quit IRC | 21:08 | |
| *** martine_ has quit IRC | 21:09 | |
| *** digitalsanctum has quit IRC | 21:10 | |
| dolphm | gyee: no, but they could have other attributes that we haven't defined yet | 21:14 |
| *** john5223 has quit IRC | 21:14 | |
| dolphm | gyee: for example, a user object could have links to "projects" and "domains" for the things they can auth against | 21:14 |
| gyee | k | 21:14 |
| dolphm | gyee: next/previous are currently only defined for paged collections though -- but we're not paging yet | 21:15 |
| gyee | I am trying to figure out how to proper handle them | 21:15 |
| *** jruzicka has quit IRC | 21:15 | |
| gyee | properly | 21:15 |
| dolphm | gyee: if they're null in JSON, they shouldn't appear in XML, as there's not really a null equivalent in xml | 21:16 |
| gyee | k | 21:16 |
| dolphm | gyee: was that the concern? | 21:16 |
| gyee | no, I am trying to figure out what the rel | 21:17 |
| gyee | like rel="previous" | 21:17 |
| gyee | besides previous, next, and self, what other rel do we need to worry about | 21:17 |
| *** sandywalsh has joined #openstack-dev | 21:18 | |
| *** jbr_ has left #openstack-dev | 21:18 | |
| *** digitals_ is now known as digitalsanctum | 21:18 | |
| gordc | dolphm,gyee: i'm sitting in on ceilometer meetings right now, would you guys know when keystoneclient release is? | 21:18 |
| *** marun has quit IRC | 21:18 | |
| *** marun has joined #openstack-dev | 21:19 | |
| gyee | gordc, I bet dolphm knows the answer | 21:19 |
| dolphm | gordc: i've been following along, i'm looking to see if there's any reason why we couldn't do one immediately | 21:19 |
| dolphm | gyee: it just won't have v3 auth support yet :( | 21:19 |
| gyee | doh! | 21:19 |
| gordc | dolphm:yeah, i was going to mention that. i don't think ceilometer uses v3 auth though (don't know a lot of the code though) | 21:20 |
| gyee | dolphm, why we need to cut a release now? | 21:20 |
| *** henrynash has joined #openstack-dev | 21:22 | |
| dolphm | gyee: we're blocking as a result of https://bugs.launchpad.net/python-keystoneclient/+bug/1151665 | 21:22 |
| uvirtbot | Launchpad bug 1151665 in python-keystoneclient "Pypi python-keystoneclient(v0.2.2) not compatible with the latest oslo.config-1.1" [Critical,Confirmed] | 21:22 |
| *** jbresnah has quit IRC | 21:22 | |
| *** digitalsanctum has quit IRC | 21:23 | |
| gyee | i c | 21:23 |
| *** stevebaker has quit IRC | 21:24 | |
| *** cp16net is now known as cp16net|away | 21:24 | |
| *** cp16net|away is now known as cp16net | 21:26 | |
| *** davidhadas has quit IRC | 21:28 | |
| *** david-lyle has quit IRC | 21:29 | |
| *** davidha has joined #openstack-dev | 21:30 | |
| *** imsplitbit has quit IRC | 21:34 | |
| dolphm | gordc: is the current oslo.config tarball sufficient, or should we merge https://review.openstack.org/#/c/24228/ prior to tagging? | 21:34 |
| *** davidha has quit IRC | 21:35 | |
| *** timello has quit IRC | 21:36 | |
| gordc | not sure myself. i know ceilometer just merged in the 1.1 final update yesterday | 21:36 |
| dolphm | gordc: it's going through gating for keystone itself as well | 21:37 |
| dolphm | nijaba: ^ | 21:38 |
| nijaba | dolphm: thanks a lot | 21:38 |
| dolphm | nijaba: i meant the question above :) | 21:38 |
| dolphm | henrynash: same as the one on keystone, but for keystoneclient https://review.openstack.org/#/c/24228/ | 21:39 |
| henrynash | dolphm: too slow :-) | 21:39 |
| dolphm | henrynash: haha thanks anyway (thanks gyee!) | 21:40 |
| gyee | henrynash, btw, did you have a chance to investigate the Swift ACL impact with non globally unique names? | 21:41 |
| henrynash | gyee: so I did, yes…and sent findings to the swift guys to get them to confirm their findings…let me copy you as I chase them for confirmation that we ware ok | 21:41 |
| gyee | I was going to take a look, but don't want to overlap with your work | 21:42 |
| gordc | dolphm,gyee,henrynash: thanks for pushing it through. | 21:42 |
| mordred | this oslo.config rollout is killin gme | 21:42 |
| *** llu-laptop has joined #openstack-dev | 21:43 | |
| llu-laptop | dolphm: could you repost your question? I just joined | 21:44 |
| *** mtreinish has quit IRC | 21:44 | |
| llu-laptop | sorry for that | 21:44 |
| dolphm | mordred: do i have permission to push a git tag, or do we need heckj for that? | 21:45 |
| dolphm | mordred: (keystoneclient) | 21:45 |
| mordred | dolphm: I believe we need heckj for that ... looking | 21:45 |
| dolphm | llu-laptop: was wondering if we needed to wait for https://review.openstack.org/#/c/24228/ | 21:45 |
| mordred | dolphm: (or try it, it won't break if you don't have permission) | 21:45 |
| mordred | dolphm: just make sure to do git tag -s to sign it | 21:45 |
| *** alop has quit IRC | 21:46 | |
| dolphm | mordred: k, i'll try once the above review merges | 21:46 |
| llu-laptop | dolphm: it's merged, right? | 21:46 |
| dolphm | llu-laptop: approved but waiting for gate | 21:47 |
| llu-laptop | dolphm: I think it'd better to include that so all openstack service are now using the same version of oslo | 21:48 |
| *** glikson has quit IRC | 21:48 | |
| dolphm | mordred: git tag -s 0.2.3 && git push gerrit master <-- is master correct for a client tag? | 21:49 |
| dolphm | llu-laptop: agree; i wasn't sure based on the discussion if ya'll were specifically looking for a release with the tarball as a dep or not | 21:50 |
| mordred | dprince: nope | 21:50 |
| mordred | dolphm: git tag -s 0.2.3 && git push --tags gerrit | 21:50 |
| ttx | mordred: if that doesn't work I can probably help | 21:50 |
| mordred | ttx: aha! | 21:51 |
| dprince | mordred: did I ask a question sir? | 21:52 |
| dolphm | ttx: thanks :) | 21:52 |
| mordred | dprince: nope. dolphm did and my tab-complete went wild | 21:52 |
| ttx | dolphm: want me to tag something ? | 21:52 |
| termie | ayoung-afk: added responses on https://review.openstack.org/#/c/24284/ | 21:52 |
| dprince | mordred: cool. carry on. | 21:52 |
| ttx | dolphm: python-keystoneclient d782a998474d92d4299b4404b69442f0288efc3b -> 0.2.3 ? | 21:53 |
| openstackgerrit | A change was merged to openstack/tempest: Fix skip formats to trigger skip_tracker. https://review.openstack.org/24349 | 21:53 |
| openstackgerrit | A change was merged to openstack/nova: Make run_instance() bail quietly if instance has been deleted https://review.openstack.org/24243 | 21:53 |
| openstackgerrit | A change was merged to openstack/nova: Fix use of potentially-stale instance_type in tenant_usage https://review.openstack.org/24343 | 21:53 |
| openstackgerrit | A change was merged to openstack/nova: Refactor work with session in db.block_device_mapping_* methods https://review.openstack.org/24147 | 21:53 |
| openstackgerrit | A change was merged to openstack/glance: Fix circular import in glance/db/sqlalchemy https://review.openstack.org/24171 | 21:54 |
| ttx | looks like quantum rc1 is still pending a few things, so I have free time tonight :) | 21:54 |
| openstackgerrit | A change was merged to openstack/python-novaclient: Use keyring for testing https://review.openstack.org/24065 | 21:54 |
| openstackgerrit | A change was merged to openstack/glance: Fix check_003 method. https://review.openstack.org/23844 | 21:54 |
| ttx | zuul in merging frenzy | 21:54 |
| *** maoy has quit IRC | 21:55 | |
| openstackgerrit | A change was merged to openstack/quantum: Replace "OpenStack LLC" with "OpenStack Foundation" https://review.openstack.org/24326 | 21:55 |
| openstackgerrit | A change was merged to openstack/glance: Declare index on ImageMember model https://review.openstack.org/24351 | 21:55 |
| ttx | dolphm: make up your mind, the shop closes in 2 min. | 21:55 |
| openstackgerrit | A change was merged to openstack/swift: Add a region tier to Swift's ring. https://review.openstack.org/23733 | 21:55 |
| mordred | dude. that's a nice pile of merge there | 21:56 |
| ttx | wow, did someone disable tests in gating or what | 21:56 |
| *** alunduil has quit IRC | 21:57 | |
| *** sacharya has quit IRC | 21:59 | |
| *** ewindisch has quit IRC | 21:59 | |
| *** colinmcnamara has joined #openstack-dev | 21:59 | |
| *** pcm__ has quit IRC | 22:00 | |
| asalkeld | ttx / mordred is this ok? https://review.openstack.org/#/c/24293 | 22:00 |
| asalkeld | should we propose a change to openstack/requirements? | 22:00 |
| openstackgerrit | A change was merged to openstack/quantum: drop rfc.sh https://review.openstack.org/24355 | 22:01 |
| openstackgerrit | A change was merged to openstack/nova: Don't load system_metadata when it isn't joined. https://review.openstack.org/24221 | 22:01 |
| vishy | jgriffith: man i'm tired of this hang on dd issue | 22:04 |
| vishy | jgriffith: can't clone a volume without freezing my vm | 22:04 |
| *** stevebaker_ has joined #openstack-dev | 22:06 | |
| termie | jeblair: heya, wanna install cassandra on our jenkins machines? | 22:07 |
| termie | jeblair: i'll give you a hug or altenatively _not_ give you a hug next time i see you | 22:07 |
| *** kbringard has quit IRC | 22:09 | |
| * ttx goes to bed | 22:09 | |
| jgriffith | vishy: I hear ya... update the LVM driver | 22:11 |
| jgriffith | vishy: That's the only reliable solution I've come up with | 22:11 |
| jgriffith | vishy: and looking at the last update from sebastian I'm not holding my breath on a better option | 22:11 |
| llu-laptop | dolphm: sorry for the late response. our bottom line would be to have a pypi verison python-keystoneclient compatible with oslo.config 1.1.0 | 22:12 |
| jgriffith | vishy: I've tried so many different things but I've had ZERO luck | 22:12 |
| *** kaushikc has quit IRC | 22:12 | |
| mordred | termie: is cassandra actually installable these days? last I checked the process was "do java things, pray, cray, drink" | 22:15 |
| *** zzs has joined #openstack-dev | 22:15 | |
| dolphm | ttx: no worries | 22:15 |
| termie | mordred: on my mac i just typed "brew install cassandra" followed by a short wait, then type "cassandra" and ran the tests | 22:15 |
| mordred | termie: interesting. | 22:15 |
| dolphm | llu-laptop: i'll try a tag as soon as https://review.openstack.org/#/c/24228/ is merged (it's in progress) | 22:16 |
| *** alop has joined #openstack-dev | 22:16 | |
| mordred | termie: that's definitely less involved than before | 22:16 |
| termie | mordred: i didn't see a package for it on my ancient debian box | 22:16 |
| termie | mordred: but i suspect there is an apt repo you add nowadays | 22:16 |
| llu-laptop | dolphm: thanks! tagging means pypi version ready? | 22:16 |
| mordred | termie: jeblair is out today - I'll poke him (or you can) tomorrow | 22:16 |
| termie | mordred: http://java.dzone.com/articles/installing-apache-cassandra looks pretty straightforward | 22:17 |
| termie | mordred: basically adding an apt source and their key | 22:17 |
| mordred | cool. that's not terrible | 22:17 |
| termie | in the meantime i am looking into a mock implementation, but the ones i've found are pretty old and don't support the features we use | 22:18 |
| termie | but since i think we are already testing multiple versions of sql this doesn't seem like a stretch | 22:18 |
| *** stevebaker_ has quit IRC | 22:20 | |
| henrynash | dolphm, gyee, young: LDAP testing change from Brad probably ready to be given the push: https://review.openstack.org/#/c/24023/ | 22:21 |
| *** gordc has quit IRC | 22:24 | |
| *** monst_ has quit IRC | 22:24 | |
| *** ayoung-afk is now known as ayoung | 22:25 | |
| *** ewindisch has joined #openstack-dev | 22:26 | |
| *** topol has quit IRC | 22:26 | |
| ayoung | dolphm, jcannava approved | 22:26 |
| *** eharney has quit IRC | 22:29 | |
| *** datsun180b has quit IRC | 22:30 | |
| *** hemna is now known as hemnafk | 22:31 | |
| *** koolhead17 has quit IRC | 22:32 | |
| *** edmund has quit IRC | 22:32 | |
| *** ewindisch has quit IRC | 22:32 | |
| *** shardy has quit IRC | 22:33 | |
| gyee | henrynash, ldap doesn't support update project name? | 22:36 |
| henrynash | gyye: I think it does, I just don't think it supports you updating it once you create a project | 22:37 |
| *** stevebaker has joined #openstack-dev | 22:38 | |
| *** monst_ has joined #openstack-dev | 22:40 | |
| *** bdpayne has quit IRC | 22:42 | |
| *** henrynash has quit IRC | 22:43 | |
| *** mkollaro has quit IRC | 22:44 | |
| *** bknudson has quit IRC | 22:47 | |
| *** alunduil has joined #openstack-dev | 22:47 | |
| *** pabelanger has quit IRC | 22:48 | |
| openstackgerrit | A change was merged to openstack/nova: Fix: Nova aggregate API throws an uncaught exception on invalid host https://review.openstack.org/24187 | 22:49 |
| openstackgerrit | A change was merged to openstack/keystone: Switch to final 1.1.0 oslo.config release https://review.openstack.org/24202 | 22:49 |
| gyee | dolphm, ayoung, https://review.openstack.org/#/c/24290/ | 22:49 |
| *** stevebaker has quit IRC | 22:50 | |
| *** jbresnah has joined #openstack-dev | 22:52 | |
| openstackgerrit | A change was merged to openstack/swift: Fix swift-ring-builder's list_parts command. https://review.openstack.org/24361 | 22:53 |
| openstackgerrit | A change was merged to openstack/swift: Add missing CONTRIBUTING.md to source tarball https://review.openstack.org/24300 | 22:53 |
| openstackgerrit | A change was merged to openstack/python-keystoneclient: Switch to final 1.1.0 oslo.config release https://review.openstack.org/24228 | 22:53 |
| *** enikanorov_ has quit IRC | 22:54 | |
| *** enikanorov has joined #openstack-dev | 22:54 | |
| *** kaushikc has joined #openstack-dev | 22:55 | |
| *** dontalton has quit IRC | 22:56 | |
| *** egallen has joined #openstack-dev | 22:57 | |
| *** kaushikc has left #openstack-dev | 22:59 | |
| *** woodspa has quit IRC | 23:02 | |
| vishy | dolphm, ayoung: ping | 23:02 |
| dolphm | vishy: o/ | 23:02 |
| vishy | dolphm: so I'm attempting to understand https://bugs.launchpad.net/nova/+bug/1154809 | 23:03 |
| uvirtbot | Launchpad bug 1154809 in nova "Volume detach fails via OSAPI: AmbiguousEndpoints" [Undecided,New] | 23:03 |
| vishy | it appears that somehow two endpoints are getting returned | 23:03 |
| vishy | did the endpoint format change somehow in the service catalog? | 23:04 |
| vishy | i see these things like legacy_endpoint_id | 23:04 |
| *** yjiang5 has quit IRC | 23:04 | |
| vishy | and i no longer see a publicURL and adminURL | 23:04 |
| *** kpavel has quit IRC | 23:04 | |
| vishy | dolphm: I"m also curious why serviceName would be None | 23:05 |
| dolphm | vishy: hmm... yes and no -- what's returned should be identical but the way endpoints are persisted has changed :( | 23:06 |
| vishy | dolphm: well something seems a bit different | 23:06 |
| dolphm | vishy: legacy_endpoint_id is an implementation detail meant to be hidden from the API | 23:06 |
| dolphm | vishy: which i just fixed ^ hopefully | 23:06 |
| dolphm | vishy: https://bugs.launchpad.net/keystone/+bug/1152635 | 23:07 |
| uvirtbot | Launchpad bug 1152635 in keystone "legacy_endpoint_id returned on v3" [Medium,Fix committed] | 23:07 |
| *** pholland has quit IRC | 23:07 | |
| vishy | ok well that catalog looks different | 23:07 |
| dolphm | vishy: but you're seeing that attribute on v2? | 23:07 |
| vishy | dolphm: well i don't know | 23:07 |
| vishy | dolphm: it looks like it might be requesting v3 api by accident | 23:07 |
| dolphm | vishy: you might be seeing this? https://bugs.launchpad.net/keystone/+bug/1152632 | 23:08 |
| uvirtbot | Launchpad bug 1152632 in keystone "null endpoints are not ignored by v3" [Low,In progress] | 23:08 |
| vishy | dolphm: the catalog is coming from auth_token_middleware | 23:08 |
| dolphm | vishy: if either of these two bugs have an impact on v2 i'm not aware of it | 23:08 |
| vishy | dolphm: does auth_token middleware still use v2? | 23:08 |
| dolphm | vishy: it was just upgraded to support v3 | 23:08 |
| vishy | dolphm: ok so that is what is causing the problem | 23:09 |
| *** dstufft has quit IRC | 23:09 | |
| vishy | how does one set it back to v2 | 23:09 |
| dolphm | vishy: https://github.com/openstack/python-keystoneclient/commit/d782a998474d92d4299b4404b69442f0288efc3b | 23:09 |
| *** dstufft has joined #openstack-dev | 23:09 | |
| vishy | dolphm: the issue is we need a service catalog in v2 format | 23:09 |
| vishy | in order to send it in properly to cinderclient | 23:10 |
| *** jaypipes has quit IRC | 23:10 | |
| dolphm | vishy: you can set auth_version in nova.conf -- not sure of the value for v2 (v2.0 / v2 / 2.0 / etc) | 23:10 |
| vishy | dolphm: unless I'm missing the problem | 23:11 |
| *** jbresnah has quit IRC | 23:11 | |
| vishy | it also seems odd that serviceName is not set | 23:11 |
| vishy | dolphm: so did the format of the serviceCatalog change from v2 -> v3? | 23:12 |
| dolphm | vishy: that definitely looks broken -- adding keystone to that bug | 23:12 |
| dolphm | vishy: yes | 23:12 |
| vishy | ok so can you find out what i need to set for v2 | 23:12 |
| vishy | like to post it as a workaround on that bug | 23:12 |
| dolphm | vishy: looks like auth_version = v2.0 | 23:12 |
| vishy | also curious why i don't see it in devstack | 23:13 |
| openstackgerrit | A change was merged to openstack/tempest: Add duplicate bug detection to skip_tracker.py https://review.openstack.org/24110 | 23:14 |
| dolphm | vishy: i'm running out the door -- gyee (implemented the v3 catalog) or henrynash (implemented the v3 auth_token support) might be able to help | 23:15 |
| dolphm | vishy: i added to keystone to the nova bug for the null service name at the very least -- i haven't noticed that | 23:15 |
| dolphm | vishy: i also have a partial fix for https://bugs.launchpad.net/keystone/+bug/1152632 -- hoping to push tonight/tomorrow | 23:16 |
| uvirtbot | Launchpad bug 1152632 in keystone "null endpoints are not ignored by v3" [Low,In progress] | 23:16 |
| vishy | dolphm: so there is a problem | 23:17 |
| vishy | ok | 23:17 |
| *** esp1 has quit IRC | 23:17 | |
| vishy | adam_g: ping | 23:19 |
| *** jbresnah has joined #openstack-dev | 23:19 | |
| *** monst_ has quit IRC | 23:20 | |
| openstackgerrit | A change was merged to openstack/oslo-config: Fix copyright - from LLC to Foundation https://review.openstack.org/24375 | 23:21 |
| *** stevebaker has joined #openstack-dev | 23:23 | |
| *** jergerber has quit IRC | 23:24 | |
| *** anteaya has left #openstack-dev | 23:25 | |
| termie | ayoung: ... wait you just added this "trust" thing? | 23:25 |
| *** riskable has quit IRC | 23:27 | |
| termie | ayoung: and a new argument to the driver driver definition? i don't think this is bueno | 23:28 |
| *** gongysh has joined #openstack-dev | 23:29 | |
| termie | ayoung: _and_ changed the behavior of null expires? | 23:29 |
| *** epim has quit IRC | 23:29 | |
| *** markwash has quit IRC | 23:29 | |
| termie | ayoung: i guess htat means i get to get rid of weird explicit edge case behavior to preserve stupid None expires | 23:30 |
| *** llu-laptop has quit IRC | 23:30 | |
| *** torandu has quit IRC | 23:31 | |
| *** dolphm has quit IRC | 23:31 | |
| *** zzs has quit IRC | 23:31 | |
| *** egallen has quit IRC | 23:32 | |
| adam_g | vishy, pong | 23:33 |
| vishy | adam_g: found a workaround for you | 23:34 |
| vishy | did you see? | 23:34 |
| adam_g | vishy, just got back to the keyboard. checking now | 23:34 |
| vishy | adam_g: workaround works for me | 23:36 |
| vishy | will propose it as a change for now | 23:36 |
| adam_g | vishy, yah, trying it now | 23:36 |
| adam_g | vishy, that works here, too. shioul | 23:37 |
| adam_g | vishy, er, i should note i'm using UUID tokens ATM, should i try PKI as well? | 23:38 |
| vishy | yeah | 23:38 |
| vishy | it doesn't affect pki right now | 23:38 |
| *** mscott has joined #openstack-dev | 23:38 | |
| *** Ritz-Mobile has quit IRC | 23:39 | |
| *** Ritz-Mobile has joined #openstack-dev | 23:40 | |
| adam_g | vishy, PKI looks okay here, too | 23:40 |
| *** cloudchimp has joined #openstack-dev | 23:40 | |
| *** woodspa has joined #openstack-dev | 23:43 | |
| *** rmohan has quit IRC | 23:43 | |
| *** markmcclain has quit IRC | 23:43 | |
| *** rmohan has joined #openstack-dev | 23:43 | |
| *** flaper87 has quit IRC | 23:43 | |
| termie | ayoung: so, how about we rip out all the trust stuff | 23:47 |
| termie | ayoung: feels last-minute | 23:49 |
| termie | ayoung: i know you're excited about it | 23:50 |
| *** Ritz-Mobile has quit IRC | 23:51 | |
| *** garyTh has quit IRC | 23:53 | |
| termie | ayoung: i think it's the wrong solution, and while i'm sure the first argument will be something along the lines of "well you should have paid attention earlier" i don't think bumping an entirely new concept in is a good idea | 23:54 |
| *** terryh has quit IRC | 23:54 | |
| *** vipul is now known as vipul|away | 23:54 | |
| *** novas0x2a|lapto1 has joined #openstack-dev | 23:56 | |
| *** novas0x2a|laptop has quit IRC | 23:56 | |
| *** gyee has quit IRC | 23:57 | |
| *** vipul|away is now known as vipul | 23:59 | |
| openstackgerrit | A change was merged to openstack/glance: Fix uniqueness constraint on image_members table. https://review.openstack.org/19547 | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
