| *** galstrom_zzz is now known as galstrom | 00:03 | |
| *** galstrom is now known as galstrom_zzz | 00:13 | |
| *** darrenc is now known as darrenc_afk | 00:14 | |
| *** galstrom_zzz is now known as galstrom | 00:16 | |
| *** galstrom is now known as galstrom_zzz | 00:19 | |
| *** darrenc_afk is now known as darrenc | 00:26 | |
| cloudnull | sigmavirus24_awa: its part of the lint check . | 00:28 |
|---|---|---|
| cloudnull | so its supposed to . | 00:28 |
| *** alop has quit IRC | 00:37 | |
| *** abitha has quit IRC | 00:44 | |
| *** sdake has joined #openstack-ansible | 00:57 | |
| *** sdake_ has quit IRC | 01:01 | |
| *** weezS has joined #openstack-ansible | 01:08 | |
| *** weezS has quit IRC | 01:13 | |
| *** sdake_ has joined #openstack-ansible | 01:19 | |
| *** sdake has quit IRC | 01:21 | |
| *** galstrom_zzz is now known as galstrom | 01:21 | |
| *** yaya has quit IRC | 01:32 | |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: Fixing protocol for UDP in horizon for the juno releases https://review.openstack.org/199603 | 01:38 |
| *** bitblt has quit IRC | 01:45 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 01:49 | |
| sigmavirus24 | cloudnull: it should have caught the problem in Hugh's review for adding federation bits then instead of mancdaz having to catch it | 01:49 |
| sigmavirus24 | Although gertty might just be broke | 01:49 |
| *** sdake_ has quit IRC | 02:07 | |
| *** galstrom is now known as galstrom_zzz | 02:10 | |
| *** sdake has joined #openstack-ansible | 02:15 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 02:17 | |
| *** yaya has joined #openstack-ansible | 02:22 | |
| cloudnull | sigmavirus24_awa: was it an issue in lint ? | 02:31 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Fixing protocol for UDP in horizon for master https://review.openstack.org/199613 | 02:33 |
| *** galstrom_zzz is now known as galstrom | 02:48 | |
| *** JRobinson__ is now known as JRobinson__afk | 02:51 | |
| *** annashen has joined #openstack-ansible | 03:17 | |
| *** jwagner is now known as jwagner_away | 03:23 | |
| *** galstrom is now known as galstrom_zzz | 04:08 | |
| *** markvoelker has quit IRC | 04:08 | |
| *** JRobinson__afk is now known as JRobinson__ | 04:43 | |
| *** sdake has quit IRC | 04:45 | |
| openstackgerrit | Charles Farquhar proposed stackforge/os-ansible-deployment: Fixed minor typo in etc/README.rst "Global Options" section https://review.openstack.org/199856 | 04:48 |
| *** annashen has quit IRC | 04:50 | |
| *** shausy has joined #openstack-ansible | 05:03 | |
| *** yaya has quit IRC | 05:41 | |
| *** fawadkhaliq has joined #openstack-ansible | 05:45 | |
| *** ig0r_ has joined #openstack-ansible | 05:52 | |
| *** ig0r__ has quit IRC | 05:55 | |
| *** markvoelker has joined #openstack-ansible | 06:00 | |
| *** sdake has joined #openstack-ansible | 06:03 | |
| *** markvoelker has quit IRC | 06:04 | |
| *** annashen_ has joined #openstack-ansible | 06:27 | |
| *** radek__ has joined #openstack-ansible | 06:31 | |
| *** annashen_ has quit IRC | 06:45 | |
| *** sdake has quit IRC | 07:03 | |
| *** JRobinson__ has quit IRC | 07:07 | |
| openstackgerrit | Darren Birkett proposed stackforge/os-ansible-deployment: allow innodb_buffer_pool_size to be configurable https://review.openstack.org/199546 | 07:35 |
| *** JRobinson__ has joined #openstack-ansible | 07:48 | |
| *** JRobinson__ has quit IRC | 07:50 | |
| *** markvoelker has joined #openstack-ansible | 08:01 | |
| *** markvoelker has quit IRC | 08:06 | |
| *** vdo has joined #openstack-ansible | 08:24 | |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Remove haproxy galera monitoring user https://review.openstack.org/199625 | 08:51 |
| *** fawadkhaliq has quit IRC | 09:22 | |
| *** markvoelker has joined #openstack-ansible | 10:02 | |
| *** radek__ has quit IRC | 10:05 | |
| *** markvoelker has quit IRC | 10:06 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Enable all services to use Keystone 'insecurely' https://review.openstack.org/199307 | 10:07 |
| *** radek__ has joined #openstack-ansible | 10:07 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Enable all services to use Keystone 'insecurely' https://review.openstack.org/199307 | 10:13 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust key distribution mechanism for Swift https://review.openstack.org/199992 | 10:14 |
| *** fawadkhaliq has joined #openstack-ansible | 10:22 | |
| *** fawadkhaliq has quit IRC | 10:23 | |
| *** fawadkhaliq has joined #openstack-ansible | 10:23 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Ensure that flake8 runs against all python files https://review.openstack.org/200006 | 10:41 |
| *** sdake has joined #openstack-ansible | 10:45 | |
| *** sdake_ has joined #openstack-ansible | 10:46 | |
| *** sdake has quit IRC | 10:50 | |
| *** fawadkhaliq has quit IRC | 11:06 | |
| *** fawadkhaliq has joined #openstack-ansible | 11:08 | |
| *** markvoelker has joined #openstack-ansible | 11:33 | |
| *** sdake_ has quit IRC | 11:33 | |
| *** markvoelker has quit IRC | 11:37 | |
| *** sdake has joined #openstack-ansible | 11:42 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Upgrade the Keystone library to use v3 https://review.openstack.org/196943 | 11:44 |
| mattt | anyone know how we're going to handle https://mail.google.com/mail/#search/label%3Alists-openstack-operators+semver/14dfdfcd2857ff14 ? | 11:44 |
| *** jaypipes has joined #openstack-ansible | 11:45 | |
| mattt | sorry http://lists.openstack.org/pipermail/openstack-operators/2015-June/007390.html is probably more appropriate | 11:45 |
| *** sdake_ has joined #openstack-ansible | 11:56 | |
| *** sdake has quit IRC | 12:00 | |
| odyssey4me | mattt destroy and rebuild the containers, and remove the packages from the hosts and re-install them... perhaps that could be done in the containers, but it seems a waste | 12:01 |
| odyssey4me | that seems to me the best way to make sure that we never have to be concerned with it again | 12:01 |
| odyssey4me | mattt hughsaunders can we get https://review.openstack.org/200006 in asap to prevent any future issues? | 12:02 |
| mancdaz | mattt http://legacy.python.org/dev/peps/pep-0440/#adding-version-epochs | 12:03 |
| mancdaz | http://legacy.python.org/dev/peps/pep-0440/#version-epochs | 12:03 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Ensure that flake8 runs against all python files https://review.openstack.org/200006 | 12:06 |
| *** markvoelker has joined #openstack-ansible | 12:12 | |
| *** fawadkhaliq has quit IRC | 12:22 | |
| *** sdake has joined #openstack-ansible | 12:28 | |
| *** sdake_ has quit IRC | 12:31 | |
| odyssey4me | mattt that should perhaps be added to the agenda for the meeting later today | 12:35 |
| mattt | can one of you raise it? i won't be attending the meeting | 12:36 |
| *** jmckind has quit IRC | 12:36 | |
| odyssey4me | mattt yeah, I'll add it cc mancdaz | 12:39 |
| mattt | thanks ! | 12:40 |
| *** fawadkhaliq has joined #openstack-ansible | 12:45 | |
| openstackgerrit | git-harry proposed stackforge/os-ansible-deployment: Restart mysql when config changed https://review.openstack.org/200054 | 12:45 |
| *** pserebryakov has joined #openstack-ansible | 12:47 | |
| *** tlian has joined #openstack-ansible | 12:51 | |
| *** jaypipes has quit IRC | 12:56 | |
| *** fawadkhaliq has quit IRC | 13:01 | |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust key distribution mechanism for Swift https://review.openstack.org/199992 | 13:09 |
| *** jaypipes has joined #openstack-ansible | 13:10 | |
| *** shausy has quit IRC | 13:16 | |
| dstanek | odyssey4me: the security comment is interesting. what is openrc used for? | 13:18 |
| odyssey4me | dstanek I would imagine it's predominantly used for administrative purposes. | 13:19 |
| odyssey4me | some of the ansible tasks use it too | 13:19 |
| *** jmckind has joined #openstack-ansible | 13:22 | |
| dstanek | odyssey4me: it's odd that it has it's own auth url because in theory it can be secure and the other two not | 13:23 |
| openstackgerrit | git-harry proposed stackforge/os-ansible-deployment: Restart mysql when config changed https://review.openstack.org/200054 | 13:24 |
| odyssey4me | dstanek sure, although all I'm trying to do there is implement a best guess based on the information at hand | 13:25 |
| odyssey4me | a deployer can easily override the setting if desired | 13:25 |
| dstanek | odyssey4me: no i think your patch does what it needs to. it's the overall setup that i find confusing | 13:26 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Enable all services to use Keystone 'insecurely' https://review.openstack.org/199307 | 13:33 |
| odyssey4me | dstanek so we drop openrc on every openstack service container and on every openstack host - this is primarily to facilitate administration, testing and troubleshooting - but we also have the bad habit (I think) of consuming the openrc in some of the tasks which execute in the container being setup | 13:35 |
| odyssey4me | does that make things clearer? otherwise perhaps you can be more specific about what you find confusing :) | 13:35 |
| dstanek | clear as mud | 13:37 |
| dstanek | since some tasks may use that openrc should the auth url be based on one of the existing vars instead or hard coded separately? | 13:38 |
| *** TheIntern has joined #openstack-ansible | 13:42 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 13:47 | |
| openstackgerrit | David Alfano proposed stackforge/os-ansible-deployment: Sort dynamic inventory files in juno https://review.openstack.org/199229 | 13:49 |
| odyssey4me | dstanek so the role has a basic default, while we override that default in user_group_vars.yml - etc/openstack_deploy/user_group_vars.yml:openrc_os_auth_url: "{{ keystone_service_internalurl }}" | 13:52 |
| odyssey4me | sigmavirus24 yeah, you're right - they actually aren't dependant... I mainly did that in order to ensure that they merge cleanly because they both affect the same files | 13:53 |
| sigmavirus24 | odyssey4me: sure, I get that, but that's not really what dependencies semantically mean | 13:54 |
| * sigmavirus24 shrugs | 13:54 | |
| sigmavirus24 | I'm not going to undo it | 13:54 |
| sigmavirus24 | I'd just rather we not repeat that in the future | 13:54 |
| odyssey4me | sure, was there a different way that I could have done the same thing? | 13:54 |
| odyssey4me | have you seen https://bugs.launchpad.net/openstack-ansible/+bug/1470635 ? | 13:54 |
| openstack | Launchpad bug 1470635 in puppet-keystone "endpoints added with v3 are not visible with v2" [High,Confirmed] | 13:54 |
| *** pserebryakov has quit IRC | 14:00 | |
| odyssey4me | sigmavirus24 due to ^ do you think that perhaps we should create all endpoints using the v2 api instead? | 14:02 |
| sigmavirus24 | odyssey4me: I had noticed that when testing | 14:02 |
| sigmavirus24 | It stems from the fact that endpoints are no longer a trio but instead are individual endpoints for (public, internal, admin) | 14:03 |
| sigmavirus24 | that's what "interface" is when you talk about an endpoint (public, internal, admin) | 14:03 |
| sigmavirus24 | I'm not sure I totally grok why needing them to be accessible via v2 is important though | 14:03 |
| cloudnull | morning | 14:03 |
| odyssey4me | morning cloudnull | 14:04 |
| *** Mudpuppy has joined #openstack-ansible | 14:05 | |
| odyssey4me | sigmavirus24 hmm, quite right - perhaps we should just document this and move on? | 14:05 |
| sigmavirus24 | odyssey4me: cloudnull is there a compelling reason for having that accessible through v2? | 14:05 |
| sigmavirus24 | Although | 14:06 |
| sigmavirus24 | I guess the way the module works makes it so that users have to have v3 enabled and the primary way they consume Keystone | 14:06 |
| cloudnull | what ? | 14:06 |
| odyssey4me | cloudnull https://bugs.launchpad.net/openstack-ansible/+bug/1470635 | 14:06 |
| openstack | Launchpad bug 1470635 in puppet-keystone "endpoints added with v3 are not visible with v2" [High,Confirmed] | 14:06 |
| odyssey4me | which affects https://review.openstack.org/196943 | 14:07 |
| alextricity | Hey guys, I won't be able to make the IRC meeting today. I just wanted to mention that I need a few more reviews for ceilometer | 14:07 |
| alextricity | https://review.openstack.org/#/c/173067/ | 14:07 |
| sigmavirus24 | L273 https://review.openstack.org/#/c/196943/27/playbooks/library/keystone only ever imports Keystone v3 (we were previously only importing v2) | 14:07 |
| stevelle | thanks alextricity | 14:07 |
| palendae | alextricity: Did you get your oslo.config problem figured out? | 14:07 |
| sigmavirus24 | If a deployer wants to primarily use v2 though, they can't use any osad that includes 196943 | 14:07 |
| alextricity | palendae: I havn't been able to look at it :/ I got pulled in to a last minute deployment | 14:08 |
| palendae | Ok | 14:08 |
| sigmavirus24 | We'd have to refactor the keystone module again in case people want to configure osad to use v2 | 14:08 |
| sigmavirus24 | Right now they'll only ever be able to use keystone v3 which all the services will be fine | 14:08 |
| odyssey4me | sigmavirus24 I think I'm with you on this one - we document that we've switched to using the v3 api and that all administrative access must now use that API and the openstack client | 14:08 |
| cloudnull | I think it has to do with the domains where v2 compat assumes the default domain. dstanek: dolphm: lbragstad: please correct me if im wrong , but i think thats whats up | 14:08 |
| sigmavirus24 | odyssey4me: yeah, that means people have to rewrite anything else they're using though | 14:08 |
| sigmavirus24 | cloudnull: no this is about endpoints | 14:09 |
| sigmavirus24 | endpoints in v3 can consume endpoints created with the v2 api, but not vice versa | 14:09 |
| sigmavirus24 | (endpoints created with the v3 API cannot be consumed via the v2 API) | 14:09 |
| cloudnull | oh. | 14:09 |
| cloudnull | i see now . | 14:09 |
| cloudnull | i have no idea | 14:10 |
| odyssey4me | it does result in the keystone CLI simply failing | 14:11 |
| odyssey4me | so it forces the use of the openstack CLI instead | 14:11 |
| cloudnull | does the openstack client work ? | 14:12 |
| odyssey4me | cloudnull yes, but only when setting the API version to 3 | 14:12 |
| odyssey4me | perhaps we should alias the CLI to something that produces a message saying that the only supported CLI is now the openstack CLI ? | 14:13 |
| cloudnull | can we use the keystone lib to return the endpoints with both v2/3 ? | 14:13 |
| odyssey4me | otherwise we'll get some very confused deployers spamming us with bugs | 14:13 |
| odyssey4me | cloudnull the issue is not the endpoint versions, but rather how the endpoints were created | 14:14 |
| cloudnull | i think current versions of the keystone client have a cli deprecation warning | 14:14 |
| lbragstad | odyssey4me: doing some digging | 14:14 |
| odyssey4me | this is what you get from keystone's CLI: | 14:14 |
| odyssey4me | 'python-keystoneclient.', DeprecationWarning) | 14:14 |
| odyssey4me | WARNING: unsupported identity-api-version 3, falling back to 2.0 | 14:14 |
| odyssey4me | Authorization Failed: The resource could not be found. (HTTP 404) (Request-ID: req-b873b20d-027f-4128-9ca0-741f4a1ffcdb) | 14:14 |
| cloudnull | odyssey4me: right im curious if its a client failing to parse data that keystone is returning | 14:14 |
| sigmavirus24 | cloudnull: a 404 would be a 404 from Keystone iirc | 14:14 |
| odyssey4me | /usr/local/lib/python2.7/dist-packages/keystoneclient/shell.py:64: DeprecationWarning: The keystone CLI is deprecated in favor of python-openstackclient. For a Python library, continue using python-keystoneclient. | 14:15 |
| lbragstad | odyssey4me: https://github.com/openstack/keystone/blob/692326610cdedbed9d66515619b668bdc16075dc/keystone/catalog/controllers.py#L74 | 14:15 |
| cloudnull | ha ^ | 14:15 |
| cloudnull | so thats why | 14:15 |
| odyssey4me | lbragstad heh, so this was definitely by design | 14:15 |
| lbragstad | odyssey4me: :) looks like it, | 14:15 |
| cloudnull | move along citizen, nothing to see here. | 14:15 |
| sigmavirus24 | lol | 14:16 |
| lbragstad | odyssey4me: doing some git blaming, hoping to find someone with a little more context | 14:16 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust swift_rings.py to work on specified regions https://review.openstack.org/200114 | 14:16 |
| lbragstad | lol | 14:16 |
| lbragstad | https://github.com/openstack/keystone/commit/2f851340ee8969193b9dcc1913401aa9b33c5d97 | 14:16 |
| lbragstad | dun dun dun! | 14:17 |
| cloudnull | dolphm: ^ | 14:17 |
| * sigmavirus24 makes a condescending "tsk tsk tsk" noise at dolphm | 14:17 | |
| sigmavirus24 | (in jest of course) | 14:17 |
| cloudnull | lololol | 14:18 |
| odyssey4me | hehehe, I would expect that operators would not agree with this design decision | 14:18 |
| cloudnull | fail forward . | 14:18 |
| odyssey4me | well, so what do we do about it? | 14:18 |
| odyssey4me | I think that perhaps we should add a follow-on patch which adds a little script that outputs a message saying naughty, naughty... move along to the openstack CLI now - then alias all the legacy CLI's to that script | 14:20 |
| lbragstad | odyssey4me: that doesn't seem like a bad idea | 14:20 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Ensure that flake8 runs against all python files https://review.openstack.org/200006 | 14:21 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Ensure that flake8 runs against all python files https://review.openstack.org/200122 | 14:22 |
| odyssey4me | cloudnull considering that we're hoping to merge the v3 API change into 11.1, do you think that doing something like that is a bit too much of a change? | 14:23 |
| *** jmckind has quit IRC | 14:23 | |
| odyssey4me | perhaps we should just have a specific known issues note and move on | 14:23 |
| cloudnull | i dont know that theres much we can do to help deployers in this space. it may simply be something that we add to the known issues. | 14:24 |
| cloudnull | i mean we can set the v3 api identity env var in the openrc | 14:25 |
| cloudnull | and doc that people should use the openstack client | 14:25 |
| odyssey4me | cloudnull part 1 done | 14:25 |
| *** jwagner_away is now known as jwagner | 14:26 | |
| *** jmckind has joined #openstack-ansible | 14:27 | |
| sigmavirus24 | odyssey4me: chat with stevemar before doing that fwiw | 14:27 |
| sigmavirus24 | he's an openstackclient core and would probably tell you which clients you can't do that with | 14:27 |
| *** jwagner is now known as jwagner_away | 14:28 | |
| *** jwagner_away is now known as jwagner | 14:31 | |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Add v3 calls for federation to keystone module https://review.openstack.org/197677 | 14:36 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Upgrade the Keystone library to use v3 https://review.openstack.org/196943 | 14:37 |
| odyssey4me | sigmavirus24 cloudnull I've related the patch to the bug and added a note in the commit message about it, and added the DocImpact tag too | 14:38 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Add v3 calls for federation to keystone module https://review.openstack.org/197677 | 14:44 |
| cloudnull | odyssey4me: LGTM | 14:48 |
| *** shausy has joined #openstack-ansible | 14:50 | |
| odyssey4me | cloudnull any thoughts on https://review.openstack.org/199307 and https://review.openstack.org/194474 | 14:55 |
| cloudnull | those are on my review list today | 14:57 |
| *** hiddentoken has joined #openstack-ansible | 14:58 | |
| *** galstrom_zzz is now known as galstrom | 14:58 | |
| cloudnull | i looked at the miguelgrinberg ssl changes and they looked good . | 14:58 |
| cloudnull | theres been some changes in the last couple of patch sets and so im going to apply / give it another go today. | 14:59 |
| cloudnull | i'ce not looked at https://review.openstack.org/#/c/199307/ yet, but today for sure. | 14:59 |
| *** jmckind has quit IRC | 15:00 | |
| *** sdake_ has joined #openstack-ansible | 15:01 | |
| odyssey4me | thanks :) | 15:04 |
| *** sdake has quit IRC | 15:05 | |
| *** sdake_ has quit IRC | 15:09 | |
| *** sdake has joined #openstack-ansible | 15:10 | |
| cloudnull | also reviews on https://review.openstack.org/#/c/173067/ need to be prioritized soon. | 15:11 |
| cloudnull | alextricity: ^ lgtm. deploys and i have pretty graphs in horizon so well done. | 15:11 |
| alextricity | woot | 15:12 |
| odyssey4me | cloudnull lemme get an AIO running for that | 15:13 |
| odyssey4me | it seems that everyone is kinda allergic to it :p | 15:13 |
| *** weezS has joined #openstack-ansible | 15:13 | |
| cloudnull | it seems that way . | 15:14 |
| cloudnull | but theres at least one new reviewer on it . | 15:15 |
| mattt | bah, i've been meaning to test this | 15:15 |
| cloudnull | so progress ? :) | 15:15 |
| cloudnull | another one that should help neutron performance https://review.openstack.org/#/c/196699/ | 15:15 |
| odyssey4me | cloudnull git-harry's queried some things on that patch | 15:17 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Add v3 calls for federation to keystone module https://review.openstack.org/197677 | 15:20 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Added openstack_kernel options for gc_thresh https://review.openstack.org/196699 | 15:20 |
| cloudnull | odyssey4me: yup. just updated that. | 15:20 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: [WIP] Keystone IdP configuration https://review.openstack.org/194259 | 15:23 |
| odyssey4me | svg can you peek at https://review.openstack.org/196699 and +1 it if you're happy that it does fix the problem you reported? | 15:27 |
| *** hiddentoken has quit IRC | 15:27 | |
| *** jmckind has joined #openstack-ansible | 15:29 | |
| *** shausy has quit IRC | 15:36 | |
| svg | odyssey4me: done | 15:37 |
| odyssey4me | svg thanks - did you apply those settings only to some hosts, or to all your hosts? | 15:38 |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: Upgrade the Keystone library to use v3 https://review.openstack.org/196943 | 15:40 |
| svg | odyssey4me: to all hosts, given that is needed - AFAIK and at least in our setup - on all (metal) hosts where a neutron agent runs, which is on all controllers and all computes | 15:40 |
| odyssey4me | thanks svg :) | 15:42 |
| svg | np :) | 15:42 |
| *** daneyon has joined #openstack-ansible | 15:48 | |
| cloudnull | tyvm svg ! | 15:51 |
| *** jaypipes has quit IRC | 15:53 | |
| cloudnull | hughsaunders: merge conflict in https://review.openstack.org/#/c/197677/ and its dep is out of date so i cant auto rebase it | 15:53 |
| hughsaunders | cloudnull: yep | 15:53 |
| hughsaunders | just testing my manual merge | 15:54 |
| cloudnull | ok | 15:54 |
| *** klindgren has joined #openstack-ansible | 15:54 | |
| *** alop has joined #openstack-ansible | 15:57 | |
| cloudnull | meeting time in a few min cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, mancdaz, dolphm, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung | 15:59 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Add v3 calls for federation to keystone module https://review.openstack.org/197677 | 15:59 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: [WIP] Keystone IdP configuration https://review.openstack.org/194259 | 16:00 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: [WIP] Keystone SP configuration https://review.openstack.org/194395 | 16:00 |
| *** bitblt has joined #openstack-ansible | 16:03 | |
| *** fawadkhaliq has joined #openstack-ansible | 16:11 | |
| *** TheIntern has quit IRC | 16:15 | |
| *** hachi has joined #openstack-ansible | 16:20 | |
| *** hachi has quit IRC | 16:27 | |
| *** vdo has quit IRC | 16:27 | |
| *** annashen has joined #openstack-ansible | 16:38 | |
| *** bitblt has quit IRC | 16:45 | |
| odyssey4me | sigmavirus24 was your agreement with git-harry including his comment here? https://review.openstack.org/#/c/199307/8/etc/openstack_deploy/user_group_vars.yml,cm | 16:49 |
| odyssey4me | I'm kinda inclined to agree. cloudnull your thoughts? | 16:49 |
| sigmavirus24 | odyssey4me: I'm not sure what harry means by that comment | 16:50 |
| odyssey4me | ie instead of using logic to determine whether the insecure flag should be set, rather set it manually | 16:51 |
| cloudnull | i agree with that | 16:51 |
| odyssey4me | alright, I'll change that up - please add your agreement to the review for record's sake | 16:53 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Added openstack_kernel options for gc_thresh https://review.openstack.org/200179 | 16:53 |
| odyssey4me | wohoo alextricity https://review.openstack.org/173067 is on its way to merging | 16:56 |
| sigmavirus24 | odyssey4me: yeah that works for me | 16:56 |
| alextricity | awesome! | 16:56 |
| sigmavirus24 | alextricity: A+ | 16:56 |
| alextricity | Thanks everyone! :) | 16:56 |
| alextricity | Does it just need a few more non-core reviews? | 16:57 |
| sigmavirus24 | For the tokyo summit I think OSAD should make badges like the ones HP made for Vancouver (but better) | 16:57 |
| sigmavirus24 | alextricity: nope | 16:57 |
| sigmavirus24 | It's on its way to merge city | 16:57 |
| sigmavirus24 | Where the grass is green and the hills are hilly | 16:57 |
| alextricity | haha | 16:57 |
| *** jmckind has quit IRC | 17:01 | |
| *** TheIntern has joined #openstack-ansible | 17:04 | |
| *** alop has quit IRC | 17:05 | |
| *** jmckind has joined #openstack-ansible | 17:06 | |
| *** jmckind has quit IRC | 17:08 | |
| *** sdake has quit IRC | 17:16 | |
| *** jmckind has joined #openstack-ansible | 17:19 | |
| *** sdake has joined #openstack-ansible | 17:20 | |
| *** sdake_ has joined #openstack-ansible | 17:26 | |
| *** sdake has quit IRC | 17:29 | |
| *** sdake has joined #openstack-ansible | 17:30 | |
| *** sdake_ has quit IRC | 17:34 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: remove conntrackd package https://review.openstack.org/199574 | 17:50 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Implement Ceilometer https://review.openstack.org/173067 | 17:50 |
| *** alop has joined #openstack-ansible | 18:13 | |
| sigmavirus24 | odyssey4me: mind if I take over your patch? | 18:14 |
| odyssey4me | sigmavirus24 which one? | 18:15 |
| odyssey4me | this is so annoying: https://github.com/ansible/ansible/blob/v1.9.2-1/lib/ansible/module_utils/basic.py#L37-L39 | 18:17 |
| odyssey4me | it doesn't recognise True/False as booleans | 18:17 |
| odyssey4me | only lowercase | 18:17 |
| alextricity | They don't use .lower() later? | 18:18 |
| odyssey4me | odd, it appears that they do: https://github.com/ansible/ansible/blob/v1.9.2-1/lib/ansible/module_utils/basic.py#L1204-L1215 | 18:19 |
| odyssey4me | so the issue is in the module choices validation | 18:19 |
| *** sdake is now known as sdae | 18:19 | |
| *** sdae is now known as sdake | 18:19 | |
| odyssey4me | sigmavirus24 if it's the insecure patch, I've already fixed most of it | 18:20 |
| odyssey4me | just working through the boolean issue | 18:20 |
| sigmavirus24 | okay odyssey4me | 18:20 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Enable all services to use Keystone 'insecurely' https://review.openstack.org/199307 | 18:21 |
| alextricity | The only reason it might not be using .lower is if type(arg) is not in types.StringTypes | 18:22 |
| *** markvoelker has quit IRC | 18:24 | |
| *** markvoelker has joined #openstack-ansible | 18:24 | |
| *** jaypipes has joined #openstack-ansible | 18:27 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Enable all services to use Keystone 'insecurely' https://review.openstack.org/199307 | 18:27 |
| openstackgerrit | Miguel Grinberg proposed stackforge/os-ansible-deployment: SSL support for haproxy https://review.openstack.org/198957 | 18:29 |
| odyssey4me | alextricity nope, that doesn't work - so I've just added to the choices which is a bit of a hack | 18:29 |
| *** sdake_ has joined #openstack-ansible | 18:33 | |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: Upgrade the Keystone library to use v3 https://review.openstack.org/196943 | 18:36 |
| *** sdake has quit IRC | 18:36 | |
| *** sdake has joined #openstack-ansible | 18:37 | |
| odyssey4me | sigmavirus24 I'm done for the day, so you're welcome to improve it from here on. :) | 18:38 |
| *** sdake_ has quit IRC | 18:40 | |
| odyssey4me | sigmavirus24 blast, pep8 failures... shall I fix them quickly? | 18:42 |
| sigmavirus24 | odyssey4me: why don't you go relax? | 18:42 |
| sigmavirus24 | i'll take care of them | 18:43 |
| odyssey4me | sigmavirus24 cool, thanks :) | 18:43 |
| sigmavirus24 | no problem | 18:45 |
| sigmavirus24 | I'm just sitting here abandoning glance reviews waiting for things to go in osad | 18:45 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Updated default fernet key usage https://review.openstack.org/196499 | 18:48 |
| *** jmckind has quit IRC | 18:50 | |
| *** jmckind has joined #openstack-ansible | 19:00 | |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: Enable all services to use Keystone 'insecurely' https://review.openstack.org/199307 | 19:03 |
| *** rromans has quit IRC | 19:08 | |
| *** annashen has quit IRC | 19:11 | |
| *** yaya has joined #openstack-ansible | 19:12 | |
| *** markvoelker has quit IRC | 19:15 | |
| *** markvoelker has joined #openstack-ansible | 19:15 | |
| miguelgrinberg | alextricity: I'm getting a mongo error on my gate job, "couldn't connect to server x.x.x.x". Known problem? | 19:16 |
| *** TheIntern has quit IRC | 19:16 | |
| alextricity | If it's a slower server it might be trying to make the ceilometer database before the service comes up. | 19:16 |
| alextricity | Are you just running the gate-check-commit script? | 19:17 |
| miguelgrinberg | it's not me, it's jenkins: http://logs.openstack.org/57/198957/5/check/os-ansible-deployment-dsvm-check-commit/fe3d3b4/console.html | 19:17 |
| *** rromans has joined #openstack-ansible | 19:17 | |
| alextricity | We saw that a few weeks back, but I thought I fixed it. | 19:19 |
| miguelgrinberg | should we see if it does it again on a recheck? | 19:19 |
| alextricity | Can you confirm that IP? | 19:19 |
| alextricity | yeah | 19:19 |
| alextricity | We've had at least 5-8 successful gates with this patch set | 19:20 |
| alextricity | Not sure why it would be failing now | 19:20 |
| odyssey4me | alextricity can you unblock the gate by pushing a patch to disable it being part of the gate check, just until it can be worked out what the issue is | 19:22 |
| miguelgrinberg | okay, I'm giving it another go | 19:22 |
| odyssey4me | otherwise it's going to hold up other work | 19:22 |
| miguelgrinberg | hey odyssey4me, your comment regarding using internaluri_proto | 19:23 |
| odyssey4me | miguelgrinberg yep? | 19:23 |
| alextricity | odyssey4me: Sure, but how will it merge? | 19:23 |
| miguelgrinberg | we really want the public proto to be https and the internal proto to be http | 19:23 |
| miguelgrinberg | that's why I added public proto to user_group_vars | 19:23 |
| odyssey4me | miguelgrinberg but they share a port, so that's not possible | 19:24 |
| odyssey4me | the admin_proto needs to be http, I agree with that | 19:24 |
| odyssey4me | internal/public = :5000 and admin = :35357 | 19:24 |
| *** sdake has quit IRC | 19:25 | |
| miguelgrinberg | so that's the particular implementation that we have currently, but nothing prevents us from bypassing haproxy for the internal uri, or having two layers of haproxy. | 19:25 |
| miguelgrinberg | if we ever need to | 19:25 |
| miguelgrinberg | or actually this can probably be done with a single haproxy, just map the internal to another port | 19:26 |
| odyssey4me | yuk, non-standard ports confuse everything | 19:27 |
| odyssey4me | I'm not bent on it, which is why I didn't -1 the patch. It was just a comment as that is the current convention. | 19:27 |
| miguelgrinberg | if it's in the catalog, you shouldn't care. I'm not saying we should do it right now, but leave the door open | 19:27 |
| *** radek__ has quit IRC | 19:28 | |
| *** sdake has joined #openstack-ansible | 19:29 | |
| miguelgrinberg | odyssey4me: another option that you may like better is to combine the public and internal proto vars into one, but do this also in the keystone defaults | 19:32 |
| miguelgrinberg | at least that way it is clear that we intend to work together | 19:32 |
| *** radek__ has joined #openstack-ansible | 19:34 | |
| odyssey4me | miguelgrinberg that's a breaking change I'd rather not try to introduce at this stage - we've spent a lot of time on this stuff and need to move on to the federation bits now | 19:34 |
| odyssey4me | almost everything keys into using the internalURL endpoint for keystone in almost every playbook, that's the reason I suggested that way of doing it | 19:34 |
| *** annashen has joined #openstack-ansible | 19:35 | |
| odyssey4me | anyway, I'm out - train's arriving at my station | 19:35 |
| odyssey4me | chat tomorrow :) | 19:35 |
| miguelgrinberg | sure, good night | 19:35 |
| miguelgrinberg | alextricity: another failure :( | 19:35 |
| alextricity | miguelgrinberg: Aw man..why would this be happening now | 19:36 |
| alextricity | lol | 19:36 |
| alextricity | I'm going to submit a patch set to turn of ceilometer in the gate scripts | 19:37 |
| miguelgrinberg | I agree with odyssey4me, if you can make that not be part of the gate job then you can investigate without urgency | 19:37 |
| alextricity | miguelgrinberg: Yup. Working on it now | 19:37 |
| miguelgrinberg | awesome, thanks! | 19:37 |
| alextricity | But how would it merge? | 19:37 |
| alextricity | Force merge? | 19:38 |
| miguelgrinberg | I guess someone with +2 power will have to force it | 19:38 |
| palendae | alextricity: I'll +2 it to unblock gate | 19:38 |
| palendae | Not sure who else is around, maybe d34dh0r53 | 19:38 |
| palendae | Assuming of course, it doesn't get stuck, too | 19:38 |
| d34dh0r53 | yep, what you need? | 19:38 |
| palendae | d34dh0r53: alextricity will be submitting a patch to unbreak the gate. Appears ceilometer stuff is failing there | 19:39 |
| d34dh0r53 | ok | 19:39 |
| alextricity | I'm testing the changes now | 19:39 |
| d34dh0r53 | cool | 19:41 |
| *** annashen has quit IRC | 19:51 | |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: Upgrade the Keystone library to use v3 https://review.openstack.org/196943 | 19:51 |
| *** annashen has joined #openstack-ansible | 19:51 | |
| *** annashen has quit IRC | 19:56 | |
| *** palendae has quit IRC | 19:58 | |
| *** palendae has joined #openstack-ansible | 19:59 | |
| *** jmckind has quit IRC | 20:05 | |
| *** jmckind has joined #openstack-ansible | 20:07 | |
| *** daneyon has quit IRC | 20:14 | |
| *** sdake_ has joined #openstack-ansible | 20:15 | |
| *** sdake has quit IRC | 20:19 | |
| *** sdake_ is now known as sdake | 20:29 | |
| *** annashen has joined #openstack-ansible | 20:31 | |
| *** prometheanfire has quit IRC | 20:42 | |
| *** prometheanfire has joined #openstack-ansible | 20:42 | |
| *** wmlynch has quit IRC | 20:44 | |
| openstackgerrit | Miguel Alejandro Cantu proposed stackforge/os-ansible-deployment: Unblock gate by removing ceilometer from script. https://review.openstack.org/200252 | 20:45 |
| alextricity | d34dh0r53: palendae ^^ | 20:45 |
| *** prometheanfire has quit IRC | 20:45 | |
| alextricity | although this gate didn't fail | 20:45 |
| alextricity | http://logs.openstack.org/43/196943/31/check/os-ansible-deployment-dsvm-check-commit/949ad3a/console.html | 20:45 |
| *** prometheanfire has joined #openstack-ansible | 20:46 | |
| alextricity | well..the mongodb bits worked just fine here: https://review.openstack.org/#/c/196943/ | 20:46 |
| palendae | alextricity: Left a comment | 20:46 |
| alextricity | oh dang | 20:47 |
| alextricity | Thanks :) | 20:47 |
| *** jaypipes has quit IRC | 20:53 | |
| openstackgerrit | Miguel Alejandro Cantu proposed stackforge/os-ansible-deployment: Unblock gate by removing ceilometer from script. https://review.openstack.org/200252 | 20:57 |
| palendae | d34dh0r53: Yeah, the intent of the change is to not test ceilometer for now | 21:00 |
| d34dh0r53 | palendae: cool, just want to make sure | 21:00 |
| palendae | alextricity: patch set 2 doesn't remove the ansible.cfg lines | 21:01 |
| *** jwagner is now known as jwagner_away | 21:01 | |
| sigmavirus24 | We do eventually want to add ceilometer to the gate but perhaps when we have more people to investigate a reliable way to do that | 21:03 |
| palendae | Right | 21:04 |
| sigmavirus24 | Or perhaps as a separate non-voting (for now) job =P | 21:05 |
| d34dh0r53 | alextricity: log_path should be removed from ansible.cfg, it's added by the gate scripts | 21:05 |
| sigmavirus24 | alextricity: would you like a hand with that? | 21:05 |
| alextricity | got it! | 21:05 |
| sigmavirus24 | k | 21:05 |
| palendae | sigmavirus24: Not a bad idea to separate out some of these more specific scenarios. Like stand alone swift | 21:05 |
| sigmavirus24 | palendae: yeah | 21:05 |
| sigmavirus24 | palendae: makes it a bit nicer than fitting everything into one gate job | 21:06 |
| palendae | Yeah | 21:06 |
| sigmavirus24 | will allow things to finish faster too | 21:06 |
| palendae | Some things | 21:06 |
| sigmavirus24 | Yeah | 21:06 |
| palendae | Overall time will probably be increased | 21:06 |
| sigmavirus24 | It will | 21:06 |
| palendae | But that's the nature of adding more tests | 21:06 |
| sigmavirus24 | Especially when zuul is under heavy load | 21:06 |
| palendae | That need to set up a whole stack | 21:06 |
| *** jmckind has quit IRC | 21:06 | |
| sigmavirus24 | YOLO | 21:07 |
| palendae | Zuul we can't do much about | 21:07 |
| openstackgerrit | Miguel Alejandro Cantu proposed stackforge/os-ansible-deployment: Unblock gate by removing ceilometer from script. https://review.openstack.org/200252 | 21:07 |
| palendae | Or, we could probably join infra | 21:07 |
| sigmavirus24 | Thanks alextricity | 21:07 |
| sigmavirus24 | palendae: I'm like 50% of the way there | 21:07 |
| palendae | Of course you are | 21:08 |
| sigmavirus24 | Infra are good people | 21:08 |
| palendae | Soon there won't be an openstack project you don't contribute to | 21:08 |
| cloudnull | sorry ive been in hell today, but why are we doing this https://review.openstack.org/#/c/200252/ | 21:08 |
| palendae | cloudnull: miguelgrinberg and odyssey4me were reporting that it was consistently breaking the gate | 21:08 |
| cloudnull | in the last for gate jobs i see that they failed but not on mongo or ceilo bits | 21:09 |
| palendae | miguelgrinberg: ^^ | 21:10 |
| *** sigmavirus24 is now known as emacsraptor | 21:10 | |
| *** emacsraptor is now known as sigmavirus24 | 21:10 | |
| alextricity | sorry guys I have to head out. Are we okay with that patch set? | 21:10 |
| cloudnull | i see memcached failures from hpb4 | 21:10 |
| sigmavirus24 | alextricity: you have my +2 | 21:10 |
| cloudnull | keystone failures for mysql | 21:10 |
| cloudnull | etc... but nothing mongo related . | 21:10 |
| palendae | http://logs.openstack.org/57/198957/5/check/os-ansible-deployment-dsvm-check-commit/fe3d3b4/console.html#_2015-07-09_18_30_40_148 is what miguelgrinberg reported | 21:11 |
| cloudnull | but in the last job for that it failed here http://logs.openstack.org/57/198957/5/check/os-ansible-deployment-dsvm-check-commit/5f64fa5/console.html#_2015-07-09_20_09_45_363 | 21:12 |
| cloudnull | and thats a keystone / galera issue | 21:13 |
| *** yaya has quit IRC | 21:15 | |
| palendae | Heh, https://review.openstack.org/#/c/200252/ is failing on mongo anyway | 21:15 |
| palendae | http://logs.openstack.org/52/200252/3/check/os-ansible-deployment-dsvm-check-commit/91bb2a2/console.html#_2015-07-09_21_11_55_764 | 21:15 |
| cloudnull | right . . . | 21:15 |
| cloudnull | so it looks like theres a bug in the gate job and we should fix that | 21:15 |
| cloudnull | but it also looks like there are plenty of other issues that need to be fixed in those prs | 21:17 |
| openstackgerrit | David Wilde proposed stackforge/os-ansible-deployment: Parameterize galera slow/unindexed logging options https://review.openstack.org/200274 | 21:26 |
| *** annashen has quit IRC | 21:26 | |
| *** annashen has joined #openstack-ansible | 21:27 | |
| *** Mudpuppy has quit IRC | 21:29 | |
| *** TheIntern has joined #openstack-ansible | 21:30 | |
| *** sdake_ has joined #openstack-ansible | 21:39 | |
| *** radek__ has quit IRC | 21:40 | |
| *** sdake has quit IRC | 21:42 | |
| *** daneyon has joined #openstack-ansible | 21:46 | |
| *** auld has joined #openstack-ansible | 21:47 | |
| *** daneyon_ has joined #openstack-ansible | 21:49 | |
| *** daneyon has quit IRC | 21:52 | |
| *** javeriak has joined #openstack-ansible | 21:55 | |
| *** daneyon has joined #openstack-ansible | 21:56 | |
| openstackgerrit | David Wilde proposed stackforge/os-ansible-deployment: Parameterize galera slow/unindexed logging options https://review.openstack.org/200274 | 21:59 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 21:59 | |
| d34dh0r53 | palendae: sigmavirus24_awa fixerated my patch | 21:59 |
| *** daneyon_ has quit IRC | 22:00 | |
| palendae | d34dh0r53: kk | 22:00 |
| *** daneyon_ has joined #openstack-ansible | 22:01 | |
| *** TheIntern has quit IRC | 22:02 | |
| *** daneyon has quit IRC | 22:04 | |
| *** daneyon_ has quit IRC | 22:11 | |
| *** galstrom is now known as galstrom_zzz | 22:24 | |
| *** fawadkhaliq has quit IRC | 22:43 | |
| *** annashen has quit IRC | 22:43 | |
| *** annashen has joined #openstack-ansible | 22:44 | |
| *** annashen has quit IRC | 22:49 | |
| *** annashen has joined #openstack-ansible | 22:51 | |
| openstackgerrit | Miguel Grinberg proposed stackforge/os-ansible-deployment: SSL support for haproxy https://review.openstack.org/198957 | 23:04 |
| *** weezS has quit IRC | 23:12 | |
| *** sdake has joined #openstack-ansible | 23:24 | |
| *** sdake__ has joined #openstack-ansible | 23:26 | |
| *** sdake_ has quit IRC | 23:27 | |
| *** sdake has quit IRC | 23:29 | |
| *** sdake__ is now known as sdake | 23:30 | |
| *** annashen has quit IRC | 23:34 | |
| *** openstack has joined #openstack-ansible | 23:41 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!