| *** saurabhs has quit IRC | 00:05 | |
| *** saurabhs has joined #tripleo | 00:12 | |
| *** saurabhs has quit IRC | 00:13 | |
| *** akuznetsov has joined #tripleo | 00:24 | |
| *** chuckC_ has quit IRC | 00:26 | |
| *** akuznetsov has quit IRC | 00:28 | |
| *** spzala has joined #tripleo | 00:30 | |
| *** ChuckC has quit IRC | 00:36 | |
| sdake | chmouel nope, check out the launchpad - just assign yourself if you want to work on it | 00:37 |
|---|---|---|
| *** saurabhs has joined #tripleo | 00:39 | |
| *** yamahata has quit IRC | 00:41 | |
| *** spzala has quit IRC | 00:49 | |
| *** rainya has quit IRC | 00:53 | |
| *** dsneddon has quit IRC | 00:58 | |
| *** dsneddon has joined #tripleo | 01:00 | |
| *** ChuckC has joined #tripleo | 01:09 | |
| *** panda has quit IRC | 01:10 | |
| *** panda has joined #tripleo | 01:11 | |
| *** chuckC_ has joined #tripleo | 01:11 | |
| *** sdake_ has joined #tripleo | 01:13 | |
| *** akuznetsov has joined #tripleo | 01:24 | |
| *** yamahata has joined #tripleo | 01:26 | |
| *** sdake_ has quit IRC | 01:26 | |
| *** saurabhs has left #tripleo | 01:27 | |
| *** akuznetsov has quit IRC | 01:29 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/diskimage-builder: Fix $DISTRO_NAME usage for centos7 element https://review.openstack.org/125538 | 01:36 |
| openstackgerrit | Ryan Brady proposed a change to openstack/diskimage-builder: Fix $DISTRO_NAME usage for centos7 element https://review.openstack.org/125538 | 01:42 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Fix DISTRO_NAME usage for centos7 element https://review.openstack.org/125539 | 01:49 |
| *** eghobo has quit IRC | 02:11 | |
| *** nosnos has joined #tripleo | 02:14 | |
| *** sdake_ has joined #tripleo | 02:17 | |
| *** akuznetsov has joined #tripleo | 02:25 | |
| *** cwolferh_ has joined #tripleo | 02:28 | |
| *** akuznetsov has quit IRC | 02:29 | |
| *** sdake_ has quit IRC | 02:46 | |
| *** andreaf has quit IRC | 02:54 | |
| *** andreaf has joined #tripleo | 02:54 | |
| *** untriaged-bot has joined #tripleo | 03:00 | |
| untriaged-bot | Untriaged bugs so far: | 03:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1375414 | 03:00 |
| uvirtbot | Launchpad bug 1375414 in tripleo "mysql/rabbit overcloud issues when the bootstrap node is removed" [Undecided,New] | 03:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1376200 | 03:00 |
| uvirtbot | Launchpad bug 1376200 in tripleo "Stale connections when VIP is moved" [High,New] | 03:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1375422 | 03:00 |
| uvirtbot | Launchpad bug 1375422 in tripleo "Vlan not recreated on overcloud controller reboot" [Undecided,In progress] | 03:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1366164 | 03:00 |
| uvirtbot | Launchpad bug 1366164 in tripleo "Add domain roles for admin user" [Undecided,New] | 03:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1367742 | 03:00 |
| uvirtbot | Launchpad bug 1367742 in tripleo "keepalived brings the VIP up in multiple places on a controller reboot" [Undecided,New] | 03:00 |
| *** untriaged-bot has quit IRC | 03:00 | |
| *** openstackgerrit has quit IRC | 03:08 | |
| *** pcrews has quit IRC | 03:10 | |
| *** openstackgerrit has joined #tripleo | 03:10 | |
| rhallisey | larsks, thanks, must've forgotten it. I'll put it up there | 03:13 |
| *** jml has quit IRC | 03:14 | |
| *** jml has joined #tripleo | 03:16 | |
| *** openstackgerrit has quit IRC | 03:18 | |
| *** openstackgerrit has joined #tripleo | 03:18 | |
| *** akuznetsov has joined #tripleo | 03:26 | |
| *** sdake_ has joined #tripleo | 03:29 | |
| *** nosnos has quit IRC | 03:29 | |
| *** nosnos has joined #tripleo | 03:30 | |
| *** akuznetsov has quit IRC | 03:30 | |
| *** nosnos has quit IRC | 03:34 | |
| *** Yanyanhu has joined #tripleo | 03:52 | |
| *** cwolferh_ has quit IRC | 04:02 | |
| *** rushiagr_away is now known as rushiagr | 04:12 | |
| *** noslzzp has quit IRC | 04:18 | |
| *** nosnos has joined #tripleo | 04:19 | |
| *** tzumainn has quit IRC | 04:20 | |
| *** akuznetsov has joined #tripleo | 04:27 | |
| *** akuznetsov has quit IRC | 04:31 | |
| *** dshulyak_ has joined #tripleo | 04:31 | |
| *** dshulyak_ has quit IRC | 04:41 | |
| *** dshulyak_ has joined #tripleo | 04:58 | |
| *** dshulyak_ has quit IRC | 05:08 | |
| openstackgerrit | Richard Su proposed a change to openstack/diskimage-builder: Update EPEL release rpm https://review.openstack.org/125552 | 05:26 |
| *** akuznetsov has joined #tripleo | 05:27 | |
| *** akuznetsov has quit IRC | 05:32 | |
| *** Yanyanhu has quit IRC | 05:59 | |
| *** mrunge has joined #tripleo | 06:01 | |
| *** Yanyanhu has joined #tripleo | 06:06 | |
| *** jtomasek has joined #tripleo | 06:07 | |
| *** Yanyanhu has quit IRC | 06:22 | |
| *** dshulyak_ has joined #tripleo | 06:23 | |
| *** ramishra has joined #tripleo | 06:25 | |
| *** jprovazn has joined #tripleo | 06:26 | |
| *** akuznetsov has joined #tripleo | 06:28 | |
| *** Yanyanhu has joined #tripleo | 06:28 | |
| *** akuznetsov has quit IRC | 06:33 | |
| *** jtomasek has quit IRC | 06:33 | |
| StevenK | yuanying: Hi, can you abandon https://review.openstack.org/#/c/120689/ ? There's a button for it | 06:37 |
| *** sdake_ has quit IRC | 06:38 | |
| yuanying | StevenK: Sorry, I press button just now. | 06:38 |
| yuanying | Is it OK? | 06:39 |
| StevenK | yuanying: Yup, thanks. :-) | 06:39 |
| *** akuznetsov has joined #tripleo | 06:42 | |
| *** ramishra has quit IRC | 06:47 | |
| *** ifarkas has joined #tripleo | 06:50 | |
| *** dtantsur|afk is now known as dtantsur | 07:00 | |
| *** athomas has joined #tripleo | 07:11 | |
| *** Yanyanhu has quit IRC | 07:12 | |
| *** Yanyanhu has joined #tripleo | 07:22 | |
| *** rdopieralski has joined #tripleo | 07:22 | |
| openstackgerrit | Giulio Fidente proposed a change to openstack/diskimage-builder: Ensure epel7 is installed only on rhel7/centos7 https://review.openstack.org/125450 | 07:26 |
| *** jcoufal has joined #tripleo | 07:29 | |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Refresh progress page with AJAX https://review.openstack.org/125064 | 07:33 |
| *** jp_at_hp has joined #tripleo | 07:34 | |
| *** ramishra has joined #tripleo | 07:38 | |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Automatic plan saving and validation https://review.openstack.org/125004 | 07:41 |
| *** jistr has joined #tripleo | 07:42 | |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Removing image forms https://review.openstack.org/125080 | 07:44 |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Refresh progress page with AJAX https://review.openstack.org/125064 | 07:57 |
| *** ramishra has quit IRC | 08:00 | |
| *** Yanyanhu has quit IRC | 08:01 | |
| *** lsmola has joined #tripleo | 08:01 | |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Fix indentation for driver info on node detail page https://review.openstack.org/125393 | 08:06 |
| *** Yanyanhu has joined #tripleo | 08:08 | |
| *** dshulyak_ has quit IRC | 08:13 | |
| *** lucasagomes has joined #tripleo | 08:15 | |
| *** andreaf has quit IRC | 08:17 | |
| *** bkero has quit IRC | 08:21 | |
| *** yuanying has quit IRC | 08:23 | |
| *** cody-somerville has quit IRC | 08:24 | |
| *** bkero has joined #tripleo | 08:25 | |
| *** rushiagr is now known as rushiagr_away | 08:27 | |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Support passing x86_64 to create-nodes https://review.openstack.org/123803 | 08:29 |
| *** regebro has joined #tripleo | 08:30 | |
| *** rushiagr_away is now known as rushiagr | 08:33 | |
| *** Yanyanhu has quit IRC | 08:37 | |
| *** Yanyanhu has joined #tripleo | 08:38 | |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Remove outdated documentation https://review.openstack.org/125571 | 08:39 |
| *** yuanying has joined #tripleo | 08:39 | |
| *** akuznetsov has quit IRC | 08:41 | |
| *** shardy has joined #tripleo | 08:44 | |
| *** Yanyanhu has quit IRC | 08:45 | |
| *** andrearosa has quit IRC | 08:46 | |
| *** cody-somerville has joined #tripleo | 08:47 | |
| *** cody-somerville has quit IRC | 08:47 | |
| *** cody-somerville has joined #tripleo | 08:47 | |
| *** Yanyanhu has joined #tripleo | 08:48 | |
| *** greghaynes has quit IRC | 08:48 | |
| openstackgerrit | Tomas Sedovic proposed a change to openstack/python-tuskarclient: Fix the homepage url in setup.cfg https://review.openstack.org/125573 | 08:48 |
| *** greghaynes has joined #tripleo | 08:51 | |
| *** xuhaiwei_ has joined #tripleo | 08:51 | |
| *** pensu has joined #tripleo | 08:54 | |
| *** dshulyak_ has joined #tripleo | 08:55 | |
| *** untriaged-bot has joined #tripleo | 09:00 | |
| untriaged-bot | Untriaged bugs so far: | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1375414 | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1376200 | 09:00 |
| uvirtbot | Launchpad bug 1375414 in tripleo "mysql/rabbit overcloud issues when the bootstrap node is removed" [Undecided,New] | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1375422 | 09:00 |
| uvirtbot | Launchpad bug 1376200 in tripleo "Stale connections when VIP is moved" [High,New] | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1366164 | 09:00 |
| uvirtbot | Launchpad bug 1375422 in tripleo "Vlan not recreated on overcloud controller reboot" [Undecided,In progress] | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1367742 | 09:00 |
| uvirtbot | Launchpad bug 1366164 in tripleo "Add domain roles for admin user" [Undecided,New] | 09:00 |
| *** untriaged-bot has quit IRC | 09:00 | |
| uvirtbot | Launchpad bug 1367742 in tripleo "keepalived brings the VIP up in multiple places on a controller reboot" [Undecided,New] | 09:00 |
| *** akuznetsov has joined #tripleo | 09:04 | |
| *** Yanyanhu has quit IRC | 09:07 | |
| *** jeevan_ullas has joined #tripleo | 09:07 | |
| *** akuznetsov has quit IRC | 09:08 | |
| *** Yanyanhu has joined #tripleo | 09:14 | |
| *** akrivoka has joined #tripleo | 09:17 | |
| *** lazy_prince has quit IRC | 09:17 | |
| *** dshulyak_ has quit IRC | 09:19 | |
| *** rushiagr is now known as rushiagr_away | 09:23 | |
| *** ramishra has joined #tripleo | 09:31 | |
| *** jtomasek has joined #tripleo | 09:31 | |
| *** rlandy has joined #tripleo | 09:32 | |
| openstackgerrit | A change was merged to openstack/python-tuskarclient: Fix the homepage url in setup.cfg https://review.openstack.org/125573 | 09:34 |
| *** ramishra has quit IRC | 09:35 | |
| *** jml has quit IRC | 09:37 | |
| *** jml has joined #tripleo | 09:39 | |
| openstackgerrit | A change was merged to openstack/tuskar: Remove outdated documentation https://review.openstack.org/125571 | 09:39 |
| *** derekh has joined #tripleo | 09:40 | |
| openstackgerrit | A change was merged to openstack/os-cloud-config: ssl_for_create_admin_client https://review.openstack.org/123806 | 09:48 |
| *** pelix has joined #tripleo | 09:51 | |
| openstackgerrit | Giulio Fidente proposed a change to openstack/tripleo-image-elements: Ensure correct permissions for /var/lib/mysql when dir is recreated https://review.openstack.org/125581 | 09:52 |
| *** jang has joined #tripleo | 09:56 | |
| *** Yanyanhu has quit IRC | 09:57 | |
| *** rushiagr_away is now known as rushiagr | 09:58 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Update EPEL release rpm https://review.openstack.org/125552 | 10:00 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Ensure correct permissions for /var/lib/mysql when dir is recreated https://review.openstack.org/125581 | 10:01 |
| *** dprince has joined #tripleo | 10:02 | |
| *** akuznetsov has joined #tripleo | 10:05 | |
| *** Yanyanhu has joined #tripleo | 10:05 | |
| *** killer_prince has joined #tripleo | 10:08 | |
| *** killer_prince is now known as lazy_prince | 10:09 | |
| *** akuznetsov has quit IRC | 10:09 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Fix $DISTRO_NAME usage for centos7 element https://review.openstack.org/125538 | 10:10 |
| openstackgerrit | Giulio Fidente proposed a change to openstack/diskimage-builder: Ensure epel7 is installed only on rhel7/centos7 https://review.openstack.org/125450 | 10:15 |
| *** dshulyak_ has joined #tripleo | 10:17 | |
| shadower | lsmola: reprod the tuskar storage bug locally, will try & fix it after lunch | 10:30 |
| *** andrearosa has joined #tripleo | 10:30 | |
| *** Dafna has quit IRC | 10:31 | |
| *** yamahata has quit IRC | 10:31 | |
| lsmola | shadower: cool | 10:32 |
| *** Dafna has joined #tripleo | 10:37 | |
| *** nosnos has quit IRC | 10:50 | |
| *** nosnos has joined #tripleo | 10:51 | |
| *** Yanyanhu has quit IRC | 10:53 | |
| *** nosnos has quit IRC | 10:55 | |
| *** xuhaiwei_ has quit IRC | 10:59 | |
| *** Yanyanhu has joined #tripleo | 11:01 | |
| *** akuznetsov has joined #tripleo | 11:05 | |
| *** akuznetsov has quit IRC | 11:10 | |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update Recommended Reading links https://review.openstack.org/125595 | 11:14 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update Tuskar Install Documentation https://review.openstack.org/125596 | 11:14 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update Tuskar Contribution Guidelines https://review.openstack.org/125597 | 11:14 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update the README and Documentation Index https://review.openstack.org/125598 | 11:14 |
| *** andreaf_ is now known as andreaf | 11:18 | |
| *** yamahata has joined #tripleo | 11:30 | |
| *** dprince has quit IRC | 11:30 | |
| *** yamahata has quit IRC | 11:39 | |
| *** ramishra has joined #tripleo | 11:40 | |
| *** yamahata has joined #tripleo | 11:40 | |
| *** ramishra has quit IRC | 11:45 | |
| *** rdopieralski has quit IRC | 11:50 | |
| *** dprince has joined #tripleo | 11:55 | |
| *** dshulyak_ has quit IRC | 12:00 | |
| *** morazi has joined #tripleo | 12:03 | |
| *** rdopieralski has joined #tripleo | 12:03 | |
| *** akuznetsov has joined #tripleo | 12:04 | |
| *** weshay has joined #tripleo | 12:05 | |
| *** jdob has joined #tripleo | 12:08 | |
| *** akuznetsov has quit IRC | 12:08 | |
| *** dshulyak_ has joined #tripleo | 12:18 | |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: Move all the overview templates https://review.openstack.org/125612 | 12:30 |
| jdob | shadower: yes :) | 12:31 |
| shadower | :-) | 12:31 |
| shadower | jdob: doing some printf debugging as we speak | 12:31 |
| shadower | oh ffs | 12:32 |
| shadower | role_name: "cinder-storage", resource name: "CinderStorage" | 12:32 |
| shadower | jdob: this name matching thing ain't a good idea | 12:33 |
| jdob | no, its not | 12:33 |
| jdob | i'm open to suggestions | 12:33 |
| *** rhallisey has quit IRC | 12:34 | |
| shadower | jdob: so renaming the resource to "Cinder-Storage" (ugh) worked so that's our band-aid to unblock the UI folks | 12:35 |
| shadower | alternatives: | 12:35 |
| shadower | 1. ignore non-alphanum characters when matching the role name to a resource ("we've dug this far, so...") | 12:36 |
| *** pensu has quit IRC | 12:36 | |
| shadower | 2. match roles based on types (will require loading the resource registry during tuskar-load-roles) | 12:36 |
| shadower | that's all I've got for now | 12:36 |
| jdob | shadower: all #1 gets us is that it would let you call it CinderStorage instead of Cinder-Storage, right? | 12:38 |
| shadower | yea | 12:39 |
| *** ramishra has joined #tripleo | 12:41 | |
| shadower | jdob: are you against #2? Sounds like the least surprising way for t-h-t template authors | 12:41 |
| shadower | because this naming assumption is comletely invisible anywhere in the templates | 12:41 |
| shadower | we'd have to keep constant vigillance over it | 12:42 |
| jdob | shadower: not sure, still thinking through it | 12:42 |
| *** dshulyak_ has quit IRC | 12:43 | |
| shadower | k | 12:43 |
| * shadower will wait | 12:43 | |
| jdob | i think it'd be a giant pain in the ass | 12:44 |
| jdob | because the flow would be | 12:44 |
| shadower | happy to dig into that code myself | 12:44 |
| jdob | start with the role name | 12:44 |
| jdob | determine the registry alias | 12:44 |
| jdob | find all resources in overcloud.yaml with that type | 12:44 |
| jdob | find those resource names | 12:44 |
| jdob | then do the translation | 12:44 |
| jdob | well, ok, not "giant" pain in the ass :) | 12:44 |
| jdob | shadower: might need you to look into it, i have something else I've been tasked with for a bit | 12:45 |
| *** dshulyak_ has joined #tripleo | 12:45 | |
| *** ramishra has quit IRC | 12:45 | |
| shadower | jdob: okay so let's do this: I'll patch Tuskar to work with the current templates and put it on gerrit. We can bikeshed the when you'll have seen the code | 12:46 |
| shadower | sounds good? | 12:46 |
| jdob | cool, works for me, lemme know when you have something you want eyes on | 12:46 |
| shadower | will do, thank | 12:46 |
| shadower | s | 12:46 |
| openstackgerrit | Alexis Lee proposed a change to openstack/tripleo-incubator: Make a sentence less ambiguous in README https://review.openstack.org/122930 | 12:49 |
| *** dshulyak_ has quit IRC | 12:50 | |
| *** dshulyak_ has joined #tripleo | 12:51 | |
| *** tzumainn has joined #tripleo | 12:55 | |
| openstackgerrit | Ana Krivokapic proposed a change to openstack/tuskar-ui: Add IPMI password to node detail page https://review.openstack.org/125392 | 12:57 |
| *** akuznetsov has joined #tripleo | 13:05 | |
| *** noslzzp has joined #tripleo | 13:06 | |
| *** akuznetsov has quit IRC | 13:10 | |
| shadower | jdob: so here's an idea: if we're passing the resource registry to tuskar-load-roles: we can just use template paths in there to load the roles themselves | 13:10 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update the README and Documentation Index https://review.openstack.org/125598 | 13:11 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update Tuskar Contribution Guidelines https://review.openstack.org/125597 | 13:11 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update Tuskar Install Documentation https://review.openstack.org/125596 | 13:11 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Rename INSTALL to install https://review.openstack.org/125622 | 13:11 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Rename CONTRIBUTING to contributing https://review.openstack.org/125623 | 13:11 |
| openstackgerrit | Jiri Tomasek proposed a change to openstack/tuskar-ui: Layout and styling tweaks https://review.openstack.org/125624 | 13:11 |
| *** ramishra has joined #tripleo | 13:13 | |
| *** lazy_prince has quit IRC | 13:14 | |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update the README and Documentation Index https://review.openstack.org/125598 | 13:15 |
| *** lazy_prince has joined #tripleo | 13:15 | |
| *** rhallisey has joined #tripleo | 13:15 | |
| * d0ugal is rebasing like a boss :) | 13:17 | |
| *** ramishra has quit IRC | 13:18 | |
| *** radez_g0n3 is now known as radez | 13:20 | |
| jdob | shadower: instead of having to specify --role for each? | 13:27 |
| shadower | jdob: ya | 13:27 |
| jdob | that makes sense for now, though in the future when we have the top level resources as their own things being loaded we might need to change it back | 13:27 |
| jdob | but for now that fits in with the "seed" concept | 13:27 |
| jdob | or the "template of templates" :) | 13:27 |
| shadower | jdob: agreed there is value with keeping --role there when you need the explicit stuff. But say if you don't pass any roles that way but do pass the registry, we'll just pick all those | 13:28 |
| jdob | makes sense | 13:28 |
| jdob | easy enough to do too, we already have stuff that parses an environment file | 13:28 |
| shadower | yea | 13:29 |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Add IPMI password to node detail page https://review.openstack.org/125392 | 13:33 |
| openstackgerrit | Peter Belanyi proposed a change to openstack/tripleo-heat-templates: Add converted version of block and object storage https://review.openstack.org/123713 | 13:34 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update the README and Documentation Index https://review.openstack.org/125598 | 13:39 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update Tuskar Contribution Guidelines https://review.openstack.org/125597 | 13:39 |
| rhallisey | larsks, added the mariadb blueprint. Thanks for the reminder | 13:41 |
| larsks | rhallisey: Thanks! | 13:41 |
| *** jehb has joined #tripleo | 13:41 | |
| rhallisey | larsks, did you want to take that one? | 13:41 |
| larsks | Yeah sure. | 13:41 |
| rhallisey | great! | 13:42 |
| *** andreaf is now known as andreaf_ | 13:49 | |
| rhallisey | larsks, do you think it's a better idea to build conf files with crudini or generate them locally and do docker ADD /my/conf/file /conf/file/location in the dockerfile | 13:50 |
| larsks | rhallisey: It depends. If they are being modified at runtime via environment variables, the answer is "use crudini". | 13:51 |
| rhallisey | gotcha | 13:51 |
| larsks | So we can certainly "bake in" reasonable defaults, I guess, and then use crudini for runtime changes (like host addresses, credentials, etc). | 13:51 |
| larsks | ...but frankly I think we are doing this in a completely wrong and hacky fashion in any case right now. | 13:52 |
| larsks | I would like to figure out a graceful way of leveraging all the puppet work to handle configuration. | 13:53 |
| openstackgerrit | Emilien Macchi proposed a change to openstack/tripleo-puppet-elements: Update repository after OpenStack migration https://review.openstack.org/125438 | 13:53 |
| larsks | So that we don't have to completely reinvent the wheel. | 13:53 |
| larsks | But that will take longer than hacking up shell scripts with crudini... :) | 13:53 |
| EmilienM | sbadia: I've updated the .gitignore with dprince suggestion. ^^^^^ | 13:53 |
| rhallisey | ya I see what you mean | 13:53 |
| sbadia | EmilienM: oh indeed | 13:54 |
| sbadia | thanks! | 13:55 |
| *** noslzzp has quit IRC | 13:58 | |
| openstackgerrit | A change was merged to openstack/tripleo-puppet-elements: Update repository after OpenStack migration https://review.openstack.org/125438 | 13:59 |
| *** ramishra has joined #tripleo | 14:01 | |
| *** akuznetsov has joined #tripleo | 14:06 | |
| *** ramishra has quit IRC | 14:06 | |
| *** hewbrocc` has joined #tripleo | 14:08 | |
| *** hewbrocca has quit IRC | 14:08 | |
| *** Yanyanhu has quit IRC | 14:10 | |
| *** akuznetsov has quit IRC | 14:10 | |
| *** mrunge has quit IRC | 14:23 | |
| *** mrunge has joined #tripleo | 14:24 | |
| *** hewbrocc` has quit IRC | 14:28 | |
| *** jprovazn is now known as jprovazn_afk | 14:31 | |
| *** pcrews has joined #tripleo | 14:36 | |
| *** noslzzp has joined #tripleo | 14:37 | |
| *** marun has joined #tripleo | 14:39 | |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Refresh progress page with AJAX https://review.openstack.org/125064 | 14:40 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Add initial svc-map support https://review.openstack.org/104196 | 14:41 |
| *** rainya has joined #tripleo | 14:41 | |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Layout and styling tweaks https://review.openstack.org/125624 | 14:41 |
| *** dvossel has joined #tripleo | 14:43 | |
| *** tteggel has quit IRC | 14:44 | |
| *** spzala has joined #tripleo | 14:45 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Add svc-map to cinder-api https://review.openstack.org/104198 | 14:46 |
| *** tteggel has joined #tripleo | 14:46 | |
| *** mrunge has quit IRC | 14:49 | |
| *** jrist has quit IRC | 14:53 | |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Refactor the Tuskar storage tests https://review.openstack.org/125660 | 14:59 |
| *** untriaged-bot has joined #tripleo | 15:00 | |
| untriaged-bot | Untriaged bugs so far: | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1375414 | 15:00 |
| uvirtbot | Launchpad bug 1375414 in tripleo "mysql/rabbit overcloud issues when the bootstrap node is removed" [Undecided,New] | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1376200 | 15:00 |
| uvirtbot | Launchpad bug 1376200 in tripleo "Stale connections when VIP is moved" [High,New] | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1375422 | 15:00 |
| uvirtbot | Launchpad bug 1375422 in tripleo "Vlan not recreated on overcloud controller reboot" [Undecided,In progress] | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1366164 | 15:00 |
| uvirtbot | Launchpad bug 1366164 in tripleo "Add domain roles for admin user" [Undecided,New] | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1367742 | 15:00 |
| uvirtbot | Launchpad bug 1367742 in tripleo "keepalived brings the VIP up in multiple places on a controller reboot" [Undecided,New] | 15:00 |
| *** untriaged-bot has quit IRC | 15:00 | |
| *** ramishra has joined #tripleo | 15:02 | |
| *** akuznetsov has joined #tripleo | 15:06 | |
| *** ramishra has quit IRC | 15:07 | |
| openstackgerrit | therese-mchale proposed a change to openstack/tripleo-image-elements: Configure logging for keepalived https://review.openstack.org/125668 | 15:11 |
| *** akuznetsov has quit IRC | 15:11 | |
| *** dshulyak_ has quit IRC | 15:14 | |
| *** ifarkas has quit IRC | 15:14 | |
| *** dshulyak_ has joined #tripleo | 15:16 | |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Rename CONTRIBUTING to contributing https://review.openstack.org/125623 | 15:17 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update the README and Documentation Index https://review.openstack.org/125598 | 15:17 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update Tuskar Contribution Guidelines https://review.openstack.org/125597 | 15:17 |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tuskar: Update Tuskar Install Documentation https://review.openstack.org/125596 | 15:17 |
| sdake | radez how did you test keystone | 15:17 |
| sdake | which permissions did you use? | 15:17 |
| radez | I came across the admin password somewhere, and got the token from the keystone config. I did a token get and added the endpoints by hand to keystone. | 15:18 |
| radez | lemme get the creds I used | 15:18 |
| radez | sdake: admin/redhat | 15:18 |
| sdake | radez cool I'll give it a go | 15:19 |
| radez | I'm currently battling a 500 in glance that looks wsgi related... | 15:19 |
| radez | trygin to figure out how to debug it deeper in an efficent way | 15:19 |
| larsks | radez: anything of interest in the glance logs? | 15:20 |
| sdake | http://paste.fedoraproject.org/138556/41226320 is this what you get with keystone radez? | 15:20 |
| sdake | I had some problem with my kernel.org version which I think is fixed now | 15:20 |
| radez | not without debug on, holdon I'll pastebin | 15:20 |
| radez | sdake: that's just because you don't have the keystone endpoint defined | 15:20 |
| radez | as soon as the endpoint is defined that error will go away | 15:21 |
| radez | I plan to work on that os_config stuff for kolla when I finish with glance if noone else has picked it up yet | 15:21 |
| radez | then the endpoints will get defined automagically | 15:21 |
| sdake | I jsut tried an endpoint create | 15:22 |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Move all the overview templates https://review.openstack.org/125612 | 15:22 |
| sdake | but not getting errors in the keystone logs | 15:22 |
| sdake | so I must be good to go | 15:22 |
| radez | larsks: http://paste.fedoraproject.org/138558/26336814/ | 15:22 |
| larsks | Looking... | 15:23 |
| radez | sdake: do your endpoint-list again and see if that works well. if so then you shoudl be good | 15:23 |
| radez | larsks: seems pretty non-descript | 15:23 |
| larsks | radez: which log is that from? | 15:23 |
| radez | api.log | 15:23 |
| larsks | Anything in registry.log? | 15:23 |
| radez | now ath you say that I guess I haven't check it. | 15:23 |
| * radez looks at registry log | 15:23 | |
| larsks | That would seem to be the place to check :) | 15:23 |
| larsks | Since the errors are coming from the registry client, it looks like... | 15:24 |
| radez | larsks: ah, as I had expected it's database related... just havn't gotten the db to sync right yet... | 15:24 |
| radez | ur my hero :) | 15:24 |
| larsks | \o/ | 15:24 |
| radez | apparently I eon't know how to seed the db for glance... db_sync by itself doesn't work | 15:26 |
| radez | larsks: is there extra stuff to pass to db_sync? | 15:26 |
| sdake | http://paste.fedoraproject.org/138561/26358914 | 15:26 |
| larsks | radez: shouldn't be. It gets all the config from the glance config files. | 15:27 |
| sdake | radez good historical info http://fedoraproject.org/wiki/Getting_started_with_OpenStack_on_Fedora_17 | 15:27 |
| radez | larsks: ack, gotta run to a meeting, I htink that gives me something to fiddle with now | 15:27 |
| radez | thx sdake | 15:27 |
| sdake | radez you may need rabbit running i'm not certain | 15:28 |
| radez | sdake: you may need a new token... but what do your endpoints look like now? | 15:28 |
| sdake | empty list radez | 15:28 |
| radez | sdake: I didn't think you did for glance | 15:28 |
| radez | sdake: yea you need to create your endpoint... do this... | 15:28 |
| radez | http://paste.fedoraproject.org/138564/14122637/ | 15:28 |
| radez | but update the ip address to one of your minions | 15:29 |
| radez | bbiab | 15:29 |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-image-elements: haproxy global maxconn is less than default maxconn https://review.openstack.org/125675 | 15:29 |
| sdake | radez still an empty keystone endpoint list | 15:29 |
| sdake | radez but I'd think the above should go in the keystone start.sh | 15:29 |
| sdake | radez how do you get a new token? | 15:30 |
| *** jrist has joined #tripleo | 15:30 | |
| larsks | sdake: you can auth with the admin token defined in the keystone.conf file. | 15:30 |
| *** dsneddon has quit IRC | 15:31 | |
| sdake | hmm well something is bust :) | 15:31 |
| sdake | and its time for a break, bbiaf ;) | 15:31 |
| sdake | jpeeler did you manage to get keystone working with our updated image? | 15:31 |
| larsks | sdake: when you're back maybe we can take a closer look. | 15:31 |
| *** dsneddon has joined #tripleo | 15:31 | |
| sdake | larsks sounds good | 15:31 |
| sdake | keystone isn't spitting out errors now which is good | 15:32 |
| jpeeler | sdake: i haven't yet rechecked honestly - in theory i should get just as far as i did previously... | 15:32 |
| jpeeler | sdake: previously using token authentication endpoint-list returned nothing, which i thought was success given that nothing was set up | 15:33 |
| larsks | jpeeler: That sounds right. Should just be a matter of creating endpoints. | 15:33 |
| *** athomas has quit IRC | 15:34 | |
| *** sdake_ has joined #tripleo | 15:37 | |
| sdake | cool | 15:40 |
| sdake | sounds like we are all functional now | 15:40 |
| sdake | although my keypoint list isn't actually creating one | 15:40 |
| sdake | endpoint-create that is | 15:40 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Add svc-map to cinder-api https://review.openstack.org/104198 | 15:41 |
| sdake | I would think keystone contianer would set all its endpoints stuff up tho right? | 15:41 |
| sdake | so we need to fix that | 15:41 |
| jpeeler | unless each service registers with keystone individually | 15:42 |
| sdake | well keystone should atleast set up its own keystone endpoints | 15:43 |
| *** sdake_ has quit IRC | 15:43 | |
| jpeeler | definitely | 15:43 |
| openstackgerrit | Tomas Sedovic proposed a change to openstack/tuskar: Use resource registry for seed role matching https://review.openstack.org/125686 | 15:45 |
| *** jcoufal has quit IRC | 15:45 | |
| shadower | jdob: https://review.openstack.org/#/c/125686/ | 15:46 |
| shadower | it's not that bad | 15:46 |
| jdob | cool, will take a look in a bit | 15:46 |
| *** yamahata has quit IRC | 15:46 | |
| shadower | jdob: I didn't add the autoloading roles from the registry yet. That feels like a separate change | 15:46 |
| *** yamahata has joined #tripleo | 15:47 | |
| shadower | but this should make tuskar work with the storage templates & be less confusing to the template authors | 15:47 |
| shadower | lsmola, rbrady: ^ | 15:47 |
| *** rdopieralski has quit IRC | 15:48 | |
| sdake | larsks any idea why my keystone cred isn't showing on endpoint-lsit? | 15:50 |
| sdake | list t hat is | 15:50 |
| lsmola | shadower: looking | 15:50 |
| rbrady | shadower: thanks. I'll test it with instack shortly | 15:50 |
| larsks | sdake: Have you run a 'keystone service-create' and 'keystone endpoint-create' to create the service catalog entries? It doesn't look like our keystone image does that right now. | 15:50 |
| larsks | The os-config stuff that we were talking about yesterday will help make that easier to implement. | 15:51 |
| sdake | larsks I ran the commands radez mentioned | 15:51 |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-image-elements: haproxy global maxconn is less than default maxconn https://review.openstack.org/125675 | 15:51 |
| larsks | sdake: Let me grab some lunch and then I will bring things up locally and see. | 15:51 |
| sdake | cool thanks | 15:52 |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Change "redhat" to "kolla" in the keystone authentication https://review.openstack.org/125694 | 15:55 |
| openstackgerrit | A change was merged to stackforge/kolla: Change "redhat" to "kolla" in the keystone authentication https://review.openstack.org/125694 | 15:55 |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-image-elements: haproxy global maxconn is less than default maxconn https://review.openstack.org/125675 | 15:55 |
| *** ramishra has joined #tripleo | 16:03 | |
| *** jang1 has joined #tripleo | 16:06 | |
| *** akuznetsov has joined #tripleo | 16:07 | |
| *** ramishra has quit IRC | 16:08 | |
| *** lucasagomes has quit IRC | 16:10 | |
| *** akuznetsov has quit IRC | 16:12 | |
| *** cwolferh_ has joined #tripleo | 16:17 | |
| *** jang1 has quit IRC | 16:18 | |
| *** jang1 has joined #tripleo | 16:19 | |
| *** Duane has joined #tripleo | 16:22 | |
| *** dtantsur is now known as dtantsur|afk | 16:24 | |
| *** sdake_ has joined #tripleo | 16:25 | |
| *** noslzzp has quit IRC | 16:29 | |
| *** dsneddon has quit IRC | 16:31 | |
| *** dsneddon has joined #tripleo | 16:32 | |
| *** derekh has quit IRC | 16:34 | |
| *** jp_at_hp has quit IRC | 16:34 | |
| *** sdake_ has quit IRC | 16:40 | |
| *** rainya has quit IRC | 16:42 | |
| *** jistr has quit IRC | 16:45 | |
| *** jang1 has quit IRC | 16:46 | |
| *** lsmola has quit IRC | 16:48 | |
| larsks | sdake: Back! | 16:48 |
| *** jang1 has joined #tripleo | 16:48 | |
| *** jtomasek has quit IRC | 16:48 | |
| *** dsneddon has quit IRC | 16:52 | |
| sdake | larsks cool just got back from a break | 16:52 |
| sdake | cold + pulled my back = not happy sdake | 16:52 |
| *** dsneddon has joined #tripleo | 16:53 | |
| *** jang1 has quit IRC | 16:53 | |
| *** jang1 has joined #tripleo | 16:54 | |
| larsks | Okay. So, the keystone start.sh script should create an admin tenant and user, but nothing else. You should be able to authenticate by grabbing the admin_token from the keystone.conf file and then running something like: | 16:58 |
| larsks | SERVICE_TOKEN=cd4d2a937ad277671c636d0c72dc52 SERVICE_ENDPOINT=http://172.17.0.50:35357/v2.0 keystone user-list | 16:58 |
| larsks | That should show you a list containing just the "admin" user. If that *doesn't* work, we start poking around inside the keystone container... | 16:59 |
| jpeeler | sdake: service token is ADMINTOKEN | 17:01 |
| larsks | jpeeler: Huh, we should consider removing that default from keystone.json :) | 17:02 |
| jpeeler | eventually, i like it for now | 17:02 |
| larsks | I know, so convenient for testing :) | 17:02 |
| sdake | or make it configurable | 17:02 |
| larsks | sdake: It is configureable. | 17:03 |
| larsks | You edit the keystone.json and put in a different value... | 17:03 |
| sdake | well that is good enough imo | 17:03 |
| sdake | no I mean by passing env | 17:03 |
| *** jang1 has quit IRC | 17:03 | |
| larsks | That's exactly what it does. | 17:03 |
| larsks | The json file is configuring the environment vars passed tot he docker container. | 17:03 |
| *** ramishra has joined #tripleo | 17:03 | |
| sdake | http://paste.fedoraproject.org/138603/14122694 | 17:04 |
| sdake | oh right | 17:04 |
| sdake | I think eventually we will want to automate the yaml creation of the containers - perhaps using a language such as python | 17:04 |
| sdake | python has rockin yaml bindings | 17:04 |
| larsks | sdake: those need to be environment vars. If you put all that on one line it will work. | 17:04 |
| jpeeler | or use export | 17:04 |
| larsks | SERVICE_TOKEN=... SERVICE_ENDPOINT=... keystone user-list | 17:04 |
| larsks | Yes, or use export. | 17:04 |
| sdake | export worked | 17:04 |
| sdake | got an admin user | 17:04 |
| larsks | Then your keystone is all up and running. | 17:05 |
| larsks | \o/ | 17:05 |
| sdake | yay | 17:05 |
| jpeeler | sdake: with optimized image right? | 17:05 |
| larsks | sdake: Now you can do this: https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux_OpenStack_Platform/3/html/Installation_and_Configuration_Guide/Creating_the_Permanent_Service_Endpoint.html | 17:05 |
| larsks | ...and then you should be able to authenticate with the admin user/password using the OS_* environment vars. | 17:05 |
| *** jang1 has joined #tripleo | 17:06 | |
| sdake | jpeeler yup opt image | 17:07 |
| sdake | wow docs | 17:07 |
| sdake | who would have thunk ;) | 17:07 |
| sdake | radez you might find some magic about glance in the docs from larsks post | 17:08 |
| *** akuznetsov has joined #tripleo | 17:08 | |
| *** ramishra has quit IRC | 17:08 | |
| sdake | larsks any thoughts on how we get 10.0.0.x into publicurl and 172.X into internalurl? | 17:10 |
| sdake | or vice versa rather ;) | 17:10 |
| *** jrist has quit IRC | 17:10 | |
| sdake | got action on endpoint-list | 17:10 |
| sdake | \o/ | 17:10 |
| sdake | i'll add that to the image | 17:11 |
| larsks | I'm not sure I understand the question...by putting them in the correct parameters to endpoint-create? | 17:11 |
| sdake | larsks I guess, but how to get the parameters in teh first place | 17:11 |
| sdake | because the internal nodes have 10.0.0.x | 17:11 |
| larsks | Right. | 17:12 |
| sdake | but to access externally as a typical user would, they would use 172.24.4.2327 | 17:12 |
| sdake | so publicurl and adminurl need to be 172.z and internalurl need to be 10.z | 17:12 |
| larsks | Let's see. There are several things to think about here. There may very well be multiple keystone instances, so you need to get the address from kubernetes somehow. | 17:12 |
| *** akuznetsov has quit IRC | 17:12 | |
| sdake | kube doesn't know the address correct? | 17:13 |
| larsks | Both the internal and external urls should probably map to things managed by k8s services. | 17:13 |
| sdake | how can it know the floating ips? | 17:13 |
| larsks | Well, kube knows *some* addresses :) | 17:13 |
| *** dsneddon has quit IRC | 17:13 | |
| sdake | then how would you use the openstack services externally? | 17:13 |
| sdake | if I put 10.z in for publicurl I doubt my system will work | 17:13 |
| sdake | although I haven't tried | 17:13 |
| *** dsneddon has joined #tripleo | 17:14 | |
| sdake | i shall try | 17:14 |
| *** yamahata has quit IRC | 17:14 | |
| larsks | But for access from "outside", I'm not sure about the right answer off the top of my head. | 17:14 |
| sdake | outside means "outside the container" | 17:14 |
| larsks | Right. | 17:14 |
| sdake | almost like we need config info from outside - eg a metadata service | 17:15 |
| sdake | oh wait... | 17:15 |
| sdake | so we are running in a heat template | 17:15 |
| sdake | when this is running on bare metal | 17:16 |
| sdake | there is no floating ip | 17:16 |
| sdake | rather heat stack | 17:16 |
| larsks | sdake: Right, I wasn't thinking about floating ips. | 17:16 |
| larsks | But for the container to know what addrses to register for publicurl, if k8s is managing services through something like a replicationcontroller, I'm not 100% sure how the *container* can find the correct external address. | 17:16 |
| sdake | external metadata would do it | 17:17 |
| sdake | but ya, seems like a gap in kube | 17:17 |
| larsks | Maybe...haven't looked closely enough yet. Possibly just a gap in my knowledge. | 17:17 |
| sdake | larsks is there any way to get rid of the floating ips and just have t he172x map into my local network address space? | 17:23 |
| sdake | (with neutronium) | 17:23 |
| sdake | nm, the 10.0.0.5 works too | 17:24 |
| larsks | sdake: possible, yes, but not so easy via a reusable template, and also not so easy on a single-interface system (because in order for neutron to connect directly to your local network space you would need to bridge your primary interface to an ovs bridge). | 17:24 |
| sdake | so 10.0.0.5 and 172.x are available from outside the vm | 17:24 |
| sdake | so we are good to go | 17:24 |
| sdake | was 10.x the public or private address range? | 17:25 |
| larsks | 10.0.0.0/24 is created by the heat templates, so that's the fixed address network. | 17:25 |
| sdake | what is the purpose of the floating ips then? | 17:26 |
| sdake | for external access? | 17:26 |
| larsks | Yes. | 17:26 |
| larsks | For some value of "external". | 17:26 |
| sdake | I see, well while we are in dev, using the fixed ips is ok, since they are locally routed atleast on my machine | 17:26 |
| *** rainya has joined #tripleo | 17:26 | |
| sdake | I dont tihnk I even need floating ips | 17:26 |
| larsks | I would not expect your host to be able to get to the 10.0.0.0/24 addresses, but if that works, go for it. | 17:27 |
| sdake | it does work | 17:27 |
| sdake | hmm maybe it was using my keystone token env | 17:27 |
| sdake | nm | 17:27 |
| sdake | is there a way to add a route to 10.z | 17:27 |
| larsks | sdake: Well, you could hack it in, but that's really what the floating ips are used for. | 17:28 |
| larsks | But in theory if you gave br-ex an address on that network you could then reach it from your host. | 17:28 |
| sdake | well the problem with floating ips is I can't access externally | 17:29 |
| larsks | sdake: "externally" meaning "from somewhere not on your host"? | 17:29 |
| sdake | because I don't know which float maps to which service | 17:29 |
| sdake | there is no kube ip <-> os IP conversion | 17:29 |
| larsks | In theory, you should be able to access keystone (for example) using any minion address, because of the proxies started by kube. | 17:29 |
| sdake | oh didn't know that | 17:30 |
| larsks | And you can get the minion floating ips from heat. | 17:30 |
| sdake | ok then that gefn | 17:30 |
| larsks | That's what "services" do -- expose a service at the same port on all the hosts in the glsuter. | 17:30 |
| larsks | Er, "cluster". | 17:30 |
| *** dshulyak_ has quit IRC | 17:31 | |
| *** dsneddon has quit IRC | 17:33 | |
| *** dsneddon has joined #tripleo | 17:34 | |
| *** cody-somerville has quit IRC | 17:36 | |
| *** lazy_prince is now known as killer_prince | 17:37 | |
| sdake | larsks well anyway I'm excited - big progress ;) | 17:38 |
| sdake | larsks the proxy is part of the services feature? | 17:38 |
| larsks | Yes. For the most part that *is* the service feature. | 17:38 |
| larsks | A proxy gets started on each host, and containers on that host talk to that proxy to access a particular service, and the proxy routes the request to the appropriate pod. | 17:39 |
| sdake | based upon port id | 17:39 |
| sdake | pretty clever | 17:39 |
| sdake | wonder why it took 90k lines of code to implement that ;-) | 17:39 |
| sdake | there is built in lb right? | 17:39 |
| larsks | I think so, via the replicationController stuff. | 17:40 |
| larsks | ...which we haven't really looked at yet. | 17:40 |
| sdake | right | 17:40 |
| sdake | well action for later I think | 17:40 |
| larsks | sdake: https://github.com/GoogleCloudPlatform/kubernetes/issues/1161 | 17:42 |
| larsks | Plans for handling external ips in kubernetes. | 17:43 |
| *** jang1 has quit IRC | 17:45 | |
| *** Duane has quit IRC | 17:46 | |
| *** rainya has quit IRC | 17:46 | |
| *** Duane has joined #tripleo | 17:47 | |
| *** noslzzp has joined #tripleo | 17:50 | |
| *** regebro has quit IRC | 17:50 | |
| *** pensu has joined #tripleo | 17:51 | |
| *** Duane has quit IRC | 17:51 | |
| *** cody-somerville has joined #tripleo | 17:51 | |
| *** rainya has joined #tripleo | 17:52 | |
| *** jang1 has joined #tripleo | 17:52 | |
| sdake | when I visualize kubernetes, I picture a borg ship | 17:53 |
| sdake | even though its 2d rather then 3d | 17:54 |
| *** pelix has quit IRC | 17:56 | |
| *** jang2 has joined #tripleo | 17:57 | |
| *** jang1 has quit IRC | 17:59 | |
| *** jang2 has quit IRC | 18:02 | |
| *** jrist has joined #tripleo | 18:07 | |
| *** akuznetsov has joined #tripleo | 18:09 | |
| *** cwolferh_ has quit IRC | 18:11 | |
| *** jprovazn_afk has quit IRC | 18:11 | |
| *** akuznetsov has quit IRC | 18:14 | |
| *** cwolferh has joined #tripleo | 18:15 | |
| sdake | larsks when I start keystone it says "Terminmated" | 18:15 |
| sdake | does that mean it exited early? | 18:15 |
| larsks | Yeah probably. | 18:16 |
| larsks | "docker ps" would confirm... | 18:17 |
| *** akrivoka has quit IRC | 18:18 | |
| sdake | what am I looking for for confirmation? | 18:18 |
| sdake | be nice if this stuff has logs{! | 18:18 |
| *** jang1 has joined #tripleo | 18:21 | |
| lifeless | how goes the spike? | 18:22 |
| openstackgerrit | A change was merged to openstack/tuskar: Update Recommended Reading links https://review.openstack.org/125595 | 18:24 |
| openstackgerrit | A change was merged to openstack/tuskar: Rename INSTALL to install https://review.openstack.org/125622 | 18:24 |
| *** pensu has quit IRC | 18:25 | |
| radez | sdake: do a docker logs {id} | 18:35 |
| radez | it will give you the output of the start script | 18:35 |
| lifeless | win 48 | 18:37 |
| sdake | there is no id listed | 18:45 |
| sdake | in docker ps | 18:45 |
| larsks | sdake: you probably want "docker ps -a" | 18:46 |
| larsks | that lists stopped containers, too. | 18:46 |
| *** jehb has quit IRC | 18:46 | |
| sdake | cool I'll try that thanks | 18:46 |
| radez | sdake: are you getting mysql terminated? | 18:47 |
| radez | I'm having trouble turning my pods over again | 18:47 |
| larsks | radez: keystone terminated | 18:47 |
| radez | yea | 18:47 |
| radez | that's what meant... mysql comes up but keystone can't connect to teh db, I think the keystoneadmin user isn't getting created right | 18:48 |
| * radez looks for where that happeneds | 18:48 | |
| larsks | radez: if you can get as far as "keystone user-list" showing you the "admin" user, that suggests that keystone was talking to the database...at least at that point. | 18:48 |
| radez | larsks: it doesn't the start.sh file's output is full of db connect errors | 18:49 |
| larsks | radez: Ah, okay. sdake was able to get working user-list. | 18:49 |
| radez | I had a working keystone... but I tore down and recreated the pods and now I can't | 18:50 |
| larsks | radez: check the environment vars passed to pid 1 and make sure they are correct. | 18:50 |
| radez | larsks: how can you do that if the container won't start | 18:51 |
| radez | isn't that command suppposed to be run in the container? | 18:51 |
| larsks | radez: oh, right, your container simply isn't starting. | 18:52 |
| larsks | Okay, cancel that idea. | 18:52 |
| radez | ah I see this in the logs now | 18:52 |
| radez | ERROR 2013 (HY000): Lost connection to MySQL server at 'reading initial communication packet', system error: 0 | 18:52 |
| sdake | I have seen this before | 18:52 |
| radez | that could be part of the issue... Maybe I'll try and rebuild the db too | 18:52 |
| radez | sdake: oh yea? | 18:52 |
| sdake | it was caused by a non-updated kernel.org | 18:52 |
| sdake | did you update all nodes? | 18:52 |
| radez | I had it all working an hour ago, now it doesn't on the same nodes | 18:53 |
| * radez double checks kernels to be sure | 18:53 | |
| sdake | http://paste.fedoraproject.org/138648/14122759 | 18:53 |
| sdake | oh | 18:53 |
| sdake | I may have busted the keystone image | 18:53 |
| sdake | forgot I was working on that | 18:53 |
| radez | heh... | 18:54 |
| radez | well that'll do it :) | 18:54 |
| *** dsneddon has quit IRC | 18:54 | |
| sdake | radez buildilng new images for you just a min | 18:55 |
| radez | cool, thx sdake | 18:55 |
| *** dsneddon has joined #tripleo | 18:55 | |
| larsks | I think maybe we need a blueprint for "image building workflow for development that doesn't break things for other people". | 18:55 |
| radez | +1 | 18:56 |
| larsks | That's something we will need to figure out especially once we have "released" images... | 18:56 |
| radez | the db issue with glance was related to it not being utf8 :( | 18:56 |
| radez | I'vw rebuilt the glance container, hope to verify it once I get the db and keystone back up | 18:56 |
| sdake | radez your good to go with keystone | 18:57 |
| * radez recreates the pod | 18:57 | |
| sdake | you may have to start fresh | 18:57 |
| jpeeler | is there not a web based interface to seeing changes made in the registry? | 18:59 |
| larsks | radez: https://blueprints.launchpad.net/kolla/+spec/image-dev-workflow | 18:59 |
| jpeeler | it's hard to see anything here https://registry.hub.docker.com/u/kollaglue/fedora-rdo-keystone/ | 18:59 |
| larsks | jpeeler: I'm not sure that would help at all. I mean, you can see how recently an image was built... | 18:59 |
| larsks | But I'm not sure you want to check that every time before deploying, and even if you do, what do you do with that information? | 19:00 |
| larsks | I am thinking that maybe some sort of tag-based namespacing may help us out here. | 19:00 |
| jpeeler | i was just wondering if there was a simple way to even see if the image had changed | 19:00 |
| jpeeler | that properties timestamp isn't all that helpful | 19:01 |
| larsks | That is, if we always push to the :dev tag for an image when developing, and pushes to :latest or (untagged) only happen via some formal workflow. | 19:01 |
| larsks | That might help. | 19:01 |
| *** rushiagr has quit IRC | 19:01 | |
| *** jrist has quit IRC | 19:03 | |
| sdake | agree larsks | 19:05 |
| sdake | just need someone to crank er out | 19:05 |
| radez | sdake: keystone is happy again for me, thx | 19:05 |
| sdake | ok i'm getting ready to bust it again | 19:06 |
| radez | cool, I won't delete my pod till it's fixed again :) | 19:06 |
| *** akuznetsov has joined #tripleo | 19:09 | |
| *** akuznetsov has quit IRC | 19:14 | |
| *** dsneddon has quit IRC | 19:16 | |
| radez | sdake: it doesn't look like the keystone contrainer creates the keystone user and ther services tenants | 19:16 |
| *** dsneddon has joined #tripleo | 19:16 | |
| radez | is that what you're working on fixing? | 19:16 |
| sdake | radez trying yes | 19:17 |
| radez | ack | 19:17 |
| sdake | something was fubar I guess in my first g o | 19:17 |
| sdake | so now I'm tryingf to fix the race between keystone nad mariadb | 19:18 |
| *** rushiagr has joined #tripleo | 19:19 | |
| radez | sdake: oh nm, there isn't supposed to be a keystone user... k so i don't have a services tenant... lemme see if I can figure out why I can't user-lit | 19:22 |
| sdake | 2014-10-02 19:18:58.889 33 ERROR keystone.common.wsgi [-] (OperationalError) (2013, "Lost connection to MySQL server at 'reading initial communication packet', system error: 0") None None | 19:23 |
| sdake | so getting this again | 19:23 |
| sdake | what is odd is | 19:23 |
| sdake | connection = mysql://keystone:2f26f4ccda8d1c0a8d0a65d3ef111c@10.0.0.4:3306/keystone | 19:23 |
| sdake | mariadb kollaglue/fedora-rdo-mariadb 10.0.0.5/ name=mariadb-master Running | 19:23 |
| sdake | maybe proxy no worky | 19:24 |
| sdake | I am not sure where it is getting 10.0.0.4 from | 19:24 |
| sdake | since 10.0.0.5 is where mariadb is running | 19:24 |
| larsks | sdake: 10.0.0.4 should be the address of the host on which the pod is running. | 19:24 |
| *** noslzzp has quit IRC | 19:25 | |
| larsks | This should be running a kube-proxy instance, bound to port 3306 on the host. | 19:25 |
| sdake | well it is definately running on 10.0.0.5 confirmed with docker ps | 19:25 |
| *** noslzzp has joined #tripleo | 19:25 | |
| sdake | the mariadb pod I assume you mean | 19:25 |
| larsks | No, the keystone pod. | 19:25 |
| sdake | but the script says "MARIADBMASTER_TCP_*" | 19:25 |
| larsks | Right, yes. This is the service stuff we were talking about earlier. | 19:26 |
| larsks | The service causes kube-proxy on each host to bind to the service port. | 19:26 |
| sdake | yes well it makes the database not work then | 19:26 |
| larsks | On that host, pods will point at the proxy address. | 19:26 |
| larsks | It does work, when it's working properly. pod -> proxy_on_host -> pod_with_service | 19:26 |
| larsks | So, is your kube-proxy instance bound to the correct port? Any errors from kube-proxy (probably via journalctl...)? | 19:27 |
| sdake | lemme check | 19:27 |
| larsks | I don't think there's a way to query the proxy for a list of bindings. That would be useful. | 19:27 |
| sdake | [1004]: E1002 19:28:36.741005 01004 etcd.go:166] Failed to get the key: registry | 19:29 |
| sdake | ts) [151] | 19:29 |
| larsks | sdake: Is that a recent error? | 19:30 |
| sdake | yup | 19:30 |
| sdake | just started fresh system | 19:31 |
| larsks | Huh. Do kubecfg commands work (like "kubecfg list pods")? | 19:31 |
| larsks | Or "... list minions"? | 19:31 |
| sdake | only on the master | 19:31 |
| sdake | [fedora@my-xdbeej5hjgdc-1-btxwrvbyu2kw-kube-node-umbmrynb3cs7 ~]$ kubecfg list pods | 19:32 |
| sdake | F1002 19:31:47.799883 01516 kubecfg.go:391] Got request error: Get http://localhost:8080/api/v1beta1/pods: dial tcp 127.0.0.1:8080: connection refused | 19:32 |
| larsks | Oh, that's a problem. | 19:32 |
| larsks | Wait, maybe, is that error on the master or on one of the minions? | 19:32 |
| sdake | minons | 19:32 |
| sdake | master works | 19:32 |
| *** marun has quit IRC | 19:32 | |
| *** jrist has joined #tripleo | 19:32 | |
| larsks | Okay, so that is a problem, because they should not be pointing at localhost. | 19:33 |
| sdake | I would think it would be more logical for the env variable to bypass the proxy | 19:33 |
| larsks | sdake: well, that's not how the services work right now. I haven't looked into the design motiviation. | 19:33 |
| larsks | But what does /etc/kubernetes/apiserver look like on the minions? | 19:33 |
| sdake | why even have an env variable :) | 19:33 |
| larsks | Specifically, does KUBE_MASTER point at the master? | 19:33 |
| openstackgerrit | Clark Boylan proposed a change to openstack/diskimage-builder: Allow for multiple image outputs from raw source https://review.openstack.org/125475 | 19:33 |
| sdake | yes points to 10.0.0.2 | 19:34 |
| clarkb | SpamapS: ^ now with backward compat. your suggestion was so simple I dunno why I didn't think of it | 19:34 |
| radez | sdake: did you push up new changes for the keystone container? I'm wanted to rebuild my pods again | 19:34 |
| sdake | radez yes they are possibly bust atm | 19:35 |
| radez | kk thx | 19:35 |
| larsks | sdake: Also KUBE_API_ADDRESS (in the same file)? | 19:35 |
| sdake | I am not sure if its my keystone change or a new version of kubernetes | 19:35 |
| sdake | yes kube-api-address is the same | 19:35 |
| sdake | give it a try radez - test for me ;) | 19:35 |
| radez | oh, ok will do | 19:36 |
| sdake | # Comma seperated list of minions | 19:36 |
| sdake | MINION_ADDRESSES="127.0.0.1" | 19:36 |
| larsks | sdake: That seems wrong. | 19:36 |
| larsks | In which file? | 19:36 |
| sdake | apiserver | 19:36 |
| larsks | Oh, in apiserver? That's okay. | 19:36 |
| larsks | That only matters on the master. | 19:36 |
| sdake | KUBE_API_ADDRESS is not localhost | 19:37 |
| openstackgerrit | Clark Boylan proposed a change to openstack/diskimage-builder: Allow for multiple image outputs from raw source https://review.openstack.org/125475 | 19:37 |
| sdake | I'm not sure why it seems to use that :( | 19:37 |
| larsks | How about /etc/kubernetes/config... | 19:37 |
| larsks | KUBE_ETCD_SERVERS? | 19:37 |
| sdake | KUBE_ETCD_SERVERS=http://10.0.0.2:4001 | 19:37 |
| larsks | Weird. Everything seems okay. | 19:37 |
| sdake | yup | 19:37 |
| sdake | still bust | 19:37 |
| sdake | time for some nictoeen back in a bit | 19:38 |
| sdake | say radez if the test works, please let me know | 19:38 |
| sdake | one thing I have goigng on different then both of you is I am using an updated kubernetes | 19:38 |
| radez | will do, just launched the db, doing keystone now | 19:38 |
| *** juice has quit IRC | 19:39 | |
| *** ccrouch has joined #tripleo | 19:41 | |
| radez | sdake: I can interact with the service token but not as the admin user | 19:41 |
| radez | all my calls say "this requires authentication" | 19:41 |
| larsks | radez: you mean, it works with the service token, but not with username/password? | 19:41 |
| radez | yup | 19:42 |
| sdake | radez double check the container started | 19:42 |
| sdake | and fpaste docker logs | 19:42 |
| larsks | Well, if it works with the token, the container started... | 19:42 |
| radez | it's started, I'm looking at the keystone logs in the conatainer | 19:42 |
| larsks | radez: what do your OS_* environment vars look like? | 19:42 |
| sdake | radez so the new container works then? | 19:42 |
| larsks | for usename/password you would need OS_AUTH_URL, OS_USERNAME, OS_PASSWORD, and OS_TENANT_NAME (or _ID). | 19:42 |
| radez | I see the admin user and tenant when I list users and tenants | 19:43 |
| radez | so it looks like they got created properly | 19:43 |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Restart container until mariadb is available https://review.openstack.org/125751 | 19:44 |
| *** juice has joined #tripleo | 19:45 | |
| openstackgerrit | A change was merged to stackforge/kolla: Restart container until mariadb is available https://review.openstack.org/125751 | 19:45 |
| jpeeler | larsks: there's no way to assign dependencies between containers? | 19:46 |
| larsks | Not right now! I think steve's just-let-kube-restart-things idea is a reasonable solution. | 19:46 |
| *** gfidente has quit IRC | 19:46 | |
| radez | sdake: yea new container works... what's the password the admin user gets? | 19:46 |
| radez | I think it's the password | 19:46 |
| radez | i did a password set and got a token, | 19:46 |
| jpeeler | how often does kube restart containers? maybe a sleep would have been good | 19:47 |
| * radez tries to do other user/password operations | 19:47 | |
| larsks | jpeeler: The more I work with kubernetes, the more I think someone should just rename it to "openstack conatiner service" and then intregate it with neutron, cinder, heat, etc. | 19:47 |
| larsks | radez: radez default admin password is "redhat", unless you change it in the keystone.json file. | 19:47 |
| larsks | (KEYSTONE_ADMIN_PASSWORD environment var) | 19:48 |
| radez | something is aout of wack with that... I had to set it to "redhat" for my keystonerc file that had redhat in it woudl owrk | 19:48 |
| radez | I'll check my env vars | 19:48 |
| jpeeler | that got changed earlier today to kolla | 19:49 |
| larsks | radez: ^^^^ | 19:49 |
| radez | haha, nice... sounds good thx jpeeler | 19:49 |
| sdake | btw radez you ahve +2 ability now | 19:58 |
| radez | oh cool, thx sdake | 19:59 |
| sdake | jpeeler it isn't the job of the container to control restart speed - ideally kubernetes would figure it out | 20:01 |
| sdake | and if it doesn't, it should :) | 20:01 |
| jpeeler | sdake: good point, i didn't think about it adjusting restart speed | 20:02 |
| sdake | restart in kube, should control restart cycle rate | 20:03 |
| sdake | if restart was done in the container, it would be a different story | 20:03 |
| larsks | radez: use your new super powers to +2 https://review.openstack.org/#/c/125416/, unless it conflicts with the changes you were making... | 20:04 |
| sdake | radez or jpeeler can one of you folks review https://review.openstack.org/#/c/125416/ | 20:04 |
| sdake | since radez already +1 | 20:05 |
| sdake | probably easiest for radez to review | 20:05 |
| *** rlandy has quit IRC | 20:05 | |
| *** radez is now known as radez_g0n3 | 20:06 | |
| *** radez_g0n3 is now known as radez | 20:06 | |
| *** akuznetsov has joined #tripleo | 20:10 | |
| sdake | larsks downgraded kube, kubecfg list pods works on minions now | 20:11 |
| larsks | interesting. I will have to investigate to see what changed. | 20:11 |
| sdake | it could have been bust by some interacvtion and not actually fixed by the downgrade | 20:12 |
| sdake | who knows | 20:12 |
| radez | just got and empty list back from glance list. need to add the services tenant to keystone and things look good | 20:12 |
| radez | sdake: are you still working on keystone? I'm gonna add the services tenant to the config script | 20:13 |
| radez | btw, what's the deal with the two pending merges? in gerrit? | 20:14 |
| sdake | which configf script | 20:14 |
| larsks | radez: which merges? | 20:14 |
| radez | https://review.openstack.org/#/c/125416/ | 20:14 |
| radez | https://review.openstack.org/#/c/125431/ | 20:14 |
| sdake | they need to go t hrough the check and gate queue | 20:14 |
| sdake | its probably quite long atm | 20:14 |
| *** akuznetsov has quit IRC | 20:14 | |
| sdake | zuul.openstack.org I think | 20:14 |
| radez | oh are they just in queue and waiting? | 20:15 |
| larsks | radez: well, the first still needs to +2 from someone not me... | 20:15 |
| larsks | and not sdake. | 20:15 |
| larsks | that's what we were asking you for a few lines back... | 20:15 |
| openstackgerrit | A change was merged to stackforge/kolla: update readme with project name https://review.openstack.org/125416 | 20:15 |
| larsks | ("unless it conflicts with the changes you were making...") | 20:15 |
| jpeeler | oh uh | 20:15 |
| larsks | Oh, hey, there it is. | 20:15 |
| jpeeler | hopefully no conflicts | 20:16 |
| larsks | rebasing is good practice :) | 20:16 |
| radez | nice | 20:16 |
| *** spzala has quit IRC | 20:16 | |
| radez | larsks: shouldn't conflict | 20:17 |
| larsks | radez: Yup, just +2d your changes. | 20:17 |
| radez | yay, we're starting to make some progress :) | 20:18 |
| sdake | when we have nova working | 20:18 |
| sdake | that will be progress ;-) | 20:18 |
| *** spzala has joined #tripleo | 20:18 | |
| radez | fair enough | 20:18 |
| radez | sdake: I was thinking in the start script for now to ad dthe services tenant | 20:18 |
| radez | where the admin user and tenant get created | 20:19 |
| sdake | which services and which start script | 20:19 |
| radez | keystone's start.sh | 20:19 |
| sdake | you mean add keystone dxyz to keystone's start.sh | 20:19 |
| radez | do a keystone tenant-create --name services | 20:19 |
| sdake | yes I have a pat ch let me try | 20:19 |
| sdake | my kube was bust | 20:19 |
| radez | that and the endpoints is what needed to be added manually for glance to function now | 20:20 |
| radez | I'm going to start looking into the os_config stuff and see if it can help us. | 20:20 |
| radez | jpeeler: what's the workflow check that you put on larsks's review? | 20:21 |
| larsks | radez: workflow +1 means "okay to merge now" | 20:21 |
| jpeeler | radez: the general openstack process is two wait for two core reviewers, so the second core checks that | 20:21 |
| larsks | You need that + passing jenkins tests for the automatic merge to happen. | 20:21 |
| radez | gotcha, core review is new to me, good to know | 20:22 |
| jpeeler | radez: i think line 112 you meant to say keystone instead of glance | 20:22 |
| radez | jpeeler: yup you're right | 20:23 |
| sdake | jpeeler kube has a 10 second delay between restarts - victory :) | 20:23 |
| *** rushiagr is now known as rushiagr_away | 20:23 | |
| jpeeler | sdake: oh cool | 20:23 |
| *** rushiagr_away is now known as rushiagr | 20:23 | |
| jpeeler | sdake: is that a setting somewhere? | 20:23 |
| sdake | no idea - I just see it through sudo docker ps | 20:24 |
| sdake | -a | 20:24 |
| sdake | my restart change appears not to work | 20:24 |
| sdake | but it appeared to work for radez | 20:24 |
| larsks | container does not exit, or kube does not reschedule? | 20:24 |
| sdake | container restarts continiously forever | 20:25 |
| sdake | option #3 ;) | 20:25 |
| larsks | restarts forever and never works? | 20:25 |
| sdake | perhaps it has old env | 20:25 |
| sdake | yup | 20:25 |
| sdake | I suspect it is not reloading the environment | 20:25 |
| sdake | if I was a go wizard I might fix that | 20:25 |
| *** andreaf has joined #tripleo | 20:27 | |
| openstackgerrit | Dan Radez proposed a change to stackforge/kolla: updating the README with documentation on launching keystone https://review.openstack.org/125431 | 20:27 |
| larsks | we might want to instrument the start scripts to dump environment on stdout so that it shows up in docker log... | 20:27 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Add unit test for cache-url https://review.openstack.org/84493 | 20:28 |
| radez | jpeeler: updated that glance to keystone | 20:28 |
| larsks | Might make this sort of thing easier to diagnose. | 20:28 |
| *** dprince has quit IRC | 20:28 | |
| openstackgerrit | A change was merged to stackforge/kolla: updating the README with documentation on launching keystone https://review.openstack.org/125431 | 20:29 |
| jpeeler | it might be better to use env anyway | 20:32 |
| sdake | still seeing this | 20:33 |
| sdake | 2014-10-02 20:31:12.972 37 TRACE keystone.common.wsgi OperationalError: (OperationalError) (2013, "Lost connection to MySQL server at 'reading initial communication packet', system error: 0") None None | 20:33 |
| morganfainberg | sdake, every time or it recovers? | 20:33 |
| *** andreaf has quit IRC | 20:33 | |
| morganfainberg | as in subsequent queries work? | 20:33 |
| sdake | I killed it then started fresh | 20:33 |
| sdake | every time getting t hat | 20:34 |
| sdake | its off and on depends on some startup relationship I think | 20:34 |
| radez | slagle: is there a copr repo for that os-cloud-config project by chance? | 20:35 |
| morganfainberg | sdake, i only ask because if it was a one off then recovered https://bugs.launchpad.net/keystone/+bug/1374497 updating oslo.db to 1.0.2 should solve it | 20:36 |
| uvirtbot | Launchpad bug 1374497 in oslo.db "change in oslo.db "ping" handling is causing issues in projects that are not using transactions" [High,Fix committed] | 20:36 |
| morganfainberg | if it was *always* having the issue, different story | 20:36 |
| *** dsneddon has quit IRC | 20:36 | |
| sdake | thanks morgan | 20:36 |
| *** dsneddon has joined #tripleo | 20:37 | |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to openstack/os-cloud-config: fix formatting issues in README.rst https://review.openstack.org/125775 | 20:40 |
| slagle | radez: there is, here: https://copr.fedoraproject.org/coprs/slagle/openstack-m/ | 20:40 |
| radez | awesome, thx slagle | 20:40 |
| slagle | radez: the same build is in koji as well if you want to install from there instead | 20:40 |
| slagle | radez: koji might be a safer bet, so you don't accidentally pull anything else from my copr repo | 20:41 |
| slagle | https://kojipkgs.fedoraproject.org//packages/os-cloud-config/0.1.11/1.fc22/noarch/os-cloud-config-0.1.11-1.fc22.noarch.rpm | 20:41 |
| radez | cool, thx | 20:41 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to openstack/os-cloud-config: fix formatting issues in README.rst https://review.openstack.org/125775 | 20:42 |
| radez | slagle: anyreason it's not built for F20? | 20:42 |
| slagle | radez: that build will work on f20. it's built from rawhide for now for rdo juno | 20:44 |
| * radez nods | 20:44 | |
| slagle | rdo juno will use the same build from master for f21 and f20 | 20:44 |
| radez | I was just thinking i could yum install it directly from fedora | 20:44 |
| sdake | [948]: E1002 20:44:01.531584 00948 etcd.go:166] Failed to get the key: registry/ | 20:44 |
| sdake | services/endpoints/mariadbmaster 100: Key not found (/registry/services/endpoint | 20:44 |
| sdake | s) [327] | 20:44 |
| radez | instead of having to pass the url :) | 20:44 |
| slagle | radez: oh, no you can't do that | 20:44 |
| sdake | larsks where do I find that registry above? ^^ | 20:44 |
| slagle | it's new in juno :) | 20:44 |
| radez | slagle: will it not work with icehoust? | 20:45 |
| slagle | it may, i don't know for sure. it uses the openstack clients | 20:45 |
| larsks | sdake: You mean from that "can't find registry" error re: etcd? | 20:45 |
| sdake | yup, trying to debug - want to look at etcd registry | 20:46 |
| radez | slagle: k, I'll give it a whirl | 20:46 |
| slagle | radez: can you see the versions required for the openstack clients in os-cloud-config's requirements.txt | 20:46 |
| radez | k | 20:46 |
| sdake | larsks but I'm unclear how precisely to do that | 20:46 |
| larsks | Right, I was asking if that was the thing to which you were referring, so that I could give you a relevant answer :) | 20:47 |
| *** rhallisey has quit IRC | 20:47 | |
| *** andreaf has joined #tripleo | 20:47 | |
| *** morgan_remote_ is now known as remote_morgan_ | 20:48 | |
| larsks | sdake: that should be a key in etcd. You can see keys with curl: curl http://KUBE_MASTER_IP:4001/v2/keys/ | 20:48 |
| larsks | That gets you the top level. | 20:48 |
| larsks | To see the registry, http://KUBE_MASTER_IP:4001/v2/keys/registry/ | 20:49 |
| larsks | If jq is installed, pipe those to "jq ." for nicer output. | 20:49 |
| larsks | sdake: E.g., on my system, just now, that gets me: http://chunk.io/f/0236bc5533bb4977a46386fb4934860e | 20:50 |
| *** remote_morgan_ is now known as morgan_remote_ | 20:50 | |
| larsks | and /registry/pods is: http://chunk.io/f/cb795378289b426197f55089044d1fb0 | 20:50 |
| sdake | {"errorCode":100,"message":"Key not found","cause":"/registry/services/endpoints","index":387} | 20:50 |
| *** morgan_remote_ is now known as morgan_remote | 20:50 | |
| *** morgan_remote is now known as remote_morgan_ | 20:51 | |
| larsks | Fwiw, I just deployed from the heat templates and was able to spin up the mariadb container without errors... | 20:51 |
| larsks | I have to run out for a bit; this is when I turn into a kid transport device. | 20:51 |
| * sdake shakes fists at computer | 20:52 | |
| radez | sdake: do you think it would be good to mark the mariadb blueprint as something other than delivery=unknown? | 20:52 |
| radez | keystone too for that matter | 20:53 |
| larsks | sdake: It looks like I'm running kubernetes-0.3-0.4.gitd381606.fc21.x86_64; is this what you've got? | 20:53 |
| *** rushiagr is now known as rushiagr_away | 20:53 | |
| larsks | radez: You could mark them as delivery = Oct. 20 to make James Lablocki happy :) | 20:54 |
| radez | lol the options are 'in progress" and "beta available and things like that... no dates | 20:55 |
| sdake | larsks yup sure do | 20:55 |
| sdake | radez sure feel free to mark it | 20:55 |
| sdake | I think its done already | 20:56 |
| sdake | but I could be wrong | 20:56 |
| sdake | we need a new blueprint with "mariadb needs persistent storage" | 20:56 |
| sdake | because that is another big chunk of work | 20:56 |
| sdake | right now if mariadb restarts = byebye storage | 20:56 |
| radez | hm, looks like you have to be assigned to the blue print to update its status | 20:57 |
| radez | sdake: everythink will need persistent storage, smae thing with the glance images right now. | 20:58 |
| *** sdake_ has joined #tripleo | 20:59 | |
| *** untriaged-bot has joined #tripleo | 21:00 | |
| untriaged-bot | Untriaged bugs so far: | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1375414 | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1376200 | 21:00 |
| uvirtbot | Launchpad bug 1375414 in tripleo "mysql/rabbit overcloud issues when the bootstrap node is removed" [Undecided,New] | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1375422 | 21:00 |
| uvirtbot | Launchpad bug 1376200 in tripleo "Stale connections when VIP is moved" [High,New] | 21:00 |
| uvirtbot | Launchpad bug 1375422 in tripleo "Vlan not recreated on overcloud controller reboot" [Undecided,In progress] | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1366164 | 21:00 |
| uvirtbot | Launchpad bug 1366164 in tripleo "Add domain roles for admin user" [Undecided,New] | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1367742 | 21:00 |
| uvirtbot | Launchpad bug 1367742 in tripleo "keepalived brings the VIP up in multiple places on a controller reboot" [Undecided,New] | 21:00 |
| *** untriaged-bot has quit IRC | 21:00 | |
| sdake_ | boy my minidell needs an ssd | 21:08 |
| *** akuznetsov has joined #tripleo | 21:11 | |
| openstackgerrit | Dan Radez proposed a change to stackforge/kolla: Updates to glance start scripts, db and user/role creation - adding db sync and db creation as utf8 - fixing user and role creation https://review.openstack.org/125791 | 21:13 |
| openstackgerrit | Clark Boylan proposed a change to openstack/diskimage-builder: Allow for multiple image outputs from raw source https://review.openstack.org/125475 | 21:14 |
| *** akuznetsov has quit IRC | 21:15 | |
| *** radez is now known as radez_g0n3 | 21:15 | |
| sdake_ | radez_g0n3 if you come back can you test the latest keystone container | 21:18 |
| sdake_ | something definatley busted in my d ev env | 21:18 |
| *** jsidhu has joined #tripleo | 21:24 | |
| *** jdob has quit IRC | 21:37 | |
| *** corvus has quit IRC | 21:40 | |
| *** corvus has joined #tripleo | 21:41 | |
| *** noslzzp has quit IRC | 21:48 | |
| openstackgerrit | Clark Boylan proposed a change to openstack/diskimage-builder: Allow for multiple image outputs from raw source https://review.openstack.org/125475 | 21:53 |
| chmouel | sdake_: i am thinking to start working on the swift implementation | 21:55 |
| chmouel | but I could not find any way to share data in kubernetes | 21:55 |
| chmouel | as this is needed for having to share the rings between the different swift services | 21:55 |
| chmouel | any ideas? | 21:56 |
| sdake_ | there is no persistent storage | 21:56 |
| sdake_ | it is in the queue - I saw a pull requewst a few days ago for one | 21:56 |
| sdake_ | do you need messaging or persistent storage? | 21:56 |
| chmouel | I can find a workaround until then i guess | 21:56 |
| chmouel | sdake_: persistent storage | 21:56 |
| chmouel | so basically you generate the rings once | 21:56 |
| *** jsidhu has quit IRC | 21:56 | |
| chmouel | and send them to all the other swift proxies | 21:57 |
| chmouel | it doesn't need to be shared storage, it just need to be the same on all host | 21:57 |
| chmouel | containers i mean | 21:57 |
| chmouel | i guess i can have swift-ring container with a small http server and the other containers linking to it and getting the rings | 21:58 |
| sdake_ | chmouel there are two optoins | 21:59 |
| sdake_ | persistent storage - once that hits the upstream | 21:59 |
| sdake_ | or use etcd | 21:59 |
| chmouel | i can't really store the rings in etcd, can I? | 21:59 |
| sdake_ | I am not certain | 21:59 |
| sdake_ | etcd can store key/value pairs | 21:59 |
| chmouel | rings are binary blobs | 21:59 |
| sdake_ | so you could store a key with the value eing the ring | 21:59 |
| sdake_ | could base 64 encode ;-) | 21:59 |
| sdake_ | probably best to wait for persistent storage | 22:00 |
| sdake_ | everyone is complaining for it so it should happen soon | 22:00 |
| chmouel | yeah i am not sure how that would work tho persistent storage in a distributed environement but i guess they will figure out :) | 22:00 |
| sdake_ | well it may not be shared | 22:00 |
| sdake_ | I am not sure | 22:00 |
| sdake_ | it may just be persistent between restarts | 22:01 |
| chmouel | humm i see, would not solve my issue tho, but i think the swift-ring container should be work (and be an elegant solution) | 22:02 |
| chmouel | fyi: those guys are doing it https://clusterhq.com/ with some zfs magics | 22:03 |
| *** rainya has quit IRC | 22:06 | |
| chmouel | sdake_: ah btw: i could not assign myself the bp, i am not sure what rights i need to do that | 22:06 |
| sdake_ | chmouel what is your launchpad login | 22:07 |
| chmouel | sdake_: chmouel | 22:07 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Save extended attributes when creating tar https://review.openstack.org/125428 | 22:08 |
| *** akuznetsov has joined #tripleo | 22:12 | |
| sdake_ | chmouel you should be able to join now | 22:12 |
| sdake_ | rather modify blueprints | 22:12 |
| openstackgerrit | Richard Su proposed a change to openstack/tripleo-image-elements: Fix heat purge deleted cron job https://review.openstack.org/121744 | 22:13 |
| *** rhallisey has joined #tripleo | 22:13 | |
| chmouel | sdake_: works, tks | 22:14 |
| *** marun has joined #tripleo | 22:15 | |
| *** akuznetsov has quit IRC | 22:16 | |
| *** sdake_ has quit IRC | 22:17 | |
| *** jang1 has quit IRC | 22:22 | |
| *** andreaf has quit IRC | 22:36 | |
| *** andreaf has joined #tripleo | 22:36 | |
| *** ccrouch has quit IRC | 22:45 | |
| *** ccrouch has joined #tripleo | 22:45 | |
| *** dvossel has quit IRC | 22:48 | |
| *** ccrouch has quit IRC | 22:50 | |
| *** marun has quit IRC | 23:01 | |
| *** yamahata has joined #tripleo | 23:04 | |
| *** nosnos has joined #tripleo | 23:06 | |
| *** akuznetsov has joined #tripleo | 23:12 | |
| *** akuznetsov has quit IRC | 23:17 | |
| *** noslzzp has joined #tripleo | 23:25 | |
| *** yamahata has quit IRC | 23:27 | |
| *** noslzzp has quit IRC | 23:38 | |
| *** ccrouch has joined #tripleo | 23:46 | |
| *** andreaf has quit IRC | 23:47 | |
| *** andreaf has joined #tripleo | 23:48 | |
| *** noslzzp has joined #tripleo | 23:48 | |
| *** rainya has joined #tripleo | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!