| *** r-mibu has quit IRC | 00:26 | |
| *** r-mibu has joined #tripleo | 00:26 | |
| *** dmacpher has joined #tripleo | 00:36 | |
| *** panda has quit IRC | 00:48 | |
| *** panda has joined #tripleo | 00:49 | |
| *** sthillma has quit IRC | 00:58 | |
| openstackgerrit | James Slagle proposed openstack/instack-undercloud: Ensure systemd-journald is running https://review.openstack.org/320643 | 01:18 |
|---|---|---|
| *** yamahata has joined #tripleo | 01:19 | |
| *** fzdarsky_ has joined #tripleo | 01:19 | |
| *** fzdarsky|afk has quit IRC | 01:23 | |
| openstackgerrit | Merged openstack/tripleo-puppet-elements: Remove puppet-qpid from module list https://review.openstack.org/317641 | 01:24 |
| *** jcoufal has quit IRC | 01:29 | |
| *** cwolferh has quit IRC | 01:36 | |
| *** diazjf has joined #tripleo | 01:54 | |
| *** diazjf has quit IRC | 01:56 | |
| *** julim has joined #tripleo | 02:04 | |
| *** rhallisey has joined #tripleo | 02:06 | |
| *** rhallisey has quit IRC | 02:07 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: WIP nuke khaleesi https://review.openstack.org/320740 | 02:22 |
| *** yamahata has quit IRC | 02:26 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: WIP nuke khaleesi https://review.openstack.org/320740 | 02:37 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: WIP nuke khaleesi https://review.openstack.org/320740 | 02:42 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: WIP nuke khaleesi https://review.openstack.org/320740 | 02:45 |
| *** penick has quit IRC | 02:45 | |
| *** TSCHAK has joined #tripleo | 02:52 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: WIP nuke khaleesi https://review.openstack.org/320740 | 02:59 |
| *** absubram has joined #tripleo | 03:06 | |
| *** absubram_ has joined #tripleo | 03:10 | |
| *** absubram has quit IRC | 03:11 | |
| *** absubram_ is now known as absubram | 03:11 | |
| *** mburned has quit IRC | 03:12 | |
| *** mburned has joined #tripleo | 03:14 | |
| *** TSCHAK has quit IRC | 03:17 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: WIP nuke khaleesi https://review.openstack.org/320740 | 03:31 |
| *** myoung has quit IRC | 03:36 | |
| *** myoung has joined #tripleo | 03:36 | |
| *** stendulker has joined #tripleo | 03:39 | |
| *** ayoung has quit IRC | 03:42 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: Remove dependency on the khaleesi project https://review.openstack.org/320740 | 03:52 |
| *** yamahata has joined #tripleo | 03:54 | |
| *** dmacpher has quit IRC | 04:00 | |
| *** cwolferh has joined #tripleo | 04:01 | |
| *** links has joined #tripleo | 04:01 | |
| *** dmacpher has joined #tripleo | 04:03 | |
| *** masco has joined #tripleo | 04:03 | |
| *** mburned has quit IRC | 04:32 | |
| *** mburned has joined #tripleo | 04:32 | |
| *** shadower has quit IRC | 04:36 | |
| *** shadower has joined #tripleo | 04:36 | |
| *** liverpooler has quit IRC | 04:39 | |
| *** jaosorior has joined #tripleo | 04:59 | |
| *** rlandy_ has quit IRC | 05:11 | |
| *** cwolferh has quit IRC | 05:19 | |
| *** cwolferh has joined #tripleo | 05:20 | |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack-infra/tripleo-ci: Replace heat commands with openstack stack https://review.openstack.org/320780 | 05:20 |
| jaosorior | coolsvap: ping | 05:23 |
| coolsvap | jaosorior, pong | 05:23 |
| coolsvap | jaosorior, good morning | 05:24 |
| jaosorior | good morning dude | 05:24 |
| jaosorior | coolsvap, did you get your environment working? | 05:25 |
| coolsvap | jaosorior, almost done | 05:25 |
| coolsvap | doing the pingtest | 05:25 |
| jaosorior | awesome! | 05:25 |
| jaosorior | coolsvap: I -1ed the commit you just did above ^^ | 05:26 |
| *** dsariel has quit IRC | 05:27 | |
| *** numans has joined #tripleo | 05:27 | |
| coolsvap | jaosorior, but if we dont backport it will it still cause problem? | 05:27 |
| coolsvap | i mean stable checks will be done from stable branches, right? | 05:27 |
| jaosorior | coolsvap, I'm just re-verifying | 05:27 |
| jaosorior | we had some wrapper already set | 05:27 |
| coolsvap | jaosorior, yeah lets see | 05:28 |
| *** rcernin has joined #tripleo | 05:29 | |
| jaosorior | coolsvap: we do have a wrapper called function "openstack" | 05:29 |
| jaosorior | which at some point caused the stable branches to fail | 05:29 |
| jaosorior | so we had removed it | 05:29 |
| jaosorior | ccamacho was working on that | 05:29 |
| jaosorior | so you might want to ping him when he's online (in a few mins) | 05:29 |
| jaosorior | (I think) | 05:29 |
| coolsvap | jaosorior, yeah sure | 05:29 |
| jaosorior | but yeah, if that works then it's all good | 05:30 |
| coolsvap | i will mark the change WIP | 05:30 |
| coolsvap | jaosorior, technically we should be able to do it | 05:30 |
| jaosorior | we should, yeah | 05:30 |
| jaosorior | coolsvap: Anyway, part of the issue is that we don't have branches for the tripleo-ci repo | 05:31 |
| jaosorior | so master is used for everything | 05:31 |
| coolsvap | jaosorior, ohk | 05:32 |
| *** dmacpher_ has joined #tripleo | 05:39 | |
| *** dmacpher has quit IRC | 05:43 | |
| *** apetrich has joined #tripleo | 05:45 | |
| *** apetrich has quit IRC | 05:49 | |
| *** apetrich has joined #tripleo | 05:50 | |
| *** cwolferh has quit IRC | 05:55 | |
| *** olap has joined #tripleo | 05:58 | |
| *** liverpooler has joined #tripleo | 05:58 | |
| *** oshvartz has joined #tripleo | 06:13 | |
| *** rcernin has joined #tripleo | 06:21 | |
| *** paramite has joined #tripleo | 06:22 | |
| *** aufi has joined #tripleo | 06:27 | |
| *** mcornea has joined #tripleo | 06:31 | |
| *** cwolferh has joined #tripleo | 06:35 | |
| *** shardy has joined #tripleo | 06:41 | |
| *** tesseract has joined #tripleo | 06:47 | |
| *** paramite is now known as paramite|afk | 06:48 | |
| *** paramite|afk is now known as paramite | 06:49 | |
| *** tremble has joined #tripleo | 06:54 | |
| openstackgerrit | Merged openstack/tripleo-common: Removes unused code https://review.openstack.org/320675 | 06:57 |
| *** jprovazn has joined #tripleo | 06:57 | |
| *** ccamacho has joined #tripleo | 07:03 | |
| *** gumbyblake has joined #tripleo | 07:06 | |
| coolsvap | ccamacho, morning :) | 07:08 |
| ccamacho | morning all! | 07:08 |
| ccamacho | hey coolsvap! | 07:08 |
| shardy | 'morning everyone | 07:09 |
| *** olap has quit IRC | 07:09 | |
| ccamacho | coolsvap, manage to have the CI up and running?! | 07:09 |
| coolsvap | ccamacho, i am almost done with the ci setup i did, its failing at the overcloud-ping | 07:09 |
| coolsvap | not able to create stack | 07:09 |
| *** ifarkas has joined #tripleo | 07:10 | |
| *** athomas has joined #tripleo | 07:10 | |
| *** gumbyblake has quit IRC | 07:10 | |
| ccamacho | coolsvap, do you have the error¿? | 07:11 |
| marios | o/ tripleo * | 07:11 |
| coolsvap | ccamacho, its failing with no valid host found | 07:11 |
| *** ebarrera has joined #tripleo | 07:12 | |
| ccamacho | shardy, quick question, why are mandatory EndpointMap, MysqlVirtualIPUri for ControllerServiceChain? do we need them for the other roles? | 07:13 |
| *** paramite is now known as paramite|afk | 07:14 | |
| *** florianf has joined #tripleo | 07:14 | |
| ccamacho | coolsvap, do you have available vm's to deploy the tenant-stack? | 07:14 |
| shardy | ccamacho: Yes, most services have to setup endpoints (which comes from the EndpointMap) | 07:14 |
| shardy | the Mysql Uri is planned to move inside the EndpointMap (gfidente has a patch posted) which will make only the EndpointMap mandatory | 07:15 |
| coolsvap | ccamacho, i belive i do, the overcloud deploy creates 1 controller 1 compute deployment, am I correct> | 07:15 |
| ccamacho | shardy, but they are mandatory, even if the service does not use it? | 07:15 |
| shardy | ccamacho: yes, because we pass them here: | 07:16 |
| shardy | https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/services/services.yaml#L31 | 07:16 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Update FormErrorList component https://review.openstack.org/320498 | 07:16 |
| shardy | therefore they are mandatory and must be defined in all service templates, even if they're not used | 07:16 |
| shardy | otherwise heat will fail to pass those properties into the nested stack, and a validation error will occur | 07:17 |
| shardy | ccamacho: think of it as the common interface for all services | 07:17 |
| *** mikelk has joined #tripleo | 07:17 | |
| shardy | it's fine if a service ignores them | 07:17 |
| ccamacho | shardy, I realized that when the patch for the compute node, I see it now "heat will fail to pass those properties into the nested stack, and a validation error will occur" thanks! | 07:18 |
| openstackgerrit | Merged openstack/python-tripleoclient: Add Heat CloudFormation service for endpoint creation https://review.openstack.org/318705 | 07:18 |
| shardy | ccamacho: np - also see https://review.openstack.org/#/c/320361/ where I wired it in for the ObjectStorage node | 07:19 |
| shardy | the main decision we have to make for compute nodes is which steps it supports | 07:19 |
| shardy | currently there is only one, which is fine as a first step I guess | 07:19 |
| *** paramite|afk is now known as paramite | 07:19 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Optimize Nova disk_cachemodes and hw_disk_discard options for RBD https://review.openstack.org/269561 | 07:20 |
| *** jtomasek has joined #tripleo | 07:20 | |
| *** cwolferh has quit IRC | 07:27 | |
| *** jpena|off is now known as jpena | 07:29 | |
| *** fzdarsky_ is now known as fzdarsky | 07:30 | |
| ccamacho | shardy, should be step 2 right (Core Services) ? | 07:30 |
| *** zoli_gone-proxy is now known as zoliXXL | 07:30 | |
| ccamacho | I added 2, 3 and 4 | 07:33 |
| shardy | ccamacho: I think most of what overcloud_compute is doing is better aligned with step 4, e.g general openstack services? | 07:35 |
| zoliXXL | good morning | 07:35 |
| shardy | adding 2, 3 and 4 sounds OK to me tho, and it aligns with what I've done on the swift role (although that was already doing things in those three steps) | 07:36 |
| shardy | eventually I think we'll want a common pattern where all roles can run all steps, but that can come later | 07:36 |
| *** zoliXXL is now known as zoli|afk | 07:37 | |
| ccamacho | shardy, ack, when have time, Could you give it a look? to https://review.openstack.org/#/c/319873/ | 07:38 |
| shardy | will do | 07:39 |
| shardy | ccamacho: Hey, you mentioned only being able to do 2 node deployments recently - what hardware do you have? | 07:39 |
| shardy | I've been running 3-node plus undercloud on a 32G desktop box OK, it does have ksm enabled and some swap. | 07:40 |
| ccamacho | shardy, is a 32GB box, but have (2+1nodes) them with 9GB each, when using less I had mem issues, how many RAM are you giving to each node? | 07:45 |
| ccamacho | To try it with your config | 07:45 |
| *** jaosorior has quit IRC | 07:45 | |
| *** chlong has quit IRC | 07:46 | |
| *** anshul has joined #tripleo | 07:46 | |
| shardy | I've got a 6G undercloud VM and three 8G VM nodes for the overcloud | 07:46 |
| openstackgerrit | Swapnil Kulkarni (coolsvap) proposed openstack-infra/tripleo-ci: Replace heat commands with openstack stack https://review.openstack.org/320780 | 07:47 |
| shardy | I did enable swap on the undercloud, but it doesn't use it much | 07:47 |
| coolsvap | ccamacho, any comments related to the above review | 07:47 |
| *** anshul is now known as Guest50431 | 07:47 | |
| shardy | ccamacho: with ksm you may be able to leave your current allocations and run three node deployments | 07:47 |
| shardy | I'd try it before changing anything :) | 07:48 |
| *** dsariel has joined #tripleo | 07:50 | |
| ccamacho | shardy, Ill check it now :) | 07:52 |
| shardy | ccamacho: your patch looks pretty close, I added a few minor comments | 07:52 |
| *** dmacpher_ has quit IRC | 07:55 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Test do not merge https://review.openstack.org/320819 | 07:56 |
| *** sshnaidm has quit IRC | 07:57 | |
| shardy | This script https://review.openstack.org/#/c/299643/ is ready to land I think, I've been using it locally and it's pretty useful for the composable services stuff where you're constantly updating puppet-tripleo profiles | 08:00 |
| ccamacho | shardy, thanks! | 08:00 |
| *** paramite is now known as paramite|afk | 08:02 | |
| *** itamarl has joined #tripleo | 08:02 | |
| openstackgerrit | Merged openstack/tripleo-docs: Document --initial-state param on baremetal import https://review.openstack.org/320001 | 08:03 |
| openstackgerrit | Merged openstack/tripleo-ui: Refactor notification for better display https://review.openstack.org/316874 | 08:04 |
| *** olap has joined #tripleo | 08:05 | |
| shardy | jistr: Hey, I added some comments to https://review.openstack.org/#/c/313544/ | 08:09 |
| shardy | not blocking, but wondering if there's some way we can get the same data with the existing interfaces | 08:09 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Test do not merge https://review.openstack.org/319135 | 08:12 |
| *** bvandenh has joined #tripleo | 08:14 | |
| openstackgerrit | Carlos Camacho proposed openstack-infra/tripleo-ci: Fix validation for stable branch https://review.openstack.org/312350 | 08:15 |
| *** paramite|afk is now known as paramite | 08:16 | |
| *** lucas-afk is now known as lucasagomes | 08:19 | |
| jistr | shardy: thanks, will check them out | 08:24 |
| openstackgerrit | Carlos Camacho proposed openstack-infra/tripleo-ci: Update deprecated references to heat in tripleo.sh https://review.openstack.org/312350 | 08:26 |
| coolsvap | ccamacho, i have abandoned the change | 08:26 |
| ccamacho | coolsvap, if have some time you might review it, I have long time trying to pass CI on this (https://review.openstack.org/312350), the patch name wasnt quite descriptive, is already working, but not able to have all green in other branches (liberty and mitaka), depending on the day... Have greens in one job, other days in other jobs.. :) | 08:28 |
| *** dmk0202 has joined #tripleo | 08:28 | |
| hewbrocca | shardy: user requests for ops-tools for the undercloud | 08:29 |
| hewbrocca | log aggregation, etc... | 08:29 |
| coolsvap | ccamacho, sure i can | 08:30 |
| *** coolsvap is now known as coolsvap_ | 08:30 | |
| *** yamahata has quit IRC | 08:30 | |
| shardy | hewbrocca: Hi, yeah there are some patches posted from EmilienM, not all of them have landed yet | 08:30 |
| shardy | the main question is how do we test it, as we can't enable it all by default for CI/dev usage | 08:31 |
| ccamacho | coolsvap, thanks! I think the CI failures are not related to the script (can't allocate memory, no hosts available, API errors in keystone) but I wanted to be 100% before breaking things again.. | 08:32 |
| hewbrocca | right, not if we ever want a job to finish | 08:32 |
| shardy | https://review.openstack.org/#/c/258653/ has the logging pieces, sensu already landed | 08:34 |
| shardy | hewbrocca: we can revive that, but there's also the resource question in general, e.g will folks complain when they deploy everything then deploy 100 node stacks and it uses a ton of ram | 08:34 |
| *** derekh has joined #tripleo | 08:34 | |
| hewbrocca | Yes | 08:35 |
| jistr | shardy: good points re service manifest. Yea i can see we could probably write the OS::TripleO::... service list instead, and try detecting service properties (e.g. from hiera as you suggested) instead of writing them explicitly. Just btw one other use i had in mind for service manifest was to provide information to upgrade tooling. We could as well output the service list there too probably. Detecting | 08:35 |
| jistr | things like pacemaker-managed then means actually connecting to nodes via ssh and "introspecting" them for particular properties, but that should also be reasonably doable e.g. with Ansible. | 08:35 |
| shardy | one option is to deploy the logging stack on a non-undercloud node, but that's more work | 08:35 |
| hewbrocca | it's worth a broader conversation | 08:35 |
| shardy | (probably not that much more, but we'd have to integrate it into the workflow) | 08:35 |
| * hewbrocca thinks | 08:35 | |
| hewbrocca | Is there any reason you wouldn't use the same log aggregator to pull logs from the undercloud as well as the overcloud machines? | 08:35 |
| shardy | hewbrocca: some of the heat resource usages issues are soon to be fixed, but it's still a question of figuring out what a reasonable requirement for the undercloud is | 08:36 |
| hewbrocca | yeah | 08:36 |
| shardy | hewbrocca: I guess not | 08:36 |
| *** fragatina has quit IRC | 08:36 | |
| hewbrocca | the deployer I'm currently talking to is very interested in being able to look at undercloud logs as well | 08:37 |
| *** dtantsur|afk is now known as dtantsur | 08:37 | |
| *** mgould|afk is now known as mgould | 08:37 | |
| shardy | hewbrocca: ack - I guess we need to decide if the answer is easily hook into an existing logging stack (for overcloud and undercloud) vs integrated all-the-things on the undercloud | 08:38 |
| shardy | I'm leaning towards the former myself | 08:38 |
| hewbrocca | *nod* | 08:38 |
| shardy | because it also means you don't lose overcloud logging etc whenever you take the undercloud down for an update | 08:38 |
| hewbrocca | Yes | 08:38 |
| hewbrocca | The *only* drawback there is that now we need another machine | 08:38 |
| shardy | and it's going to be more scalable without forcing us to do fully-composable undercloud | 08:38 |
| shardy | Yeah | 08:39 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: Merge with fuel-infra/puppet-pacemaker https://review.openstack.org/296440 | 08:39 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge. https://review.openstack.org/310713 | 08:39 |
| shardy | I mean folks could probably deploy the undercloud and the logging stack in VMs on one big machine | 08:39 |
| hewbrocca | But, if users want proper monitoring, they need that machine anyway | 08:39 |
| hewbrocca | Yeah | 08:40 |
| hewbrocca | yeah that's the right approach I guess | 08:40 |
| *** devvesa has joined #tripleo | 08:40 | |
| shardy | A lot of the undercloud scalability issues have been related to the monolithic deployment - so it's be good to break out obviously-resource-intensive pieces vs compounding that problem | 08:40 |
| shardy | The other thing I've heard suggested is deploying management services actually *on* the overcloud, e.g storage management nodes etc, but again that has some negative implications in terms of maintenance outages etc | 08:41 |
| *** zoli|afk is now known as zoli | 08:42 | |
| *** zoli is now known as zoliXXL | 08:42 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: Merge with fuel-infra/puppet-pacemaker https://review.openstack.org/296440 | 08:42 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge. https://review.openstack.org/310713 | 08:44 |
| *** saneax_AFK is now known as saneax | 08:44 | |
| jistr | shardy: also given that we want to only pursue the improved HA architecture and not do dual arch, we probably won't need the pacemaker-managed information anywhere, as we'll always be able to assume the single expected config | 08:46 |
| jistr | i'll WIP the manifest patches and try the approach you suggested | 08:47 |
| *** jaosorior has joined #tripleo | 08:47 | |
| shardy | jistr: ack - thanks, I'm open to ideas, but it'd be good to avoid too much duplication of data | 08:47 |
| *** panda has quit IRC | 08:48 | |
| jistr | yea, agreed :) | 08:49 |
| *** panda has joined #tripleo | 08:49 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-tripleo: WIP: integration of the new puppet pacemaker. https://review.openstack.org/309069 | 08:49 |
| *** sshnaidm has joined #tripleo | 08:51 | |
| *** mbound has joined #tripleo | 08:55 | |
| *** paramite is now known as paramite|afk | 08:59 | |
| *** numans has quit IRC | 09:05 | |
| openstackgerrit | Imre Farkas proposed openstack/python-tripleoclient: Add Ironic password generation during deployment https://review.openstack.org/316146 | 09:13 |
| *** gfidente has joined #tripleo | 09:14 | |
| *** gfidente has quit IRC | 09:14 | |
| *** gfidente has joined #tripleo | 09:14 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: Merge with fuel-infra/puppet-pacemaker https://review.openstack.org/296440 | 09:22 |
| *** sambetts|afk is now known as sambetts | 09:23 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge. https://review.openstack.org/310713 | 09:24 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge. https://review.openstack.org/310713 | 09:32 |
| *** bvandenh has quit IRC | 09:34 | |
| *** numans has joined #tripleo | 09:35 | |
| openstackgerrit | Liu Qing proposed openstack/diskimage-builder: Fix variable unbound error while REPOREF="*" https://review.openstack.org/320859 | 09:39 |
| *** paramite|afk is now known as paramite | 09:47 | |
| *** zoliXXL is now known as zoli|lunch | 09:49 | |
| *** coolsvap_ is now known as coolsvap | 09:50 | |
| *** akrivoka has joined #tripleo | 09:55 | |
| *** fragatina has joined #tripleo | 09:55 | |
| *** fragatina has quit IRC | 09:59 | |
| *** devvesa has quit IRC | 10:09 | |
| *** milan has joined #tripleo | 10:16 | |
| *** devvesa has joined #tripleo | 10:24 | |
| openstackgerrit | Steven Hardy proposed openstack-infra/tripleo-ci: Make deploy validation errors/warnings fatal for CI https://review.openstack.org/320883 | 10:27 |
| *** tosky has joined #tripleo | 10:34 | |
| *** electrofelix has joined #tripleo | 10:35 | |
| *** devvesa has quit IRC | 10:39 | |
| shardy | derekh: Hey FYI I'm trying ^^ as I'm wondering if the lack of image building is causing us to see more no-valid-host-found errors | 10:47 |
| shardy | e.g if we're racing nova to make the newly registered nodes available | 10:47 |
| shardy | I guess if it shows errors related to that we could put a loop in somewhere to poll for the expected node count | 10:48 |
| *** zoli|lunch is now known as zoli | 10:49 | |
| *** zoli is now known as zoliXXL | 10:49 | |
| openstackgerrit | yolanda.robla proposed openstack/diskimage-builder: Fix yum-minimal locale config https://review.openstack.org/319909 | 10:50 |
| yolanda | ianw, greghaynes , SpamapS ^ | 10:51 |
| *** mburned has quit IRC | 10:51 | |
| derekh | shardy: the image build part (or lack of when using the cache) happens before we register nodes, so it shouldn't change anything there | 10:53 |
| derekh | shardy: and we ahve a sleep 60 after that | 10:54 |
| *** fragatina has joined #tripleo | 10:55 | |
| shardy | derekh: Hmm, OK - well I'm looking at e.g http://logs.openstack.org/14/308514/10/check-tripleo/gate-tripleo-ci-f22-nonha/156d183/console.html | 10:56 |
| shardy | we finish introspection, then about a minute later we start the overcloud deploy | 10:56 |
| shardy | I incorrectly assumed we used to do the image build before we setup the flavors | 10:56 |
| shardy | 2016-05-23 20:55:44.156 | tripleo.sh -- Introspect nodes - DONE. | 10:56 |
| jaosorior | shardy: Hey, what's the plan for the service loadbalancing in the composable roles world? Will each service profile have a map of the IPs (or the nodes) that are hosting the service and we will pass that to HAProxy? | 10:57 |
| shardy | 2016-05-23 20:56:48.187 | + /opt/stack/new/tripleo-ci/scripts/tripleo.sh --overcloud-deploy | 10:57 |
| shardy | that looked to me pretty close to the nova periodic job interval (1min, right?) | 10:57 |
| shardy | maybe the sleep 60 should come after the introspection? | 10:59 |
| shardy | (and/or we should poll at that point to check nova hypervisor stats) | 10:59 |
| shardy | I guess --validation-errors-fatal will tell us anyway :) | 10:59 |
| shardy | jaosorior: Yeah, we need a way to generate that list, I don't have an example of how that will be wired in yet | 11:00 |
| *** fragatina has quit IRC | 11:00 | |
| shardy | it's related to the service manifest I was discussing earlier with jistr | 11:00 |
| dtantsur | shardy, we do poll nova in tripleoclient iirc | 11:01 |
| dtantsur | shardy, https://github.com/openstack/python-tripleoclient/blob/master/tripleoclient/v1/overcloud_deploy.py#L621 | 11:01 |
| shardy | dtantsur: I looked for that, does it actually poll, or just check them and add an error if they're not what we expect? | 11:02 |
| shardy | https://github.com/openstack/python-tripleoclient/blob/master/tripleoclient/utils.py#L208 | 11:03 |
| dtantsur | iirc polls | 11:03 |
| dtantsur | ouch | 11:03 |
| * shardy doesn't see any polling | 11:03 | |
| derekh | shardy: the sleep 60 is after the introspection, | 11:03 |
| dtantsur | shardy, you're right. this should poll indeed. | 11:03 |
| * dtantsur facepalms | 11:03 | |
| dtantsur | derekh, lets kill it with fire for the all kittens sake? especially since nova poll interval is 2 minutes iirc | 11:04 |
| dtantsur | we need actually wait for that before deploying anything | 11:04 |
| derekh | shardy: checking hypervisor stats might help | 11:04 |
| shardy | derekh: in this case, the timestamp for starting introspection and setting the nodes available is the same, so I assume we're not actually doing it | 11:04 |
| shardy | which is confusing given the log output.. | 11:05 |
| shardy | http://logs.openstack.org/14/308514/10/check-tripleo/gate-tripleo-ci-f22-nonha/156d183/console.html#_2016-05-23_20_55_44_107 | 11:05 |
| shardy | http://logs.openstack.org/14/308514/10/check-tripleo/gate-tripleo-ci-f22-nonha/156d183/console.html#_2016-05-23_20_55_44_197 | 11:05 |
| shardy | Ah there's the sleep | 11:05 |
| shardy | Ok, maybe there is some other race then - I'll see what that --validation-error-fatal patch says | 11:07 |
| shardy | it definitely seems like we're seeing a lot more no valid host errors recently tho | 11:07 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Pass MysqlVirtualIP via EndpointMap https://review.openstack.org/305751 | 11:07 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Make service chain to depend on Networks resource https://review.openstack.org/320901 | 11:07 |
| derekh | shardy: your right though, the introspection seem too fast, actually maybe the timestamp is getting set when stdout is flushed, "tripleo.sh --introspect-nodes" is being called 2 minutes earlier | 11:08 |
| shardy | Yeah, some of those timestamps seem a little off | 11:09 |
| openstackgerrit | Merged openstack/tripleo-quickstart: Remove dependency on the khaleesi project https://review.openstack.org/320740 | 11:09 |
| *** stendulker has quit IRC | 11:10 | |
| *** Goneri has quit IRC | 11:13 | |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Prevent python from buffering stdout https://review.openstack.org/320905 | 11:16 |
| derekh | shardy: ^^ might fix the timestamps .... maybe | 11:16 |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci: Add md5 files to images upload https://review.openstack.org/320906 | 11:17 |
| derekh | dtantsur: so you think we should set it to 2 minutes? | 11:17 |
| *** openstackgerrit has quit IRC | 11:18 | |
| dtantsur | derekh, I think we should poll for hypervisor stats in 'openstack overcloud deploy' | 11:18 |
| *** openstackgerrit has joined #tripleo | 11:18 | |
| derekh | dtantsur: yup, would probably be better | 11:19 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-image-elements: Remove tuskar-ui element https://review.openstack.org/320907 | 11:19 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-image-elements: Remove tuskar element https://review.openstack.org/320908 | 11:22 |
| openstackgerrit | Jiri Stransky proposed openstack/tripleo-heat-templates: Write a list of services on the controllers https://review.openstack.org/320909 | 11:23 |
| jistr | jaosorior, shardy ^^ that could be a simpler alternative to the service manifest | 11:24 |
| openstackgerrit | Merged openstack/instack-undercloud: Ensure systemd-journald is running https://review.openstack.org/320643 | 11:28 |
| EmilienM | shardy: /me working on what? my backlog is too short | 11:29 |
| *** morazi has joined #tripleo | 11:29 | |
| openstackgerrit | James Slagle proposed openstack/instack-undercloud: Only run mistral commands if mistral is enabled https://review.openstack.org/320035 | 11:30 |
| *** thrash|g0ne is now known as thrash | 11:33 | |
| *** pkovar has joined #tripleo | 11:37 | |
| *** fragatina has joined #tripleo | 11:37 | |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Initial zaqar websockets https://review.openstack.org/319070 | 11:38 |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Initial zaqar websockets https://review.openstack.org/319070 | 11:38 |
| *** fultonj has joined #tripleo | 11:42 | |
| *** fragatina has quit IRC | 11:42 | |
| *** dsariel has quit IRC | 11:46 | |
| *** numans has quit IRC | 11:46 | |
| openstackgerrit | Jiri Stransky proposed openstack/tripleo-heat-templates: Write a list of services on the controllers https://review.openstack.org/320909 | 11:48 |
| sshnaidm | derekh, hi | 11:49 |
| derekh | sshnaidm: hi | 11:49 |
| sshnaidm | derekh, thanks for guidelines, going to do it now | 11:49 |
| derekh | sshnaidm: ok | 11:50 |
| sshnaidm | derekh, is it OK to do ha job though? I think it will be more interesting, isn't it? | 11:50 |
| derekh | sshnaidm: sure, if you think there will be enough time to run it | 11:50 |
| sshnaidm | derekh, not 100% sure, but it worth a try | 11:51 |
| *** MaxPC has joined #tripleo | 11:56 | |
| *** jaosorior has quit IRC | 11:56 | |
| *** jaosorior has joined #tripleo | 11:57 | |
| *** zoliXXL is now known as zoli|afk | 11:57 | |
| openstackgerrit | Merged openstack/diskimage-builder: Remove deprecated overriding of cloud-init defaults https://review.openstack.org/320021 | 11:58 |
| *** jaosorior has quit IRC | 11:58 | |
| *** athomas has quit IRC | 11:58 | |
| *** jaosorior has joined #tripleo | 11:59 | |
| *** noslzzp has joined #tripleo | 11:59 | |
| *** noslzzp has quit IRC | 12:01 | |
| *** fzdarsky is now known as fzdarsky|lunch | 12:01 | |
| *** noslzzp has joined #tripleo | 12:01 | |
| *** lucasagomes is now known as lucas-hungry | 12:03 | |
| *** rbrady has joined #tripleo | 12:03 | |
| *** athomas has joined #tripleo | 12:03 | |
| *** fragatina has joined #tripleo | 12:08 | |
| *** fragatina has quit IRC | 12:12 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Pass MysqlVirtualIP via EndpointMap https://review.openstack.org/305751 | 12:12 |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Make service chain to depend on Networks resource https://review.openstack.org/320901 | 12:12 |
| *** jayg|g0n3 is now known as jayg | 12:13 | |
| *** devvesa has joined #tripleo | 12:15 | |
| gfidente | jaosorior the hiera function | 12:18 |
| gfidente | we could probably parse puppetversion | 12:18 |
| gfidente | and behave differently basing on that? | 12:18 |
| *** fragatina has joined #tripleo | 12:18 | |
| gfidente | jaosorior I'll update it to do so ... | 12:19 |
| jaosorior | gfidente: We could try that | 12:19 |
| openstackgerrit | James Slagle proposed openstack/python-tripleoclient: Make node count a validation https://review.openstack.org/306060 | 12:19 |
| jaosorior | but I'm not sure how that's gonna work. As even calling the hiera function from lib/puppet/parser/functions will raise an error | 12:19 |
| jaosorior | in puppet 4.X | 12:19 |
| *** derekh has quit IRC | 12:19 | |
| *** dprince has joined #tripleo | 12:20 | |
| openstackgerrit | Juan Antonio Osorio Robles proposed openstack/tripleo-heat-templates: Pass MysqlVirtualIP via EndpointMap https://review.openstack.org/305751 | 12:23 |
| *** fragatina has quit IRC | 12:24 | |
| *** fragatina has joined #tripleo | 12:29 | |
| *** paramite has quit IRC | 12:29 | |
| openstackgerrit | Merged openstack/puppet-pacemaker: Merge with fuel-infra/puppet-pacemaker https://review.openstack.org/296440 | 12:30 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert SNMP configuration to composable services format https://review.openstack.org/320597 | 12:31 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert swift storage to composable services https://review.openstack.org/320580 | 12:31 |
| *** fragatina has quit IRC | 12:33 | |
| *** numans has joined #tripleo | 12:33 | |
| *** pkovar has quit IRC | 12:36 | |
| openstackgerrit | Emilien Macchi proposed openstack-infra/tripleo-ci: CI test - never merge https://review.openstack.org/320959 | 12:36 |
| *** jpena is now known as jpena|lunch | 12:42 | |
| jokke_ | Hi, Anyone here who would be willing to explain to me how upgrade with tripleO works atm.? | 12:46 |
| *** absubram has quit IRC | 12:47 | |
| *** pradk has quit IRC | 12:48 | |
| *** panda has quit IRC | 12:48 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Configure ComputeServices via resource chains https://review.openstack.org/319873 | 12:49 |
| *** panda has joined #tripleo | 12:49 | |
| *** absubram has joined #tripleo | 12:50 | |
| openstackgerrit | Martin André proposed openstack/tripleo-validations: Initial Cookiecutter Commit. https://review.openstack.org/320973 | 12:51 |
| ansiwen | shardy: hi, I have my tripleo running on my machine now. any advice how to get started into the composable roles topic efficiently? especially I would like to have an overview of the work that has to be done for nova. | 12:52 |
| ansiwen | I would need a introduction at what composable roles exactly are, and how to transfer to them | 12:52 |
| ansiwen | (anyone else here who want to support me is welcomed as well, of course ;-) ) | 12:54 |
| *** pkovar has joined #tripleo | 12:54 | |
| *** Goneri has joined #tripleo | 12:56 | |
| marios | ansiwen: hey man, ccamacho has some docs in progress you might find helpful https://review.openstack.org/#/c/311512/ | 12:56 |
| thrash | marios: you beat me to it. :P | 12:56 |
| marios | :) | 12:56 |
| *** zoli|afk is now known as zoliXXL | 12:57 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add lookup_hiera_hash function https://review.openstack.org/320112 | 12:57 |
| ccamacho | ansiwen, also requests for the docs are very welcome | 12:57 |
| *** tzumainn has quit IRC | 12:57 | |
| *** tzumainn has joined #tripleo | 12:58 | |
| ansiwen | marios: thanks! | 12:58 |
| *** rlandy has joined #tripleo | 12:58 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Configure ComputeServices via resource chains https://review.openstack.org/319873 | 13:00 |
| *** ayoung has joined #tripleo | 13:01 | |
| jaosorior | For adding neutron-server specific hieradata, should I add them in neutron-base.yaml? | 13:02 |
| shardy | ansiwen: Hey, I'd suggest starting with reviews from https://etherpad.openstack.org/p/tripleo-composable-services | 13:02 |
| shardy | and also potentially pulling some or experimenting with what's already landed to see how it fits together | 13:03 |
| shardy | re nova, you may want to sync with EmilienM as he's mentioned possibly tackling that | 13:03 |
| jaosorior | gfidente: hahaha dammit, I saw the call_function from puppet 4.X but didn't think it the way you did. That's some smarts right there dude. Did you get to test it locally? | 13:03 |
| *** lucas-hungry is now known as lucasagomes | 13:03 | |
| shardy | ccamacho already posted a patch that enables the StepConfig interface for the Compute role, which will be a prerequisite to decomposing that role | 13:03 |
| openstackgerrit | Martin André proposed openstack/tripleo-puppet-elements: Fix puppet modules in mixed src/pkg environments https://review.openstack.org/285860 | 13:04 |
| ansiwen | shardy: ok, thanks | 13:05 |
| ccamacho | shardy, can you give me the link? | 13:05 |
| *** rbrady has quit IRC | 13:05 | |
| *** mikelk has quit IRC | 13:05 | |
| shardy | ccamacho: I was referring to https://review.openstack.org/#/c/319873/ (your patch :) | 13:06 |
| *** mikelk has joined #tripleo | 13:06 | |
| shardy | obviously there are also nova services on the controller that don't depend on that landing | 13:06 |
| ccamacho | shardy, aaaaaaa sorry... need some coffee man. | 13:07 |
| shardy | hehe :) | 13:07 |
| *** trown|outtypewww is now known as trown | 13:08 | |
| *** numans has quit IRC | 13:09 | |
| *** numans has joined #tripleo | 13:09 | |
| ansiwen | shardy: with reviewing you mean actually giving improvement comments? or you mean looking into it how others did it? | 13:10 |
| shardy | ansiwen: both! :) | 13:10 |
| shardy | ansiwen: reviews are a great way to see the patterns and how it all is wired together | 13:11 |
| ansiwen | shardy: without knowing what composable roles are, I will have my difficulties to give good advice :-) | 13:11 |
| shardy | you could also consider pulling some of the patches locally (even if they're failing CI) and testing them | 13:11 |
| gfidente | jaosorior ack locally works with 4.x | 13:12 |
| gfidente | jaosorior but I didn't test it locally with 3.x | 13:12 |
| shardy | ansiwen: in summary, the composable-services-within-roles blueprint is about decomposing our large monolithic puppet manifests into a standardized per-service format | 13:12 |
| jaosorior | lets see | 13:12 |
| shardy | it's just about breaking things up so we don't say "deploy a controller" with a fixed mixture of services | 13:12 |
| ansiwen | shardy: regarding testing: how can I test it in a consitent way, do I have to run quickstart again? for me it's still a bit unclear, how to test an installer framework. :-) | 13:13 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: A helper script to launch quickstart from a usbkey https://review.openstack.org/319345 | 13:13 |
| shardy | intead, say "deploy services x, y and z on controller, but also deploy service a on CephStorage nodes" | 13:13 |
| shardy | ansiwen: no, when you have a running environment, you can just git clone tripleo-heat-templates, then cherry-pick the patch into the local tree (or check out the branch), see the download options on the top-right of the gerrit UI | 13:14 |
| ansiwen | shardy: yes, the basic idea, to break things up to get more granularity, is clear to me. just how it actually looks like, I have no clue. | 13:14 |
| shardy | ansiwen: well reviewing all the existing patches (and the WIP doc from ccamacho) is a good first step and should help solve the lack-of-clue'ness | 13:15 |
| shardy | :) | 13:15 |
| ansiwen | shardy: ok, I will try that :-) | 13:15 |
| bandini | is the nonha mitaka job working these days? Keeps failing on an ha-only patch here | 13:17 |
| bandini | specifically this one https://review.openstack.org/#/c/319430/ | 13:18 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Configure ComputeServices via resource chains https://review.openstack.org/319873 | 13:18 |
| ansiwen | shardy: so, I deployed the whole environment until the overcloud-validate.sh step. so if I clone the heat templates, I will repeat which steps for testing? or I will run heat directly on them? | 13:19 |
| ansiwen | sorry, for the noob questions... | 13:20 |
| shardy | ansiwen: you do openstack overcloud deploy --templates path/to/templates | 13:20 |
| shardy | ansiwen: personally I don't use quickstart, so I'm not familiar with the wrapper scripts you refer to | 13:20 |
| ansiwen | I see | 13:20 |
| shardy | larsks: Hey, is there a quickstart doc, or a tripleo-docs patch, that describes developer workflow via quickstart? | 13:21 |
| ansiwen | I wonder if it is wise to use quickstart.sh... | 13:21 |
| openstackgerrit | Martin André proposed openstack/tripleo-validations: Fixed pep8 issues https://review.openstack.org/320984 | 13:21 |
| jistr | jokke_: hi, WIP documentation is available at https://review.openstack.org/#/c/308985 -- however as we progress towards overcloud composability (selecting which services to deploy), the upgrades will change quite a bit too. It's still being designed and will change a lot, a tripleo spec review for the new upgrades is available here https://review.openstack.org/#/c/319264/ | 13:21 |
| trown | ansiwen: you would just `heat stack-delete overcloud` and then re-run the overcloud-deploy.sh script to redeploy | 13:21 |
| trown | ansiwen: it is very wise :) | 13:21 |
| shardy | trown: where does that pull the t-h-t tree from th? | 13:21 |
| jokke_ | thanks a million jistr I will have a look on those two | 13:21 |
| ansiwen | trown: hehe | 13:21 |
| ansiwen | shardy: that would also be my next question :-) | 13:22 |
| shardy | ansiwen: a lot of folks are very happy with quickstart, I just personally am sticking with the "old" way for now, due to familiarity and because it gets me closer to what we CI | 13:22 |
| ansiwen | trown: if I wanna test heat templates, I have to point the script to it, right? | 13:22 |
| trown | oh right ya | 13:23 |
| *** jaosorior has quit IRC | 13:23 | |
| shardy | trown: FWIW I dislike that we're wrapping the CLI in scripts for quickstart, as it makes it less easy to follow our docs | 13:23 |
| trown | sorry did not scroll up enough for context | 13:23 |
| ansiwen | shardy: I wondered if for understanding the internals the "old" way would be more educational | 13:23 |
| leifmadsen | The script is nice, but I personally would prefer to be pointed at just the direct Ansible commands to run in the docs | 13:23 |
| shardy | ansiwen: well, right now the old way is more closely aligned with our docs, but slightly more effort to set up | 13:23 |
| leifmadsen | since it makes it more flexible to modify | 13:23 |
| trown | shardy: well the alternative for ansible is to put them only in ansible tasks... which is even harder to see what is happening for a non-ansible person | 13:24 |
| shardy | trown: can't we just document how to log onto the undercloud, source stackrc and run them? | 13:25 |
| shardy | e.g just like the old flow? | 13:25 |
| trown | shardy: that is documented now, and should also work | 13:25 |
| shardy | ansiwen: FWIW this is how I set up my dev environment: http://paste.fedoraproject.org/370693/14641826/ | 13:26 |
| trown | shardy: the scripts are so that someone *could* run exactly what ran in CI | 13:26 |
| shardy | trown: ack, maybe we just need ansiwen directed to those docs then | 13:26 |
| *** lblanchard has joined #tripleo | 13:26 | |
| trown | ya, quickstart.sh instructions clearly point to tripleo-docs as one of the ways to continue | 13:26 |
| shardy | trown: what may be useful, if they don't already exist, is some docs on developer workflow, and in particular if e.g you want to pull an under review t-h-t change, combined with a puppet-tripleo change, then apply both to a new overcloud deployment | 13:27 |
| trown | it is either "run these handy scripts that are exactly what CI did" or "follow these doc (which may be different then what CI did) | 13:27 |
| *** fzdarsky|lunch is now known as fzdarsky | 13:27 | |
| *** akshai has joined #tripleo | 13:27 | |
| sshnaidm | larsks, hi, can you take a look please? https://review.openstack.org/#/c/309995/ | 13:27 |
| shardy | trown: ack, so basically folks can use their existing workflows, cool | 13:28 |
| *** akshai_ has joined #tripleo | 13:28 | |
| shardy | ansiwen: if quickstart is working for you, probably you can ignore my dev setup stuff and just log on to it, then use the tripleoclient CLI directly | 13:29 |
| trown | shardy: wrt developer workflow, that is still pretty WIP, but getting close: https://review.gerrithub.io/#/c/277418/ | 13:29 |
| *** pkovar has quit IRC | 13:29 | |
| *** fragatina has joined #tripleo | 13:29 | |
| trown | once we have that ability (to pass in arbitrary list of packaging/code changes), documenting a developer workflow would be next | 13:29 |
| *** dmacpher has joined #tripleo | 13:30 | |
| shardy | trown: ack, that looks good | 13:30 |
| shardy | for now we probably just want to land https://review.openstack.org/#/c/299643/ and document how to use upload-puppet-modules (which is what I've been using locally) | 13:31 |
| *** akshai has quit IRC | 13:32 | |
| ansiwen | shardy, trown: ok, for now I'm overwhelmed with info, thanks a lot. :-) | 13:35 |
| *** fragatina has quit IRC | 13:35 | |
| ansiwen | shardy: well, quickstart worked for me in a way that I have an under- and overcloud running. just was wondering how to inject changes and test them... | 13:36 |
| *** myoung has quit IRC | 13:38 | |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-docs: Composable roles within services Tutorial https://review.openstack.org/311512 | 13:39 |
| trown | ansiwen: ya, the primary use case for quickstart is still RDO (ie. near-guaranteed success for first-time users of tripleo), but my main focus is on making it more useful for tripleo development | 13:39 |
| trown | ansiwen: that said, I use it 100% of the time to test out tripleo patches... just currently requires some hacking | 13:40 |
| openstackgerrit | Imre Farkas proposed openstack/tripleo-specs: Metal to tenant https://review.openstack.org/320995 | 13:40 |
| ifarkas | shardy, the spec for ironic ingegration ^ | 13:40 |
| ifarkas | also lucasagomes, dtantsur, mgould, milan ^ | 13:41 |
| milan | ifarkas, thx | 13:41 |
| *** derekh has joined #tripleo | 13:41 | |
| *** pkovar has joined #tripleo | 13:41 | |
| slagle | we're getting lots of sporadic CI failures with the No valid Host error, https://bugs.launchpad.net/tripleo/+bug/1585641 | 13:41 |
| openstack | Launchpad bug 1585641 in tripleo "CI: jobs failing with Message: No valid host was found. There are not enough hosts available., Code: 500"" [High,Triaged] | 13:41 |
| *** milan has quit IRC | 13:42 | |
| *** itamarl has quit IRC | 13:42 | |
| EmilienM | bnemec, dtantsur: I found out why ironic/api in wsgi does not work in undercloud | 13:42 |
| EmilienM | it's a bug I found several months ago and was mostly ignored | 13:42 |
| EmilienM | https://bugs.launchpad.net/ironic/+bug/1564075 | 13:43 |
| openstack | Launchpad bug 1564075 in Ironic "ironic-conductor does not retry to connect to rpc if connection failure" [Undecided,Incomplete] - Assigned to Haomeng,Wang (whaom) | 13:43 |
| EmilienM | I think it's related | 13:43 |
| *** liverpooler has quit IRC | 13:43 | |
| lucasagomes | ifarkas, awesome! will take a look | 13:43 |
| derekh | slagle: long shot, but any chance the failed jobs may all be using the same testenv's (i.e. dodgy testenv or something, it wouldn't be the first time) | 13:43 |
| ansiwen | trown: ok, and that hacking is documented in which of the links you posted? | 13:44 |
| openstackgerrit | Imre Farkas proposed openstack/tripleo-specs: Metal to tenant https://review.openstack.org/320995 | 13:45 |
| slagle | derekh: on this particular job i'm looking at, this might be the root cause: | 13:46 |
| slagle | 2376 Stderr: u'Calling LC_ALL=C /usr/bin/virsh --connect qemu:///system start baremetal1brbm_one1_0 \nerror: Failed to start domain baremetal1brbm_one1_0\nerror: Unable to add port vnet26 to O VS bridge 1brbm_four1: Operation not permitted\n'. | 13:46 |
| slagle | havent we seen that before? | 13:46 |
| derekh | shardy: timestamps seem more realistic on this patch https://review.openstack.org/#/c/320905/1 | 13:46 |
| derekh | slagle: yes, thats the reason we added the cron job to the testenvs hosts to reset ovs nightly | 13:47 |
| trown | ansiwen: that is the part that is not well documented :P, http://tripleo.org/advanced_deployment/in_progress_review.html should at least have some details about using tripleo.sh to build dlrn packages, but it does not | 13:47 |
| *** links has quit IRC | 13:48 | |
| derekh | slagle: this https://review.openstack.org/#/c/292480/ | 13:48 |
| trown | ansiwen: I am working on adding functionality directly to tripleo-quickstart to do that, and will document that once it exists | 13:48 |
| ansiwen | trown: ok, so for me the advice to proceed is to ask you all about it, or go with the non-quickstart approach? | 13:48 |
| ansiwen | trown: I mean for now | 13:48 |
| dtantsur | EmilienM, oh hmm.. yeah, pretty possible | 13:50 |
| *** egafford has joined #tripleo | 13:50 | |
| trown | ansiwen: I could put up an etherpad that shows how to pull in the in-progress work for quickstart and how to use it | 13:50 |
| *** oshvartz has quit IRC | 13:50 | |
| openstackgerrit | Emilien Macchi proposed openstack/instack-undercloud: Deploy and Upgrade Ironic to run in mod_wsgi https://review.openstack.org/309140 | 13:51 |
| trown | ansiwen: I dont use tripleo.sh, so I am not keen to update the current docs which should describe using it | 13:51 |
| derekh | slagle: in the past when this was happening I used to identify the problem testenv hosts a rebuild it, but I thought the ovs reset had sorted out the problem, it certainly didn't seem to be happening as much | 13:51 |
| ansiwen | trown: well, I don't want to bother you, I have no problem with going the other way | 13:52 |
| slagle | derekh: yea i think that might be a red herring | 13:52 |
| slagle | derekh: i dont think it's responsible for all these falures | 13:52 |
| trown | ansiwen: it is more if you are willing to be a guinea pig for me, and less about bothering me :) | 13:52 |
| *** jpena|lunch is now known as jpena | 13:52 | |
| ansiwen | trown: well, my goal is very clear: I want to put myself as fast as possible into the position to work on the composable roles for nova. I will do (almost) whatever you guys tell me to do for that. :-) | 13:54 |
| ansiwen | trown: assume close to zero pre-knowledge | 13:55 |
| ccamacho | ansiwen, this might be helpful (https://review.openstack.org/#/c/311512/19/doc/source/tht_walkthrough/tips.rst) | 13:55 |
| trown | ansiwen: ok, I am filling out https://etherpad.openstack.org/p/tripleo-quickstart-dev-workflow now | 13:55 |
| derekh | slagle: ok, it may be causing a few of them though | 13:55 |
| derekh | slagle: also the current-tripleo pin moved last night which may be relevant | 13:56 |
| ansiwen | trown: my current status is: quickstart.sh used until overcloud-validate.sh | 13:58 |
| ansiwen | trown: so under and overcloud is running | 13:58 |
| openstackgerrit | Leif Madsen proposed openstack/tripleo-quickstart: A helper script to launch quickstart from a usbkey https://review.openstack.org/319345 | 13:59 |
| *** noslzzp has quit IRC | 13:59 | |
| trown | ansiwen: ya, you can just delete the overcloud in that case, and hack directly on tripleo-heat-templates | 13:59 |
| sshnaidm | derekh, can you take a look please? I'm not sure about the job names.. https://review.openstack.org/#/c/320957/ | 14:00 |
| *** paramite has joined #tripleo | 14:00 | |
| trown | ansiwen: but I am documenting more of how to start from scratch and run tripleo-quickstart with in progress reviews | 14:00 |
| ansiwen | trown: with 'openstack overcloud delete' | 14:01 |
| trown | ansiwen: indeed | 14:01 |
| *** fragatina has joined #tripleo | 14:01 | |
| ansiwen | trown: thanks | 14:01 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: Inject gating repo in overcloud-full image as well https://review.openstack.org/319901 | 14:02 |
| *** bvandenh has joined #tripleo | 14:03 | |
| derekh | sshnaidm: done, besides a minor tweak it lgtm | 14:04 |
| gfidente | EmilienM so the unittests failing for 4.0 | 14:04 |
| EmilienM | gfidente: you want me to look? | 14:04 |
| gfidente | but not with <4 and not with latest | 14:04 |
| EmilienM | oh | 14:04 |
| EmilienM | I already hit that | 14:04 |
| EmilienM | they fixed a bug in 4.5 | 14:05 |
| slagle | derekh: some of the clocks are way off on the testenv hosts | 14:05 |
| gfidente | I think the problem with call_function is fixed in 4.10 | 14:05 |
| gfidente | sorry, 4.1.0 | 14:05 |
| sshnaidm | derekh, great, will rename it now.. although I think we can do much more than tempest there, but let's start from it :) | 14:05 |
| EmilienM | gfidente: let me 2 min to dig | 14:05 |
| slagle | derekh: just fyi, dont know if that's related to anything | 14:05 |
| gfidente | but I don't know how to dig into 4.0 specific issue | 14:05 |
| *** paramite is now known as paramite|afk | 14:05 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-tripleoclient: Updated from global requirements https://review.openstack.org/300044 | 14:05 |
| gfidente | or if we can put a workaround for it | 14:05 |
| derekh | slagle: I'll fix them now, don't think it causes a problem these days, it used to when we ran more from the jenkins slaves | 14:06 |
| derekh | slagle: done | 14:07 |
| EmilienM | gfidente: I think they fixed it in 4.5 with https://github.com/puppetlabs/puppet/commit/14174311d3533ad186a33f99ab011d12a589de44#diff-4f09a82a8885a4bfdf97c548c86e5c00 | 14:07 |
| *** fragatina has quit IRC | 14:08 | |
| *** egafford has quit IRC | 14:08 | |
| EmilienM | gfidente: you're hitting a critical bug I think, let me continue to dig | 14:09 |
| sshnaidm | derekh, I hoe it will not be a confusion when we have jobs "ha-mitaka", "ha-liberty" and "ha-tempest"? So it will not be considered as a branch :) | 14:09 |
| sshnaidm | s/hoe/hope | 14:09 |
| EmilienM | gfidente: an example of some workaround https://github.com/example42/puppi/blob/master/lib/puppet/functions/params_lookup.rb | 14:10 |
| *** pradk has joined #tripleo | 14:12 | |
| EmilienM | gfidente: see https://www.mail-archive.com/puppet-users@googlegroups.com/msg58173.html | 14:12 |
| *** radez has joined #tripleo | 14:12 | |
| radez | where does tftp traffic get logged to? I don't see any log file or log entries in var log messages | 14:13 |
| radez | do I have to look in journal ctl? | 14:13 |
| derekh | sshnaidm: it should be ok, so your next step will be to add support for tempest in this loop if "tempest" is in TOCI_JOBTYPE , http://git.openstack.org/cgit/openstack-infra/tripleo-ci/tree/toci_gate_test.sh#n81 | 14:13 |
| sshnaidm | derekh, sure, do you want the pingtest stay there as well? | 14:14 |
| *** rhallisey has joined #tripleo | 14:15 | |
| derekh | sshnaidm: whatever you think is fine by me, we'll still be doing it in the other tests anyways | 14:15 |
| sshnaidm | derekh, ok | 14:16 |
| *** yamahata has joined #tripleo | 14:18 | |
| openstackgerrit | Gabriele Cerami proposed openstack-infra/tripleo-ci: Add md5 files to images upload https://review.openstack.org/320906 | 14:18 |
| larsks | shardy: did you get your answer to your question re: developer workflow? | 14:21 |
| shardy | larsks: thanks, yep I think trown is working with ansiwen to generate some notes/docs | 14:22 |
| larsks | Awesome. | 14:22 |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: Add tempest tests to periodic and experimental job https://review.openstack.org/321019 | 14:22 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert SNMP configuration to composable services format https://review.openstack.org/320597 | 14:24 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert swift storage to composable services https://review.openstack.org/320580 | 14:24 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Configure ObjectStorage services via resource chains https://review.openstack.org/320361 | 14:24 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Convert Swift proxy to composable services format https://review.openstack.org/319959 | 14:24 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-heat-templates: Fix inconsistency with ringbuilder/storage steps https://review.openstack.org/320351 | 14:24 |
| *** myoung has joined #tripleo | 14:24 | |
| pkovar | any idea why this jenkins check is failing? http://logs.openstack.org/71/265271/9/check/gate-tripleo-docs-docs/145fb4b/console.txt | 14:26 |
| pkovar | this is for https://review.openstack.org/#/c/265271/ | 14:26 |
| snecklifter | Hello. I'm getting "No package python-hardware-detect available." when trying to build images, any clues? | 14:31 |
| snecklifter | I have it installed on the undercloud | 14:31 |
| *** masco has quit IRC | 14:32 | |
| *** noslzzp has joined #tripleo | 14:33 | |
| *** Goneri has quit IRC | 14:35 | |
| openstackgerrit | Jiri Tomasek proposed openstack/tripleo-ui: Update FormErrorList component https://review.openstack.org/320498 | 14:35 |
| openstackgerrit | Steven Hardy proposed openstack/puppet-tripleo: Add puppet profile for snmp https://review.openstack.org/320596 | 14:35 |
| *** Goneri has joined #tripleo | 14:36 | |
| openstackgerrit | Marios Andreou proposed openstack/tripleo-heat-templates: Initial rework of pacemaker_common_functions for M..N upgrades https://review.openstack.org/321027 | 14:40 |
| *** hoobaman has joined #tripleo | 14:41 | |
| hoobaman | hi! :) | 14:41 |
| hoobaman | i am a bit confused about the new default external network model since kilo (horizon dashboard and pubabi in floating ip space) | 14:41 |
| hoobaman | i do not see anymore ovs bridge mappins in neutron plugin.ini on controller | 14:42 |
| hoobaman | the floating ip space is available on a virtual tagged vlan interface | 14:43 |
| hoobaman | how does the b-ex mapping work now? | 14:43 |
| hoobaman | br-ex | 14:43 |
| *** oshvartz has joined #tripleo | 14:44 | |
| *** mgarciam has joined #tripleo | 14:45 | |
| *** yamahata has quit IRC | 14:47 | |
| hoobaman | or does this new model require the --provider:network_type vlan provider? | 14:47 |
| trown | ansiwen: ok, I just copy pasted https://etherpad.openstack.org/p/tripleo-quickstart-dev-workflow and it successfully built and deployed the nova and puppet-nova changes I used as an example | 14:48 |
| trown | mgould: you may be interested in trying out https://etherpad.openstack.org/p/tripleo-quickstart-dev-workflow as well | 14:48 |
| sshnaidm | trown, maybe you know, I'm trying to build images on undercloud machine, but it fails, where should I look into? https://paste.fedoraproject.org/370721/14641876/ | 14:49 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-docs: Update baremetal import to not use --json option https://review.openstack.org/291147 | 14:49 |
| openstackgerrit | Steven Hardy proposed openstack/tripleo-docs: Add link from node import to where file format is documented https://review.openstack.org/321035 | 14:49 |
| *** mcornea has quit IRC | 14:50 | |
| trown | sshnaidm: something failed in 10-enable-os-collect-config so the first step would be to track down what element that comes from | 14:51 |
| openstackgerrit | Florian Fuchs proposed openstack/tripleo-ui: Fixes plan file update https://review.openstack.org/321037 | 14:52 |
| mgould | trown: thanks, will take a look | 14:52 |
| trown | sshnaidm: why are you building images though? maybe you don't need to | 14:52 |
| openstackgerrit | Florian Fuchs proposed openstack/tripleo-ui: Fixes plan file update https://review.openstack.org/321037 | 14:52 |
| sshnaidm | trown, I changed the delorean.repo to some particular one and want to build new images with it | 14:52 |
| sshnaidm | trown, including reinstall of all delorean packages.. | 14:53 |
| sshnaidm | trown, until there is another way to get undercloud and overcloud images from particular delorean repo version.. | 14:54 |
| *** tserong has quit IRC | 14:56 | |
| trown | sshnaidm: hmm, I tend to use virt-customize for that, rather than build images from scratch | 14:56 |
| trown | image building is such a PITA... I just let CI do that :P | 14:56 |
| hoobaman | uname -a | 14:59 |
| *** cwolferh has joined #tripleo | 15:00 | |
| *** milan has joined #tripleo | 15:01 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/os-cloud-config: Updated from global requirements https://review.openstack.org/313748 | 15:01 |
| sshnaidm | trown, could you point me to relevant lines? you mean in quckstart, right? | 15:01 |
| *** dc has quit IRC | 15:02 | |
| *** oshvartz has quit IRC | 15:02 | |
| *** egafford has joined #tripleo | 15:02 | |
| *** dc has joined #tripleo | 15:03 | |
| *** paramite|afk is now known as paramite | 15:03 | |
| *** ebarrera has quit IRC | 15:03 | |
| *** fragatina has joined #tripleo | 15:03 | |
| trown | sshnaidm: basically just `sudo LIBGUESTFS_BACKEND=direct virt-customize -a overcloud-full.qcow2 --run 'script-with-commands-get-repo-and-update'` | 15:04 |
| trown | sshnaidm: just need to create a script that pulls down the repo you want and runs yum update | 15:04 |
| dtantsur | trown, I think there is a flag to say LIBGUESTFS_BACKEND=direct (-d?) | 15:04 |
| trown | dtantsur: really, that would be great | 15:04 |
| sshnaidm | trown, and what about all these ironic-python-agent* ? | 15:06 |
| pino|work | dtantsur: no | 15:06 |
| tosky | trown: why sudo? | 15:06 |
| EmilienM | dtantsur: I'm a bit lost about the wsgi/rabbit error | 15:07 |
| EmilienM | dtantsur: even if rabbit starts before, I still have the error: http://logs.openstack.org/40/309140/8/check-tripleo/gate-tripleo-ci-f22-nonha/b312ed3/logs/undercloud/var/log/ironic/app.txt.gz#_2016-05-25_14_43_49_759 | 15:07 |
| EmilienM | you can see the ERROR report in http://logs.openstack.org/40/309140/8/check-tripleo/gate-tripleo-ci-f22-nonha/b312ed3/logs/undercloud/var/log/rabbitmq/rabbit@instack.txt.gz at 14:43:26 | 15:07 |
| trown | tosky: from the undercloud setup via quickstart that is the only way it works for me, I have not looked into how to fix that | 15:07 |
| tosky | trown: unless the image is owned by root, you don't need sudo | 15:07 |
| pino|work | trown: just run libguestfs with the same user that owns the image | 15:08 |
| dtantsur | EmilienM, ehmmm.. weird. did you try to reproduce it locally? | 15:08 |
| trown | tosky: pino|work, the stack user did not have access to qemu://system I think though | 15:08 |
| openstackgerrit | Gabriele Cerami proposed openstack/tripleo-quickstart: Add options to download and handle overcloud images https://review.openstack.org/321057 | 15:08 |
| dtantsur | it looks like some network issue between these 2 | 15:08 |
| *** fragatina has quit IRC | 15:08 | |
| trown | s/did not/does not/ | 15:08 |
| pino|work | trown: does not matter for local images | 15:08 |
| *** eggmaster has joined #tripleo | 15:08 | |
| pino|work | trown: it would if you were trying to edit a libvirt guest (-d/--domain) | 15:09 |
| trown | pino|work: hmm, ok I will try again shortly, but I am pretty sure it failed for me without sudo | 15:09 |
| *** tserong has joined #tripleo | 15:09 | |
| *** tserong has joined #tripleo | 15:09 | |
| pino|work | trown: most probably the issue you hit is http://libguestfs.org/guestfs-faq.1.html#permission-denied-when-running-libguestfs-as-root , but cannot say more without other info | 15:09 |
| pino|work | if qemu can read/write the image, then you don't need sudo | 15:10 |
| *** mburned has joined #tripleo | 15:10 | |
| *** yolanda_ has joined #tripleo | 15:10 | |
| panda | trown: ^^ I thought a review was better than a etherpad, for the overcloud. Still WIP though | 15:10 |
| trown | sshnaidm: you are trying to build a new IPA image? That one won't work via virt-customize, but it should rarely *require* update unless you are testing something specific to ironic/ironic-inspector | 15:11 |
| dtantsur | correct | 15:11 |
| trown | panda: awesome will check it out | 15:11 |
| *** dsariel has joined #tripleo | 15:11 | |
| dtantsur | I can tell you how to pack/unpack it ofc | 15:11 |
| * dtantsur probably needs to document it | 15:12 | |
| *** aufi has quit IRC | 15:12 | |
| trown | we do have an example now buried in tripleo-ci, as part of the image caching | 15:13 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Testing stable/liberty without making changes -DO NOT MERGE- https://review.openstack.org/321060 | 15:13 |
| *** mburned has left #tripleo | 15:13 | |
| *** mburned has joined #tripleo | 15:13 | |
| trown | https://github.com/openstack-infra/tripleo-ci/blob/master/scripts/common_functions.sh#L73-L162 | 15:14 |
| sshnaidm | trown, I wanted to build everything from scratch: "openstack overcloud image build --all" to be sure I get exactly the required version. Seems like I have to settle for overcloud-full image only | 15:18 |
| *** chem` has joined #tripleo | 15:18 | |
| *** tserong has quit IRC | 15:19 | |
| bnemec | sshnaidm: I wouldn't use that command directly on master tripleo. There's some funny business going on in tripleo.sh to make everything work right, so tripleo.sh --overcloud-images is the safest way to build images. | 15:19 |
| *** tserong has joined #tripleo | 15:20 | |
| *** yolanda has quit IRC | 15:20 | |
| *** chem has quit IRC | 15:20 | |
| sshnaidm | bnemec, oh, I'm looking for a way to build it in quickstart undercloud machine, tripleo.sh is not there unfortunately :) | 15:21 |
| sshnaidm | this example is not trivial, though.. | 15:21 |
| trown | sshnaidm: you can always git clone tripleo-ci to get tripleo.sh (it is done for the ping test already) | 15:22 |
| trown | I don't love that we can't even build images without using tripleo.sh, but I prefer letting CI do that anyways | 15:22 |
| sshnaidm | trown, is it cloned there already? | 15:22 |
| trown | sshnaidm: only if you ran a full deploy, since it is cloned in the validate step | 15:23 |
| sshnaidm | trown, I see | 15:23 |
| trown | sshnaidm: but the validate script has the clone command right in it :) | 15:23 |
| *** numans has quit IRC | 15:24 | |
| larsks | trown: is artifacts.ci.centos.org still the site from which I should be fetching images? Because I have basically lost the ability to download images from there in anything resembling a reasonable amount of time. | 15:24 |
| derekh | why can't we build images without tripleo.sh ? | 15:25 |
| trown | larsks: ya, http://artifacts.ci.centos.org/artifacts/rdo/images/master/delorean/stable/ is for master images | 15:26 |
| larsks | I was afraid of that. | 15:26 |
| trown | larsks: at least we fixed caching :P | 15:26 |
| hoobaman | qg interface/port is being added on br-ex bridge, NAT iptables rules are in place, however no ip on the interface? | 15:27 |
| larsks | Yeah, but I kind of want a new image :). Plus if other folks are seeing this it will suck for first-time deployers. | 15:27 |
| larsks | I am currently trying with wget and the ETA is measured in *days* right now. | 15:27 |
| trown | ugh | 15:28 |
| sshnaidm | trown, I think myoung created a role for image building and it could be a part of quickstart, just yesterday was a talk | 15:28 |
| trown | sshnaidm: that role was split out from tripleo-quickstart and is what we use to build the image in RDO CI | 15:29 |
| sshnaidm | oh, then I misunderstood this | 15:29 |
| trown | it builds overcloud IPA and undercloud directly on the virthost though | 15:30 |
| larsks | trown: running since 11:27: http://chunk.io/f/48d9542eebf940ae982d060983eb51d1 | 15:32 |
| trown | larsks: hmm is http://buildlogs.centos.org/centos/7/cloud/x86_64/tripleo_images/mitaka/delorean/ any better? | 15:32 |
| trown | mitaka images are mirrored, master images are not (yet) | 15:33 |
| larsks | I will check. I am wondering also if proxying through the corporate vpn would actually be faster. | 15:33 |
| trown | there is also the problem that the way CDN works, the md5 ends up propogating faster, so we can't switch quickstart to use the mirror location by default | 15:34 |
| trown | since during the time a new md5 shows up, but before the new image shows up, quickstart will just fail on the md5 not matching | 15:34 |
| trown | I need to follow up with centos folks to see if they have any ideas for that | 15:35 |
| *** dsariel has quit IRC | 15:37 | |
| shardy | EmilienM: Hi - looking for some puppet advice when you get a moment | 15:38 |
| shardy | EmilienM: https://review.openstack.org/#/c/321075 | 15:38 |
| EmilienM | shardy: ok looking in a few min | 15:38 |
| shardy | I want to add support for a new option in heat and I'm not sure how to implement the test so it won't break when heat changes the service default | 15:38 |
| larsks | trown: fetching from the cnd is way faster. Feching via the vpn also works, but slower than the cdn. | 15:38 |
| shardy | EmilienM: thanks, not urgent | 15:39 |
| larsks | s/cnd/cdn/ | 15:39 |
| *** rbrady has joined #tripleo | 15:40 | |
| *** mikelk has quit IRC | 15:41 | |
| *** Guest50431 has quit IRC | 15:41 | |
| larsks | trown: It looks like I lied about the cdn. I see similar behavior there that I see with the artifacts site: initial download speed seems reasonable, but then it simply cuts out and stops responding. | 15:42 |
| trown | well that is less then ideal | 15:43 |
| trown | than... | 15:43 |
| * trown needs a nap | 15:43 | |
| * pino|work throws trown a pillow | 15:44 | |
| *** masco has joined #tripleo | 15:44 | |
| sshnaidm | derekh, should I add all project-config core team for reviewing this patch or somebody particular? https://review.openstack.org/#/c/320957/ | 15:45 |
| sshnaidm | trown, panda sorry if I missed, but is there something new about puppet sources vs packages...? | 15:46 |
| panda | sshnaidm: since there is automation in place and packages follow sources more closely, the plan is to use only packages now | 15:48 |
| trown | sshnaidm: ya in master there are per module packages, that follow source just like everything else in DLRN | 15:48 |
| sshnaidm | cool | 15:49 |
| *** sambetts is now known as sambetts|afk | 15:49 | |
| *** dmk0202 has quit IRC | 15:49 | |
| shardy | sshnaidm: I'd avoid adding the whole project-config-core yet, it's not even been posted a day | 15:50 |
| shardy | reviews on that repo aren't that fast, but they do generally happen after a few days | 15:50 |
| *** eggmaster has quit IRC | 15:50 | |
| sshnaidm | shardy, ok, let's wait then | 15:51 |
| *** ebarrera has joined #tripleo | 15:54 | |
| EmilienM | shardy: https://github.com/openstack/puppet-heat/blob/master/spec/classes/heat_engine_spec.rb#L73 | 15:55 |
| EmilienM | the default value is tested this way ^ | 15:55 |
| openstackgerrit | Steven Hardy proposed openstack/instack-undercloud: Disable Heat convergence pending further testing https://review.openstack.org/321090 | 15:57 |
| shardy | EmilienM: Ok, thanks, so do I need to add convergence_engine mapped to <SERVICE DEFAULT> in the let: default_params block? | 15:58 |
| shardy | https://github.com/openstack/puppet-heat/blob/master/spec/classes/heat_engine_spec.rb#L5 | 15:58 |
| EmilienM | shardy: can I push over so I show you the diff? | 15:58 |
| shardy | EmilienM: Heat plans to change this soon, so I wanted to ensure we don't break when that happens | 15:58 |
| shardy | EmilienM: sure! | 15:58 |
| shardy | I also pushed a corresponding instack-undercloud patch to test it ^^ | 15:59 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Configure CephStorage services via resource chains https://review.openstack.org/321092 | 16:00 |
| *** liverpooler has joined #tripleo | 16:01 | |
| *** fragatina has joined #tripleo | 16:02 | |
| *** rcernin has quit IRC | 16:04 | |
| EmilienM | shardy: https://review.openstack.org/#/c/321075/1..2/spec/classes/heat_engine_spec.rb | 16:05 |
| EmilienM | shardy: so we test default value & overriding | 16:05 |
| EmilienM | shardy: no worries for the param, using os_service_default makes it unset if no value is set for the param | 16:05 |
| EmilienM | which means, by default the param will be absent in heat.conf | 16:06 |
| EmilienM | shardy: I already ran 'recheck' on your uc patch | 16:06 |
| *** fragatina has quit IRC | 16:06 | |
| shardy | EmilienM: great, thanks! | 16:07 |
| EmilienM | shardy: I dropped a comment even | 16:08 |
| *** athomas has quit IRC | 16:08 | |
| *** athomas has joined #tripleo | 16:09 | |
| openstackgerrit | Merged openstack/tripleo-heat-templates: Tighten the access rules for galera https://review.openstack.org/319430 | 16:09 |
| shardy | EmilienM: re your comment, Heat is about to switch the default to true | 16:09 |
| shardy | which is why I want to set the TripleO default to false now | 16:10 |
| shardy | I know it will break us otherwise from local testing | 16:10 |
| EmilienM | ahhhh | 16:10 |
| *** athomas has quit IRC | 16:10 | |
| EmilienM | +2 then | 16:10 |
| shardy | thanks! | 16:10 |
| *** links has joined #tripleo | 16:10 | |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: Add tempest tests to periodic and experimental job https://review.openstack.org/321019 | 16:11 |
| *** ifarkas has quit IRC | 16:12 | |
| *** devvesa has quit IRC | 16:12 | |
| gfidente | ah the steps thing is interesting | 16:13 |
| *** olap has quit IRC | 16:13 | |
| *** myoung has quit IRC | 16:13 | |
| *** tesseract has quit IRC | 16:13 | |
| gfidente | though we'll have to wait for cephstorage to go through 4 steps | 16:14 |
| gfidente | when there is no need to | 16:14 |
| gfidente | but on the other hand, we need the steps for the service to be pluggable on say, the controller or the compute nodes | 16:14 |
| gfidente | where there is other stuff happening on steps < 4 | 16:15 |
| openstackgerrit | Carlos Camacho proposed openstack/tripleo-heat-templates: Configure ComputeServices via resource chains https://review.openstack.org/319873 | 16:16 |
| *** mbound has quit IRC | 16:18 | |
| ccamacho | should we agreed about the config steps? | 16:18 |
| *** akrivoka has quit IRC | 16:19 | |
| shardy | gfidente: Yeah, for ObjectStorage I only wired in the steps it actually uses, but I agree we want to move to a model where all roles support the same steps | 16:19 |
| shardy | it would be interesting to see if we could have one global -post.yaml that applies to all roles | 16:20 |
| *** ccamacho is now known as ccamacho|hungry | 16:20 | |
| shardy | that would enable the depends_on optimization you're looking for, but potentially could have other performance costs | 16:20 |
| gfidente | ack | 16:20 |
| *** dtantsur is now known as dtantsur|afk | 16:21 | |
| *** devvesa has joined #tripleo | 16:22 | |
| *** derekh has quit IRC | 16:22 | |
| *** devvesa has quit IRC | 16:23 | |
| *** MaxPC has quit IRC | 16:24 | |
| *** athomas has joined #tripleo | 16:24 | |
| *** penick has joined #tripleo | 16:25 | |
| *** paramite is now known as paramite|afk | 16:25 | |
| openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart: WIP: Tempest support for quickstart https://review.openstack.org/318751 | 16:25 |
| openstackgerrit | Sagi Shnaidman proposed openstack/tripleo-quickstart: Tempest support for quickstart https://review.openstack.org/318751 | 16:27 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: move most things out from the playbooks dir, and use config for release info https://review.openstack.org/313116 | 16:28 |
| *** sshnaidm has quit IRC | 16:32 | |
| *** masco has quit IRC | 16:33 | |
| *** devvesa has joined #tripleo | 16:34 | |
| *** devvesa has quit IRC | 16:34 | |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Configure CephStorage services via resource chains https://review.openstack.org/321092 | 16:35 |
| *** ebarrera has quit IRC | 16:35 | |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Nothing to see here https://review.openstack.org/321113 | 16:36 |
| *** florianf has quit IRC | 16:37 | |
| adarazs | trown: hi! I think I have a the jenkins/gerrit dependency stuff done as well... now I'm testing the gating role and I fail to build puppet-tripleo with dlrn. have you seen problems like this? | 16:38 |
| *** tremble has quit IRC | 16:38 | |
| trown | adarazs: hmm, nope in fact I tested just this morning with https://etherpad.openstack.org/p/tripleo-quickstart-dev-workflow | 16:39 |
| trown | works pretty great | 16:39 |
| adarazs | cool to hear. the dep resolution actually worked pretty well for me too. | 16:40 |
| adarazs | trown: I see this in the rpmbuild.log: https://paste.fedoraproject.org/370783/64194417/ | 16:40 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Nothing to see here https://review.openstack.org/321113 | 16:40 |
| trown | adarazs: ah, I have a fix for that in my WIP on top of yours | 16:42 |
| adarazs | oh, cool. | 16:42 |
| adarazs | I'll rebase yours on the new change and try. | 16:42 |
| trown | adarazs: for some reason package install test in delorean is not working for puppet modules, even though installing them works fine on the actual undercloud | 16:42 |
| trown | so I just removed it :) | 16:42 |
| adarazs | awesome :) | 16:42 |
| adarazs | trown: the workflow etherpad looks okay. | 16:43 |
| trown | adarazs: ya obviously super WIP for the setup part since so much is yet to be merged, but the examples I think show off how easy it is to use | 16:44 |
| trown | I have a full deploy going with the heat composable roles example now | 16:44 |
| *** zoliXXL is now known as zoli|relocating | 16:44 | |
| *** zoli|relocating is now known as zoli_gone-proxy | 16:45 | |
| adarazs | trown: btw https://review.gerrithub.io/277358 is feature complete, I just need to update the docs, and test it a tiny bit more, but it seems to be working with all three case. | 16:46 |
| adarazs | (manual/jenkins/zuul) | 16:46 |
| adarazs | so if you have time, take a look and let me know if you see some room for improvement :) | 16:47 |
| openstackgerrit | Miles Gould proposed openstack/python-tripleoclient: Document --initial-state param on baremetal import https://review.openstack.org/321116 | 16:47 |
| trown | adarazs: awesome, will do | 16:47 |
| openstackgerrit | Miles Gould proposed openstack/python-tripleoclient: Document --initial-state param on baremetal import https://review.openstack.org/321117 | 16:47 |
| gfidente | shardy in the new comment you mean start from Step2 and in CephStorage case only keep 2/3 ? | 16:48 |
| openstackgerrit | Miles Gould proposed openstack/instack-undercloud: Add option to enable introspection of UEFI nodes https://review.openstack.org/321118 | 16:49 |
| openstackgerrit | Miles Gould proposed openstack/instack-undercloud: Add option to enable introspection of UEFI nodes https://review.openstack.org/321119 | 16:49 |
| openstackgerrit | Miles Gould proposed openstack/instack-undercloud: Document "enable UEFI" undercloud.conf option https://review.openstack.org/321120 | 16:49 |
| openstackgerrit | Giulio Fidente proposed openstack/tripleo-heat-templates: Configure CephStorage services via resource chains https://review.openstack.org/321092 | 16:56 |
| openstackgerrit | Miles Gould proposed openstack/instack-undercloud: Document "enable UEFI" undercloud.conf option https://review.openstack.org/321125 | 16:56 |
| openstackgerrit | Darragh Bailey proposed openstack/diskimage-builder: Handle file magic type varying order matching https://review.openstack.org/321126 | 16:56 |
| openstackgerrit | Miles Gould proposed openstack/instack-undercloud: Document "enable UEFI" undercloud.conf option https://review.openstack.org/321120 | 16:58 |
| *** dmacpher is now known as dmacpher-afk | 16:59 | |
| *** MaxPC has joined #tripleo | 16:59 | |
| *** athomas has quit IRC | 17:00 | |
| *** liverpooler has quit IRC | 17:01 | |
| *** fragatina has joined #tripleo | 17:03 | |
| *** noslzzp has quit IRC | 17:07 | |
| *** fragatina has quit IRC | 17:07 | |
| *** sshnaidm has joined #tripleo | 17:13 | |
| *** lucasagomes is now known as lucas-dinner | 17:13 | |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: Add tempest tests to periodic and experimental job https://review.openstack.org/321019 | 17:14 |
| *** trown is now known as trown|lunch | 17:15 | |
| *** myoung has joined #tripleo | 17:17 | |
| openstackgerrit | Pradeep Kilambi proposed openstack/tripleo-heat-templates: Fix ceph keyring setting for gnocchi https://review.openstack.org/321133 | 17:18 |
| *** jpena is now known as jpena|pto | 17:22 | |
| trozet | Hi All, can someone please tell me why heat parameters for disabling/enabling Neutron Agents were removed from overcloud.yaml, but still exist in controller.yaml? | 17:30 |
| trozet | dsneddon, dprince:^ ? | 17:32 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates: WIP: integration of the new puppet pacemaker. https://review.openstack.org/302409 | 17:38 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-tripleo: WIP: integration of the new puppet pacemaker. https://review.openstack.org/309069 | 17:40 |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates: WIP: integration of the new puppet pacemaker. https://review.openstack.org/302409 | 17:40 |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add lookup_hiera_hash function https://review.openstack.org/320112 | 17:42 |
| *** gfidente has quit IRC | 17:42 | |
| *** Goneri has quit IRC | 17:43 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/tripleo-heat-templates: WIP: integration of the new puppet pacemaker. https://review.openstack.org/302409 | 17:44 |
| *** Goneri has joined #tripleo | 17:45 | |
| *** oshvartz has joined #tripleo | 17:47 | |
| *** links has quit IRC | 17:52 | |
| *** pkovar has quit IRC | 17:53 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-tripleo: WIP: integration of the new puppet pacemaker. https://review.openstack.org/309069 | 17:54 |
| *** saneax is now known as saneax_AFK | 17:55 | |
| openstackgerrit | Athlan-Guyot sofer proposed openstack/puppet-pacemaker: WIP: integrate PCS provider in the merge. https://review.openstack.org/310713 | 17:56 |
| *** fzdarsky is now known as fzdarsky|afk | 18:02 | |
| *** fragatina has joined #tripleo | 18:03 | |
| dprince | trozet: the plan was to cleanup controller.yaml parameters once all the network services have been decomposed | 18:06 |
| dprince | trozet: getting CI passing on the ML2 agent and server patches is proving slow | 18:06 |
| dsneddon | trozet: I wasn't aware that there were any changes to heat parameters for disabling/enabling Neutron Agents that removed them from overcloud.yaml | 18:06 |
| *** Tristitia has quit IRC | 18:06 | |
| trozet | dprince: so in mitaka, what is the correct to disable agents now? | 18:06 |
| trozet | dprince: We were setting parameters in environment files. Do parameter_defaults just set the defaults for every template? Will that work if the parameter is missing in overcloud.yaml? | 18:07 |
| dprince | trozet: http://git.openstack.org/cgit/openstack/tripleo-heat-templates/tree/environments/neutron-plumgrid.yaml#n6 | 18:07 |
| *** fragatina has quit IRC | 18:08 | |
| trozet | dprince: but thats composable services right? | 18:08 |
| dprince | trozet: sorry, I missed that you said Mitaka | 18:09 |
| dprince | trozet: so for the Mitaka release I think we still have Enable/Disable boolean parameters to control the various neutron services | 18:09 |
| dprince | trozet: some of them anyway | 18:09 |
| trozet | dprince: not in overcloud.yaml | 18:09 |
| *** zoli_gone-proxy is now known as zoliXXL | 18:09 | |
| dprince | trozet: right, so use parameter_defaults and it'll work | 18:09 |
| dprince | trozet: parameter_defaults get passed to all the nested stacks | 18:10 |
| trozet | dprince: so shouldn't they be added back in mitaka? Or is parameter_defaults an acceptable way to set parameters? | 18:10 |
| dprince | trozet: parameter_defaults is acceptable. We've switch python-tripleoclient so it always uses parameter_defaults for everything | 18:10 |
| trozet | dprince: I don't really get the difference... is parameter_defaults just searching for every template and replacing the parameter's default value with what you provide? | 18:11 |
| dprince | trozet: I don't like the name 'parameter_defaults'... but I think we like the way it behaves | 18:11 |
| dprince | trozet: it is providing a way to set the default value for each nested stack if it has a parameter of that name | 18:12 |
| trozet | dprince: ok just the default value | 18:13 |
| trozet | dprince: ok got it thanks | 18:13 |
| dprince | trozet: sure, was there a specific patch you wanted help with or anything? | 18:13 |
| EmilienM | FYI all puppet CI broken, major outage - see progress on https://review.openstack.org/#/c/321155/ | 18:13 |
| trozet | dprince: we have our own fork for OPNFV with ODL, ONOS SDN contoller patches, and we just migrated from liberty -> mitaka THT, and I noticed those are broken because we set NeutronL3Agent: false in our parameters for those | 18:14 |
| trozet | dprince: so easy fix, just change the parameters to defaults :) | 18:15 |
| dprince | trozet: yep, that should do it I think | 18:15 |
| slagle | trown|lunch: hey was wondering if have any log of when we promote a dlrn repo? | 18:16 |
| *** fragatina has joined #tripleo | 18:17 | |
| *** penick has quit IRC | 18:18 | |
| *** sthillma has joined #tripleo | 18:18 | |
| *** noslzzp has joined #tripleo | 18:19 | |
| *** electrofelix has quit IRC | 18:21 | |
| *** trown|lunch is now known as trown | 18:22 | |
| trown | slagle: specifically current-tripleo promotion? | 18:22 |
| slagle | trown: yes | 18:23 |
| slagle | trown: i'm wondering if we promoted a repo around the time https://bugs.launchpad.net/tripleo/+bug/1585641 started | 18:23 |
| openstack | Launchpad bug 1585641 in tripleo "CI: jobs failing with Message: No valid host was found. There are not enough hosts available., Code: 500"" [High,Triaged] | 18:23 |
| trown | slagle: it could use some timestamps... but https://trunk.rdoproject.org/centos7-master/promote-current-tripleo.log | 18:23 |
| *** fragatina has quit IRC | 18:24 | |
| trown | slagle: looks like we promoted this morning a repo from yesterday, just looking at timestamp of that log being modified and timestamp of https://trunk.rdoproject.org/centos7-master/current-tripleo/delorean.repo | 18:25 |
| slagle | i think i can track by what i need based on the log | 18:27 |
| *** penick has joined #tripleo | 18:29 | |
| *** fragatina has joined #tripleo | 18:30 | |
| slagle | based on that log, i think this is repo where the issue started: http://trunk.rdoproject.org/centos7/39/d0/39d0fdf0de72a3e76716948497b2b5009d1c64b6_8df7385a/ | 18:31 |
| *** apetrich has quit IRC | 18:32 | |
| *** apetrich has joined #tripleo | 18:32 | |
| *** fragatina has quit IRC | 18:34 | |
| *** zoliXXL is now known as zli|afk | 18:37 | |
| *** zli|afk is now known as zoli|afk | 18:37 | |
| *** zoli|afk is now known as zoli | 18:51 | |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Nothing to see here https://review.openstack.org/321113 | 18:55 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Nothing to see here https://review.openstack.org/321113 | 18:56 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Nothing to see here https://review.openstack.org/321113 | 18:56 |
| *** noslzzp has quit IRC | 18:57 | |
| *** liverpooler has joined #tripleo | 18:57 | |
| *** mgarciam has left #tripleo | 19:06 | |
| *** fragatina has joined #tripleo | 19:07 | |
| *** fragatina has quit IRC | 19:07 | |
| *** fragatina has joined #tripleo | 19:08 | |
| *** liverpooler has quit IRC | 19:13 | |
| larsks | trown: what sets up the masquerade rules on the undercloud for the controlplane network (192.0.2.0/24)? There are two rules, one in POSTROUTING and one in BOOTSTACK_MASQ (referenced by POSTROUTING)? | 19:30 |
| trown | larsks: not sure... I know instack-undercloud is setting them up, but not so much on why | 19:32 |
| larsks | trown: okay. Here's the thing: when network_isolation=True, the overcloud hosts are using the "external" network for outbound access, but there are not NAT rules for that, so it's a dead end...so the overcloud hosts can't install packages, etc. | 19:33 |
| trown | slagle: bnemec do you know what the purpose of those rules are ^ | 19:33 |
| larsks | trown: I'm not asking the *purpose* of the rules :) | 19:33 |
| trown | oh | 19:33 |
| larsks | I was more curious about whether we should be adding rules for the external network for the quickstart... | 19:34 |
| larsks | ...in order to enable outbound access from the overcloud hosts. And whether or not this is a virt-only problem. | 19:34 |
| larsks | I sort of think it is, since the network_isolation stuff just creates a fake vlan that's not actually connected to anything. | 19:34 |
| slagle | for virt, you need to add an interface on the undercloud for the external network | 19:35 |
| slagle | assign it an IP, and set the IP as the default route in your net iso template | 19:36 |
| trown | larsks: shouldnt we route overcloud traffic through the undercloud | 19:36 |
| trown | ah ya that ^ | 19:36 |
| slagle | if you look in tripleo-ci or khaleesi, you will see it does that | 19:36 |
| larsks | trown: we *are*. slagle are you familiar with how the undercloud is confiugred by the quickstart? | 19:36 |
| slagle | no, i'm not familiar with quickstart | 19:36 |
| larsks | The network_isolation configuration (contributed by matbu) sets up a vlan interface and gives it an address, and this becomes the "external" network. | 19:37 |
| larsks | It's not actually connected to a "real" physical network. | 19:37 |
| larsks | So one way of solving that would simply be to set up NAT rules on the undercloud. | 19:37 |
| larsks | trown: we are routing through the undercloud; it's just that the undercloud is right now where packets go to die. | 19:37 |
| larsks | matbu: any changes you're around? | 19:38 |
| larsks | *chance | 19:39 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: Explode loadbalancer role in 2 sub-roles https://review.openstack.org/321201 | 19:40 |
| slagle | larsks: ok, looking at some of my notes, i do add a masquerade rule for the external network subnet as well on the undercloud | 19:41 |
| slagle | this is what i do: http://paste.openstack.org/show/505446/ | 19:42 |
| *** chem`` has joined #tripleo | 19:42 | |
| *** oshvartz has quit IRC | 19:42 | |
| larsks | slagle: ok, yeah, that is pretty much exactly what I was proposing. trown: I guess that should go in our overcloud-post script? | 19:43 |
| slagle | i'm surprised you're getting passed the network validations w/o that | 19:43 |
| trown | larsks: ah ya, makes sense | 19:43 |
| slagle | or i guess those just ping the external route, nothing else | 19:43 |
| *** chem` has quit IRC | 19:43 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: move most things out from the playbooks dir, and use config for release info https://review.openstack.org/313116 | 19:44 |
| openstackgerrit | yolanda.robla proposed openstack/diskimage-builder: Fix yum-minimal locale config https://review.openstack.org/319909 | 19:47 |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: loadbalancer: make sure controller_* are array-typed https://review.openstack.org/321206 | 19:53 |
| *** pkovar has joined #tripleo | 19:55 | |
| *** pkovar has quit IRC | 19:56 | |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: Explode loadbalancer role in 2 sub-roles https://review.openstack.org/321201 | 19:56 |
| *** lblanchard has quit IRC | 19:58 | |
| *** ayoung has quit IRC | 20:04 | |
| *** jprovazn has quit IRC | 20:05 | |
| openstackgerrit | Emilien Macchi proposed openstack/instack-undercloud: loadbalancer: explode hieradata to separate HAproxy & keepalived https://review.openstack.org/321210 | 20:10 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: loadbalancer: update hiera parameters for HAproxy/keepalived split https://review.openstack.org/320411 | 20:18 |
| *** jayg is now known as jayg|g0n3 | 20:19 | |
| *** noslzzp has joined #tripleo | 20:22 | |
| *** jcoufal has joined #tripleo | 20:23 | |
| *** paramite|afk is now known as paramite | 20:26 | |
| openstackgerrit | Emilien Macchi proposed openstack/puppet-tripleo: Update profiles for loabalancer split https://review.openstack.org/320410 | 20:26 |
| openstackgerrit | Emilien Macchi proposed openstack/tripleo-heat-templates: loadbalancer: update hiera parameters for HAproxy/keepalived split https://review.openstack.org/320411 | 20:27 |
| shardy | slagle, EmilienM: Hey, I'm about to drop off post-PTO, are you happy to go ahead and release all-the-things sometime next week based on a good periodic job promotion? | 20:27 |
| shardy | I did't get around to writing a script to automate it yet, so it'll be a bit of cut/paste from the delorean repo hashes | 20:28 |
| shardy | And ensuring we bump or complete what remains on https://launchpad.net/tripleo/+milestone/newton-1 | 20:28 |
| EmilienM | shardy: like a first newton? | 20:28 |
| EmilienM | ure | 20:28 |
| EmilienM | sure* | 20:28 |
| EmilienM | shardy: I'm also working on neutron-1 for all our puppet modules | 20:29 |
| shardy | EmilienM: yup - the idea was to release all of the tripleo related repos around n-1 with known-good versions, so we can claim that to be our n-1 milestone release | 20:29 |
| EmilienM | i'll do tripleo in same time and ask reviews from slagle | 20:29 |
| EmilienM | shardy: no worries | 20:29 |
| shardy | EmilienM: great, if you guys can work together that would be great | 20:29 |
| EmilienM | shardy: enjoy your well-deserved PTO | 20:29 |
| EmilienM | we'll do it together | 20:29 |
| shardy | I'll sort out the launchpad stuff when I get back - I was planning to run a script over the tripleo bugs and switch all Fix Committed things to Fix Released | 20:30 |
| EmilienM | shardy: did you find someone to lead the weekly meeting? | 20:30 |
| shardy | (or you can do it if you like) | 20:30 |
| EmilienM | I can do it if you want | 20:30 |
| *** paramite has quit IRC | 20:30 | |
| shardy | EmilienM: Not yet, that would be great, thanks! | 20:30 |
| EmilienM | cool I'll do it | 20:30 |
| shardy | Thanks, sorry for the bad timing wrt the milestone, I booked this before realizing the conflict | 20:31 |
| EmilienM | no worries | 20:31 |
| slagle | yea i can help out | 20:31 |
| shardy | Great, thanks guys, much appreciated | 20:32 |
| openstackgerrit | Andreas Florath proposed openstack/diskimage-builder: Refactor: block-device handling https://review.openstack.org/319591 | 20:34 |
| *** shardy has quit IRC | 20:35 | |
| openstackgerrit | Gabriele Cerami proposed openstack/tripleo-quickstart: Add options to download and handle overcloud images https://review.openstack.org/321057 | 20:37 |
| panda | trown: is it normal that I have to install tripleoclient before launching any scripts, even with the rdo image ? | 20:40 |
| *** dsariel has joined #tripleo | 20:41 | |
| trown | panda: hmm, no that should be on the undercloud image | 20:41 |
| trown | oh, right but maybe not on the tripleo instack.qcow2 | 20:42 |
| trown | forgot what you were working on for a second | 20:42 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: move most things out from the playbooks dir, and use config for release info https://review.openstack.org/313116 | 20:43 |
| panda | trown: hm, no, to test overcloud injection I'm using rdo images (no md5 yet on tripleo-ci) | 20:44 |
| trown | panda: oh... the undercloud.qcow2 from RDO should most definitely have python-tripleoclient pre-installed | 20:45 |
| panda | then I'm messing things up... | 20:45 |
| trown | its ok I do it all the time :P | 20:47 |
| *** panda has quit IRC | 20:48 | |
| *** panda has joined #tripleo | 20:48 | |
| *** dprince has quit IRC | 20:54 | |
| trown | larsks: ok https://review.openstack.org/313116 is legitimately passing now (pass on PS11 was a false positive which is a bit disconcerting) mind if I turn all open patches into merge conflicts? | 20:55 |
| *** julim has quit IRC | 20:56 | |
| *** penick has quit IRC | 21:03 | |
| *** rhallisey has quit IRC | 21:10 | |
| openstackgerrit | Merged openstack/tripleo-quickstart: move most things out from the playbooks dir, and use config for release info https://review.openstack.org/313116 | 21:12 |
| *** fultonj has quit IRC | 21:14 | |
| panda | oh | 21:15 |
| panda | .... | 21:15 |
| panda | /o\ | 21:16 |
| panda | how much do I have to rebase now ? | 21:16 |
| trown | panda: it touches just about every file, but it is mostly just a rename so not too bad | 21:17 |
| *** rhallisey has joined #tripleo | 21:23 | |
| *** trown is now known as trown|outtypewww | 21:23 | |
| *** penick has joined #tripleo | 21:25 | |
| *** penick has quit IRC | 21:26 | |
| *** Goneri has quit IRC | 21:30 | |
| *** sthillma has quit IRC | 21:32 | |
| *** penick has joined #tripleo | 21:37 | |
| *** morazi has quit IRC | 21:37 | |
| *** myoung has quit IRC | 21:39 | |
| *** penick has quit IRC | 21:50 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add lookup_hiera_hash function https://review.openstack.org/320112 | 21:51 |
| *** apetrich has quit IRC | 21:54 | |
| *** morazi has joined #tripleo | 21:55 | |
| openstackgerrit | Giulio Fidente proposed openstack/puppet-tripleo: Add lookup_hiera_hash function https://review.openstack.org/320112 | 21:55 |
| *** dbecker has quit IRC | 21:56 | |
| *** penick has joined #tripleo | 22:00 | |
| *** dbecker has joined #tripleo | 22:07 | |
| *** morazi has quit IRC | 22:10 | |
| openstackgerrit | Gabriele Cerami proposed openstack/tripleo-quickstart: Add options to download and handle overcloud images https://review.openstack.org/321057 | 22:10 |
| *** noslzzp has quit IRC | 22:16 | |
| *** dbecker has quit IRC | 22:26 | |
| *** gumbyblake has joined #tripleo | 22:26 | |
| *** jcoufal has quit IRC | 22:27 | |
| *** ayoung has joined #tripleo | 22:30 | |
| *** egafford has quit IRC | 22:37 | |
| *** pradk has quit IRC | 22:40 | |
| *** zoli is now known as zoli_gone-proxy | 22:42 | |
| *** akshai_ has quit IRC | 22:42 | |
| EmilienM | slagle: I realize we have no release notes in tripleo, right? | 22:47 |
| *** dkelson has joined #tripleo | 22:56 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: Fix full deploy CI https://review.openstack.org/321273 | 22:58 |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: Fix full deploy CI https://review.openstack.org/321273 | 23:01 |
| *** dkelson has quit IRC | 23:01 | |
| openstackgerrit | John Trowbridge proposed openstack/tripleo-quickstart: Fix full deploy CI https://review.openstack.org/321273 | 23:04 |
| *** chlong has joined #tripleo | 23:05 | |
| *** dsariel has quit IRC | 23:07 | |
| *** tzumainn has quit IRC | 23:08 | |
| openstackgerrit | Merged openstack/tripleo-quickstart: Fix full deploy CI https://review.openstack.org/321273 | 23:13 |
| *** tzumainn has joined #tripleo | 23:18 | |
| *** MaxPC has quit IRC | 23:24 | |
| *** Goneri has joined #tripleo | 23:24 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!