| opendevreview | OpenStack Proposal Bot proposed openstack/openstack-ansible master: Imported Translations from Zanata https://review.opendev.org/c/openstack/openstack-ansible/+/846871 | 03:44 |
|---|---|---|
| *** ysandeep|out is now known as ysandeep | 04:34 | |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Bump roles and services for Yoga RC1 https://review.opendev.org/c/openstack/openstack-ansible/+/846797 | 06:06 |
| noonedeadpunk | snadge: could it be some race condition, that check was faster then service actually came up? | 06:14 |
| jrosser_ | would that be https in that case? | 06:18 |
| jrosser_ | hmm | 06:19 |
| noonedeadpunk | we defaulted internal to https only in CI | 06:20 |
| noonedeadpunk | it's done in bootstrap-host | 06:20 |
| noonedeadpunk | https://opendev.org/openstack/openstack-ansible/src/branch/master/tests/roles/bootstrap-host/templates/user_variables.aio.yml.j2#L262 | 06:21 |
| noonedeadpunk | so for normal deployments, internal is not protected by default | 06:21 |
| jrosser_ | oh right yes we did talk about this before | 06:22 |
| noonedeadpunk | I guess we did that because of non-trivial upgrade path | 06:22 |
| jrosser_ | and if we needed to drop CI back to http in order to then test upgrades | 06:22 |
| noonedeadpunk | for upgrades we actually test http indeed now | 06:23 |
| noonedeadpunk | I wonder if for Z we should just enable https by default and create an upgrade playbook that would preserve http in variables | 06:24 |
| noonedeadpunk | Or we can land that for Y :D | 06:25 |
| noonedeadpunk | But I'd rather not | 06:25 |
| opendevreview | Merged openstack/openstack-ansible master: Imported Translations from Zanata https://review.opendev.org/c/openstack/openstack-ansible/+/846871 | 06:38 |
| *** ysandeep is now known as ysandeep|afk | 06:42 | |
| noonedeadpunk | https://review.opendev.org/c/openstack/ansible-config_template/+/846391 finally passed. I wonder if it will just work as expected | 06:59 |
| opendevreview | Damian Dąbrowski proposed openstack/openstack-ansible-os_tempest master: Define fixed_network_name when dynamic credentials are not used https://review.opendev.org/c/openstack/openstack-ansible-os_tempest/+/846954 | 07:31 |
| snadge | i dont think so.. it retries the test 10 times or something then gives up | 07:43 |
| noonedeadpunk | but manually it works? | 07:45 |
| snadge | i haven't looked into it much further than that.. other than trying to connect to that ip and port, which it does | 07:45 |
| snadge | the error isn't connection refused though.. its service unavailable | 07:46 |
| snadge | i just used telnet | 07:46 |
| noonedeadpunk | well it sounds like haproxy marked backend as down for some reason | 07:46 |
| noonedeadpunk | aha | 07:46 |
| noonedeadpunk | so yeah, worth checking why haproxy marks it as down and then check keystone itself. I guess you mentioned Xena, so it should be apache status and keystone-api service | 07:47 |
| jrosser_ | snadge: try curl/wget against the backend rather than telnet, haproxy is looking for a proper http response rather than just if the port is connectable | 07:51 |
| snadge | yeah its coming back 503 so if i dig through the logs i should find out why | 07:59 |
| anskiy | sounds like that old error when mariadb on debian-based systems couldn't properly start after installation | 08:00 |
| jrosser_ | snadge: you mean 503 from the keystone backend or 503 from haproxy? | 08:01 |
| noonedeadpunk | I recently catched issue when tmpfiles were not created due to race condition on systemd-tmpfiles service startup. that was on centos though. But that lead to apache not being able to start | 08:02 |
| noonedeadpunk | I believe 503 can go only from haproxy | 08:02 |
| snadge | from the address it was trying to test which is the br-mgmt address on the controller port 5000 | 08:02 |
| noonedeadpunk | try curl to backend directly | 08:03 |
| snadge | keystone_service-front-2 its configured as in haproxy.cfg | 08:04 |
| noonedeadpunk | 503 is no available backends error iirc | 08:05 |
| noonedeadpunk | so look on backends not frontends :) | 08:05 |
| jrosser_ | "from the address it was trying to test which is the br-mgmt address on the controller port 5000" <- this is the frontend, that HAProxy presents the loadbalanced service on | 08:05 |
| anskiy | snadge: there is a block keystone_service-back right after the one you've mentioned: try to curl one of those hosts | 08:05 |
| jrosser_ | you need to look in the keystone backend config, it will list the actual keystone service IP/ports and it's those you want to test with curl | 08:06 |
| snadge | yeah thats what i just did and its not connecting.. just timing ou | 08:07 |
| snadge | ok so the keystone container should have this 172.29.239.91:5000 | 08:09 |
| jrosser_ | the next thing to do would be check on that container for the status of apache and the keystone api service | 08:09 |
| snadge | apache2 is loaded and running and so is keystone-wsgi-public.service | 08:13 |
| opendevreview | Danny Nelhams proposed openstack/openstack-ansible-os_magnum stable/xena: Fixed dest typo in config_template https://review.opendev.org/c/openstack/openstack-ansible-os_magnum/+/846932 | 08:14 |
| snadge | i can ping that ip address but port 5000 isn't listening | 08:14 |
| jrosser_ | there should be some config a bit like this https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_073/845684/1/gate/openstack-ansible-deploy-aio_lxc-ubuntu-focal/0739c11/logs/etc/openstack/aio1_keystone_container-36e9f45e/apache2/sites-available/keystone-httpd.conf.txt | 08:21 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Bump roles and services for Yoga RC1 https://review.opendev.org/c/openstack/openstack-ansible/+/846797 | 08:23 |
| snadge | yep i see it, theres a listen on that address and a virtualhost directive that passes it to 127.0.0.1:35358 | 08:24 |
| snadge | ahh you're kidding me i just restarted apache | 08:26 |
| opendevreview | Danny Nelhams proposed openstack/openstack-ansible-os_magnum stable/wallaby: Fixed dest typo in config_template https://review.opendev.org/c/openstack/openstack-ansible-os_magnum/+/846964 | 08:33 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Add timestamp to written journal file https://review.opendev.org/c/openstack/openstack-ansible/+/846439 | 08:47 |
| jrosser_ | snadge: apache should have been restarted during the installation automatically | 09:06 |
| jrosser_ | like this (and a few other places) https://github.com/openstack/openstack-ansible-os_keystone/blob/e26aabe440b75d134ee90907656207445ddfcf64/tasks/keystone_apache.yml#L56 | 09:07 |
| snadge | yeah odd.. then i had another error.. restarted it and its going past now.. im not too fussed about this, as its a "test" ubuntu installation before we actually install it on production hardware | 09:09 |
| snadge | theres issue with licensing, drivers.. and it seems easier to go this way | 09:09 |
| snadge | centos hasn't been much fun lately | 09:09 |
| jrosser_ | if you ctrl-c then you can end up in a situation where the config files have been dropped but the handlers wont run | 09:09 |
| snadge | i think at one point i tried blowing away the containers and starting the install from scratch | 09:10 |
| *** ysandeep|afk is now known as ysandeep | 09:10 | |
| snadge | i have one of those environments where everything is behind a proxy so i have to use config to assist with that | 09:15 |
| jrosser_ | snadge: same here re. proxy, if theres anything we don't cover properly for you or it's difficult, let me know | 09:29 |
| snadge | Slurp up the constraints file for later re-deployment is failing now | 09:32 |
| snadge | its saying it cant find "/var/www/repo/os-releases/24.2.0/ubuntu-20.04-x86_64/requirements/placement-24.2.0-constraints.txt" | 09:37 |
| snadge | looks like this is building the placement container so at least im past the keystone issue now | 09:38 |
| opendevreview | Merged openstack/openstack-ansible-os_tempest master: Check return value for router create https://review.opendev.org/c/openstack/openstack-ansible-os_tempest/+/846802 | 09:38 |
| jrosser_ | snadge: it helps if you can paste the actual errors, and say which playbook it is, etc | 09:41 |
| jrosser_ | paste.opendev.org of course, not into IRC | 09:41 |
| jrosser_ | the repo server should be hosting those files (placement-24.2.0-constraints.txt) after the python wheels for that service have been built | 09:42 |
| jrosser_ | similar thing with checking that you can get those with curl, should be port 8181 on the internal vip i think | 09:43 |
| anskiy | snadge: this could be because of interrupted execution some time ago. There are four files per each virtualenv in /var/www/repo/os-releases/24.2.0/ubuntu-20.04-x86_64/requirements/ and it could be so you're missing some. Deleting all of them like '/var/www/repo/os-releases/24.2.0/ubuntu-20.04-x86_64/requirements/placement-* could solve the issue, effectively causing rebuild of all the dependent packages for that env, AFAIR. | 09:46 |
| noonedeadpunk | jrosser_: as you have some insight about zun, I have question about etcd there | 09:48 |
| * jrosser_ summons andrewbonney :) | 09:49 | |
| noonedeadpunk | so we install etcd as part of zun https://opendev.org/openstack/openstack-ansible-os_zun/src/branch/master/tasks/main.yml#L188-L196 | 09:49 |
| noonedeadpunk | But we also have playbook for standalone etcd installation https://opendev.org/openstack/openstack-ansible/src/branch/master/playbooks/etcd-install.yml | 09:49 |
| noonedeadpunk | I'm not really aware about etcd functionality, but why we can't leverage the one installed with infra? | 09:50 |
| noonedeadpunk | Asking mostly as trying to triage https://bugs.launchpad.net/openstack-ansible/+bug/1979208 and way to solve that | 09:51 |
| noonedeadpunk | And I believe this issue indeed happens because of serial | 09:51 |
| andrewbonney | I can't remember a specific reason it couldn't be changed to use an external cluster, but I clearly didn't get that far when trying to tidy the role up | 09:52 |
| noonedeadpunk | But likely to solve that we need to install etcd outside of zun role. And it's added right at the end of main.yml... | 09:53 |
| noonedeadpunk | so sounds like zun even don't need it on config stage. | 09:53 |
| jrosser_ | we had quite a big look at zun in V(?) but eventually did not deploy it | 09:58 |
| noonedeadpunk | I believe I broke things on X for real deployments | 10:04 |
| noonedeadpunk | with changes to playbooks regarding serial/facts gathering | 10:04 |
| noonedeadpunk | eventually etcd role deployment isn't designed to run in serial... | 10:05 |
| jrosser_ | it does seem that the use of etcd in the zun role is really minimal | 10:05 |
| jrosser_ | tbh not sure why this is not using the infra role | 10:06 |
| noonedeadpunk | the only reason I see is different cluster name... But I'm not sure about etcd overall - like can it be used by different services at same time... | 10:08 |
| noonedeadpunk | never looked into it deeply | 10:08 |
| jrosser_ | the other use case was calico, which i guess we are really close to deprecating anyway | 10:17 |
| jrosser_ | and also we never addressed needing a lock manager for stuff like designate | 10:17 |
| jrosser_ | noonedeadpunk: we could move the calling of the etcd role out of the os_zun role into the playbook | 10:20 |
| jrosser_ | a bit like we do for installing glusterfs ahead of the repo server role, thats also a thing which just won't work for serial=1 | 10:20 |
| noonedeadpunk | yeah. but still I'd say it's worth fixing running role in serial manner... | 10:21 |
| noonedeadpunk | and I don't like idea to run some change across ha cluster at same time... | 10:21 |
| noonedeadpunk | or dunno | 10:21 |
| jrosser_ | is it more that it needs to be two plays "do etcd first in serial" "do zun service in serial" | 10:22 |
| noonedeadpunk | for lock driver etcd feature-wise sounds like not best option tbh | 10:25 |
| jrosser_ | yup | 10:25 |
| noonedeadpunk | but that's different topic anyway | 10:25 |
| noonedeadpunk | so the problem I see that etcd in serial on itself breaks now | 10:26 |
| noonedeadpunk | because it expects for other members to exists I believe and they're not yet | 10:26 |
| noonedeadpunk | But need to read more about etcd to understand it more. I bet there should be some flag to start regardless of other members... | 10:27 |
| *** dviroel|out is now known as dviroel | 11:30 | |
| mgariepy | good morning | 13:15 |
| jrosser_ | hello | 13:22 |
| mgariepy | anything need a push this morning ? | 13:23 |
| jrosser_ | well - this is top priority 846797 | 13:24 |
| jrosser_ | oh | 13:24 |
| jrosser_ | https://review.opendev.org/c/openstack/openstack-ansible/+/846797 | 13:24 |
| mgariepy | hmm c9s failing.. | 13:25 |
| mgariepy | surprising ;) | 13:25 |
| mgariepy | https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_22b/846797/4/check/openstack-ansible-deploy-aio_lxc-centos-9-stream/22b5585/logs/ara-report/results/2968.html | 14:11 |
| mgariepy | https://paste.openstack.org/show/bBifbZuzcQDJ21e6tNoB/ | 14:11 |
| mgariepy | seems to be kinda ok on the cinder logs.. | 14:11 |
| jrosser_ | i looked in the haproxy log and it just says the backend is down | 14:11 |
| jrosser_ | i wonder if it took too long to create that volume or something and didnt respond to the healthcheck in time | 14:12 |
| *** ysandeep is now known as ysandeep|afk | 14:19 | |
| mgariepy | can this be an issue with the thread/process config ? | 14:34 |
| fanfi | lj5p;< zin 8uy h aioik,,mda< J<Z NNSSOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOONNNNNNNNNNNNNNNNNNNNNKOOOOOOOOOACK LEEEEBZ cvvxsaassdczdcd4, | 14:37 |
| fanfi | 900 | 14:37 |
| fanfi | 999999999999999999999999999n | 14:37 |
| fanfi | i | 14:37 |
| fanfi | j | 14:37 |
| spatel | Look like keyboard issue.. lol | 14:48 |
| mgariepy | or an app focus one. | 14:48 |
| noonedeadpunk | #startmeeting openstack_ansible_meeting | 15:02 |
| opendevmeet | Meeting started Tue Jun 21 15:02:38 2022 UTC and is due to finish in 60 minutes. The chair is noonedeadpunk. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:02 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:02 |
| opendevmeet | The meeting name has been set to 'openstack_ansible_meeting' | 15:02 |
| noonedeadpunk | #topic roll call | 15:02 |
| noonedeadpunk | hey o/ sorry for being late | 15:02 |
| NeilHanlon | o/ | 15:03 |
| mgariepy | hello | 15:03 |
| jrosser_ | o/ hello | 15:04 |
| noonedeadpunk | #topic office hours | 15:05 |
| noonedeadpunk | so. We almost branched Yoga for integrated repo and branched all repos | 15:05 |
| noonedeadpunk | except it failing in CI :) | 15:06 |
| noonedeadpunk | After we branch we have to drop centos 8 | 15:06 |
| noonedeadpunk | and Rocky 8 as well as a result mostly because py36 being unsupported by Zed and no python-libvirt bindings for py38 | 15:07 |
| noonedeadpunk | NeilHanlon: any info about Rocky 9 ? | 15:07 |
| jrosser_ | NeilHanlon: ^ | 15:07 |
| jrosser_ | :) | 15:07 |
| noonedeadpunk | ok, next to that I wonder how you feel about publishing config_template collection on galaxy? | 15:10 |
| noonedeadpunk | should we drop symlinks created for compatability as module for that? | 15:11 |
| noonedeadpunk | releases team suggested to tag it as 2.0.0 regardless of publishing | 15:12 |
| noonedeadpunk | because we dropped py27 support basically | 15:12 |
| jrosser_ | that would be good | 15:12 |
| jrosser_ | then its very easily consumable | 15:12 |
| NeilHanlon | jrosser_: we have a release candidate! I should have an image available soon-ish and am going to put in a few commits for dib, too | 15:12 |
| *** ysandeep|afk is now known as ysandeep | 15:12 | |
| noonedeadpunk | NeilHanlon: these are very good news! | 15:12 |
| jrosser_ | fun upgrade path | 15:13 |
| noonedeadpunk | then I believe we should backport Rocky 9 support to Y... | 15:13 |
| noonedeadpunk | should be quite straightforward? | 15:13 |
| noonedeadpunk | given no huge surprises there.... | 15:13 |
| NeilHanlon | i can help with the python-libvirt bindings for python3.8, if that would help. but also just as happy to move towards 9 and revisit 8 if I have time | 15:13 |
| jrosser_ | should be - i expect maybe a few places we specifically check for centos-9 or named vars files | 15:14 |
| jrosser_ | but that can be fixed | 15:14 |
| NeilHanlon | totally, i kept a list of the places I had to touch last time so hopefully I'll catch them before review ;) | 15:14 |
| jrosser_ | i expect it to be much much less work than the original rocky or centos-9 | 15:14 |
| noonedeadpunk | I believe we could be more carefull now as well | 15:14 |
| noonedeadpunk | I guess moving towards 9 might be more perspecitve path. And libvirt thing is just first I recalled but there were likely more | 15:16 |
| NeilHanlon | undoubtedly | 15:16 |
| jrosser_ | if we backport 9 to Y we can drop 8 from master | 15:16 |
| jrosser_ | that would be nice | 15:16 |
| jrosser_ | then a bunch more logic / ternary / when: can get removed | 15:17 |
| NeilHanlon | that sounds like a reasonable plan. it'd be nice to only support 9 which ideally has less issues than 8 with dependencies | 15:17 |
| noonedeadpunk | I bet devstack and projects already dropped 8 for master | 15:17 |
| noonedeadpunk | ok, great that Rocky 9 is coming! | 15:17 |
| noonedeadpunk | I want to return back to config template. jrosser_ "that would be good" was it about publishing or removing legacy ?:) | 15:18 |
| noonedeadpunk | I think for publishing we just need to add this job https://review.opendev.org/c/openstack/ansible-config_template/+/846391/6/zuul.d/project.yaml | 15:19 |
| jrosser_ | oh well all of it really :) | 15:19 |
| noonedeadpunk | I'm not sure how tripleo consumes it actually.... | 15:19 |
| noonedeadpunk | Won't be surprised if with python setup.py install .... | 15:20 |
| noonedeadpunk | But I can recall pushing smth there to switch to collection | 15:20 |
| jrosser_ | i think that got undone | 15:20 |
| jrosser_ | when the setup.py modules stuff all went odd | 15:20 |
| noonedeadpunk | Ok, i abandoned it https://review.opendev.org/c/openstack/tripleo-ansible/+/820140 | 15:21 |
| noonedeadpunk | arxcruz|rover: do you recall how you install/consume config_template collection now? | 15:22 |
| noonedeadpunk | do you checkout to some specific tag at least? | 15:22 |
| arxcruz|rover | not really | 15:22 |
| noonedeadpunk | well.. I can propose patch to CI and witness how it fails... | 15:24 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/ansible-config_template master: Remove possibility to install as plugin https://review.opendev.org/c/openstack/ansible-config_template/+/847045 | 15:37 |
| noonedeadpunk | here we go | 15:38 |
| noonedeadpunk | I kind of wonder if we should publish it in galaxy first and then tripleo case would be solved | 15:39 |
| noonedeadpunk | as we have that https://opendev.org/openstack/openstack-ansible-os_tempest/src/branch/master/meta/main.yml#L44-L45 | 15:39 |
| noonedeadpunk | which kind of should pull collection in if it's absent? | 15:39 |
| noonedeadpunk | We also have interesting bug that I spoilered during the day regarding zun and etcd | 15:43 |
| noonedeadpunk | #link https://bugs.launchpad.net/openstack-ansible/+bug/1979208 | 15:43 |
| noonedeadpunk | the issue there is for some reason we have etcd installation inside zun role while we also have etcd in playbooks and even have env.d file for it | 15:44 |
| noonedeadpunk | So likely this needs some love and refactoring | 15:44 |
| noonedeadpunk | will try to look into that | 15:44 |
| noonedeadpunk | I beleive that's it for today from myself | 15:45 |
| noonedeadpunk | and thanks everybody for working on Y! | 15:46 |
| noonedeadpunk | I know last week was tough on amount of things we worked on | 15:46 |
| jrosser_ | i don't remeber why we have a separate etcd | 15:47 |
| noonedeadpunk | we'll figure that out :) | 15:47 |
| noonedeadpunk | maybe it's really necessity as in if they can't share the one | 15:47 |
| noonedeadpunk | *as in case | 15:47 |
| noonedeadpunk | *in case | 15:48 |
| noonedeadpunk | like trove is good example why you might want to have another rabbitmq cluster | 15:50 |
| noonedeadpunk | maybe here it's the same. | 15:50 |
| *** dviroel is now known as dviroel|lunch | 15:52 | |
| noonedeadpunk | #endmeeting | 15:54 |
| opendevmeet | Meeting ended Tue Jun 21 15:54:51 2022 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:54 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/openstack_ansible_meeting/2022/openstack_ansible_meeting.2022-06-21-15.02.html | 15:54 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/openstack_ansible_meeting/2022/openstack_ansible_meeting.2022-06-21-15.02.txt | 15:54 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/openstack_ansible_meeting/2022/openstack_ansible_meeting.2022-06-21-15.02.log.html | 15:54 |
| odyssey4me | noonedeadpunk jrosser_ o/ I think the etcd was extracted because there was a network service using it... Logan Vig was using it that way IIRC | 15:56 |
| odyssey4me | ie it was used across multiple services | 15:57 |
| noonedeadpunk | yeah, it was calico | 15:57 |
| noonedeadpunk | but can't etcd be shared across services by design? | 15:58 |
| jrosser_ | the odd thing is that the zun role came after calico iirc but makes it's own etcd cluster using the same role | 15:59 |
| jrosser_ | *same etcd role as calico | 16:00 |
| noonedeadpunk | it has different cluster name though | 16:00 |
| *** dviroel|lunch is now known as dviroel | 16:38 | |
| mgariepy | woohoo rc1 passed.. | 17:02 |
| jrosser_ | lets try for 2 in a row | 17:07 |
| mgariepy | should i go get lottery ? | 17:08 |
| jrosser_ | we need some luck | 17:10 |
| jrosser_ | over in #qa they discuss the value of having centos-9 jobs when they're constantly breaking | 17:11 |
| mgariepy | wonder about the thread/process issue tho. | 17:11 |
| *** ysandeep is now known as ysandeep|out | 17:37 | |
| noonedeadpunk | I can't say we have tons of issues with centos rather then with post_failures as of today | 18:27 |
| noonedeadpunk | what a surprise - https://review.opendev.org/c/openstack/ansible-config_template/+/847045 failed on our functional jobs | 18:33 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/ansible-config_template master: Remove possibility to install as plugin https://review.opendev.org/c/openstack/ansible-config_template/+/847045 | 18:37 |
| jrosser_ | oh well like I found the other day I think the functional config template jobs are very odd | 18:37 |
| jrosser_ | I don’t think we have anything to symlink to the zuul repo in the tox env | 18:37 |
| jrosser_ | or otherwise install from there | 18:38 |
| noonedeadpunk | yeah.... I think this is true | 18:39 |
| noonedeadpunk | We indeed need to proceed with molecule | 18:40 |
| noonedeadpunk | I got a bit hands-on with it :D | 18:40 |
| noonedeadpunk | We just need good way of managing variables for it in one place | 18:40 |
| jrosser_ | maybe this is silly but can it go in plugins collection | 18:41 |
| noonedeadpunk | tbh I'd rather did that in integrated repo. As plugins we likely need to sort out one day. | 18:42 |
| noonedeadpunk | for example I was thinking recently about moving service_setup to ansible-collections-openstack. | 18:42 |
| jrosser_ | we just need to make sure that the workflow out of zuul is sensible | 18:43 |
| noonedeadpunk | as what I found here tells me we do exactly smae thing in 2 repos https://github.com/openstack/bifrost/blob/7271695714739af10d741a39dc4acd5e68465cb5/playbooks/roles/bifrost-ironic-install/tasks/keystone_setup.yml | 18:45 |
| noonedeadpunk | I evene started a bit wondering why bifrost and os_ironic even co-exist, but I'd rather not go there now :D | 18:47 |
| jrosser_ | one step at a time | 18:54 |
| noonedeadpunk | yeah | 19:31 |
| mgariepy | look like it's going to merge now :D | 19:35 |
| opendevreview | Merged openstack/openstack-ansible master: Bump roles and services for Yoga RC1 https://review.opendev.org/c/openstack/openstack-ansible/+/846797 | 19:47 |
| *** dviroel is now known as dviroel|afk | 20:06 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!