| opendevreview | Boxiang Zhu proposed openstack/kolla-ansible master: [CI] Add skyline scenario https://review.opendev.org/c/openstack/kolla-ansible/+/861687 | 02:43 |
|---|---|---|
| opendevreview | Boxiang Zhu proposed openstack/kolla-ansible master: [CI] Add skyline scenario https://review.opendev.org/c/openstack/kolla-ansible/+/861687 | 05:19 |
| shyamb | Hi | 05:26 |
| shyamb | Does Kolla yoga supports binary and source installation modes? or just source? | 05:26 |
| hrw | shyamb: yoga supports binary but it is deprecated there already | 05:49 |
| shyamb | hrw: Okay.thanks. Is this deprecation started from Xena release? | 05:51 |
| shyamb | Any rough plan of when we are planning to stop complete binary support in future? | 05:52 |
| shyamb | Yoga release notes does not have this information: https://docs.openstack.org/releasenotes/kolla-ansible/yoga.html | 05:54 |
| hrw | shyamb: kolla does: https://docs.openstack.org/releasenotes/kolla/yoga.html#prelude | 05:57 |
| hrw | shyamb: Zed has only one type of images | 05:58 |
| hrw | Yoga will be supported for some time | 05:58 |
| shyamb | hrw: Okay, thank you. | 06:00 |
| opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: CI: cephadm - use download.ceph.com packages for Ubuntu and Rocky https://review.opendev.org/c/openstack/kolla-ansible/+/861701 | 06:07 |
| mnasiadka | morning | 06:10 |
| shyamb | hrw: In source images, can we install rpm, debian packages or only pip packages are supported? | 07:21 |
| shyamb | We are building a container for our service - triliovault | 07:21 |
| shyamb | we have only rpm and debian packages | 07:21 |
| hrw | shyamb: binary/source describes where from we take openstack components. and only that | 07:31 |
| hrw | shyamb: mariadb container image is same in both cases | 07:31 |
| shyamb | hrw: Okay | 07:31 |
| hrw | as at one point in time I moved few vars from base image into macro to make our images more 'the same' | 07:32 |
| hrw | shyamb: there was a time when we planned to add 3rd type - 'infra' | 07:33 |
| hrw | shyamb: I wrote stuff for it but we never enabled that by default | 07:33 |
| hrw | shyamb: so you would get debian-infra-mariadb + debian-source-nova-compute + debian-infra-nova-libvirt | 07:33 |
| shyamb | hrw: What would be our recommendation for third party applications running on kolla openstack? Should they also move their images to 'source' format? | 07:34 |
| hrw | shyamb: there is no binary/source anymore | 07:34 |
| shyamb | hrw: Okay | 07:34 |
| shyamb | I am from TrilioVault, we have only debian and rpm packages | 07:35 |
| hrw | shyamb: if your app is not openstack component then install as you like. if it is then show where source is | 07:35 |
| hrw | "TrilioVault is the only OpenStack-native data backup and recovery solution that gives cloud and backup administrators the ability to restore entire workloads in one click." one? | 07:35 |
| shyamb | yes | 07:36 |
| shyamb | hrw: So, I wanted to check if we can keep rpm, debian based container images instead of moving to pip format and virtual env | 07:36 |
| hrw | you can | 07:36 |
| shyamb | hrw: Okay | 07:37 |
| hrw | shyamb: I see that this is comercial product, right? | 07:38 |
| hrw | 'free trial' gives 30 days | 07:38 |
| hrw | 'free basic edition' is 16 nodes or 160 vcpus | 07:38 |
| hrw | hm. those are for k8s edition | 07:39 |
| hrw | 'triliovault for openstack' only has 'request a demo' | 07:40 |
| shyamb | hrw: Hi, I will check about licensing and will let you know | 07:40 |
| shyamb | There should be some free trial license. | 07:40 |
| shyamb | hrw: Yes, it's commercial product. | 07:41 |
| hrw | I will leave the decision to mnasiadka as PTL then. | 07:41 |
| mnasiadka | including a commercial licensed product in Kolla upstream open source repositories? I don't think that is possible | 07:43 |
| hrw | cause now we do not have images with 'binary only' software. even elasticsearch, with their weird licensing etc, has sources available | 07:43 |
| mnasiadka | But I remember at some point in time we wanted to enable users to build ,,bring your own container images'' - maybe that's a solution to that problem | 07:43 |
| hrw | we have --docker-dir option for adding external ones | 07:44 |
| shyamb | hrw: mnasiadka: We have not committed our code in upstream repos. We keep all our repos separate | 07:46 |
| mnasiadka | so again - source images mean we install OpenStack projects code from source, nothing else - all the other services are installed from rpm/deb | 07:51 |
| mnasiadka | as Marcin mentioned | 07:52 |
| shyamb | mnasiadka: Okay, thanks. | 08:47 |
| amorin | hello kolla team, is there any reason not starting neutron as a wsgi process behind apache2, like nova? | 08:51 |
| amorin | https://github.com/openstack/kolla/blob/master/docker/neutron/neutron-server/Dockerfile.j2 | 08:51 |
| mnasiadka | https://review.opendev.org/c/openstack/kolla/+/749601 | 09:11 |
| mnasiadka | see last comment | 09:11 |
| mnasiadka | I think nobody tried again since 2020 | 09:11 |
| amorin | ack | 09:12 |
| amorin | thanks | 09:12 |
| amorin | wsgi was introduced in 2019 AFAIKS on neutron side | 09:12 |
| amorin | we are facing some issue on our deployment using neutron in wsgi mod as well (not using kolla), so I was wondering if the community solved the issue | 09:13 |
| opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Add max_replication_lag to proxysql template https://review.opendev.org/c/openstack/kolla-ansible/+/861724 | 09:20 |
| kevko | amorin: what is the issue ...because i wanted to reimplement neutron in kolla-ansible | 09:22 |
| amorin | so, | 09:23 |
| kevko | because I think neutron should have neutron-rpc-server and neutron-api containers | 09:23 |
| amorin | when using wsgi, you have a choice to disable lazy loading of wsgi | 09:23 |
| kevko | and i wanted to implement neutron-api as other wsgi services in kolla-ansible | 09:23 |
| amorin | when we disable lazy loading, some neutron rpc messages sent by api are lost | 09:23 |
| amorin | I am not 100% sure that upstream is affected, we are running a custom stein deployment | 09:23 |
| kevko | hmm | 09:31 |
| opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Add heartbeat config option to rabbitmq.conf https://review.opendev.org/c/openstack/kolla-ansible/+/861727 | 09:37 |
| opendevreview | Boxiang Zhu proposed openstack/kolla-ansible master: Add skyline service https://review.opendev.org/c/openstack/kolla-ansible/+/828464 | 09:48 |
| opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Add max_message_size to rabbitmq config https://review.opendev.org/c/openstack/kolla-ansible/+/861728 | 09:57 |
| opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Add heartbeat config option to rabbitmq.conf https://review.opendev.org/c/openstack/kolla-ansible/+/861727 | 09:59 |
| opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Add heartbeat config option to rabbitmq.conf https://review.opendev.org/c/openstack/kolla-ansible/+/861727 | 10:01 |
| opendevreview | Boxiang Zhu proposed openstack/kolla-ansible master: [CI] Add skyline scenario https://review.opendev.org/c/openstack/kolla-ansible/+/861687 | 10:01 |
| opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Add heartbeat config option to rabbitmq.conf https://review.opendev.org/c/openstack/kolla-ansible/+/861727 | 10:03 |
| opendevreview | Boxiang Zhu proposed openstack/kolla-ansible master: [CI] Add skyline scenario https://review.opendev.org/c/openstack/kolla-ansible/+/861687 | 10:06 |
| opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Add max_replication_lag to proxysql template https://review.opendev.org/c/openstack/kolla-ansible/+/861724 | 10:08 |
| kevko | mnasiadka, yoctozepto >> https://review.opendev.org/c/openstack/kolla-ansible/+/861724 , https://review.opendev.org/c/openstack/kolla-ansible/+/861727, https://review.opendev.org/c/openstack/kolla-ansible/+/861728 ...can you then check after zuul will pass ? | 10:15 |
| kevko | it's only some config options | 10:15 |
| kevko | rabbitmq config tuning - heartbeat configuration which we needed to modify, and max_message_size ...(we found that some messages from nova or neutron were very big and rabbitmq was dropping this) | 10:16 |
| opendevreview | Boxiang Zhu proposed openstack/kolla-ansible master: [CI] Add skyline scenario https://review.opendev.org/c/openstack/kolla-ansible/+/861687 | 10:17 |
| kevko | these options are set to default rabbitmq values of course | 10:17 |
| kevko | proxysql - max_replication_lag ...because default value is 0 ..which is sometimes not what you want .. | 10:18 |
| opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: openvswitch: switch bridge setup to Ansible modules https://review.opendev.org/c/openstack/kolla-ansible/+/843460 | 11:07 |
| hrw | amorin: stein is a bit old | 11:19 |
| frickler | .oO(straight from the stone age) | 11:21 |
| mnasiadka | Stein? From the stone age? come on! | 11:21 |
| frickler | that's literally what it means | 11:22 |
| frickler | also for reference https://bugs.launchpad.net/neutron/+bug/1912359 was mentioned in #-neutron regarding OVN+wsgi | 11:22 |
| frickler | so no progress yet | 11:22 |
| opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: Fix post-config of OVS for manila-share servers https://review.opendev.org/c/openstack/kolla-ansible/+/824707 | 11:30 |
| mnasiadka | I started abandoning some really old patches that have been already covered or not applicable anymore. | 11:31 |
| mnasiadka | I'll prepare a list of maybe-relevant for next weeks meeting, so we can decide what to do with them. | 11:31 |
| mnasiadka | (kolla-ansible only) | 11:31 |
| frickler | sounds great, thx mnasiadka | 11:34 |
| mnasiadka | kevko: with your patches I have a bit of an issue - what happens if rabbitmq/proxysql changes the default? do we need to track it? it would be better to put it in the config only if the user defines it (well actually rabbitmq's conf can be easily overridden by a user without those variables). | 11:36 |
| kevko | mnasiadka: proxysql - in both cases this value is set and it's default is = 0 ...you are right that i put default 5, i will change to = 0 | 11:37 |
| hrw | kevko: -1 on one patch just after seeing commit message | 11:38 |
| mnasiadka | just saying if proxysql in a years time changes the default to 1, we're still setting 0, just set it to an empty variable in defaults/main.yml and check if it's defined by user in the code, and set it only then. | 11:39 |
| hrw | maybe I am too picky but commit message is a place to write "what changes" and "why it is changed" rather than "a few words which tell you to read whole patch and understand whole project" | 11:41 |
| kevko | mnasiadka: rabbitmq - yes you are right - but unfortunatelly rabbitmq can change between versions ..so that means i every version need to copy template from kolla | 11:41 |
| kevko | hrw: ok, i will change commit message | 11:42 |
| hrw | thx | 11:43 |
| hrw | kevko: when you also write release note then read it again and consider adding it into commit message | 11:45 |
| hrw | "rabbitmq: add ability to configure heartbeat" looks (imho) nicer than "Add heartbeat config option to rabbitmq.conf" as former one informs that it is configurable now. Second one just adds an option, but can it be altered/configured? | 11:47 |
| mnasiadka | kevko: right, rabbitmq.conf is nor ini, neither yaml, so we're not merging them - include that in the commit message - so everybody understands your motivation? | 11:50 |
| kevko | mnasiadka: ok | 11:50 |
| mnasiadka | Another stupid idea - RMQ 3.8 is unsupported now | 11:52 |
| mnasiadka | Should we think about raising the version in stable branches somehow? | 11:52 |
| mnasiadka | RabbitMQ 3.8 was released in October 2019, and reached end of life on July 31st, 2022, we strongly discourage its use | 11:53 |
| mnasiadka | https://www.rabbitmq.com/which-erlang.html | 11:53 |
| hrw | mnasiadka: rmq 3.10.8 + erlang 25.1.1? | 11:53 |
| hrw | mnasiadka: upgrade to rmq 3.10.8 + erlang 25.1.1? | 11:53 |
| mnasiadka | I guess so - question if we don't break anything by doing so | 11:54 |
| mnasiadka | seems 3.8 has a feature flag for rolling upgrades and allows interoperability with newer versions | 11:54 |
| mnasiadka | https://www.rabbitmq.com/upgrade.html#rolling-upgrades | 11:54 |
| mnasiadka | hrw/kevko/frickler: willing to +2 https://review.opendev.org/c/openstack/kolla/+/861355 ? | 11:55 |
| hrw | done | 11:56 |
| hrw | mnasiadka: or 3.8 -> 3.9.23 as they do not mention 3.8 -> 3.10 upgrade yet | 11:57 |
| hrw | mnasiadka: still erlang 25.1.1 everywhere (needs my patches merge to master) | 11:58 |
| frickler | mnasiadka: hrw: reno question on the above | 11:59 |
| hrw | frickler: we need one of those 'fix renos' patch before release to unify naming etc | 12:00 |
| mnasiadka | yes, that's right | 12:00 |
| mnasiadka | 3.11 is out, but let's maybe not push it :) | 12:01 |
| hrw | mnasiadka: 'check do we need to bump rmq' is probably good add to release cycle notes | 12:02 |
| mnasiadka | yes, basically we should list components that need bumping - let me check our current docs | 12:02 |
| opendevreview | Michal Nasiadka proposed openstack/kolla master: release-docs: Add infra components bump to R-5 https://review.opendev.org/c/openstack/kolla/+/861745 | 12:08 |
| opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: ovn: Break out role into ovn-db and ovn-controller roles https://review.opendev.org/c/openstack/kolla-ansible/+/847769 | 12:11 |
| opendevreview | Michal Nasiadka proposed openstack/kolla master: release-docs: Add infra components bump to R-5 https://review.opendev.org/c/openstack/kolla/+/861745 | 12:14 |
| hrw | mnasiadka: abandoned 3.10.8 patch - we move to newer erlang instead so no problem | 12:22 |
| mnasiadka | good, I worried it's some maintenance burden attached ;) | 12:22 |
| hrw | mnasiadka: we need to merge https://review.opendev.org/c/openstack/kolla/+/861096 https://review.opendev.org/c/openstack/kolla/+/861327 so we have newest erlang everywhere. And then backport and backport | 12:23 |
| hrw | mnasiadka: and then backport after each release again probably ;D | 12:24 |
| mnasiadka | first one is merging | 12:24 |
| mnasiadka | frickler: can you look into the second one? | 12:24 |
| mnasiadka | well, I think we should bump up the version in stable branches to 3.9 instead of 3.10 and test that we don't break anything with it | 12:25 |
| mnasiadka | 3.9 supports erlang 25 or only 24? | 12:25 |
| mnasiadka | seems it supports 25 | 12:26 |
| hrw | both | 12:26 |
| kevko | guys, I have a question regarding cinder upgrade - We are going to upgrade OS from victoria to wallaby, kolla-ansible is restarting all services, then it waits for configurable time (not ideal - but OK) and then online db migration is running ...problem is that we have two cinder volumes (clustered with redis coordination) ..but cinder-volume is reporting ONLINE after 25 minutes (because tousands and tousands volumes) ... we can't bump count of | 12:30 |
| kevko | cinder-volume services ..because then ceph is killed by cinder-volumes ... | 12:30 |
| amorin | I know stein is old, but hey, I do with what I have :) | 12:30 |
| kevko | so, we will be offline for 25 minutes | 12:30 |
| kevko | any suggestion how to deal with this ? | 12:31 |
| hrw | amorin: you also waste your time a bit as upstream will tell 'rebase to master' anyway | 12:32 |
| kevko | i think in kolla-ansible there should be a task which will check if cinder-volume is ONLINE ..then kolla-ansible will continue and restart other cinder-volumes to avoid time when we will be offline | 12:32 |
| frickler | how do you handle this during normal maintenance? also not everyone identifies as "guys" | 12:32 |
| hrw | 'folks' is often considered as more neutral | 12:34 |
| kevko | frickler: sorry, no problem :) , from today i will use 'folks' | 12:34 |
| amorin | this is not a waste IMO, most of the issues I currently have in stein are usually in master, but anyway, agree that it's old, I cant update yet unfortunately | 12:34 |
| kevko | frickler: normal maintance ? normal maintance ussually not includes DB upgrades, API upgrades, RPC version upgrades ..etc | 12:35 |
| hrw | amorin: do work on stein, compare with master, check which parts can be done by backporting from stein+ | 12:36 |
| mnasiadka | kevko: ceph killer | 12:38 |
| kevko | ? | 12:39 |
| kevko | did you get my idea right ? :D | 12:39 |
| mnasiadka | no, I didn't - you said you had a question, but didn't really ask the question :) | 12:45 |
| mnasiadka | ah, you did, a couple lines later | 12:45 |
| mnasiadka | so, cinder-volume will be down for 25 minutes - is that a problem? | 12:45 |
| opendevreview | Dr. Jens Harbott proposed openstack/kolla master: Fix incorrect type conversion for cmd option '--docker-dir' https://review.opendev.org/c/openstack/kolla/+/656603 | 12:45 |
| opendevreview | Marcin Juszkiewicz proposed openstack/kolla-ansible master: Update RHEL family information https://review.opendev.org/c/openstack/kolla-ansible/+/860075 | 12:46 |
| hrw | rebased to pass linters | 12:46 |
| mnasiadka | frickler: do we actually need it? ;-) | 12:46 |
| opendevreview | Dr. Jens Harbott proposed openstack/kolla master: Fix incorrect type conversion for cmd option '--docker-dir' https://review.opendev.org/c/openstack/kolla/+/656603 | 12:47 |
| kevko | mnasiadka: well, it is problem ...because of 700+ kubernetes pernamently wants from cinder something | 12:48 |
| mnasiadka | ah, that type of problem | 12:48 |
| kevko | yeah :D | 12:48 |
| kevko | and it is big issue if it is not working ... | 12:48 |
| mnasiadka | so, upgrade cinder-volume one by one? | 12:49 |
| kevko | in kolla-ansible there should be a mechanism to wait for at least one cinder-volume in cluster for backend to be up | 12:49 |
| mnasiadka | yes, that's what I meant | 12:49 |
| opendevreview | Marcin Juszkiewicz proposed openstack/kolla-ansible master: Update RHEL family information https://review.opendev.org/c/openstack/kolla-ansible/+/860075 | 12:50 |
| opendevreview | Marcin Juszkiewicz proposed openstack/kolla-ansible master: Default to Rocky Linux instead of CentOS Stream https://review.opendev.org/c/openstack/kolla-ansible/+/860692 | 12:50 |
| kevko | mnasiadka: yes, but upgrade will also upgrade DB ..so | 12:50 |
| frickler | mnasiadka: I'm a big fan of considering warnings fatal, so yes, this should be fixed | 12:50 |
| mnasiadka | frickler: I haven't seen this warning, but maybe I haven't been paying attention to tox output :) | 12:50 |
| frickler | mnasiadka: I verified the bug still applies to master, yes | 12:51 |
| opendevreview | Dr. Jens Harbott proposed openstack/kolla master: Fix incorrect type conversion for cmd option '--docker-dir' https://review.opendev.org/c/openstack/kolla/+/656603 | 12:59 |
| opendevreview | wu.chunyang proposed openstack/kolla-ansible master: Switch trove-api to wsgi running under apache. https://review.opendev.org/c/openstack/kolla-ansible/+/854759 | 12:59 |
| opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: openvswitch: switch bridge setup to Ansible modules https://review.opendev.org/c/openstack/kolla-ansible/+/843460 | 13:00 |
| opendevreview | Maksim Malchuk proposed openstack/kolla-ansible master: Fix freezer deployment https://review.opendev.org/c/openstack/kolla-ansible/+/861756 | 13:15 |
| opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: ovn: Break out role into ovn-db and ovn-controller roles https://review.opendev.org/c/openstack/kolla-ansible/+/847769 | 13:18 |
| opendevreview | wu.chunyang proposed openstack/kolla-ansible master: Switch trove-api to wsgi running under apache. https://review.opendev.org/c/openstack/kolla-ansible/+/854759 | 13:22 |
| opendevreview | Marcin Juszkiewicz proposed openstack/kolla master: rabbitmq/aarch64: use COPR for Erlang https://review.opendev.org/c/openstack/kolla/+/861327 | 13:23 |
| hrw | s/tabtab/spacesspaces/ | 13:23 |
| opendevreview | Merged openstack/kolla master: Bump node_exporter to 1.4.0 https://review.opendev.org/c/openstack/kolla/+/861355 | 13:25 |
| kevko | mnasiadka: ok, i probably found a solution regarding cinder volumes | 13:51 |
| kevko | mnasiadka: and kolla SHOULD provide configurable option to this variable | 13:52 |
| kevko | mnasiadka: # Set to False if the pool is shared with other usages. On exclusive use driver | 13:52 |
| kevko | # won't query images' provisioned size as they will match the value calculated | 13:52 |
| kevko | # by the Cinder core code for allocated_capacity_gb. This reduces the load on | 13:52 |
| kevko | # the Ceph cluster as well as on the volume service. On non exclusive use | 13:52 |
| kevko | # driver will query the Ceph cluster for per image used disk, this is an | 13:52 |
| kevko | # intensive operation having an independent request for each image. (boolean | 13:52 |
| kevko | # value) | 13:52 |
| kevko | #rbd_exclusive_cinder_pool = true | 13:52 |
| kevko | mnasiadka: https://github.com/openstack/cinder/blob/ad184ca698d4b9cce912fcc1e936ef02c44b47d9/cinder/volume/drivers/rbd.py#L773-L777 | 13:53 |
| kevko | we have tens tousands ceph volumes ..and cinder volume is starting for 25 minutes only because of default cinder value true ...this should be kolla configurable | 13:53 |
| opendevreview | Merged openstack/kolla master: rabbitmq: use Erlang from PPA on Debian/Ubuntu https://review.opendev.org/c/openstack/kolla/+/861096 | 13:57 |
| opendevreview | Merged openstack/kolla master: release-docs: Add infra components bump to R-5 https://review.opendev.org/c/openstack/kolla/+/861745 | 13:57 |
| mnasiadka | kevko: nice, add it to external ceph guide with a variable - I'm all in :) | 14:00 |
| kevko | mnasiadka: hmm, but it looks like default is good (but it was changed from version to version - so anyway kolla should set this or document it ) ...because conditional is negation | 14:02 |
| kevko | conditional in cinder | 14:03 |
| kevko | mnasiadka: ok, kolla should definitively set this option ... because default was changed during releases and it was also backported | 14:11 |
| kevko | I32c7746fa9149bce6cdec96ee9aa87b303de4271 | 14:11 |
| opendevreview | Marcin Juszkiewicz proposed openstack/kolla stable/yoga: rabbitmq: use Erlang from PPA on Debian/Ubuntu https://review.opendev.org/c/openstack/kolla/+/861762 | 14:13 |
| opendevreview | Marcin Juszkiewicz proposed openstack/kolla stable/yoga: rabbitmq: use Erlang from PPA on Debian/Ubuntu https://review.opendev.org/c/openstack/kolla/+/861762 | 14:19 |
| mnasiadka | kevko: should help in your case | 14:19 |
| opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: openvswitch: switch bridge setup to Ansible modules https://review.opendev.org/c/openstack/kolla-ansible/+/843460 | 14:23 |
| mnasiadka | hrw: https://review.opendev.org/c/openstack/kolla-ansible/+/860075 - planning to update per Radek's comments, or should I? | 14:27 |
| atmark | why is there no enable_horizon_barbican? | 14:35 |
| atmark | nvm, i found the answer. looks like barbican-ui is abandoned | 14:36 |
| wuchunyang | Hi, everyone, I find some project's keystone endpoint was defined in all.yml, others was defined in default/main.yml. which one should be recommend? | 14:49 |
| mnasiadka | wuchunyang: depends if multiple roles need it, or just one | 14:57 |
| wuchunyang | mnasiadka: But it doesn't seem to be that way. e.g. octavia_public_endpoint, this is only needed by octavia role. | 15:03 |
| wuchunyang | I was working on trove role, and I am not sure whether i should remove the endpoint to all.yml. | 15:04 |
| wuchunyang | If this does not affect anything, i will keep it there. | 15:07 |
| hrw | mnasiadka: https://review.opendev.org/c/openstack/kolla-ansible/+/860692 does that | 15:42 |
| * hrw out | 15:42 | |
| mnasiadka | hrw: ah, thanks | 15:46 |
| opendevreview | Merged openstack/kolla master: rabbitmq/aarch64: use COPR for Erlang https://review.opendev.org/c/openstack/kolla/+/861327 | 15:53 |
| mnasiadka | Meeting in 5 minutes for Kolla Operator Hour is here: https://meetpad.opendev.org/kolla-antelope-ptg | 15:55 |
| frickler | mnasiadka: if we use the other etherpad then this is confusing | 15:59 |
| mnasiadka | well, it was precreated - and I was asked to use the precreated one - what can I tell | 15:59 |
| opendevreview | Michal Arbet proposed openstack/kolla-ansible master: RBD: Don't query Ceph on stats for exclusive pools https://review.opendev.org/c/openstack/kolla-ansible/+/861774 | 16:00 |
| kevko | mnasiadka: https://review.opendev.org/c/openstack/kolla-ansible/+/861774 | 16:01 |
| mnasiadka | let me copy the content to kolla-antelope-ptg | 16:01 |
| opendevreview | Ebbex proposed openstack/kayobe master: Use dnf module to get rid of warning https://review.opendev.org/c/openstack/kayobe/+/861779 | 16:43 |
| dardelean | what is redis used for in kolla? | 17:00 |
| dardelean | as support for what? | 17:00 |
| kevko | coordination url for cinder ha | 17:09 |
| kevko | or designate cooridenation | 17:09 |
| kevko | or osprofiler | 17:10 |
| kevko | dardelean: ^ | 17:10 |
| dardelean | eye see, thanks kekvo | 17:10 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!