| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-rabbitmq_server master: Move RabbitMQ restart to handlers https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/931974 | 07:04 |
|---|---|---|
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-rabbitmq_server master: Move RabbitMQ restart to handlers https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/931974 | 07:04 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-rabbitmq_server master: Use rabbitmq_cluster_state task instead of command https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/931975 | 07:05 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-rabbitmq_server master: Move RabbitMQ restart to handlers https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/931974 | 07:18 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Bump ansible version to 2.17.5 https://review.opendev.org/c/openstack/openstack-ansible/+/932067 | 13:08 |
| jrosser | ahha this is running on a noble node now - thats why it breaks https://zuul.opendev.org/t/openstack/build/a52393e39f3c4cfe815ea3f1453b8b1b | 13:25 |
| noonedeadpunk | oh | 13:37 |
| noonedeadpunk | good catch | 13:37 |
| noonedeadpunk | probably we should pin back to jammy explicitly? | 13:37 |
| noonedeadpunk | would make sense for stable | 13:37 |
| jrosser | sounds good | 13:56 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Ensure that the inventory tox job runs on an ubuntu-jammy node https://review.opendev.org/c/openstack/openstack-ansible/+/932077 | 14:17 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/2023.2: Use jammy for tox jobs https://review.opendev.org/c/openstack/openstack-ansible/+/932079 | 14:21 |
| noonedeadpunk | jrosser: I think this affecting only 2023.2 and 2023.1 though | 14:22 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/2023.1: Ensure that the inventory tox job runs on an ubuntu-jammy node https://review.opendev.org/c/openstack/openstack-ansible/+/932080 | 14:22 |
| jrosser | do you think the one on master should fail? | 14:23 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/2023.2: Ensure that the inventory tox job runs on an ubuntu-jammy node https://review.opendev.org/c/openstack/openstack-ansible/+/932082 | 14:23 |
| jrosser | i was going to make a follow up to bump all the things on master to noble | 14:23 |
| noonedeadpunk | nah, but it should not be needed either. | 14:23 |
| noonedeadpunk | as it should be fine to run on noble | 14:24 |
| noonedeadpunk | ah | 14:24 |
| jrosser | well, we do have a mix of how things are set in the jobs definition right now | 14:24 |
| jrosser | most jobs define the node type, it was just the inventory one that didnt | 14:24 |
| noonedeadpunk | ok, yeah, then makes sense | 14:24 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/2023.1: Remove upgrade jobs for 2023.1 https://review.opendev.org/c/openstack/openstack-ansible/+/931813 | 14:25 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/2023.1: Bump SHAs for 2023.1 https://review.opendev.org/c/openstack/openstack-ansible/+/931742 | 14:25 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/2023.2: Bump SHAs for 2023.2 https://review.opendev.org/c/openstack/openstack-ansible/+/931741 | 14:26 |
| luvn | hey, just a quick update on the issue i had yesterday | 14:27 |
| luvn | it's related to ceph thin volumes (ceph1.img, ceph2.img, etc...) | 14:27 |
| noonedeadpunk | as they just grow but never get discarded? | 14:27 |
| noonedeadpunk | (ie fstrimm) | 14:28 |
| luvn | yeah, that's right | 14:28 |
| noonedeadpunk | iirc trimming works only if you use virtio_scsi | 14:28 |
| luvn | the problem is that when it reaches this point, the whole openstack is unresponsive | 14:28 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Move all anciallary zuul jobs to ubuntu noble https://review.opendev.org/c/openstack/openstack-ansible/+/932084 | 14:29 |
| luvn | how can i change ceph's replication factor in an osa deployment? can i do that after osa is deployed? | 14:31 |
| noonedeadpunk | so you can set following properties for your images - `hw_disk_bus=scsi hw_scsi_model=virtio-scsi` | 14:31 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Move all ancilliary zuul jobs to ubuntu noble https://review.opendev.org/c/openstack/openstack-ansible/+/932084 | 14:31 |
| noonedeadpunk | we're managing ceph through ceph-ansible project | 14:31 |
| noonedeadpunk | or well, we're leveraging the project | 14:31 |
| jrosser | maybe we should set those image properties in the AIO by default? | 14:31 |
| noonedeadpunk | might make sense... | 14:33 |
| noonedeadpunk | though I'm still avoiding that in production, except magnum images | 14:33 |
| noonedeadpunk | (where you might need more then 24 volumes attached) | 14:34 |
| jrosser | i think we switched to it generally for something to do with rescue iirc | 14:34 |
| luvn | with this property set, the idea is to trim these images every time i reach 100% use%? | 14:35 |
| jrosser | luvn: to understand how OSA interacts with ceph-ansible, and how the pools are set up, it would be worth looking at this https://github.com/openstack/openstack-ansible/commit/120d6be99290259c560efcd46a8c1eefb4a55a2b | 14:38 |
| jrosser | that has the often repeated caveat that it's only a test fixture :) | 14:38 |
| jrosser | but also shows you where the pools are defined and how the `size` could be overridden to the `ceph_pool` ansible module | 14:39 |
| jrosser | in general all the hooks are there for you to override everything, but most of the time you wont find a specific variable for every single thing, just like there isnt for the ceph pool replication | 14:40 |
| jrosser | but it is totally possible to customise that through your user_variables/group_vars | 14:40 |
| luvn | that's interesting! i'll definitely check out this commit | 14:44 |
| jrosser | key thing is we just wrap ceph-ansible | 14:45 |
| luvn | yeah, i don't intend to use this setup in production, i'm just trying to benchmark some volume stuff | 14:45 |
| jrosser | so "how to change $ceph-thing in OSA" usual == "how to change ceph thing in ceph-ansible" | 14:45 |
| luvn | makes sense! i'll try to customize ceph's deployment by using their variables then | 14:46 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-rabbitmq_server master: Use rabbitmq_cluster_state task instead of command https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/931975 | 14:50 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-rabbitmq_server master: Move RabbitMQ restart to handlers https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/931974 | 14:55 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Move all ancilliary zuul jobs to ubuntu noble https://review.opendev.org/c/openstack/openstack-ansible/+/932084 | 15:10 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-rabbitmq_server master: Use rabbitmq_cluster_state task instead of command https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/931975 | 15:13 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!