| opendevreview | OpenStack Proposal Bot proposed openstack/openstack-ansible master: Imported Translations from Zanata https://review.opendev.org/c/openstack/openstack-ansible/+/967794 | 03:15 |
|---|---|---|
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible master: Add minimal AIO files for k8s clusters https://review.opendev.org/c/openstack/openstack-ansible/+/966666 | 08:25 |
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible master: ci: add nova console log collection and fix repo file collection https://review.opendev.org/c/openstack/openstack-ansible/+/967961 | 08:25 |
| opendevreview | Merged openstack/openstack-ansible-rabbitmq_server master: Switch erlang repo to deb1.rabbitmq.com https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/967523 | 11:28 |
| andrewbonney | noonedeadpunk: I've just been debugging why nova VMs in an AIO don't have working Internet connectivity. I think this is down to https://github.com/openstack/openstack-ansible/commit/31f12f54eaccd3ff5a458cb8593897a46b2d923a but I'd appreciate a sanity check | 11:33 |
| andrewbonney | Comparing a master and 2025.1 AIO, 2025.1 has a nat postrouting iptables rule, but master doesn't. It appears it's maybe getting applied to the wrong interface in the aio networks template | 11:34 |
| noonedeadpunk[e] | I think there should be a Nat rule allowing access, and a network IP assigned on fake vlan | 11:34 |
| noonedeadpunk[e] | Magnum needs an internet access I guess? | 11:36 |
| andrewbonney | Yeah that's right | 11:36 |
| andrewbonney | I seem to have to apply the missing postrouting rule to the management interface rather than the public one to make it work, and that doesn't sound quite right | 11:38 |
| noonedeadpunk | hm, it does not for sure.... | 11:45 |
| noonedeadpunk | I could totally introduce some bug in there.... | 11:45 |
| noonedeadpunk | I'd need to spawn some fresh env to check on that... | 11:46 |
| andrewbonney | Sorry, I'm wrong there, using the public interface is fine, but the rule didn't seem to apply | 11:50 |
| andrewbonney | I'll keep digging anyway | 11:51 |
| noonedeadpunk | I was thinking that adding a gateway IP on the vlan was also needed | 11:56 |
| andrewbonney | Could it be to do with the trigger? I see the command in networking-post-up.service, but that doesn't seem to be triggered when (for example) restarting systemd-networkd | 11:58 |
| noonedeadpunk | ah | 12:01 |
| noonedeadpunk | indet is wrong https://opendev.org/openstack/openstack-ansible/src/branch/master/tests/roles/bootstrap-host/templates/aio1_networks.yml.j2#L167-L170 | 12:02 |
| noonedeadpunk | andrewbonney: it should be https://paste.openstack.org/show/bO9wBYSMFIL6Nik9A7ro/ | 12:03 |
| andrewbonney | Oh | 12:03 |
| noonedeadpunk | as it ends up adding DEFAULT section and fail to load the unit at all | 12:04 |
| noonedeadpunk | `networking-post-up.service:1: Unknown section 'DEFAULT'. Ignoring` | 12:04 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/2024.1: Bump rabbitmq SHA and stop using infra mirror for mariadb https://review.opendev.org/c/openstack/openstack-ansible/+/967864 | 12:38 |
| noonedeadpunk | andrewbonney: should I propose the patch or you on it? | 12:57 |
| andrewbonney | I'm happy to do it - just been on lunch | 12:58 |
| noonedeadpunk | ah, ok :D | 12:58 |
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible master: Fix indentation for aio networking-post-up service https://review.opendev.org/c/openstack/openstack-ansible/+/967986 | 13:01 |
| noonedeadpunk | andrewbonney: huh.... I think this is another mistake? | 13:06 |
| noonedeadpunk | ie `- "-{{ bootstrap_host_ip_path }} link set eth13 up"` | 13:06 |
| noonedeadpunk | why 2 `-`? | 13:06 |
| andrewbonney | I did wonder, but that seemed to be in the original code pre-change | 13:07 |
| noonedeadpunk | well... systemd seems to be able to handle it indeed.... | 13:08 |
| noonedeadpunk | and remove this second - | 13:08 |
| andrewbonney | I can remove if there's no reason for it though | 13:09 |
| noonedeadpunk | I don;t think there's any? | 13:09 |
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible master: Fix indentation for aio networking-post-up service https://review.opendev.org/c/openstack/openstack-ansible/+/967986 | 13:14 |
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible-ops master: Update mcapi to test against latest versions https://review.opendev.org/c/openstack/openstack-ansible-ops/+/966563 | 13:21 |
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible-ops master: mcapi_proxy: notify handler on k8s config change https://review.opendev.org/c/openstack/openstack-ansible-ops/+/968015 | 13:21 |
| opendevreview | Damian Dąbrowski proposed openstack/ansible-role-pki master: Add hashi_vault backend https://review.opendev.org/c/openstack/ansible-role-pki/+/948881 | 13:24 |
| opendevreview | Damian Dąbrowski proposed openstack/openstack-ansible master: Add support for hashi_vault PKI backend https://review.opendev.org/c/openstack/openstack-ansible/+/948888 | 13:24 |
| noonedeadpunk | damiandabrowski: sooo. what are we going to do with https://review.opendev.org/c/openstack/openstack-ansible/+/948886/8 ? | 13:46 |
| damiandabrowski | ahh sorry, I missed this comment | 13:48 |
| damiandabrowski | well, this patch just changes the format of this variable | 13:49 |
| damiandabrowski | adding ansible_facts['fqdn'] seems like a good idea, but it's probably something for a separate patch | 13:49 |
| noonedeadpunk | so the point is, that alternative could be to drop this var instead of changing its format | 13:50 |
| noonedeadpunk | As now it's not different from in-role defaults | 13:50 |
| noonedeadpunk | So just changing the format does not make much sense to me | 13:51 |
| damiandabrowski | is it really different from in-role defaults? | 13:53 |
| noonedeadpunk | now it's exactly the same. so why to have it? | 13:54 |
| damiandabrowski | ahh, "not different" sorry | 13:54 |
| damiandabrowski | well, to let customers override just `openstack_pki_san` instead of `<service>_pki_san` for all services | 13:55 |
| damiandabrowski | that's how I see it | 13:55 |
| noonedeadpunk | so what prevents them to do so, if it's not defined? | 13:56 |
| noonedeadpunk | in integrated repo | 13:56 |
| damiandabrowski | ahhh, that's what you mean. I thought that we're talking about completely dropping a support for openstack_pki_san | 13:57 |
| noonedeadpunk | I was talking about this specific patch :) | 13:57 |
| damiandabrowski | i think it makes sense then | 13:57 |
| noonedeadpunk | As imo it should either do some smart thing, or remove smth we don't need | 13:58 |
| noonedeadpunk | if it's actually the same, and then it makes zookeeper not to use this var | 13:58 |
| noonedeadpunk | (https://review.opendev.org/c/openstack/ansible-role-zookeeper/+/949422/3/defaults/main.yml) | 13:59 |
| noonedeadpunk | and same for ovn certs | 13:59 |
| noonedeadpunk | so if we can't or don't want to do smart, I think we still should use `openstack_pki_san` and let users to implement smart things if they want to.... but document a note that there're cases like this, they need to think about | 14:00 |
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible master: ci: add nova console log collection and fix repo file collection https://review.opendev.org/c/openstack/openstack-ansible/+/967961 | 14:12 |
| opendevreview | Andrew Bonney proposed openstack/openstack-ansible master: Add minimal AIO files for k8s clusters https://review.opendev.org/c/openstack/openstack-ansible/+/966666 | 14:12 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/ansible-role-systemd_networkd master: Restart systemd-udev on link changes https://review.opendev.org/c/openstack/ansible-role-systemd_networkd/+/954876 | 15:15 |
| damiandabrowski | ack, I'll work on that | 15:47 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/ansible-role-systemd_networkd master: Add Debian 13 support https://review.opendev.org/c/openstack/ansible-role-systemd_networkd/+/968034 | 16:04 |
| opendevreview | Ivan Anfimov proposed openstack/openstack-ansible master: docs: Installing with limited connectivity - updates URLs for mirrors https://review.opendev.org/c/openstack/openstack-ansible/+/967921 | 17:26 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!