| hamburgler | Does anyone know of octavia is ready to be deployed with ovn and openstack-ansible? | 02:18 |
|---|---|---|
| jrosser | good morning | 08:22 |
| noonedeadpunk | o/ | 08:25 |
| noonedeadpunk | hamburgler: in terms of using ovn driver - I don't think it is. we\re having WIP patch under review now, prometheanfire likely used it in their env | 08:26 |
| noonedeadpunk | but it might need some polishing | 08:26 |
| noonedeadpunk | not sure how good this morning is though... Hopefully better then yestardays evening for sure :D | 08:52 |
| noonedeadpunk | seems we have some green results of https://review.opendev.org/q/topic:osa%252Frdo_ovs_3.1 | 08:53 |
| noonedeadpunk | I think neutron can be recheked once plugins repo will merge | 08:53 |
| admin1 | \o | 08:54 |
| admin1 | hamburgler, i have a working dev platform with ovn + osa + octavia + magnum | 08:55 |
| admin1 | so far have not encountered any issues | 08:55 |
| opendevreview | Damian Dąbrowski proposed openstack/openstack-ansible master: Implement separated haproxy service config https://review.opendev.org/c/openstack/openstack-ansible/+/871189 | 09:11 |
| opendevreview | Damian Dąbrowski proposed openstack/openstack-ansible master: Update letsencrypt docs https://review.opendev.org/c/openstack/openstack-ansible/+/878939 | 09:12 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible-openstack_hosts master: Bump UCA release to Zed for ubuntu focal https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/879033 | 09:52 |
| jrosser | ^ oops - was that meant to stay at yoga? | 09:52 |
| noonedeadpunk | yup, there's no zed for focal | 09:53 |
| admin1 | is it safe to do-release-upgrade and reboot ? | 09:53 |
| jrosser | aaaahhhhh | 09:54 |
| noonedeadpunk | usually it is :D | 09:54 |
| admin1 | ok | 09:54 |
| noonedeadpunk | I have long-standing debate with myself if there's any good reason to do that though | 09:54 |
| admin1 | "yup, there's no zed for focal" -- | 09:54 |
| admin1 | for that ... | 09:55 |
| noonedeadpunk | I meant specifically UCA repo in that case | 09:55 |
| noonedeadpunk | But eventually it's usually faster to re-setup | 09:55 |
| admin1 | noonedeadpunk, in terms of your ovn remark that its not ready wtih octavia .. i am running my dev env on that .. and it works fine .. should I be worried that it works :D ? | 09:55 |
| noonedeadpunk | In cases with controllers - you would need to re-create all containers sooner or later | 09:55 |
| jrosser | ok, so then for UCA can we now bump this to Antelope? https://opendev.org/openstack/openstack-ansible-openstack_hosts/src/branch/master/defaults/main.yml#L22 | 09:56 |
| noonedeadpunk | For computes - you still evacuate all VMs before doing upgrade. So given you have some OS provisioning stuff - it's jsut faster to re-setup | 09:56 |
| noonedeadpunk | jrosser: yeah, well. We should. I'd wait for RDO though | 09:57 |
| jrosser | admin1: did you manage to add your compute node yesterday - noonedeadpunk said that the error you see happens after the compute is deployed? | 09:57 |
| admin1 | i ran it multiple times and i i see the compute service and neutron working ( i use compute also as network ) , so it added fine | 09:57 |
| anskiy | noonedeadpunk: wait, I was thinking that Z was the release, when you can have focal and jammy, isn't it so? | 09:58 |
| admin1 | just that the playbooks exit and does not end up in a success state | 09:58 |
| jrosser | anskiy: there is no release of the ubuntu cloud archive repo for Zed, on focal | 09:58 |
| admin1 | i ran the whole setup hosts followed by setup infra and then setup-nova and setup-neutron | 09:58 |
| noonedeadpunk | anskiy: yep you can. We're using UCA repo for distro installs and using it for "fresher" libvirt, ovs and qemu for source installs | 09:58 |
| admin1 | the playbooks end us broken, but does it job .. | 09:58 |
| admin1 | ends up* | 09:59 |
| noonedeadpunk | So we still need it, but using Yoga version on Zed and antelope. | 09:59 |
| anskiy | ah, I see. So that's possible, thank you | 09:59 |
| noonedeadpunk | provided versions still satisfy minimum versions for these releases | 09:59 |
| jrosser | i saw this too which we might need http://ubuntu-cloud.archive.canonical.com/ubuntu/dists/focal-updates/ovn-22.03/ | 09:59 |
| noonedeadpunk | But for distro path Zed is marked as unsupported for 20.04 | 09:59 |
| noonedeadpunk | So basically if you will attempt - it won't fail, but wil ljsut install Yoga | 10:00 |
| anskiy | noonedeadpunk: nah, I'm on source | 10:01 |
| noonedeadpunk | jrosser: hm, we're not enabling it right now, are we? | 10:01 |
| jrosser | no, we're not | 10:01 |
| jrosser | idk if newer version on focal becomes a requirement at some point | 10:01 |
| * jrosser not familiar with required versions of OVN | 10:02 | |
| noonedeadpunk | 22.03 looks like smth old to be frank... | 10:04 |
| noonedeadpunk | looking at http://ubuntu-cloud.archive.canonical.com/ubuntu/dists/focal-updates/yoga/main/binary-amd64/Packages and it also seems to contain same 22.03 | 10:05 |
| noonedeadpunk | at least for Yoga. So likely we don't need that | 10:06 |
| opendevreview | Merged openstack/openstack-ansible-plugins master: Ensure systemd-udev is installed for gluster https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/878803 | 10:16 |
| noonedeadpunk | wait, wf | 11:14 |
| noonedeadpunk | Looks like ^ should not be needed at all | 11:15 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-plugins master: Revert "Ensure systemd-udev is installed for gluster" https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/878842 | 11:17 |
| jrosser | oh wow it doesnt even do what the commit message says :( | 11:17 |
| jrosser | sorry to not spot that, i confused it with the other one that *actually* does install it | 11:17 |
| noonedeadpunk | So most confusing thing, is that I've triggered recheck for neutron after this one landed. And it's still not applied | 11:18 |
| noonedeadpunk | https://zuul.opendev.org/t/openstack/build/e427299af98245138927ad0781a6f405 | 11:18 |
| noonedeadpunk | BUT, I really struggle to understand how it doesn't work as is | 11:18 |
| noonedeadpunk | https://zuul.opendev.org/t/openstack/build/e427299af98245138927ad0781a6f405/log/logs/openstack/aio1_repo_container-3fda0517/dnf.rpm.log.txt | 11:19 |
| noonedeadpunk | systemd-udev is just being ignored... Or it's "already" installed but smth deeper is off | 11:19 |
| jrosser | err what is going on here | 11:20 |
| jrosser | what is this for? https://review.opendev.org/c/openstack/ansible-role-systemd_mount/+/878851/2 | 11:20 |
| noonedeadpunk | yeah, it probably won't help with what we've faced in centos. I need really fresh image I guess first... | 11:20 |
| jrosser | but systemd_mount feels like kind of when glusterfs server is already set up "mount it here" | 11:21 |
| jrosser | and in the client place (which is co-inidentally also the server for repo hosts) | 11:21 |
| noonedeadpunk | Yeah, so it might be indeed needed there, but it's not systemd-udev that provides the service in topic I assume | 11:23 |
| noonedeadpunk | anymore | 11:23 |
| noonedeadpunk | as it's obviously would be installed as usual, but there's no trace if installation, which means that it's still present in image | 11:25 |
| noonedeadpunk | I just checked older centos image and systemd-udev indeed does provide `systemd-tmpfiles-setup-dev`. I'm uploading latest now to check what has changed | 11:27 |
| noonedeadpunk | spotz: hey! regarding us being picky about CentOS. I dunno where to report that, maybe it's intended, but I've downloaded latest centos image (https://cloud.centos.org/centos/9-stream/x86_64/images/CentOS-Stream-GenericCloud-9-20230313.0.x86_64.qcow2) and it does not have openssh enabled. So I see that cloud-init works, but nothing responds on 22 port. | 12:59 |
| noonedeadpunk | No idea if that's how CentOS seeing things should be though or it's a bug... I'm getting confused between these 2 lately | 12:59 |
| noonedeadpunk | systemctl enable sshd does the trick. If you have a root password for console... | 13:04 |
| * jrosser reads 4th gen xeon product guide, what is all this wierdness.... | 14:14 | |
| jrosser | high / low priority cores blah blah omg | 14:14 |
| noonedeadpunk | trying to compete with cortex or what? | 14:18 |
| jrosser | wanting to benchmark some pcie5 and ddr5 | 14:20 |
| jrosser | very quickly starting to look at amd 9004 stuff instead | 14:20 |
| noonedeadpunk | Eventually I'd love just to switch to AMDs. And we even got agreement for that. But mixing up deployments of amd/intel is kinda messy | 14:29 |
| noonedeadpunk | maybe it's not _that_ messy after all... | 14:29 |
| prometheanfire | hamburgler: I've tried octavia with ovn, but traffic wasn't passing, this could have been because of how not normal the env is (it needs to be rebuilt) | 14:32 |
| noonedeadpunk | prometheanfire: fwiw, admin1 has it working apparently using your patch? | 14:34 |
| NeilHanlon | noonedeadpunk, jrosser: sorry for the bad patch | 14:36 |
| jrosser | oh its ok don't worry | 14:37 |
| noonedeadpunk | Nah, it was good | 14:37 |
| jrosser | i had my brain scrambled yesterday and it was all a it of a blur | 14:37 |
| NeilHanlon | also, and unrelatedly, Rocky 8.8 beta is out so I am going to probably mess around with that at some point to prepare :) | 14:37 |
| NeilHanlon | err. 9.2, rather | 14:37 |
| * NeilHanlon has too many things | 14:37 | |
| prometheanfire | noonedeadpunk: ya, I saw that, so that's encouraging | 14:37 |
| noonedeadpunk | can anybody explain to me wtf is that https://paste.openstack.org/show/b8u9QH5L05jXJJ3amNjI/ ? | 16:03 |
| noonedeadpunk | NeilHanlon: ? :D | 16:03 |
| opendevreview | Merged openstack/openstack-ansible master: Add documentation for multi-architecture deployments. https://review.opendev.org/c/openstack/openstack-ansible/+/877954 | 16:04 |
| noonedeadpunk | Just adding on top of my frustration https://paste.openstack.org/show/bo2AaMplXu1Y8riIrWXP/ | 16:36 |
| noonedeadpunk | Can anybody teach me on how to use dnf to install stuff I'm asking it to install? | 16:36 |
| NeilHanlon | wtf | 16:39 |
| NeilHanlon | i... think I know what that is related to.. | 16:40 |
| noonedeadpunk | I'm about to start googling "how to install package with dnf" at this point | 16:42 |
| noonedeadpunk | oh, that's interesting https://bugzilla.redhat.com/show_bug.cgi?id=2051019 | 16:45 |
| NeilHanlon | where did we get systemd-boot-unsigned from ? | 16:46 |
| noonedeadpunk | @System | 16:49 |
| noonedeadpunk | https://paste.openstack.org/show/bZtEwJ9njqNCANHfnb1u/ | 16:51 |
| noonedeadpunk | So `https://mirrors.centos.org/metalink?repo=centos-crb-$stream&arch=$basearch` | 16:52 |
| opendevreview | Merged openstack/openstack-ansible-openstack_hosts master: Pin openvswitch package on RHEL to 2.17 https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/878926 | 16:57 |
| NeilHanlon | noonedeadpunk: for now... let's try and exclude systemd-boot-unsigned from the `crb` repo? I'm opening a bug referencing the one you found | 16:59 |
| NeilHanlon | something in the depsolver has marked that systemd-boot-unsigned provides systemd-udev... and that's clearly wrong | 16:59 |
| noonedeadpunk | dohhhhh | 17:01 |
| noonedeadpunk | I'm about to just set centos lxc to NV | 17:10 |
| NeilHanlon | I wouldn't blame you. | 17:10 |
| noonedeadpunk | given they're have "fix" somewhere already | 17:10 |
| * prometheanfire is using rocky-9-lxc, not so fun | 17:19 | |
| prometheanfire | it seems like people are moving away from lxc in general? | 17:23 |
| mgariepy | not me. | 17:25 |
| noonedeadpunk | neither me | 17:26 |
| prometheanfire | ok | 17:26 |
| noonedeadpunk | prometheanfire: also rocky is not affected | 17:26 |
| noonedeadpunk | only centos 9 stream is | 17:26 |
| prometheanfire | cool, rocky did get hit by the ovs-3.1 thing, that's fun | 17:27 |
| noonedeadpunk | yeah.... | 17:30 |
| noonedeadpunk | but it's kinda not rocky failt, as this is coming from RDO... | 17:31 |
| prometheanfire | yep | 17:33 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Disable CentOS LXC jobs due to the bug in systemd packaging https://review.opendev.org/c/openstack/openstack-ansible/+/879069 | 17:43 |
| noonedeadpunk | jrosser: I think it's still worth having https://review.opendev.org/c/openstack/ansible-role-systemd_mount/+/878851 to be frank | 17:44 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_neutron master: Disable CentOS LXC jobs due to the bug in systemd packaging https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/879070 | 17:47 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_neutron master: Workaround ovs bug that resets hostname with add command https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/878911 | 17:47 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_neutron master: Workaround ovs bug that resets hostname with add command https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/878911 | 17:48 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-plugins master: Revert "Ensure systemd-udev is installed for gluster" https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/878842 | 17:50 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Disable CentOS LXC jobs due to the bug in systemd packaging https://review.opendev.org/c/openstack/openstack-ansible/+/879069 | 18:19 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Define individual haproxy services for each console type https://review.opendev.org/c/openstack/openstack-ansible/+/878083 | 18:25 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-ceph_client master: Add EPEL GPG key for RHEL 9 https://review.opendev.org/c/openstack/openstack-ansible-ceph_client/+/879081 | 19:02 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-ceph_client master: Add thrift to includepkgs from EPEL https://review.opendev.org/c/openstack/openstack-ansible-ceph_client/+/879082 | 19:05 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-ceph_client master: Unify EPEL gpg key and repo provisioning https://review.opendev.org/c/openstack/openstack-ansible-ceph_client/+/879084 | 19:16 |
| opendevreview | Damian Dąbrowski proposed openstack/openstack-ansible master: Add support for TLS backends https://review.opendev.org/c/openstack/openstack-ansible/+/879085 | 19:52 |
| opendevreview | Merged openstack/ansible-role-systemd_mount master: Ensure systemd-udev is installed for gluster https://review.opendev.org/c/openstack/ansible-role-systemd_mount/+/878851 | 20:09 |
| jrosser | damiandabrowski: could you take a quick look at https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/879070/1 | 20:15 |
| damiandabrowski | sure, done ;) | 20:23 |
| opendevreview | Merged openstack/openstack-ansible-os_neutron master: Disable CentOS LXC jobs due to the bug in systemd packaging https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/879070 | 22:36 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!