| *** johanssone has quit IRC | 00:16 | |
| *** tosky has quit IRC | 00:17 | |
| *** johanssone has joined #openstack-ansible | 00:19 | |
| *** manti has quit IRC | 00:19 | |
| *** manti has joined #openstack-ansible | 00:23 | |
| jamesdenton | jrosser Looks like the ODL role for Ansible hasn't been updated in > 18 months: https://github.com/opendaylight/integration-packaging-ansible-opendaylight. I'm not gonna spend much more time on it and maybe we should consider next steps | 00:30 |
|---|---|---|
| *** macz_ has quit IRC | 00:49 | |
| *** spatel has joined #openstack-ansible | 00:52 | |
| *** spatel has quit IRC | 00:57 | |
| *** spatel has joined #openstack-ansible | 01:07 | |
| *** Underknowledge has quit IRC | 01:08 | |
| *** Underknowledge has joined #openstack-ansible | 01:08 | |
| *** gyee has quit IRC | 01:17 | |
| *** Underknowledge2 has joined #openstack-ansible | 01:51 | |
| *** Underknowledge has quit IRC | 01:54 | |
| *** Underknowledge2 is now known as Underknowledge | 01:54 | |
| *** admin0 has quit IRC | 02:04 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Bump ansible-base version and collections https://review.opendev.org/c/openstack/openstack-ansible/+/776581 | 02:29 |
| *** Underknowledge has quit IRC | 02:38 | |
| *** macz_ has joined #openstack-ansible | 02:46 | |
| *** macz_ has quit IRC | 02:50 | |
| *** jamesdenton has quit IRC | 02:58 | |
| *** jamesdenton has joined #openstack-ansible | 02:58 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Configure my.cnf for server when galera_root_user is not root https://review.opendev.org/c/openstack/openstack-ansible-galera_server/+/775893 | 03:07 |
| *** dasp has quit IRC | 03:34 | |
| *** dasp has joined #openstack-ansible | 03:45 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi master: Drop distro tests https://review.opendev.org/c/openstack/openstack-ansible-os_gnocchi/+/776425 | 04:02 |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi master: Delete gnocchi_identity_setup https://review.opendev.org/c/openstack/openstack-ansible-os_gnocchi/+/775918 | 04:02 |
| *** shyamb has joined #openstack-ansible | 04:16 | |
| openstackgerrit | James Denton proposed openstack/openstack-ansible-os_neutron master: Add integrated tests various Neutron deployment scenarios https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/778015 | 04:21 |
| *** spatel has quit IRC | 04:48 | |
| *** evrardjp has quit IRC | 05:33 | |
| *** evrardjp has joined #openstack-ansible | 05:33 | |
| *** shyamb has quit IRC | 06:24 | |
| *** jamesdenton has quit IRC | 06:27 | |
| *** jamesdenton has joined #openstack-ansible | 06:27 | |
| *** pto has joined #openstack-ansible | 06:54 | |
| *** pto has joined #openstack-ansible | 06:56 | |
| pto | mgariepy: Yes! I did. The OpenStack part was straight forward, but I had some problems with Ubuntu 20.04 which have changed the way to blacklist pci devices. Here are user_variables.ini part: http://paste.openstack.org/show/803172/ and the playbook i use to configure the computes http://paste.openstack.org/show/803173/ - Any suggestions on how to embed the compute part in openstack ansible? | 07:02 |
| *** luksky has joined #openstack-ansible | 07:19 | |
| *** waxfire0 has joined #openstack-ansible | 07:35 | |
| *** waxfire has quit IRC | 07:35 | |
| *** waxfire0 is now known as waxfire | 07:35 | |
| *** miloa has joined #openstack-ansible | 07:36 | |
| *** miloa has quit IRC | 07:45 | |
| *** rpittau|afk is now known as rpittau | 07:52 | |
| *** andrewbonney has joined #openstack-ansible | 08:14 | |
| *** jamesgibo has joined #openstack-ansible | 08:19 | |
| *** MickyMan77 has quit IRC | 08:30 | |
| *** gokhani has joined #openstack-ansible | 08:31 | |
| *** tosky has joined #openstack-ansible | 08:34 | |
| gokhani | Hi folks, ı have a development environment which is installed by OSA Ussuri. ı have problems on nova-api side. I can observe rabbitmq timeout errors > http://paste.openstack.org/show/803175/ and rarely I am getting connection failed errors to rabbitmq on api side > http://paste.openstack.org/show/803176/. And ıt is strange behaviour now we don't | 08:37 |
| gokhani | have any uwsgi processes on nova side. I doubt about uwsgi + nova. this is expected behaviour or there is a problem on nova, I am not sure. do you have any ideas about this ? | 08:37 |
| gokhani | it opens rabbitmq uwsgi connection and after 3 minutes it gives timeout errors and last it creates a new connection | 08:58 |
| *** Underknowledge has joined #openstack-ansible | 09:11 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts for dynamic_address_fact https://review.opendev.org/c/openstack/openstack-ansible/+/777989 | 09:12 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts for dynamic_address_fact https://review.opendev.org/c/openstack/openstack-ansible/+/777989 | 09:12 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts in common playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777990 | 09:12 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777996 | 09:12 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in group_vars https://review.opendev.org/c/openstack/openstack-ansible/+/777997 | 09:13 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in tests https://review.opendev.org/c/openstack/openstack-ansible/+/777998 | 09:13 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in AIO config https://review.opendev.org/c/openstack/openstack-ansible/+/777999 | 09:13 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in docs https://review.opendev.org/c/openstack/openstack-ansible/+/778000 | 09:14 |
| *** spatel has joined #openstack-ansible | 09:15 | |
| *** shyamb has joined #openstack-ansible | 09:16 | |
| *** shyam89 has joined #openstack-ansible | 09:16 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts for dynamic_address_fact https://review.opendev.org/c/openstack/openstack-ansible/+/777989 | 09:38 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts in common playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777990 | 09:38 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777996 | 09:38 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in group_vars https://review.opendev.org/c/openstack/openstack-ansible/+/777997 | 09:38 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in tests https://review.opendev.org/c/openstack/openstack-ansible/+/777998 | 09:39 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in AIO config https://review.opendev.org/c/openstack/openstack-ansible/+/777999 | 09:39 |
| *** spatel has quit IRC | 09:39 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in docs https://review.opendev.org/c/openstack/openstack-ansible/+/778000 | 09:39 |
| *** SiavashSardari has joined #openstack-ansible | 09:52 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: WIP - Gather minimal facts and disable fact variables https://review.opendev.org/c/openstack/openstack-ansible/+/778396 | 09:53 |
| *** sshnaidm|afk is now known as sshnaidm | 10:06 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_gnocchi master: Provide different package lists for install and wheel build https://review.opendev.org/c/openstack/openstack-ansible-os_gnocchi/+/778222 | 10:10 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_gnocchi master: Move gnocchi pip packages from constraints to requirements https://review.opendev.org/c/openstack/openstack-ansible-os_gnocchi/+/772281 | 10:10 |
| *** shyam89 has quit IRC | 10:11 | |
| *** shyamb has quit IRC | 10:11 | |
| *** gshippey has joined #openstack-ansible | 10:15 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: WIP - Gather minimal facts and disable fact variables https://review.opendev.org/c/openstack/openstack-ansible/+/778396 | 10:29 |
| *** shyam89 has joined #openstack-ansible | 10:37 | |
| *** shyamb has joined #openstack-ansible | 10:37 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts for dynamic_address_fact https://review.opendev.org/c/openstack/openstack-ansible/+/777989 | 10:50 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts in common playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777990 | 10:51 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts for dynamic_address_fact https://review.opendev.org/c/openstack/openstack-ansible/+/777989 | 10:55 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777996 | 10:55 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts in common playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777990 | 10:56 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in group_vars https://review.opendev.org/c/openstack/openstack-ansible/+/777997 | 10:57 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in tests https://review.opendev.org/c/openstack/openstack-ansible/+/777998 | 11:00 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts for dynamic_address_fact https://review.opendev.org/c/openstack/openstack-ansible/+/777989 | 11:00 |
| *** luksky has quit IRC | 11:07 | |
| *** luksky has joined #openstack-ansible | 11:08 | |
| *** luksky has quit IRC | 11:09 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts for dynamic_address_fact https://review.opendev.org/c/openstack/openstack-ansible/+/777989 | 11:23 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts in common playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777990 | 11:24 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in AIO config https://review.opendev.org/c/openstack/openstack-ansible/+/777999 | 11:24 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777996 | 11:25 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in docs https://review.opendev.org/c/openstack/openstack-ansible/+/778000 | 11:25 |
| *** luksky has joined #openstack-ansible | 11:25 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in group_vars https://review.opendev.org/c/openstack/openstack-ansible/+/777997 | 11:26 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: WIP - Gather minimal facts and disable fact variables https://review.opendev.org/c/openstack/openstack-ansible/+/778396 | 11:26 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in tests https://review.opendev.org/c/openstack/openstack-ansible/+/777998 | 11:26 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: WIP - Gather minimal facts and disable fact variables https://review.opendev.org/c/openstack/openstack-ansible/+/778396 | 11:27 |
| jonher | for masakari to install on AIO stable/victoria i had to install pkg-config and libvirt-devel on the aio host manually, why is that not failing in ci? did i miss something in my setup? | 11:30 |
| jrosser | jonher: is yours an lxc install? | 11:31 |
| jonher | standard aio, just copied over the aio -> /etc/openstack_deploy/conf.d/masakari.yml libvirt seems to be running on the host and not in lxc | 11:32 |
| jrosser | jonher: if you go here https://review.opendev.org/c/openstack/openstack-ansible-os_masakari/+/772288 and press the 'Zuul Summary' tab you'll see that for CI all the job names include 'metal' which means that the LXC part is disabled - thats to keep the CI time as low as possible | 11:33 |
| jonher | ah | 11:34 |
| jrosser | unfortunately that does introduce a fairly big gap as all the services get installed on the the same host | 11:34 |
| jrosser | so if a dependancy for X was previously installed by Y they all get mixed and it wont fail | 11:34 |
| jrosser | however in LXC they are separated rootfs so those kind of errors can then show up | 11:34 |
| jrosser | did it want pkg-config and libvirt-devel when it was building the python wheels for masakari? | 11:36 |
| jonher | i'm trying to paste a log on paste.openstack.org but it keeps on giving me a captcha and then cutting parts of the content | 11:37 |
| jonher | https://pastebin.com/a15URKpX | 11:38 |
| jonher | so playbook fails on the step of installing python packages in venv, masakari container seems to succeed building what it needs to, then on the host it fails to libvirt-python/setup.py | 11:39 |
| jrosser | i would expect that any wheel build would happen on the repo container rather than the host, just looking at the code | 11:44 |
| jrosser | there is also another complexity here, that the libvirt python bindings should be taken from the host rather than compiled i think | 11:46 |
| jrosser | https://github.com/openstack/openstack-ansible-os_masakari/blob/master/vars/debian.yml#L31-L32 | 11:46 |
| jonher | oh i see | 11:51 |
| jrosser | so imho there seem to be several things going on here | 11:55 |
| jrosser | first i'd expect the building to be done on the repo server | 11:55 |
| jrosser | and something is kind of messed up with masakari-monitors vs. libvirt-python | 11:56 |
| *** shyam89 has quit IRC | 12:00 | |
| *** shyamb has quit IRC | 12:00 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: WIP - Gather minimal facts and disable fact variables https://review.opendev.org/c/openstack/openstack-ansible/+/778396 | 12:36 |
| jonher | i don't have enough knowledge yet to make changes to that process, is there a place to log this as a known issue? or could anyone have a look? i'm guessing it's kind of a edge-case though so might not be very important/prioritized | 12:52 |
| jrosser | you can file a bug here, would be really helpful https://bugs.launchpad.net/openstack-ansible | 12:53 |
| jrosser | including the pasted info would be good too | 12:53 |
| jrosser | i did have a quick look and the trouble with masakari-monitors is it directly has a python requirement on libvirt-python | 12:54 |
| jrosser | so that tries to be installed earlier in the process than we symlink the distro provided version of that package into the venv | 12:55 |
| jrosser | so i can see how this goes wrong, not exactly sure what to do about it though | 12:55 |
| jrosser | other things like cinder which use the ceph rbd python bindings load those at runtime, so theres no issue with ordering when the venv is installed | 12:56 |
| jonher | this looks related to the same issues i had https://bugs.launchpad.net/openstack-ansible/+bug/1886719 | 12:56 |
| openstack | Launchpad bug 1886719 in openstack-ansible "masakari-monitors wheel building failed for libvirt-python" [Undecided,New] | 12:56 |
| jrosser | yes that looks very similar | 12:59 |
| *** rh-jlabarre has quit IRC | 13:01 | |
| *** rh-jelabarre has joined #openstack-ansible | 13:11 | |
| *** rh-jelabarre has quit IRC | 13:11 | |
| *** rh-jelabarre has joined #openstack-ansible | 13:12 | |
| mgariepy | morning. | 13:30 |
| mgariepy | jrosser, my small patch for neutron is failing with The new implementation of the resolver no longer supports these forms. A possible replacement is replacing the constraint with a requirement.. You can find discussion regarding this at https://github.com/pypa/pip/issues/8210.\nERROR: Links are not allowed as constraint | 13:33 |
| mgariepy | have you see that ? | 13:33 |
| mgariepy | it's only faling on localhost for the check, all the containers seems to be installing just fine. | 13:33 |
| mgariepy | https://zuul.opendev.org/t/openstack/build/88e05286c67e4f079b5372f096a55dc3 | 13:33 |
| mgariepy | https://zuul.opendev.org/t/openstack/build/88e05286c67e4f079b5372f096a55dc3/log/logs/host/python_venv_build.log.txt#11 | 13:34 |
| *** spatel has joined #openstack-ansible | 13:40 | |
| mgariepy | pto, glad it work. | 13:44 |
| mgariepy | i usually configure the iommu stuf prior to run osa. | 13:44 |
| pto | I think osa will overwrite some parts - e.g. sysctl.conf? | 13:44 |
| jamesdenton | mgariepy i think everything fails right now in os_neutron - i have patches to change from functional -> integrated and don't see those errors | 13:45 |
| jamesdenton | Could use some reviews on https://review.opendev.org/c/openstack/openstack-ansible/+/778008/. Will need that to merge first | 13:45 |
| jamesdenton | these are the neutron patches: https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/778015 | 13:45 |
| *** pto has quit IRC | 13:52 | |
| mgariepy | pto i didn't configure passwtrough on 20.04 yet. only 16 and 18. | 13:55 |
| mgariepy | but i only had vendor_iommu=on in grub. i do update the initramfs tool with pci_stub /etc/initramfs-tools/modules << pci_stub ids=10de:17c8,10de:0fb0 | 13:57 |
| mgariepy | and some blacklist for nouveau and nvidiafb | 13:57 |
| *** ajg20 has joined #openstack-ansible | 14:17 | |
| *** watersj has quit IRC | 14:23 | |
| openstackgerrit | Ebbex proposed openstack/openstack-ansible-ops master: apm-server gets fussy about file-permissions https://review.opendev.org/c/openstack/openstack-ansible-ops/+/778481 | 14:55 |
| *** SiavashSardari has quit IRC | 14:56 | |
| openstackgerrit | Jonathan Rosser proposed openstack/ansible-hardening master: Use ansible_facts[] instead of fact variables https://review.opendev.org/c/openstack/ansible-hardening/+/778483 | 15:02 |
| *** juanoterocas has joined #openstack-ansible | 15:02 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Convert host and container journal collection to python https://review.opendev.org/c/openstack/openstack-ansible/+/734883 | 15:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts for dynamic_address_fact https://review.opendev.org/c/openstack/openstack-ansible/+/777989 | 15:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Gather minimal facts in common playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777990 | 15:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in AIO config https://review.opendev.org/c/openstack/openstack-ansible/+/777999 | 15:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777996 | 15:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in docs https://review.opendev.org/c/openstack/openstack-ansible/+/778000 | 15:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in group_vars https://review.opendev.org/c/openstack/openstack-ansible/+/777997 | 15:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in tests https://review.opendev.org/c/openstack/openstack-ansible/+/777998 | 15:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: WIP - Gather minimal facts and disable fact variables https://review.opendev.org/c/openstack/openstack-ansible/+/778396 | 15:03 |
| *** luksky has quit IRC | 15:19 | |
| *** luksky has joined #openstack-ansible | 15:19 | |
| *** luksky has quit IRC | 15:19 | |
| spotz | About for reveiws just ping | 15:32 |
| *** miloa has joined #openstack-ansible | 15:33 | |
| *** miloa has quit IRC | 15:36 | |
| *** luksky has joined #openstack-ansible | 15:36 | |
| openstackgerrit | Jonathan Rosser proposed openstack/ansible-hardening master: Use ansible_facts[] instead of fact variables https://review.opendev.org/c/openstack/ansible-hardening/+/778483 | 15:38 |
| jrosser | jamesdenton: do you think there is value in keeping the more esoteric neutron scenarios? | 15:51 |
| jamesdenton | TBD. At this point, probably not. | 15:52 |
| jamesdenton | i think ODL support needs to be deprecated, since it appears to be unmaintained at a lot of levels (outside of us) | 15:52 |
| jamesdenton | the NSH and SFC scenarios can probably also be removed | 15:53 |
| jamesdenton | tests, rather | 15:53 |
| jrosser | SFC seems to be just another neutron plugin, so it's not really a big deployment deal? | 15:53 |
| jamesdenton | right | 15:53 |
| jamesdenton | and we're not testing anything except to add a plugin | 15:53 |
| jamesdenton | so, happy to make that change now and simplify our lives. They all set to non-voting now | 15:54 |
| jrosser | yeah i was just looking at the results with one of those NV jobs timing out at 3hours | 15:54 |
| jrosser | thats kind of sad as the rest are passing in 1.5hours-ish | 15:54 |
| jamesdenton | yeah, that one seemed to hang on galera for a long time | 15:55 |
| jrosser | oh is that focal | 15:55 |
| jamesdenton | yes | 15:55 |
| jrosser | galera 10.5.8 + focal == oddness | 15:55 |
| jamesdenton | good to know | 15:55 |
| jrosser | https://jira.mariadb.org/browse/MDEV-24829 | 15:55 |
| jamesdenton | i bumped everything to focal except those ODL jobs. | 15:56 |
| jamesdenton | those are all non-value-add at this point, IMO | 15:56 |
| jrosser | i think you're right that so much 3rd party stuff is broken for ODL we shouldnt really spend time on it | 15:57 |
| jrosser | SFC is a bit pointless | 15:57 |
| jrosser | and NSH wants ODL it seems? so thats kind of decided by default | 15:58 |
| jamesdenton | the NSH bit is a special version of OVS stuck at 2.9 | 15:58 |
| jamesdenton | which, looking at OVS, that may have all been rolled in | 15:58 |
| jrosser | i was looking in here https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/778015/8/tests/neutron-overrides-ovs-nsh.yml | 15:59 |
| jamesdenton | so, again, maybe unnecessary t this point | 15:59 |
| jrosser | references to ODL? | 15:59 |
| jamesdenton | yep, it was there before, too | 15:59 |
| jrosser | huh | 15:59 |
| jamesdenton | NSH is for service chaining | 15:59 |
| jamesdenton | but there's no real 'test' here for that, so unecessary | 15:59 |
| jamesdenton | want me to just clean those up and toss a release note in here? | 16:01 |
| jrosser | i think so | 16:02 |
| jrosser | even the NSH stuff is pointing to some ancient PPA, not so sure whats going on there either | 16:03 |
| jrosser | like you say it might be all in stock OVS these days | 16:03 |
| jamesdenton | right. so i will tweak the jobs, remove the cruft, and then circle back on those scenarios in a different patch | 16:03 |
| openstackgerrit | Jonathan Rosser proposed openstack/ansible-hardening master: Use ansible_facts[] instead of fact variables https://review.opendev.org/c/openstack/ansible-hardening/+/778483 | 16:07 |
| *** gokhani has quit IRC | 16:19 | |
| openstackgerrit | Jonathan Rosser proposed openstack/ansible-hardening master: Use ansible_facts[] instead of fact variables https://review.opendev.org/c/openstack/ansible-hardening/+/778483 | 16:22 |
| *** macz_ has joined #openstack-ansible | 16:25 | |
| openstackgerrit | James Denton proposed openstack/openstack-ansible master: Add Neutron Role Testing Overrides https://review.opendev.org/c/openstack/openstack-ansible/+/778008 | 16:33 |
| *** macz_ has quit IRC | 16:46 | |
| *** macz_ has joined #openstack-ansible | 16:47 | |
| openstackgerrit | James Denton proposed openstack/openstack-ansible-os_neutron master: Add integrated tests various Neutron deployment scenarios https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/778015 | 16:54 |
| *** luksky has quit IRC | 16:59 | |
| *** luksky has joined #openstack-ansible | 17:00 | |
| *** klamath_atx has quit IRC | 17:01 | |
| *** openstackgerrit has quit IRC | 17:02 | |
| *** spatel has quit IRC | 17:03 | |
| *** klamath_atx has joined #openstack-ansible | 17:07 | |
| *** spatel has joined #openstack-ansible | 17:07 | |
| *** macz_ has quit IRC | 17:27 | |
| jrosser | jamesdenton: did you have an OVN AIO around? | 17:34 |
| jrosser | i'd be interested to know what `ansible aio1 -m debug -a "var=hostvars['aio1']['container_networks']"` looks like for OVN (and i guess thats generically the same for OVS?) | 17:38 |
| jrosser | ultimately what i'm trying to understand is what should happen here for OVN https://github.com/openstack/openstack-ansible/blob/master/playbooks/common-playbooks/neutron.yml#L47-L52 | 17:51 |
| *** admin0 has joined #openstack-ansible | 18:08 | |
| *** rpittau is now known as rpittau|afk | 18:11 | |
| *** andrewbonney has quit IRC | 18:23 | |
| *** jamesgibo has quit IRC | 18:40 | |
| *** juanoterocas has quit IRC | 18:46 | |
| *** sshnaidm is now known as sshnaidm|afk | 19:00 | |
| *** fresta has quit IRC | 19:08 | |
| admin0 | my bootstrap is stuck on TASK [Install collection requirements with ansible galaxy] .. is there a way to debug and see what exactly that script does ? | 19:08 |
| admin0 | i figured out the debug part | 19:10 |
| *** juanoterocas has joined #openstack-ansible | 19:14 | |
| *** mmercer has quit IRC | 19:39 | |
| *** mmercer has joined #openstack-ansible | 19:40 | |
| jamesdenton | jrosser i do not, sorry | 19:45 |
| jamesdenton | actuall, i lied | 19:45 |
| jamesdenton | one sec | 19:45 |
| jamesdenton | jrosser http://paste.openstack.org/show/803201/ | 19:46 |
| jrosser | interesting | 19:47 |
| jrosser | no tunnel_addresss | 19:47 |
| jamesdenton | the IP configurations is handled here: https://github.com/openstack/openstack-ansible-os_neutron/blob/master/tasks/providers/setup_ovs_ovn.yml#L56 | 19:50 |
| jrosser | where does neutron_local_ip come from? | 19:52 |
| jrosser | I guess what I mean is how that different from the tunnel dynamic address fact | 19:53 |
| *** spatel has quit IRC | 19:58 | |
| jamesdenton | jrosser https://github.com/openstack/openstack-ansible/blob/master/playbooks/common-playbooks/neutron.yml#L99 | 20:00 |
| *** spatel has joined #openstack-ansible | 20:00 | |
| jrosser | oh! | 20:01 |
| jrosser | so i guess this kind of comes full circle :) | 20:02 |
| jamesdenton | and i think i'm missing some context here :) | 20:03 |
| jrosser | it's why I was wanting to know if tunnel_address was defined on an OVN AIO | 20:03 |
| jrosser | so i've been working on patches to gather the absolute smallest set of facts necessary as for compute/network nodes that can save a lot of runtime | 20:04 |
| jrosser | hence this https://review.opendev.org/c/openstack/openstack-ansible/+/777989 | 20:04 |
| jamesdenton | fancy | 20:05 |
| *** klamath_atx has quit IRC | 20:05 | |
| *** klamath_atx has joined #openstack-ansible | 20:07 | |
| jrosser | the only problem is that is failing on just the OVN jobs like this https://zuul.opendev.org/t/openstack/build/22b6dd92d4524bb4b288aa30fbd1bbf7/log/job-output.txt#25076 | 20:07 |
| jrosser | and i'm kind of unsure if thats missing tunnel_address being legit and i need to allow for that, or its missing becasue of a bug | 20:07 |
| *** ajg20 has quit IRC | 20:14 | |
| jamesdenton | well, curious how your task would fail but the subsequent one wouldn't (or hasn't) ? | 20:23 |
| jrosser | the use of json_query in the second task with the || can be resistant to an undefined var | 20:25 |
| jrosser | but the syntax of the second task is kind of obtuse and i'm not sure i understand the behaviour under all defined/undefined vars conditions | 20:26 |
| jamesdenton | i was wondering about that | 20:26 |
| jrosser | do you think its correct that tunnel_address is undefined in this case? | 20:27 |
| jamesdenton | it would seem that way? | 20:29 |
| jamesdenton | ultimately what is being used in thew aio is 172.29.236.100 | 20:29 |
| *** jamesgibo has joined #openstack-ansible | 20:30 | |
| jrosser | oh, this maybe? https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/openstack_user_config.yml.aio.j2#L70 | 20:31 |
| jamesdenton | mayyyybe. i'd have to see how that stanza is used here | 20:32 |
| *** jamesgibo has quit IRC | 20:32 | |
| jamesdenton | in the OVN deploy that's defined on o_u_c but set to neutron_linuxbridge_agent | 20:32 |
| jamesdenton | so, ignored | 20:32 |
| jamesdenton | overriding those anyway in /etc/openstack_deploy/user_variables_neutron_ovn.yml | 20:33 |
| *** spatel has quit IRC | 20:35 | |
| *** spatel has joined #openstack-ansible | 20:38 | |
| jamesdenton | jrosser this might be relevant: https://github.com/openstack/openstack-ansible/commit/b8802f5ce1dc922ff20cab0a46cd7d715323c461 | 20:38 |
| jamesdenton | brb | 20:39 |
| jrosser | sure | 20:39 |
| *** gyee has joined #openstack-ansible | 21:01 | |
| spatel | hey! folks | 21:04 |
| spatel | look like lots of goodness going on | 21:05 |
| *** fresta has joined #openstack-ansible | 21:20 | |
| *** gary_perkins has joined #openstack-ansible | 21:27 | |
| *** irclogbot_2 has quit IRC | 21:33 | |
| *** partlycloudy has quit IRC | 21:33 | |
| *** odyssey4me has quit IRC | 21:33 | |
| *** irclogbot_2 has joined #openstack-ansible | 21:34 | |
| *** partlycloudy has joined #openstack-ansible | 21:34 | |
| *** odyssey4me has joined #openstack-ansible | 21:34 | |
| jrosser | jamesdenton: ah i see you're removing neutron_local_ip override here https://review.opendev.org/c/openstack/openstack-ansible/+/778008/6/tests/roles/bootstrap-host/templates/user_variables_neutron_ovn.yml.j2 | 21:36 |
| jrosser | somehow i'm not understanding "where do I specify the physical interface to use for tunnels" | 21:37 |
| jamesdenton | yeah, because it turns out, those variables are not available at that stage | 21:37 |
| jamesdenton | wait, i'm thinking of the opendaylight studfd | 21:37 |
| * jrosser very confused :) | 21:37 | |
| jamesdenton | lol | 21:37 |
| jamesdenton | i was getting an error, somewhere, ansible_host may have been coming back as localhost (name) and not ip and breaking something | 21:38 |
| jrosser | somehow we've missed defining what should be in o_u_c for OVN as 'sensible defaults" | 21:39 |
| jrosser | i was expecting to find an interface name that would be used to look up the local ip, like for linuxbridge | 21:39 |
| jamesdenton | now, that could be true | 21:39 |
| jamesdenton | https://docs.openstack.org/openstack-ansible-os_neutron/latest/app-ovn.html | 21:40 |
| jamesdenton | you will see there are some examples provided | 21:40 |
| jamesdenton | the AIO overrides may not follow that guidance, however :D | 21:40 |
| jrosser | oh that makes sense now | 21:41 |
| jamesdenton | and it very well could have been a case of, "Well it works and I don't know that it's broken" | 21:42 |
| jrosser | i guess we need to update the template here https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/openstack_user_config.yml.aio.j2 | 21:42 |
| jrosser | thats what created o_u_c for the AIO | 21:42 |
| jrosser | and br-vxlan is kind of an unhelpful name too | 21:43 |
| jamesdenton | well, yes, but the neutron_provider_networks are overwritten by this: https://review.opendev.org/c/openstack/openstack-ansible/+/778008/6/tests/roles/bootstrap-host/templates/user_variables_neutron_ovn.yml.j2#19 | 21:43 |
| jamesdenton | yes, i've changed sall my stuff to br-overlay | 21:43 |
| jrosser | feels like these overrides are a hang-over from the functional tests | 21:44 |
| jrosser | where you didnt have the integrated repo playbooks to orchestrate all this stuff | 21:44 |
| jamesdenton | maybe. we could simply setup a new template per scenario and copy accordingly | 21:44 |
| jrosser | we already do things like this https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/openstack_user_config.yml.aio.j2#L71-L83 | 21:45 |
| jrosser | could have a conditional ovn section there | 21:45 |
| jamesdenton | i mean, if you don't mind blowing up that file a little bit, thats fine | 21:46 |
| jamesdenton | there'd be one for OVS, too | 21:46 |
| *** devtolu1__ has joined #openstack-ansible | 21:47 | |
| jrosser | i think the value is in generating an o_u_c that someone can pretty much take and use, based on building an AIO for a secenario | 21:47 |
| jamesdenton | sure, i can see that | 21:47 |
| jamesdenton | do you want me to rework 778008, then? | 21:48 |
| jrosser | thanks for helping me understand, i was thinking there was some OVN magic that i wanst grasping | 21:48 |
| jamesdenton | oh, don't rely on me to remember or help with anything. :D | 21:48 |
| jamesdenton | but you're welcome | 21:49 |
| *** juanoterocas has quit IRC | 21:49 | |
| jrosser | up to you on 778008, we could put another patch on top? | 21:50 |
| jamesdenton | if we can get the existing thing to pass, maybe we close the loop on that | 21:51 |
| *** juanoterocas has joined #openstack-ansible | 21:51 | |
| jamesdenton | (nothing more irritating than watching normally reliable jobs fail and the failing ones pass) | 21:52 |
| jrosser | looks like debian has failed on 778015 | 21:53 |
| jamesdenton | test_create_list_show_delete_endpoint | 21:53 |
| jamesdenton | among others | 21:53 |
| *** devtolu1__ has quit IRC | 21:54 | |
| jrosser | i've been trying to catalog these random fails https://etherpad.opendev.org/p/osa-ci-failures | 21:55 |
| jrosser | if you find something specific then make a note of it | 21:55 |
| jamesdenton | added it to the tempest identity line | 21:58 |
| jrosser | it's useful to see which are hurting us the most - focal/mariadb is a big problem right now | 21:58 |
| *** jamesdenton has quit IRC | 22:15 | |
| *** jamesdenton has joined #openstack-ansible | 22:15 | |
| *** gshippey has quit IRC | 22:25 | |
| *** spatel has quit IRC | 22:28 | |
| *** spatel has joined #openstack-ansible | 22:39 | |
| *** spatel has quit IRC | 22:39 | |
| *** juanoterocas has quit IRC | 22:52 | |
| *** luksky has quit IRC | 23:39 | |
| *** mgariepy has quit IRC | 23:42 | |
| *** mgariepy has joined #openstack-ansible | 23:55 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!