opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_horizon master: Ensure that selected Apache MPM is enforced https://review.opendev.org/c/openstack/openstack-ansible-os_horizon/+/929695 | 07:54 |
---|---|---|
jrosser | o/ morning | 08:06 |
noonedeadpunk | o/ | 08:09 |
jrosser | do you have a test environment with skyline? | 08:10 |
noonedeadpunk | I think only realtively old one | 08:10 |
noonedeadpunk | but I'm looking at skyline haproxy definition now | 08:11 |
noonedeadpunk | if that's what you're checking as well :) | 08:11 |
jrosser | yeah that was it - i just don't have anywhere to test it at all (other than spawn AIO) | 08:23 |
jrosser | saw the errors in another job for that | 08:23 |
opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Convert Skyline HAProxy httpcheck https://review.opendev.org/c/openstack/openstack-ansible/+/929892 | 08:28 |
opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_skyline master: Ensure that selected Apache MPM is enforced https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/929697 | 08:29 |
noonedeadpunk | I hope that would make it ^ | 08:29 |
jrosser | oh | 08:30 |
jrosser | :) https://review.opendev.org/c/openstack/openstack-ansible/+/929812 | 08:30 |
jrosser | i'll abandon mine as you've setup the depends-on with yours | 08:31 |
noonedeadpunk | oh sorry | 08:43 |
noonedeadpunk | I haven;t seen it :( | 08:43 |
noonedeadpunk | I can abandon mine as well :p | 08:44 |
jrosser | i forgot you had patches to skyline - makes sense to use the depends-on to test it | 08:49 |
gokhan | hello folks, in caracal deployment, the deployment is stuck in TASK [systemd_networkd : Create systemd-networkd network link(s)]. If we remove containers and rerun lxc deploy, it is working. why it can stuck in this task. it doesn't continue after this task.https://paste.openstack.org/show/btKPRvXheIucl644rw5y/ | 09:13 |
gokhan | if I look at servers, onyl ı see openstackansible@notty process. there is no ongoing task. | 09:15 |
noonedeadpunk | could it be that you're re-using same network twice, or try to re-configure network you're using for connection? | 09:39 |
gokhan | noonedeadpunk, we are using a vm as deployment host. can you give an example for using same network twice? | 10:17 |
noonedeadpunk | well, I had a cases where a network for my VM was from the same subnets as are configured for AIO | 10:27 |
noonedeadpunk | then I was loosing coonection to VM when configuring networks for containers | 10:27 |
jrosser | it would be worth trying to figure out if that task jams up on one if it's iterations | 10:29 |
jrosser | or if running the next task is jammed up | 10:29 |
jrosser | some -vvvvv might tell you that | 10:29 |
jrosser | that would also tell you if some attempt to ssh to the target was failing | 10:30 |
noonedeadpunk | well, I was not shown all output as well when that was happening... | 10:30 |
noonedeadpunk | but then the only way for me was to re-create the VM | 10:30 |
gokhan | noonedeadpunk, jrosser it is waiting but there is no attempt to ssh any host. https://paste.openstack.org/show/bNqqnTY0pgMVOdQmOfHG/ | 10:39 |
gokhan | I also tried with -vvv , there is no attempt to ssh any target | 10:40 |
jrosser | what release is this? | 10:40 |
gokhan | it is 29.0.1 | 10:41 |
noonedeadpunk | it could lost connection by then... | 10:42 |
noonedeadpunk | what is the VM network when it's empty? | 10:42 |
noonedeadpunk | like - `ip a` output? | 10:42 |
jrosser | what is `Collection ansible.utils does not support Ansible version 2.13.8` all about? | 10:42 |
jrosser | thats a very old ansible | 10:43 |
noonedeadpunk | O_O | 10:43 |
jrosser | and would not really make sense at all for 29.0.1 | 10:44 |
gokhan | ok I will check ansible version | 10:44 |
jrosser | gokhan: i would say there is something wrong with the setup on your deploy host | 10:44 |
gokhan | jrosser, this is ansible version https://paste.openstack.org/show/b91BO9pt2PC56qlFlTT8/ | 10:47 |
jrosser | i would guess that bootstrap-ansible.sh has not been run | 10:48 |
noonedeadpunk | or you have some ENV var set which backs ansible to older version | 10:50 |
noonedeadpunk | like `ANSIBLE_PACKAGE` | 10:51 |
gokhan | ok ı will check ENV variables and rerun bootstrap ansible.sh | 10:54 |
gokhan | jrosser, I rerun bootstrap-ansible.sh but again getting [WARNING]: Collection ansible.utils does not support Ansible version 2.13.8 | 11:05 |
gokhan | ansible.utils version is v3.0.0 | 11:06 |
jrosser | so what version does ansible —version give you? | 11:06 |
gokhan | ansible [core 2.13.8] | 11:09 |
gokhan | in this commit https://github.com/openstack/openstack-ansible/commit/4130be36d07e89d36c30c3d8477d35fc65935569 it must be 2.15.3 | 11:12 |
gokhan | sorry now it is 2.15.9 | 11:13 |
kleini_ | Is that true, that I need to set ANSIBLE_INJECT_FACT_VARS=true to be able to use openstack-ansible-ops/prometheus? The prometheus.prometheus collection otherwise always fails with missing ansible_architecture. | 13:03 |
noonedeadpunk | if that's true, we need to fix that | 13:27 |
noonedeadpunk | it should be ansible_facts['architecture'] | 13:27 |
gokhan | I sometimes get he user requested python-openstackclient", " The user requested (constraint) python-openstackclient===6.6.1 error. what is the solution of this conflict ? | 13:33 |
gokhan | https://paste.openstack.org/show/bZ1my18QucdF27NrEOoB/ | 13:33 |
noonedeadpunk | hm, it's weird - contraint should not conflict with requirement in this case... | 13:37 |
gokhan | in upper constraints python-openstackclient version is 6.6.1 but in utility requirements file python openstackclient version is not defined | 13:45 |
noonedeadpunk | you got very interesting issues I never saw :D | 13:45 |
noonedeadpunk | might be worth re-running repo-server playbook to re-fetch u-c cache? | 13:46 |
noonedeadpunk | as if you had old ansible, I assume you're upgrading... | 13:46 |
noonedeadpunk | so that could be the thing... | 13:46 |
opendevreview | Jonathan Rosser proposed openstack/openstack-ansible-ops master: Test all supported versions of k8s workload cluster with magnum-cluster-api https://review.opendev.org/c/openstack/openstack-ansible-ops/+/916649 | 14:09 |
jrosser | i wonder if we should remove ironic-inspector | 14:38 |
jrosser | https://docs.openstack.org/releasenotes/ironic-inspector/unreleased.html | 14:39 |
gokhani | noonedeadpunk: ı am not upgrading. I am working with local repos and I am making some arrangements. It seems there are missing things I need to do. I resolved old ansible things and rerun bootstrap-ansible.sh. ı also destroyed previosly created containers and redeployed again. | 14:41 |
noonedeadpunk | jrosser: well, yes, but we need to re-implement the thing natively now I guess | 14:42 |
jrosser | hmmm yes we don't really have a proper aio test setup for it either | 14:47 |
gokhani | noonedeadpunk: I find my issue. devpi is running in offline mode and it didn't find the python-openstackclient 6.6.1 version. thanks :) | 15:00 |
noonedeadpunk | I don't have even ironic running anywhere.... | 15:01 |
jrosser | i guess i could fiddle with our caracal dev environment | 15:03 |
noonedeadpunk | so yeah, we need to do that, but I'm potentially not a big helper there | 15:06 |
jrosser | i guess the question is if we can leave it as is and remove next cycle | 15:08 |
jrosser | or if it really must be done for dalmation | 15:09 |
jrosser | one of the difficulties is that the ironic config is based on what actually works for us, rather than anything we have in tests | 15:09 |
noonedeadpunk | from what I can tell - there's gonna be a release? https://review.opendev.org/c/openstack/releases/+/929764 | 15:17 |
noonedeadpunk | so we can do that for epoxy I'd say | 15:19 |
noonedeadpunk | but no branching makes it weird | 15:19 |
noonedeadpunk | and also no major version bump | 15:19 |
noonedeadpunk | so... dunno | 15:19 |
noonedeadpunk | https://review.opendev.org/c/openstack/releases/+/929409 | 15:20 |
jrosser | i was just digging into whats needed to make some actually useful tests for os_ironic | 15:23 |
jrosser | and it's reasonable amount of work | 15:23 |
noonedeadpunk | ah, ok https://review.opendev.org/c/openstack/releases/+/929409/2/deliverables/dalmatian/ironic-inspector.yaml | 15:23 |
noonedeadpunk | so they branch out of tag | 15:23 |
opendevreview | Merged openstack/openstack-ansible-galera_server master: Manage apt repositores and keys using deb822_repository module https://review.opendev.org/c/openstack/openstack-ansible-galera_server/+/907752 | 15:28 |
opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-ceph_client master: Unify approaches for repo setup between apt/dnf https://review.opendev.org/c/openstack/openstack-ansible-ceph_client/+/929645 | 15:32 |
opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-ceph_client master: Improve handling of ceph_pkg_source changing https://review.opendev.org/c/openstack/openstack-ansible-ceph_client/+/929648 | 15:32 |
opendevreview | Merged openstack/openstack-ansible-os_keystone master: Ensure correct order for credential rotate/migrate https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/925673 | 15:42 |
opendevreview | Merged openstack/openstack-ansible-rabbitmq_server master: Improve handling of rabbitmq_install_method changing https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/927906 | 16:11 |
opendevreview | Merged openstack/openstack-ansible-galera_server master: Improve handling of galera_install_method changing https://review.opendev.org/c/openstack/openstack-ansible-galera_server/+/927901 | 16:45 |
opendevreview | Merged openstack/openstack-ansible-repo_server master: Improve role idempotency https://review.opendev.org/c/openstack/openstack-ansible-repo_server/+/929657 | 19:48 |
opendevreview | Merged openstack/openstack-ansible-ops master: mcapi_proxy: ensure proxy service restarts when venv changes https://review.opendev.org/c/openstack/openstack-ansible-ops/+/927472 | 19:55 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!