| *** MickyMan77 has quit IRC | 00:12 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Add infra testing scenario https://review.opendev.org/755497 | 00:26 |
|---|---|---|
| *** spatel has joined #openstack-ansible | 00:50 | |
| *** gyee has quit IRC | 00:54 | |
| *** spatel has quit IRC | 00:54 | |
| *** spatel has joined #openstack-ansible | 00:59 | |
| *** cshen has joined #openstack-ansible | 01:26 | |
| *** MickyMan77 has joined #openstack-ansible | 01:28 | |
| *** cshen has quit IRC | 01:30 | |
| *** MickyMan77 has quit IRC | 01:37 | |
| *** MickyMan77 has joined #openstack-ansible | 02:06 | |
| *** MickyMan77 has quit IRC | 02:14 | |
| *** MickyMan77 has joined #openstack-ansible | 02:42 | |
| *** MickyMan77 has quit IRC | 02:51 | |
| *** cshen has joined #openstack-ansible | 03:10 | |
| *** cshen has quit IRC | 03:14 | |
| *** MickyMan77 has joined #openstack-ansible | 03:20 | |
| *** MickyMan77 has quit IRC | 03:28 | |
| *** MickyMan77 has joined #openstack-ansible | 04:01 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-ansible | 04:33 | |
| *** miloa has joined #openstack-ansible | 04:36 | |
| *** MickyMan77 has quit IRC | 05:04 | |
| *** spatel has quit IRC | 05:07 | |
| *** cshen has joined #openstack-ansible | 05:10 | |
| *** cshen has quit IRC | 05:14 | |
| *** cshen has joined #openstack-ansible | 05:25 | |
| *** cshen has quit IRC | 05:30 | |
| *** djhankb has quit IRC | 05:55 | |
| *** djhankb has joined #openstack-ansible | 05:55 | |
| *** cshen has joined #openstack-ansible | 06:00 | |
| *** cshen has quit IRC | 06:04 | |
| *** cshen has joined #openstack-ansible | 06:24 | |
| jrosser | morning | 06:39 |
| *** cshen has quit IRC | 06:39 | |
| *** cshen has joined #openstack-ansible | 06:41 | |
| *** cshen has quit IRC | 06:46 | |
| jrosser | noonedeadpunk: can you take a look at this https://review.opendev.org/#/c/756279/ | 06:59 |
| jrosser | really hoping that doesnt break things :/ | 06:59 |
| jrosser | but it's blocking python_venv_build for sure | 06:59 |
| openstackgerrit | wu.shiming proposed openstack/openstack-ansible master: Use unittest.mock instead of mock https://review.opendev.org/757490 | 06:59 |
| *** andrewbonney has joined #openstack-ansible | 07:08 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_placement stable/ussuri: Trigger service restart https://review.opendev.org/757710 | 07:18 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_cinder stable/ussuri: Trigger uwsgi restart https://review.opendev.org/757712 | 07:19 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_cinder stable/train: Trigger uwsgi restart https://review.opendev.org/757713 | 07:20 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_nova stable/train: Trigger uwsgi restart https://review.opendev.org/757714 | 07:21 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_nova stable/ussuri: Trigger uwsgi restart https://review.opendev.org/757715 | 07:21 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_aodh stable/train: Trigger uwsgi restart https://review.opendev.org/757716 | 07:21 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_aodh stable/ussuri: Trigger uwsgi restart https://review.opendev.org/757717 | 07:21 |
| *** cshen has joined #openstack-ansible | 07:22 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_aodh stable/train: Trigger uwsgi restart https://review.opendev.org/757716 | 07:23 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Add rsync to container base image https://review.opendev.org/757612 | 07:23 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_heat stable/train: Trigger uwsgi restart https://review.opendev.org/757721 | 07:24 |
| jrosser | i just run a focal AIO and somehow the stackviz install in os_tempest doesnt use the repo container to build the wheels, thats happened on the utility container | 07:24 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_heat stable/ussuri: Trigger uwsgi restart https://review.opendev.org/757722 | 07:24 |
| *** tosky has joined #openstack-ansible | 07:26 | |
| *** rpittau|afk is now known as rpittau | 07:27 | |
| noonedeadpunk | I'm also wondering how `venv_build_host` is being defined here https://opendev.org/openstack/openstack-ansible-os_tempest/src/branch/master/tasks/tempest_install_source.yml#L37 | 07:29 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cloudkitty master: Define condition for the first play host one time https://review.opendev.org/751261 | 07:47 |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: Use pip and virtualenv from host packages rather than get-pip.py https://review.opendev.org/756279 | 07:56 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_gnocchi master: Updated from OpenStack Ansible Tests https://review.opendev.org/749337 | 08:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_senlin master: Updated from OpenStack Ansible Tests https://review.opendev.org/752892 | 08:03 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_placement stable/train: Trigger service restart https://review.opendev.org/757745 | 08:19 |
| *** pmannidi has quit IRC | 08:24 | |
| *** pmannidi has joined #openstack-ansible | 08:24 | |
| noonedeadpunk | would be also awesome to get https://review.opendev.org/#/c/753376/ merged | 08:28 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add centos-8 support https://review.opendev.org/739646 | 08:29 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: [DNM] Decrease murano tempest concurrency https://review.opendev.org/743798 | 08:30 |
| recyclehero | morning guys | 08:38 |
| recyclehero | jrosser: I didnt see how it was a bug what u have sent last night? | 08:40 |
| recyclehero | https://opendev.org/openstack/openstack-ansible-os_neutron/src/branch/master/vars/main.yml#L325-L340 | 08:40 |
| jrosser | the filtered list of services will not include the metadata service when you remove it from the neutron_plugin_base | 08:40 |
| jrosser | so i think thats why you saw the service keep running even though it was removed from the list | 08:41 |
| jrosser | i don't really know what the user expectation should be here | 08:43 |
| recyclehero | opendev is not opening for some reason , I will checkout in a min. will u fix the bug so I can cherry pick or on my own? | 08:44 |
| recyclehero | jrosser: filtered_services are the services that should be deployed or the other way around? I think filtered_neutron_services incldue metering_agent based on the service_plugin. I am tring to setup ansible debug for me. | 08:52 |
| jrosser | yes i think that filtered services are the ones that should be deployed | 08:53 |
| jrosser | but i think theres nothing that can account for taking something out of the list | 08:53 |
| recyclehero | jrosser: something like this for debugging, am I on the correct path? | 09:02 |
| recyclehero | http://paste.openstack.org/show/798977/ | 09:02 |
| recyclehero | I run it with ansible-playbook -i /opt/openstack-ansible/inventory/dynamic_inventory.py AlistarAnsible | 09:03 |
| jrosser | not quite | 09:03 |
| recyclehero | oh | 09:03 |
| jrosser | are you trying to see the value of this? https://github.com/openstack/openstack-ansible-os_neutron/blob/c4d5649b11912637c8ca4753358747fc1b695d3e/vars/main.yml#L144 | 09:06 |
| recyclehero | yes | 09:08 |
| jrosser | sorry meetings time now | 09:17 |
| openstackgerrit | Merged openstack/openstack-ansible-os_manila master: Fix linter errors https://review.opendev.org/755570 | 09:24 |
| noonedeadpunk | whaaat | 09:24 |
| noonedeadpunk | ah, ok, it was just linters | 09:25 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Simplify scheduler filter additions https://review.opendev.org/650144 | 09:35 |
| *** yann-kaelig has joined #openstack-ansible | 10:18 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_nova master: Define condition for the first play host one time https://review.opendev.org/756332 | 10:26 |
| *** shyamb has joined #openstack-ansible | 11:26 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Fix focal ceph jobs https://review.opendev.org/753376 | 11:32 |
| ebbex | noonedeadpunk, jrosser https://github.com/openstack/openstack-ansible-os_nova/blob/master/vars/main.yml#L104 is it supposed to be default_filters or just `filters`? | 11:35 |
| jrosser | urgh | 11:36 |
| ebbex | I take that as a no, I'll submit a fix :) | 11:36 |
| noonedeadpunk | yeah.... | 11:38 |
| *** shyamb has quit IRC | 11:39 | |
| ebbex | should it perhaps be worded differently, so we set filters = nova_scheduler_default_filters and then set default_filters = filters.split(',') so the last line makes sense like it does now? | 11:41 |
| ebbex | nah, I'm leaving it as is. this will be deprecated/simplified in a release or two right? | 11:45 |
| noonedeadpunk | ebbex: that's the plan | 11:46 |
| noonedeadpunk | idea was just to leave {{ nova_scheduler_default_filters + nova_scheduler_extra_filters }} at the end of the day | 11:46 |
| openstackgerrit | Erik Berg proposed openstack/openstack-ansible-os_nova master: use correct filters variable https://review.opendev.org/757844 | 11:48 |
| noonedeadpunk | gerrit doesn't load for me :( | 11:50 |
| ebbex | me neither. | 11:50 |
| *** shyamb has joined #openstack-ansible | 11:58 | |
| *** jbadiapa has joined #openstack-ansible | 12:06 | |
| *** rfolco has joined #openstack-ansible | 12:10 | |
| *** MickyMan77 has joined #openstack-ansible | 12:20 | |
| MickyMan77 | I'm not able to deploy openstack-ansible with ceph-client. I failes during TASK ceph_client : Add OpenStack service to cephkeys_access_group group "The error was: TypeError: getpwnam() argument must be str, not None fatal" | 12:22 |
| MickyMan77 | or if i the disable ceph_keyrings_dir: function in user_variables.yml and try to use ssh. | 12:23 |
| MickyMan77 | if failes when it try to ssh mon nodes. (not able to lookup host). | 12:24 |
| ebbex | can we set this to something? https://github.com/openstack/openstack-ansible/blob/stable/ussuri/ansible-role-requirements.yml#L275 | 12:25 |
| MickyMan77 | does someone have any updated guides to get ceph-client to work with openstack-ansible deployments ? | 12:26 |
| ebbex | and aslo, which repo contains the release-notes for osa? | 12:27 |
| jrosser | release notes are aggregated from all roles into the integrated repo iirc | 12:27 |
| jrosser | and as far as the ceph-ansible trackbranch goes, i think thats because they start to put breaking changes on the stable-<N> branches for ceph-ansible so we have to pin to a SHA now | 12:28 |
| jrosser | beyond that SHA it's all broken | 12:28 |
| ebbex | uffda. | 12:28 |
| jrosser | broken in the sense that the data we have in group_vars will suddlenly no longer be OK for a later version of the same ceph-ansible stable branch | 12:29 |
| jrosser | which is really quite uncool | 12:29 |
| ebbex | i can agree with that. | 12:29 |
| jrosser | MickyMan77: it would help if you could share some more details, maybe at paste.openstack.org of what you're trying to do | 12:30 |
| jrosser | MickyMan77: we run a CI job on every commit that uses OSA + ceph-ansible together, so one of those code paths is checked regularly | 12:31 |
| jrosser | it looks like you are trying to integrate with an external ceph cluster though? | 12:31 |
| jrosser | it's really helpful to know which branch you are using, paste the output from the failing tasks (and a few before) and if possible share the relevant config | 12:32 |
| MickyMan77 | jrosser: here you go... http://paste.openstack.org/show/798986/ | 12:45 |
| recyclehero | jrosser: little help here | 12:51 |
| recyclehero | http://paste.openstack.org/show/798987/ | 12:51 |
| jrosser | MickyMan77: the first error you got in the glance container is probably becasue of this setting openstack_service_system_user: null | 12:51 |
| jrosser | recyclehero: ok, so you can't really write an independant playbook like tht | 12:52 |
| jrosser | the variables you are trying to print only exist in the scope of executing the os_neutron role | 12:53 |
| *** shyamb has quit IRC | 12:53 | |
| recyclehero | so how can I see if neutron_metering is on/off and also what I should do about the bug | 12:53 |
| jrosser | you should add a debug: task to somewhere in the os_neutron role, maybe really early on in main.yml | 12:53 |
| jrosser | you can even follow it directly with a fail: task to make things just stop | 12:54 |
| jrosser | so really it's bug vs. expectations | 12:54 |
| recyclehero | ok. but your take on it was neutron_metering cant be turned off just by removeing it from neutron_base_plugins ? | 12:55 |
| jrosser | as far as i can see (having looked very briefly yesterday), it should remove it from the list of things to deploy | 12:56 |
| jrosser | but if it's already deployed from a previous run then it's likley not to be touched | 12:56 |
| recyclehero | got it, thanks. | 12:57 |
| jrosser | if you take metering out of neutron_plugins_base then hopefully it will not be redeployed | 12:57 |
| jrosser | and you could stop/disable the service with systemctl and it shouldnt be restarted if you re-run the os_neutron playbook | 12:57 |
| jrosser | all subject to testing though :) | 12:58 |
| *** shyamb has joined #openstack-ansible | 12:58 | |
| *** shyamb has quit IRC | 12:58 | |
| recyclehero | that sounded like a legal disclaimer :D | 12:58 |
| jrosser | well it kind of is | 12:59 |
| jrosser | everything is volunteer effort | 12:59 |
| recyclehero | thats the context here. thank for ur efforts. its helps us move forward. | 13:00 |
| recyclehero | I ahve some volume moved from a failed openstack setup to this new one by converting them to qcow2 by qemu-img and uploading to the newly deployed glance. | 13:02 |
| recyclehero | now when I lunch them in the log they try for 120s to get to metadata agents but cant | 13:03 |
| recyclehero | thats not a problem with cirros image | 13:03 |
| recyclehero | any idea what could be the cause of that? | 13:03 |
| recyclehero | I can connect to them after the 120s period | 13:05 |
| jrosser | it feels like the metadata service is timing out | 13:06 |
| recyclehero | http://paste.openstack.org/show/798988/ | 13:16 |
| *** openstackgerrit has quit IRC | 13:17 | |
| recyclehero | [HTTPConnectionPool(host='200.200.200.220', port=80): Max retries exceeded with url: /latest/meta-data/instance-id (Caused by NewConnectionError('<requests.packages.urllib3.connection.HTTPConnection object at 0x7fc447fd2cc0>: Failed to establish a new connection: [Errno 111] Connection refused',))] | 13:18 |
| recyclehero | this is the error seen in the log of the instance | 13:18 |
| recyclehero | why is it calling the ip of the dhcp itself and not 169.254.169.254 which it has a route for? | 13:19 |
| jamesdenton | there is a fallback mechanism built into cloud-init to fall back to the DHCP server IIRC | 13:20 |
| jamesdenton | what happens in your cirros instance if you curl http://169.254.169.254/latest/meta-data ? | 13:20 |
| recyclehero | jamesdenton: I will recreate the instance and report back | 13:22 |
| jamesdenton | thx | 13:22 |
| *** klamath_atx has quit IRC | 13:23 | |
| *** klamath_atx has joined #openstack-ansible | 13:24 | |
| recyclehero | jamesdenton: a list I guess instance-id, ami-id .... | 13:28 |
| noonedeadpunk | jrosser: what do you think about http://bit.ly/osa-review-board ? (I've just stolen that from barbican folks) | 13:29 |
| jrosser | could be total lifesaver :) | 13:29 |
| recyclehero | jamesdenton: I think I shoud do that with the instance that I have imported from the other enviroment | 13:29 |
| jrosser | noonedeadpunk: the specific question today is "what do we need to merge for the V release" | 13:30 |
| jrosser | the etherpad has become a bit out of date for that becasue it was the things we needed to fix/merge when the pad was filled out | 13:31 |
| jrosser | theres now a huge pile of stuff from the last two weeks or so | 13:31 |
| noonedeadpunk | I will go through it now | 13:31 |
| jrosser | thats a really useful view | 13:34 |
| jrosser | makes the patch backlog look manageable now rather than just seeming to be gigantic | 13:34 |
| jrosser | noonedeadpunk: gshippey did some more work on the galera backup patch, it might be useful to try it out if you have a similar requirement | 13:35 |
| jrosser | same sort of thing as with the letsencrypt stuff, i think it maybe is requiring actual road-test | 13:35 |
| recyclehero | jamesdenton: this time I gave it a key upon lunch and it worked right away. I didnt add a key cause I injected a key upon transferring them and didnt needed one. I think no key triggered the fail saif machanism in it | 13:37 |
| jamesdenton | well, the key is only available upon successful connectivity to the metadata service. if the 169 IP isn't available after so many tries, cloud init will try other IPs it knows. This used to work but is now broken (by design) | 13:39 |
| jamesdenton | https://bugs.launchpad.net/neutron/+bug/1870228 | 13:40 |
| openstack | Launchpad bug 1870228 in neutron "cloud-init metadata fallback broken" [Medium,New] | 13:40 |
| recyclehero | jamesdenton: thank u | 13:42 |
| noonedeadpunk | #topic Launchpad: https://launchpad.net/openstack-ansible || Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Review Dashboard: http://bit.ly/osa-review-board-v1 | 13:44 |
| noonedeadpunk | TOPIC openstack-ansible Launchpad: https://launchpad.net/openstack-ansible || Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Review Dashboard: http://bit.ly/osa-review-board-v1 | 13:46 |
| recyclehero | jamesdenton: can u explain what is the purpose of this metadata, I read it but cant fully understand whts it does exactly | 13:46 |
| noonedeadpunk | uh | 13:46 |
| *** noonedeadpunk[h] has joined #openstack-ansible | 13:47 | |
| *** spatel has joined #openstack-ansible | 13:47 | |
| *** noonedeadpunk[h] has quit IRC | 13:47 | |
| *** openstackgerrit has joined #openstack-ansible | 13:48 | |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Update graylog-forwarder to use py3 https://review.opendev.org/755461 | 13:48 |
| jamesdenton | metadata is information about your instance that the instance can then use, including a RSA key that is then used for logging in. The cloud-init service on vm image knows how to reach metadata service and what to pull down. But you can also execute scripts and other cloud-init specific actions | 13:49 |
| *** cshen has quit IRC | 13:51 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-plugins master: Move provider_networks module into os_neutron https://review.opendev.org/658130 | 14:00 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_magnum master: Support docker registry config options https://review.opendev.org/652648 | 14:01 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: IP addresses in examples were changed https://review.opendev.org/572935 | 14:07 |
| openstackgerrit | Merged openstack/openstack-ansible-nspawn_hosts master: Add vars file for ubuntu focal https://review.opendev.org/756341 | 14:14 |
| *** alvinstarr has joined #openstack-ansible | 14:15 | |
| *** nurdie has joined #openstack-ansible | 14:15 | |
| *** klamath_atx has quit IRC | 14:20 | |
| *** redrobot has quit IRC | 14:22 | |
| *** klamath_atx has joined #openstack-ansible | 14:23 | |
| ebbex | jrosser: Could you have a quick peek at this http://paste.openstack.org/show/798993/ , is there any obvious action i might be missing for pci-passthrough? | 14:23 |
| *** klamath_atx has quit IRC | 14:25 | |
| *** jamesdenton has quit IRC | 14:31 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: Remove support for lxc2 config keys https://review.opendev.org/756255 | 14:35 |
| *** cshen has joined #openstack-ansible | 14:36 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder stable/train: Trigger uwsgi restart https://review.opendev.org/757713 | 14:40 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_nova master: Define condition for the first play host one time https://review.opendev.org/756332 | 14:40 |
| *** jamesdenton has joined #openstack-ansible | 14:42 | |
| jrosser | ebbex: you mean it's not working, or just check what you have? | 14:45 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ironic master: Define condition for the first play host one time https://review.opendev.org/756340 | 14:45 |
| jrosser | ebbex: maybe missing pci stub stuff on the boot parameters, i have GRUB_CMDLINE_LINUX_DEFAULT="intel_iommu=on modprobe.blacklist=nouveau pci-stub.ids=10de:1eb8,10de:12a2" | 14:48 |
| jrosser | actually jamesdenton already wrote about all this https://www.jimmdenton.com/gpu-offloading-openstack/ | 14:49 |
| noonedeadpunk | jrosser: answering question about what to merge for release... I think we're good enough for what we have already pushed, but I think we need to make our roles passing. I'm going to spend some time for fixing manila/murano, which were broken for U... | 14:50 |
| jrosser | yeah, i put a list of whats broken on the etherpad | 14:50 |
| noonedeadpunk | And we need to ensure that after merging things, ironic, octavia at least will be also passing | 14:50 |
| noonedeadpunk | as things are unhappy without public vip in tempest | 14:51 |
| noonedeadpunk | magnum is bad as well... | 14:51 |
| jrosser | yes - i wonder if the haproxy/metal patch is going to help with thtat | 14:51 |
| jrosser | we maybe discover more tempest plugins with no config for endpoint though | 14:51 |
| noonedeadpunk | from what I saw in tempest logs for some of the patches - it was claiming on publicurl, which should help | 14:52 |
| noonedeadpunk | ironic broken with tempestconf, which requires public endpoint iirc | 14:52 |
| noonedeadpunk | octavia is almost happy with https://review.opendev.org/#/c/755737/ except upgrade jobs... | 14:52 |
| noonedeadpunk | yeah, I agree with the list, except I moved senlin to W | 14:58 |
| noonedeadpunk | since we obvoiously won't have needed options for V | 14:58 |
| jrosser | did we ever merge the patches to make upgrade jobs work properly | 15:00 |
| jrosser | i was wondering if thats why the octavia patch breaks for upgrades | 15:00 |
| noonedeadpunk | I think we did, and they were behaving nicely lately | 15:00 |
| noonedeadpunk | but they break during tempest... | 15:00 |
| noonedeadpunk | but upgrade does not do bootstrap-host anyway 2nd time | 15:01 |
| noonedeadpunk | and that's what we need to get updated tempest conf | 15:01 |
| *** macz_ has joined #openstack-ansible | 15:04 | |
| *** rh-jelabarre has quit IRC | 15:06 | |
| Adri2000 | hello | 15:07 |
| Adri2000 | I wonder if in that patch https://review.opendev.org/#/c/673878/, there was an oversight on designate? i.e. notifications should be enabled if we have ceilometer enabled *or* designate enabled. I've just upgraded to stein and hit this problem (I use designate but don't use ceilometer). mnaser? | 15:09 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder stable/train: Explicitly enable/disable active/active https://review.opendev.org/752704 | 15:15 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: Add rsync to required packages for redhat based OS https://review.opendev.org/756651 | 15:15 |
| openstackgerrit | Merged openstack/openstack-ansible master: IP addresses in examples were changed https://review.opendev.org/572935 | 15:24 |
| openstackgerrit | Merged openstack/openstack-ansible master: Add link to configuration example for Ceph configuration from files. https://review.opendev.org/755797 | 15:24 |
| ebbex | jrosser: thanks, there's something not quite working on my part yet, the denton stuff seems like a good read :) | 15:24 |
| noonedeadpunk | jrosser: huh, manila self-repaired.... | 15:26 |
| jrosser | that is the best kind of bugfix | 15:26 |
| noonedeadpunk | just needed this https://review.opendev.org/#/c/757854/1/tests/roles/bootstrap-host/templates/user_variables_manila.yml.j2 | 15:26 |
| jrosser | thats good | 15:27 |
| jrosser | we should probably merge this https://review.opendev.org/#/c/733408/ | 15:27 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/ussuri: Fix focal ceph jobs https://review.opendev.org/757896 | 15:28 |
| jrosser | theres a risk of changing things in all the role metal jobs with that | 15:28 |
| noonedeadpunk | I hope it will just fix everything for us lol | 15:29 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_adjutant master: Make role fit to the OSA standards https://review.opendev.org/756313 | 15:40 |
| *** nurdie has quit IRC | 15:49 | |
| *** nurdie has joined #openstack-ansible | 15:50 | |
| *** cshen has quit IRC | 15:50 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/train: Trigger uwsgi restart https://review.opendev.org/757714 | 15:52 |
| *** gyee has joined #openstack-ansible | 15:54 | |
| openstackgerrit | Adrien Cunin proposed openstack/openstack-ansible-os_nova master: Enable notifications when Designate is enabled https://review.opendev.org/757904 | 15:56 |
| Adri2000 | follow up to my earlier message ^ (cc mnaser) | 15:57 |
| *** rh-jelabarre has joined #openstack-ansible | 16:03 | |
| noonedeadpunk | Adri2000: I'm trying to understand how designate will be reading nova vhost and can't really find way to configure that... | 16:05 |
| noonedeadpunk | https://docs.openstack.org/designate/latest/admin/config.html#handler-nova-fixed doesn't contain neither creds for connection nor really vhost or anything actually... | 16:06 |
| noonedeadpunk | Adri2000: and btw the same thing is here https://opendev.org/openstack/openstack-ansible-os_neutron/src/branch/master/templates/neutron.conf.j2#L233 | 16:07 |
| noonedeadpunk | Adri2000: to tbh, I'd set another topic for designate as I think then designate will fight with ceilometer for notifications | 16:08 |
| noonedeadpunk | ah well, sorry, it;s already there.... | 16:09 |
| noonedeadpunk | but then we need to use notifications topic only when ceilometer is enabled.... | 16:10 |
| noonedeadpunk | well, I'm not rerally sure | 16:10 |
| noonedeadpunk | #startmeeting openstack_ansible_meeting | 16:11 |
| openstack | Meeting started Tue Oct 13 16:11:10 2020 UTC and is due to finish in 60 minutes. The chair is noonedeadpunk. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:11 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:11 |
| *** openstack changes topic to " (Meeting topic: openstack_ansible_meeting)" | 16:11 | |
| noonedeadpunk | #topic office hours | 16:11 |
| openstack | The meeting name has been set to 'openstack_ansible_meeting' | 16:11 |
| *** openstack changes topic to "office hours (Meeting topic: openstack_ansible_meeting)" | 16:11 | |
| MickyMan77 | jrosser: do you know what user i should use for the setting openstack_service_system_user: XXXX ?? | 16:12 |
| jrosser | MickyMan77: i don't think you should set that at all | 16:12 |
| jrosser | that variable will be set to the appropriate value in the cinder/glance/etc roles as here http://codesearch.openstack.org/?q=openstack_service_system_user | 16:13 |
| noonedeadpunk | so, I don't have really much to discuss I guess this time, as we're moving forward and things continue merging (pretty slowly though) | 16:14 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Add infra scenario CI jobs https://review.opendev.org/756440 | 16:15 |
| noonedeadpunk | except I wanted to share this fancy link to the review board | 16:16 |
| noonedeadpunk | #link http://bit.ly/osa-review-board | 16:16 |
| jrosser | i am liking that a lot | 16:16 |
| noonedeadpunk | it is including non-osa projects at the moment, but I hope I will clean it out | 16:17 |
| noonedeadpunk | Yeah, it has at least some structure. not sure how good it works in terms of `2 week without feedback` | 16:18 |
| noonedeadpunk | but yeah | 16:18 |
| * jrosser just looks at the galera bump | 16:19 | |
| noonedeadpunk | oh, yes, I will test it out for sure as have a use case | 16:20 |
| ebbex | Has anyone looked through https://docs.openstack.org/releasenotes/openstack-ansible/ussuri.html | 16:20 |
| * noonedeadpunk writes down in todo list | 16:20 | |
| noonedeadpunk | ebbex: well, I did... | 16:20 |
| noonedeadpunk | with release team | 16:20 |
| ebbex | it's a bit much, no? | 16:21 |
| noonedeadpunk | and we were unable to figurre out wtf it was... | 16:21 |
| noonedeadpunk | it includes all train and ussuri | 16:21 |
| ebbex | yeah, it seems to have carried stuff all the way from rocky. | 16:21 |
| noonedeadpunk | https://storyboard.openstack.org/#!/story/2008139 | 16:21 |
| *** rpittau is now known as rpittau|afk | 16:21 | |
| noonedeadpunk | the only workaround I was suggested is to exclude manually every reno that shouldn't be there | 16:22 |
| noonedeadpunk | which is.... | 16:22 |
| ebbex | 21.0.0 requires ceph-ansible 4.0 (but that upgrade would have happened in train right?) | 16:22 |
| noonedeadpunk | and what also we can't do in no way - backport release notes.... | 16:23 |
| noonedeadpunk | like we did here https://review.opendev.org/#/c/752704/ :((( | 16:23 |
| noonedeadpunk | it is | 16:23 |
| noonedeadpunk | in 21.0.0 we require 5.0 | 16:24 |
| jrosser | i am looking at https://review.opendev.org/#/c/757851 | 16:25 |
| jrosser | doesnt that make N^2 tasks for N compute nodes, each one needing the key for all the others? | 16:26 |
| *** ChosSimbaOne has quit IRC | 16:27 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-os_nova master: Fix for scheduler filters list construction https://review.opendev.org/757909 | 16:27 |
| noonedeadpunk | jrosser: it is:( | 16:29 |
| *** ChosSimbaOne has joined #openstack-ansible | 16:29 | |
| noonedeadpunk | well, I was trying to find the way how to store authorized keys somewhere, but didn't come up with good solution either... | 16:30 |
| noonedeadpunk | jrosser: what can be done better is to use some var I guess which will collect all keys for all nodes | 16:31 |
| *** rpittau|afk has quit IRC | 16:31 | |
| jrosser | or we do the ssh certs thing | 16:31 |
| noonedeadpunk | I guess it's in terms of https://etherpad.opendev.org/p/osa-certificates-refactor ? | 16:32 |
| jrosser | yeah, if we have a CA there is already an ansible module to create the ssh credential | 16:34 |
| noonedeadpunk | but ssh certs sounds really good | 16:34 |
| *** rpittau|afk has joined #openstack-ansible | 16:34 | |
| noonedeadpunk | well ok, let's live with it till W | 16:36 |
| noonedeadpunk | and do certs afterwards | 16:36 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Add CADF notifications for federated keystone https://review.opendev.org/741946 | 16:39 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/ansible-config_template master: Missing document start "---" https://review.opendev.org/736466 | 16:41 |
| *** recyclehero has quit IRC | 16:53 | |
| noonedeadpunk | #endmeeting | 16:54 |
| *** openstack changes topic to "Launchpad: https://launchpad.net/openstack-ansible || Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Review Dashboard: https://bit.ly/2SAcGAn" | 16:54 | |
| openstack | Meeting ended Tue Oct 13 16:54:50 2020 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:54 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2020/openstack_ansible_meeting.2020-10-13-16.11.html | 16:54 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2020/openstack_ansible_meeting.2020-10-13-16.11.txt | 16:54 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2020/openstack_ansible_meeting.2020-10-13-16.11.log.html | 16:54 |
| *** recyclehero has joined #openstack-ansible | 17:00 | |
| recyclehero | gonna do a backup restore and then nuke the whole thing. if you have any experiment for 21.0.1 hit me. | 17:04 |
| recyclehero | going to deploy production | 17:04 |
| * recyclehero excited | 17:04 | |
| recyclehero | guys this is in the docs: | 17:11 |
| recyclehero | # ansible galera_container -m shell -a "mysql -h 127.0.0.1 \ | 17:11 |
| recyclehero | -e 'show status like \"%wsrep_cluster_%\";'" | 17:11 |
| recyclehero | where is the inventory in this command, its for finding a failed node | 17:12 |
| recyclehero | aha galera container should be replaced with the ip of the glarea container. noob | 17:14 |
| *** miloa has quit IRC | 17:17 | |
| rh-jelabarre | anyone watching AnsibleFest breakout sessions and getting a lot of buffering? | 17:17 |
| *** Guest75569 has joined #openstack-ansible | 17:21 | |
| *** Guest75569 is now known as redrobot | 17:23 | |
| *** vesper11 has joined #openstack-ansible | 17:25 | |
| *** andrewbonney has quit IRC | 17:28 | |
| *** cshen has joined #openstack-ansible | 17:29 | |
| *** cshen has quit IRC | 17:35 | |
| *** cshen has joined #openstack-ansible | 17:46 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Do not manage /etc/ssl or /etc/pki directories or symlinks https://review.opendev.org/754092 | 17:48 |
| *** cshen has quit IRC | 17:51 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat stable/ussuri: Trigger uwsgi restart https://review.opendev.org/757722 | 18:01 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat stable/train: Trigger uwsgi restart https://review.opendev.org/757721 | 18:01 |
| *** sshnaidm is now known as sshnaidm|afk | 18:24 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_neutron master: Enable notifications when Designate is enabled https://review.opendev.org/757937 | 19:08 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_neutron master: Enable notifications when Designate is enabled https://review.opendev.org/757937 | 19:10 |
| MickyMan77 | jrosser: I have solve the first issue, now I get this error msg... TabError: inconsistent use of tabs and spaces in indentation --> http://paste.openstack.org/show/799008/ | 19:24 |
| jrosser | MickyMan77: which release of OSA is this? | 19:25 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Added Openstack Adjutant role deployment https://review.opendev.org/756310 | 19:26 |
| MickyMan77 | Git tag 21.0.1 | 19:26 |
| noonedeadpunk | o_O | 19:26 |
| noonedeadpunk | oh well | 19:27 |
| noonedeadpunk | MickyMan77: rename/remove /root/rbd.py | 19:27 |
| jrosser | how does that even exist/ | 19:28 |
| jrosser | ? | 19:28 |
| *** gshippey has quit IRC | 19:28 | |
| noonedeadpunk | corner case ¯\(◉◡◔)/¯ | 19:28 |
| MickyMan77 | # locate rbd.py | 19:28 |
| MickyMan77 | # /usr/local/lib/python3.6/site-packages/rbd.py | 19:28 |
| noonedeadpunk | well, I think MickyMan77has created some script in root which is named rbd.py, and eventually it got imported instead of real rbd library | 19:28 |
| MickyMan77 | I dont have anyting in root | 19:29 |
| noonedeadpunk | MickyMan77: on controller? | 19:29 |
| MickyMan77 | I check | 19:29 |
| MickyMan77 | aha... | 19:29 |
| MickyMan77 | it's i file there. | 19:29 |
| noonedeadpunk | today I was really about to do the same :p | 19:30 |
| noonedeadpunk | maybe we should chdir to /tmp or dunno | 19:30 |
| MickyMan77 | i run the deployments again now... lets see if it will work now.. :) | 19:31 |
| recyclehero | If I want to test DR scenario wuth a 1 node galera deployment. for the backup I should just do a mariadb full backup? | 19:35 |
| *** cshen has joined #openstack-ansible | 19:47 | |
| recyclehero | as mentioned in https://docs.openstack.org/openstack-ansible/ussuri/admin/backup-restore.html | 19:50 |
| recyclehero | /etc/openstack_deploy contains a live inventory | 19:50 |
| recyclehero | say one of the physical hosts die, and I am going to change it with another physical host with diffrent spec | 19:51 |
| *** cshen has quit IRC | 19:51 | |
| recyclehero | what should one do with the gatherd ansible_facts | 19:51 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Added Openstack Adjutant role deployment https://review.opendev.org/756310 | 19:54 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_adjutant master: Make role fit to the OSA standards https://review.opendev.org/756313 | 19:56 |
| openstackgerrit | Merged openstack/openstack-ansible-os_placement stable/ussuri: Trigger service restart https://review.opendev.org/757710 | 20:01 |
| jrosser | recyclehero: the facts in /etc/openstack-deploy can be deleted should it be needed | 20:07 |
| jrosser | it's just a cache | 20:08 |
| recyclehero | great. I just read about that. jrosser when I am assigning variables in user_variables for variables with hierarchies like nova_virt_type If I want to only change the bottom one I should change all other variables too? | 20:10 |
| recyclehero | I changed nova_reserverd_host_memory_mb and the playbook was complaing about not defined nova_compute_driver | 20:10 |
| recyclehero | https://docs.openstack.org/openstack-ansible-os_nova/latest/ | 20:10 |
| jrosser | i don't think you need to touch nova_virt_types at all, theres a comment above which explains | 20:15 |
| jrosser | depending on nova_virt_type, one set of values out of nova_virt_types is set as ansible facts during the run | 20:16 |
| jrosser | those facts are then used in the template for the config file https://opendev.org/openstack/openstack-ansible-os_nova/src/branch/master/templates/nova.conf.j2#L15 | 20:17 |
| jrosser | ansible has a variable precedence https://docs.ansible.com/ansible/latest/user_guide/playbooks_variables.html#ansible-variable-precedence | 20:18 |
| jrosser | extra vars (anything you put in user_variables.yml) is the highest precedence, so will override any previously set fact for nova_reserverd_host_memory_mb | 20:19 |
| jrosser | so just setting nova_reserverd_host_memory_mb: <value> in user_variables.yml will hopefully do what you need | 20:19 |
| *** jbadiapa has quit IRC | 20:20 | |
| recyclehero | jrosser: I just put nova_virt_type cuz I thought nova_reserved_host_memory_db alone wihtout it parents doesnt make sense | 20:47 |
| jrosser | yeah, understood | 20:48 |
| jrosser | ultimately the thing you will be overrding is whatever is used in the template | 20:48 |
| recyclehero | nova_cpu_allocation_ratio: 2.0, If I want to use placement api later I should override these. but when I set these to 'none' I will get runtime error that it couldnt convert string to float | 20:49 |
| jrosser | and those precendence rules i linked can be useful when deciding if to put things in group/host vars or user_variables | 20:49 |
| jrosser | not sure i follow - placement is mandatory now i think | 20:50 |
| recyclehero | link with #ID | 20:51 |
| recyclehero | https://docs.openstack.org/nova/ussuri/admin/configuration/schedulers.html#usage-scenarios | 20:51 |
| recyclehero | I want to be able to set allocation ratios not as hardcoded but with apis. I think they way OSA default variables sets it take that option out of picture | 20:52 |
| recyclehero | I want to scenario #2 | 20:53 |
| recyclehero | maybe I could omit them from os_nova/defaults | 20:56 |
| recyclehero | or define them in user_variables with no value | 20:57 |
| recyclehero | ? | 20:57 |
| recyclehero | jrosser: I read the link. for overridng nova_virt_type['kvm']['nova_reserverd_host_memory_mb'] in user_variables I should just write nova_reserved_host_memory_mb? | 21:12 |
| *** rfolco has quit IRC | 21:12 | |
| jrosser | yes but do understand that it's not the inner variable in that dict you are actually overriding | 21:13 |
| *** rfolco has joined #openstack-ansible | 21:14 | |
| jrosser | it's an ansible fact of the same name set to the value of that inner variable, by virtue of the virt_type being kvm | 21:14 |
| jrosser | what you are overriding is the variable used in the template | 21:15 |
| jrosser | when you tried to override nova_cpu_allocation_ratio, did you use none or 'none' | 21:16 |
| recyclehero | 'None' :D | 21:18 |
| jrosser | intersting | 21:19 |
| recyclehero | so which one should I use? | 21:19 |
| recyclehero | or u think I shoud use | 21:19 |
| recyclehero | cause for now leaving them empty and doing successful api call to placement actually doesnt change the vcpus in horizon dashboard | 21:20 |
| jrosser | if you get a runtime error then thats really a nova issue i guess, would have to look at the code | 21:21 |
| jrosser | python None is not the same as a string containing 'None' so i wonder if thats even possible via the config file | 21:21 |
| recyclehero | what if i delete them from default/main | 21:22 |
| recyclehero | its like they never existed | 21:22 |
| jrosser | no, the template will still try to put them in the file | 21:22 |
| recyclehero | #cpu_allocation_ratio = <None> | 21:22 |
| recyclehero | this is from https://docs.openstack.org/nova/latest/configuration/sample-config.html | 21:23 |
| jrosser | this is the thing here https://github.com/openstack/nova/blob/a4068802ef2eca5568a58b8f961a521a3a3ddcb5/nova/conf/compute.py#L422 | 21:25 |
| jrosser | the nova code wants you to supply a float | 21:25 |
| *** yann-kaelig has quit IRC | 21:26 | |
| jrosser | and if you don't supply anything the value is None <- that is not a string | 21:26 |
| recyclehero | you mean like "nova_cpu_allocation_ration: " | 21:27 |
| *** spatel has quit IRC | 21:27 | |
| jrosser | no, if the option is not present | 21:28 |
| jrosser | if it is present it has to be a float | 21:28 |
| jrosser | it's an ini format file key=value | 21:28 |
| jrosser | anyway, its late here, enough for today | 21:29 |
| recyclehero | ok have a good night | 21:30 |
| MickyMan77 | jrosser: some update for you... "Starting cinder-volume cluster ceph@ceph-ssd (version 16.1.1)" | 21:30 |
| MickyMan77 | first node is working... :) | 21:30 |
| MickyMan77 | thx for you input.. | 21:31 |
| jrosser | no problem - glad it is fixed! | 21:31 |
| *** cshen has joined #openstack-ansible | 21:47 | |
| *** cshen has quit IRC | 21:51 | |
| *** rfolco has quit IRC | 22:10 | |
| *** rfolco has joined #openstack-ansible | 22:53 | |
| *** macz_ has quit IRC | 23:06 | |
| *** rfolco has quit IRC | 23:13 | |
| *** nurdie has quit IRC | 23:18 | |
| *** jamesdenton has quit IRC | 23:20 | |
| *** jamesdenton has joined #openstack-ansible | 23:20 | |
| *** zigo has quit IRC | 23:23 | |
| *** mgariepy has quit IRC | 23:24 | |
| *** watersj has joined #openstack-ansible | 23:32 | |
| *** pmannidi has quit IRC | 23:33 | |
| *** cshen has joined #openstack-ansible | 23:33 | |
| *** pmannidi has joined #openstack-ansible | 23:33 | |
| *** mgariepy has joined #openstack-ansible | 23:34 | |
| *** cshen has quit IRC | 23:37 | |
| *** MickyMan77 has quit IRC | 23:38 | |
| *** MickyMan77 has joined #openstack-ansible | 23:38 | |
| *** tosky has quit IRC | 23:40 | |
| *** rfolco has joined #openstack-ansible | 23:41 | |
| *** MickyMan77 has quit IRC | 23:47 | |
| *** nurdie has joined #openstack-ansible | 23:52 | |
| *** nurdie has quit IRC | 23:56 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!