| *** jamesden_ has quit IRC | 00:56 | |
| *** jamesdenton has joined #openstack-ansible | 00:57 | |
| *** hindret has quit IRC | 01:52 | |
| *** hindret has joined #openstack-ansible | 01:56 | |
| *** rohit02 has joined #openstack-ansible | 02:21 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: Remove neutron_keepalived_no_track variable https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/770808 | 02:31 |
|---|---|---|
| *** evrardjp has quit IRC | 02:33 | |
| *** evrardjp has joined #openstack-ansible | 02:33 | |
| *** rohit02 has quit IRC | 02:41 | |
| *** rohit02 has joined #openstack-ansible | 02:51 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in playbooks https://review.opendev.org/c/openstack/openstack-ansible/+/777996 | 03:12 |
| openstackgerrit | zhangboye proposed openstack/openstack-ansible-os_nova master: Update master for stable/wallaby https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/784722 | 03:22 |
| openstackgerrit | zhangboye proposed openstack/openstack-ansible-os_neutron master: Update master for stable/wallaby https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/784726 | 04:18 |
| *** bradm has quit IRC | 04:39 | |
| *** logan- has quit IRC | 04:39 | |
| *** jamesdenton has quit IRC | 04:42 | |
| *** logan- has joined #openstack-ansible | 04:42 | |
| *** jamesdenton has joined #openstack-ansible | 04:42 | |
| *** raukadah is now known as chkumar|ruck | 04:49 | |
| *** jamesdenton has quit IRC | 04:53 | |
| *** jamesden_ has joined #openstack-ansible | 04:54 | |
| *** yasemind has joined #openstack-ansible | 05:23 | |
| *** jamesden_ has quit IRC | 05:37 | |
| *** jamesdenton has joined #openstack-ansible | 05:38 | |
| openstackgerrit | zhangboye proposed openstack/openstack-ansible-tests master: Update master for stable/wallaby https://review.opendev.org/c/openstack/openstack-ansible-tests/+/784738 | 06:38 |
| openstackgerrit | zhangboye proposed openstack/openstack-ansible-rsyslog_server master: Update master for stable/wallaby https://review.opendev.org/c/openstack/openstack-ansible-rsyslog_server/+/784739 | 06:41 |
| openstackgerrit | zhangboye proposed openstack/openstack-ansible-repo_server master: Update master for stable/wallaby https://review.opendev.org/c/openstack/openstack-ansible-repo_server/+/784741 | 06:48 |
| openstackgerrit | zhangboye proposed openstack/openstack-ansible-rsyslog_client master: Update master for stable/wallaby https://review.opendev.org/c/openstack/openstack-ansible-rsyslog_client/+/784742 | 06:48 |
| noonedeadpunk | morning | 08:07 |
| CeeMac | morning | 08:08 |
| noonedeadpunk | wtf has happened with collections installation https://zuul.opendev.org/t/openstack/build/ac4dce3eb7ec44f0b2e0e86baf042f90/log/job-output.txt#2173 Feels like related with new ansible release, and that we install ansible along with ansible base somehow... | 08:17 |
| jrosser | i think we might need to update ansible-lint | 08:25 |
| jrosser | https://github.com/ansible-community/ansible-lint/blob/v5.0.6/setup.cfg#L88-L97 | 08:25 |
| jrosser | and only install the 'core' extras, rather than 'community' | 08:25 |
| jrosser | well, maybe that means now you don't get either ansible/ansible-base by default, not sure how extras worked | 08:26 |
| noonedeadpunk | ah | 08:27 |
| noonedeadpunk | ansible-lint[core] | 08:27 |
| noonedeadpunk | ansible-lint[core,yamllint] I guess | 08:27 |
| * jrosser horror last time i updated ansible lint, serious brokenness everywhere :( | 08:28 | |
| noonedeadpunk | yes, I also pretty much afraid of that | 08:28 |
| noonedeadpunk | from other side, my sandbox just passed | 08:29 |
| noonedeadpunk | so, um... maybe broken galaxy mirror? | 08:30 |
| *** sshnaidm|off is now known as sshnaidm | 08:36 | |
| noonedeadpunk | but yeah, sounds like good idea anyway to update ansible-lint. because by default it won't install any ansible which I guess what we need, since we install it anyway inside venv. | 08:38 |
| noonedeadpunk | but our custom rules are broken with it | 08:42 |
| zbr | jrosser: noonedeadpunk: in case you need help with ansible-lint, i am the current maintainer, ping me if needed. | 08:45 |
| noonedeadpunk | zbr: sure:) But I think we just need courage :P | 08:45 |
| jrosser | last time took literally weeks to get all our repos passing again after bumping the version | 08:46 |
| openstackgerrit | Dmitriy Rabotyagov proposed openstack/openstack-ansible-tests master: Bump ansible-lint version https://review.opendev.org/c/openstack/openstack-ansible-tests/+/784751 | 08:49 |
| noonedeadpunk | I think we might also face circulare dependency because of having it both in tests and integrated repo at once (and using both requirements, which will conflict) | 08:49 |
| noonedeadpunk | but maybe it won't be circular, not sure.... | 08:50 |
| zbr | v5 became quite picky about code layout, but is very friendly if you follow official guidelines, knowing to auto-install deps. | 08:53 |
| zbr | for sure that is not enough, you must specify core or community extra, or manually mention ansible version to install. | 08:54 |
| zbr | it may worth doing the same I did with other projects: adopting your custom rules into the linter itself. | 08:56 |
| noonedeadpunk | yeah, already pushed https://review.opendev.org/c/openstack/openstack-ansible-tests/+/784751 - we jsut used old import style | 08:57 |
| zbr | i did this for two zuul rules and one from tripleo, makes maintenance much easier. | 08:57 |
| noonedeadpunk | But what broke for us - were new implemented rules:) ofc it's great to fix things, jsut matter of we ready for that amount of work or not:) | 08:57 |
| noonedeadpunk | our rules are pretty specific and doesn't make sense for anybody | 08:58 |
| zbr | that is why you can always use "skip_list" and mention that it was skipped during upgrade. | 08:58 |
| zbr | you can always delay implementing a rule, in fact that is what I recommend | 08:58 |
| jrosser | there were also some really unhelpful contradictory rules | 08:59 |
| noonedeadpunk | any changelog to know what rules have been implemented? | 08:59 |
| zbr | otherwise the bumping efforth can overgrown | 08:59 |
| zbr | https://github.com/ansible-community/ansible-lint/releases | 08:59 |
| jrosser | trouble is we bump the version in one place in one repo | 08:59 |
| jrosser | then that applies to like 50 other repos linter jobs | 08:59 |
| zbr | well, that will never work well | 09:00 |
| zbr | one of the reasons why I use pre-commit to manage linters, including ansible-lint in almost all repos. | 09:00 |
| noonedeadpunk | otherwise we would have to bump requirements in 50 places everytime :) | 09:01 |
| zbr | yes, and that is ok. usually you bump linter when you need, not on each version release. | 09:01 |
| zbr | now you have a case where you may never be able to upgrade it due to the amount of side effects. | 09:02 |
| noonedeadpunk | I think it's really a tradeoff... | 09:02 |
| noonedeadpunk | because amount of work is really comparable in every scenario | 09:03 |
| zbr | yes it is, and for years I did had the same opinion that centralized is better. | 09:03 |
| zbr | i think that the author of pre-commit tool have some good talks/presentations on why local is better. | 09:04 |
| zbr | usually I bump all linters at once per repo. so the CR-economy is done in a different way | 09:05 |
| zbr | if 10 out of your 50 repos are (less/un)maintained or even broken, who is going to do the leg. work to fix them? | 09:05 |
| zbr | you endup having a case were less important repos to block progress on critical ones | 09:06 |
| noonedeadpunk | I'm not sure that's the case for us | 09:07 |
| zbr | there is also another aspect, more of a guess: my impression is that while one person may know how to fix issues with some repo, probably nobody knows very well all of them, so they may not be able to finish the work. | 09:07 |
| zbr | anyway, if you find bugs or need some changes that could ease your work let me know. | 09:08 |
| noonedeadpunk | yeah, well, I agree here, and local testing is really more clear, but we can't test role anyway without dependencies for the others | 09:08 |
| noonedeadpunk | yeah, sure! | 09:09 |
| noonedeadpunk | thanks for the talk! | 09:09 |
| zbr | with v5 you have to options for deps: you can mock them in config or install them using requirements.yml | 09:09 |
| noonedeadpunk | zbr: btw, I was wondering regarding ansible plugin for vscode you're maintaing. While I have it installed and enabled, I don't see it doing anything. | 09:22 |
| noonedeadpunk | the only thing I see in the output is `Ansible Language extension is now active` | 09:22 |
| noonedeadpunk | `"ansible.validate.executablePath"` set to ansible-lint binary in venv | 09:23 |
| zbr | likely the playbooks are not inside playbooks/ folder. | 09:23 |
| zbr | usually if you can run ansible-lint from vscode terminal, it will work fine. | 09:23 |
| noonedeadpunk | um, so it's not testing like tasks like dropped ansible extension for microsoft did? | 09:24 |
| zbr | it does if file pattern matches one from https://github.com/ansible-community/vscode-ansible/blob/master/package.json#L103-L163 | 09:25 |
| zbr | this extension relies on YAML by RedHat one to do the work but for the moment the file patterns are limited. I am waiting for them to release a fix. | 09:26 |
| zbr | mainly they need to switcht the pattern matching library as the original one bit limited. | 09:26 |
| noonedeadpunk | I think they should match... I tried to do stupid change in https://opendev.org/openstack/openstack-ansible-os_trove/src/branch/master/tasks/main.yml#L95 (by replacing import_tasks with import just for test) | 09:27 |
| zbr | let me test myself... | 09:27 |
| zbr | it did pick the correct schema for this file, you can see it at the top of the file | 09:30 |
| zbr | but the tasks schema is incomplete at this moment | 09:30 |
| noonedeadpunk | hm, may it be because I installed ansible-lint inside venv? And it tries to launch system ansible for test? | 09:30 |
| zbr | it does not matter how you install ansible-lint as long you give the path to it. | 09:31 |
| zbr | indeed, you must be sure that you also have ansible installed inside the same venv. | 09:32 |
| noonedeadpunk | I do have it... | 09:32 |
| zbr | having it in system should not be an issue,... in theory. | 09:32 |
| noonedeadpunk | well, when I just running ~/.virtualenvs/ansible/bin/ansible-lint I get `ERROR FATAL: Ansible CLI (2.9.6) and python module (2.10.7) versions do not match.` | 09:33 |
| zbr | probably I should add some code to enforce running ansible from the same path as ansible-lint, to avoid this kind of accident | 09:33 |
| noonedeadpunk | (without actiuvating venv) | 09:33 |
| noonedeadpunk | so wondered if vscode may face the same thing | 09:34 |
| zbr | in theory calling a command installed inside a venv, should activate the venv transparently, so a subprocess would run inside one. | 09:34 |
| noonedeadpunk | Yeah, I'd expect that :) | 09:34 |
| zbr | based on the error you are reporting, this seems to be the issue | 09:34 |
| noonedeadpunk | 2.9.6 is /usr/bin/ansible version provided by system package | 09:35 |
| zbr | so you say that running lint with manually activating venv works, but directly does not due to error above. | 09:35 |
| noonedeadpunk | yep | 09:35 |
| zbr | which tool/version have you used to create the venv? that may be essential. | 09:36 |
| zbr | it can be virtualenv, venv,... | 09:36 |
| noonedeadpunk | um... I think `python3 -m venv` | 09:36 |
| zbr | can you afford to test virtualenv instead? | 09:36 |
| noonedeadpunk | sure | 09:36 |
| zbr | i was told that venv may be lacking lots of things, especially when compared with the rewritten virtualenv | 09:37 |
| *** SiavashSardari has joined #openstack-ansible | 09:38 | |
| zbr | added https://github.com/ansible-community/vscode-ansible/issues/57 | 09:40 |
| noonedeadpunk | same thing for virtualenv | 09:40 |
| zbr | please raise a bug on the linter itself, i will try to fix it tomorrow. | 09:42 |
| zbr | today i am i vacation and decided to only work on https://github.com/pycontribs/mk :D | 09:44 |
| zbr | this bug is in the linter but is more likely to impact vscode-ansible users than others. should be easy to fix anyway | 09:45 |
| noonedeadpunk | zbr: https://github.com/ansible-community/ansible-lint/issues/1507 | 09:52 |
| noonedeadpunk | sure, no hurry here I guess | 09:53 |
| noonedeadpunk | you just that that we can ping you, so I decided not to hesitate:) | 09:53 |
| zbr | no worry, that is a serious bug. | 09:54 |
| noonedeadpunk | zbr: maybe you have a quick solution here. How *lintables should look like here https://github.com/ansible-community/ansible-lint/blob/v5.0.6/src/ansiblelint/runner.py#L39 ? Asking in context of the unit test https://opendev.org/openstack/openstack-ansible-tests/src/branch/master/ansible-lint/test/TestAPTRepositoryCacheUpdateRule.py#L16 where we are missing to pass some arguments | 10:40 |
| noonedeadpunk | aha, found https://github.com/ansible-community/ansible-lint/blob/master/test/TestLocalContent.py | 10:43 |
| openstackgerrit | Dmitriy Rabotyagov proposed openstack/openstack-ansible-tests master: Bump ansible-lint version https://review.opendev.org/c/openstack/openstack-ansible-tests/+/784751 | 10:46 |
| *** luksky has joined #openstack-ansible | 10:46 | |
| *** tosky has joined #openstack-ansible | 11:29 | |
| *** jamesdenton has quit IRC | 11:36 | |
| *** jamesden_ has joined #openstack-ansible | 11:36 | |
| *** yasemind has quit IRC | 11:51 | |
| openstackgerrit | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Bump ansible version to 2.10.7 https://review.opendev.org/c/openstack/openstack-ansible/+/781505 | 12:00 |
| openstackgerrit | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Remove note about metal/horizon compatability https://review.opendev.org/c/openstack/openstack-ansible/+/771573 | 12:02 |
| openstackgerrit | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Adjust magnum CI image https://review.opendev.org/c/openstack/openstack-ansible/+/774243 | 12:07 |
| openstackgerrit | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Remove duplicated defenition of tempest_run_concurrency https://review.opendev.org/c/openstack/openstack-ansible/+/767430 | 12:08 |
| *** rh-jelabarre has joined #openstack-ansible | 12:10 | |
| *** tosky has quit IRC | 12:13 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Fix facts collection for LXC hosts https://review.opendev.org/c/openstack/openstack-ansible/+/784112 | 12:18 |
| openstackgerrit | Merged openstack/openstack-ansible master: Install wget binary https://review.opendev.org/c/openstack/openstack-ansible/+/781506 | 12:18 |
| openstackgerrit | Merged openstack/openstack-ansible master: Use ansible_facts[] instead of fact variables in docs https://review.opendev.org/c/openstack/openstack-ansible/+/778000 | 12:19 |
| *** tosky has joined #openstack-ansible | 12:19 | |
| *** d34dh0r53 has joined #openstack-ansible | 12:26 | |
| *** jbadiapa has joined #openstack-ansible | 12:35 | |
| *** spatel_ has joined #openstack-ansible | 13:08 | |
| *** spatel_ is now known as spatel | 13:08 | |
| *** jamesden_ has quit IRC | 13:08 | |
| *** jamesdenton has joined #openstack-ansible | 13:09 | |
| *** SiavashSardari has quit IRC | 13:29 | |
| openstackgerrit | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_trove master: Update trove configuration https://review.opendev.org/c/openstack/openstack-ansible-os_trove/+/784571 | 13:33 |
| spatel | jamesdenton morning | 13:43 |
| *** tosky has quit IRC | 13:43 | |
| jamesdenton | hi | 13:44 |
| spatel | I need your help to understand L3 routing in OVN | 13:44 |
| spatel | I am trying to create DVR where all router get schedule on all compute nodes and traffic will go in/out directly from compute | 13:45 |
| jamesdenton | k | 13:45 |
| *** tosky has joined #openstack-ansible | 13:45 | |
| spatel | I am seeing strange behavior where my traffic going out/in using only 1 compute node out of two | 13:45 |
| jamesdenton | IIRC, default behavior for OSA to is make all compute nodes gateway nodes | 13:45 |
| spatel | its acting like centralized gateway | 13:46 |
| spatel | My routers got scheduled on both compute nodes but only 1 compute gateway is active and other is standby | 13:46 |
| spatel | i want true ACTIVE-ACTIVE | 13:47 |
| jamesdenton | do you see it listed active-standby somewhere? or is that just the behavior you're seeing | 13:47 |
| spatel | Behavior | 13:47 |
| spatel | when i attached floating IP to vm (FIP) then it should set external_mac address which is empty in my case. (document saying if its empty then you have centralized gateway routing) | 13:48 |
| spatel | If i manually set external_mac address then everything works fine like distributed gateway DVR. so question is how that external_mac will get setup itself.. | 13:49 |
| spatel | Redhat document saying that patch that issue but not showing how (i don't have redhat support to look) | 13:49 |
| jamesdenton | what is the url for redhat, i can check | 13:49 |
| spatel | check this Redhat ticket - https://bugzilla.redhat.com/show_bug.cgi?id=1700043 | 13:50 |
| openstack | bugzilla.redhat.com bug 1700043 in python-networking-ovn "[RHOSP13] OVN-DVR Floating IP is not distributed in each compute nodes and external_mac column not binded with MAC address" [Urgent,Closed: duplicate] - Assigned to amuller | 13:50 |
| spatel | May be we are running older version of network-ovn ? | 13:50 |
| *** tosky has quit IRC | 13:52 | |
| *** tosky has joined #openstack-ansible | 13:54 | |
| jamesdenton | "DVR will be used for floating IPs if the ovn / enable_distributed_floating_ip flag is configured to True in the neutron server configuration" | 13:55 |
| jamesdenton | https://docs.openstack.org/networking-ovn/latest/faq/index.html | 13:55 |
| jamesdenton | can you check to see if that is set? and maybe try setting it? | 13:55 |
| spatel | :) | 14:03 |
| spatel | i am talking to someone on neutron channel and same thing.. | 14:03 |
| spatel | i don't have that option set in neutron | 14:04 |
| spatel | jamesdenton look like we need to add that in OSA as a variable :) | 14:05 |
| spatel | or did i miss something | 14:05 |
| spatel | also we don't have OVN section in neutron.conf file should i be doing in ml2.conf ? | 14:06 |
| jamesdenton | the ovn implementation in OSA is a few years old now, so there's likely been changes/enhancements not accounted for | 14:08 |
| jamesdenton | you can try adding a neutron.conf override with an [ovn] section and the config option described there | 14:08 |
| spatel | i am adding in ml2.conf | 14:09 |
| spatel | jamesdenton that works!!! external_mac : "fa:16:3e:2d:7e:fa" | 14:19 |
| jamesdenton | woohoo! | 14:20 |
| jamesdenton | so, that should probably be a configurable option added to the template ,and default to true/false | 14:20 |
| spatel | Yes do you want me to cut that patch ? | 14:21 |
| jamesdenton | go for it | 14:21 |
| spatel | i am planning to add that option in ml2.conf because that is the bridge of OVN | 14:22 |
| spatel | trying to keep all OVN config in one place instead of multiple location | 14:22 |
| spatel | do you have any other thought ? | 14:22 |
| jamesdenton | well, that FAQ seems to imply it should be in the neutron.conf since it is a 'site wide' config not local, and ml2_conf is more localized IMO | 14:24 |
| spatel | okay! will add in neutron then | 14:24 |
| spatel | are you checking #neutron channel.. i am talking there. I am still seeing odd behavior | 14:30 |
| spatel | jamesdenton ^ | 14:30 |
| spatel | even after setting that option behavior is still centralized :( | 14:39 |
| spatel | let me debug and see | 14:39 |
| jamesdenton | yeah, not sure what to say about that right now | 14:41 |
| spatel | jamesdenton problem solved :) | 14:48 |
| spatel | you are right, that option need to be set in neutron.conf | 14:48 |
| jamesdenton | oh alright then | 14:48 |
| jamesdenton | glad its working now | 14:49 |
| spatel | let me cut patch before i get busy somewhere else | 14:49 |
| *** macz_ has joined #openstack-ansible | 14:49 | |
| openstackgerrit | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Listen RabbitMQ on all available container networks https://review.opendev.org/c/openstack/openstack-ansible/+/784780 | 14:50 |
| *** macz_ has quit IRC | 14:51 | |
| *** macz_ has joined #openstack-ansible | 14:51 | |
| openstackgerrit | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_trove master: [doc] Document how to use separate RabbitMQ cluster https://review.opendev.org/c/openstack/openstack-ansible-os_trove/+/784781 | 14:52 |
| spatel | jamesdenton do you think we should keep default option true? | 15:14 |
| spatel | Let me keep false because that is legacy way (centralized FIP) | 15:17 |
| jamesdenton | let's keep it false, and update the docs. i think there are some that like the centralized approach | 15:18 |
| spatel | Yes | 15:20 |
| openstackgerrit | Satish Patel proposed openstack/openstack-ansible-os_neutron master: add enable/disable option for OVN-DVR https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/784782 | 15:25 |
| spatel | jamesdenton let me know if anything missing in this patch | 15:26 |
| jamesdenton | just some doc edits | 15:26 |
| noonedeadpunk | any idea why I don't see such option/section in https://docs.openstack.org/neutron/latest/configuration/neutron.html ? | 15:27 |
| noonedeadpunk | I think config reference is just missing it? | 15:28 |
| *** dmsimard6 has joined #openstack-ansible | 15:28 | |
| noonedeadpunk | but eventually, I don't see there a need to implement that variable, unless we want non-default behaviour to become default | 15:28 |
| *** dmsimard has quit IRC | 15:29 | |
| *** dmsimard6 is now known as dmsimard | 15:29 | |
| jamesdenton | so, rather than set a var just hardcode to false and let the user override it? | 15:29 |
| noonedeadpunk | yep | 15:29 |
| noonedeadpunk | and document it? | 15:29 |
| jamesdenton | sure, thats fine too | 15:29 |
| jamesdenton | yep | 15:29 |
| jamesdenton | spatel you get that? | 15:29 |
| noonedeadpunk | default is false btw https://opendev.org/openstack/neutron/src/branch/master/neutron/conf/plugins/ml2/drivers/ovn/ovn_conf.py#L118 | 15:30 |
| noonedeadpunk | but I'm not sure that is should be put in neutron.conf rather then plugins/ml2? | 15:31 |
| noonedeadpunk | are you sure folks about neutron.conf? | 15:31 |
| noonedeadpunk | https://docs.openstack.org/neutron/latest/configuration/ovn.html#ovn.enable_distributed_floating_ip | 15:32 |
| jamesdenton | Well, he tried ml2_conf first and it didn't work as expected, but in neutron.conf it did? The FAQ here implies neutron.conf, but i don;'t know for sure: https://docs.openstack.org/neutron/latest/ovn/faq/index.html | 15:32 |
| jamesdenton | AFAIK we do not have an ovn.ini | 15:33 |
| noonedeadpunk | in config reference it seems to be ovn.ini but we really don't have it... | 15:33 |
| spatel | hmmm | 15:34 |
| noonedeadpunk | well, if it worked in neutron.conf - then fine... | 15:34 |
| * noonedeadpunk has no idea | 15:34 | |
| jamesdenton | well, we may want to revisit the way ovn conf is laid down if the new way is an independent config file, then fine | 15:34 |
| spatel | yes i don't think we need more cookie cut (keep everything in one place is good) | 15:34 |
| jamesdenton | well, it's nice to be consistent with upstream if possible | 15:35 |
| noonedeadpunk | I'm fine with it until it works:) | 15:35 |
| jamesdenton | spatel you have time to deploy a devstack? :) | 15:35 |
| spatel | sure i can do that | 15:35 |
| noonedeadpunk | we can look in neutron logs :p | 15:35 |
| spatel | jamesdenton what do you want me to test with devstack? | 15:36 |
| jamesdenton | just wanted to see what files get laid down for ovn config. since it's been merged with neutron | 15:37 |
| noonedeadpunk | https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_847/784058/1/check/neutron-ovn-tempest-ovs-release/847178f/controller/logs/etc/neutron/ | 15:37 |
| jamesdenton | we have stuff in ml2_conf that now may be better suited for ovn.ini or whatever | 15:37 |
| spatel | I will try devstack and see how does it work | 15:37 |
| noonedeadpunk | are we not good enough with how neutron does CI? I believe that's devstack anyway? | 15:38 |
| jamesdenton | well, it would appear CI doesn't match docs | 15:39 |
| jamesdenton | i don't see why the option wouldn't have worked in ml2_conf, honestly | 15:39 |
| spatel | look like OVN has tons of knobs, may be someday we have to do ovn.ini to keep them in one place - https://docs.openstack.org/neutron/ussuri/configuration/ovn.html | 15:41 |
| spatel | or do it today | 15:41 |
| spatel | question is are these option specific for ovn-northd or neutron-server also? | 15:42 |
| jamesdenton | they should get combined, since you invoke neutron-server w/ the plugin config file(s) | 15:42 |
| *** rohit02 has quit IRC | 15:44 | |
| *** rohit02 has joined #openstack-ansible | 15:44 | |
| spatel | anyway i will do devstack and see how does it laid out structure | 15:44 |
| spatel | I also need to test ovsdb-server clusting | 15:46 |
| spatel | noonedeadpunk in your comments what document you want me to write up? | 15:47 |
| spatel | release notes? | 15:47 |
| noonedeadpunk | jamesdenton: once you;re here, can I ask you to vote on https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/783047 ? | 15:47 |
| noonedeadpunk | spatel: um, no. eventually this patch is not needed at all, since you can easily use overrides to set that variable | 15:48 |
| noonedeadpunk | probably some note about that option why and how to set it in https://docs.openstack.org/openstack-ansible-os_neutron/latest/app-ovn.html ? | 15:48 |
| spatel | oh! i know what you saying | 15:48 |
| spatel | make sense | 15:49 |
| spatel | I will add something in NOTES section, is that ok or you want specific section for DVR? | 15:50 |
| spatel | I think new DVR section would be good to catch eyeballs :) | 15:52 |
| *** rohit02 has quit IRC | 16:22 | |
| *** jamesdenton has quit IRC | 16:33 | |
| *** jamesden_ has joined #openstack-ansible | 16:33 | |
| spotz | spatel noonedeadpunk - add me to the docs when the patch is up and I'll look | 16:49 |
| spatel | spotz sure will do | 16:49 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Do not use imports inside included task https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/783047 | 17:19 |
| *** spatel has quit IRC | 18:11 | |
| *** spatel_ has joined #openstack-ansible | 18:22 | |
| *** spatel_ is now known as spatel | 18:22 | |
| *** jamesden_ has quit IRC | 18:23 | |
| *** jamesdenton has joined #openstack-ansible | 18:23 | |
| *** dmsimard has quit IRC | 19:19 | |
| *** dpawlik0 has quit IRC | 19:42 | |
| *** dpawlik3 has joined #openstack-ansible | 19:42 | |
| *** dmsimard has joined #openstack-ansible | 19:47 | |
| *** dpawlik3 has quit IRC | 19:57 | |
| *** dpawlik0 has joined #openstack-ansible | 19:58 | |
| *** dmsimard has quit IRC | 20:00 | |
| *** dpawlik6 has joined #openstack-ansible | 20:22 | |
| *** dpawlik0 has quit IRC | 20:23 | |
| *** rh-jelabarre has quit IRC | 20:27 | |
| *** rh-jelabarre has joined #openstack-ansible | 20:27 | |
| *** LowKey has quit IRC | 20:38 | |
| *** LowKey has joined #openstack-ansible | 20:46 | |
| *** dmsimard has joined #openstack-ansible | 21:05 | |
| *** dpawlik6 has quit IRC | 21:26 | |
| *** dpawlik5 has joined #openstack-ansible | 21:26 | |
| *** spatel has quit IRC | 21:28 | |
| *** dpawlik5 has quit IRC | 21:56 | |
| *** dpawlik0 has joined #openstack-ansible | 21:56 | |
| *** dpawlik5 has joined #openstack-ansible | 22:07 | |
| *** dpawlik0 has quit IRC | 22:08 | |
| *** dpawlik5 has quit IRC | 22:11 | |
| *** dpawlik9 has joined #openstack-ansible | 22:11 | |
| *** luksky has quit IRC | 22:42 | |
| *** bradm has joined #openstack-ansible | 23:05 | |
| *** tosky has quit IRC | 23:19 | |
| *** dpawlik4 has joined #openstack-ansible | 23:26 | |
| *** dpawlik9 has quit IRC | 23:26 | |
| *** macz_ has quit IRC | 23:35 | |
| *** dpawlik4 has quit IRC | 23:41 | |
| *** dpawlik5 has joined #openstack-ansible | 23:41 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!