| opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: Fix Skyline API Server TLS configuration https://review.opendev.org/c/openstack/kolla-ansible/+/909329 | 06:40 |
|---|---|---|
| guesswhat[m] | Is magnum-cluster-api-proxy implemented in Kolla ? For reaching clusters without floating ip ? OSA and Amosphere have this implementation. Thanks | 07:54 |
| mnasiadka | guesswhat[m]: no, but feel free to contribute that :) | 07:59 |
| opendevreview | Marius proposed openstack/kolla master: Update prometheus-v2-server to version 2.50.1. https://review.opendev.org/c/openstack/kolla/+/913329 | 08:05 |
| guesswhat[m] | mnasiadka: found this https://opendev.org/openstack/openstack-ansible-ops/commit/e983053d9808ef41aad519ed2c975fe088cf5767 , but looks complicated tho | 08:07 |
| opendevreview | Marius proposed openstack/kolla master: Updated letsencrypt-lego to version v4.15.0. https://review.opendev.org/c/openstack/kolla/+/913316 | 08:12 |
| opendevreview | Michal Nasiadka proposed openstack/kolla-ansible master: Run ML2/OVS agents processes in separate containers https://review.opendev.org/c/openstack/kolla-ansible/+/864780 | 08:14 |
| opendevreview | Merged openstack/kolla-ansible master: Allow customizing of Skyline configuration https://review.opendev.org/c/openstack/kolla-ansible/+/905861 | 08:24 |
| opendevreview | Dr. Jens Harbott proposed openstack/kolla-ansible stable/2023.2: octavia: add missing bool to the enable_octavia_jobboard parameter https://review.opendev.org/c/openstack/kolla-ansible/+/912759 | 09:14 |
| opendevreview | Merged openstack/kolla-ansible master: CI: Increase galera node timeouts https://review.opendev.org/c/openstack/kolla-ansible/+/912725 | 09:22 |
| opendevreview | Matt Crees proposed openstack/kayobe master: CI: Only migrate RMQ queues during SLURP https://review.opendev.org/c/openstack/kayobe/+/912778 | 09:30 |
| opendevreview | Matt Crees proposed openstack/kolla-ansible master: CI: Only migrate RMQ queues during SLURP https://review.opendev.org/c/openstack/kolla-ansible/+/909971 | 09:31 |
| opendevreview | Uwe Jäger proposed openstack/kolla-ansible master: Update configuration to enable more services in Skyline Console https://review.opendev.org/c/openstack/kolla-ansible/+/909482 | 09:59 |
| opendevreview | Sven Kieske proposed openstack/kolla-ansible master: fix confusing jammy upgrade docs https://review.opendev.org/c/openstack/kolla-ansible/+/913426 | 10:16 |
| opendevreview | Merged openstack/kayobe stable/2023.2: Fix link in release note https://review.opendev.org/c/openstack/kayobe/+/912749 | 10:20 |
| opendevreview | Merged openstack/kayobe stable/2023.1: Fix link in release note https://review.opendev.org/c/openstack/kayobe/+/912750 | 10:20 |
| opendevreview | Merged openstack/kayobe stable/zed: Fix link in release note https://review.opendev.org/c/openstack/kayobe/+/912751 | 10:20 |
| SvenKieske | guys please, bash has no real boolean value support like you think it does. the only boolean values it has are literally "0" and "1". strings are strings in bash, even if they contain the letters "yes, no, true, false" | 10:43 |
| SvenKieske | for context, I'm talking about this review: https://review.opendev.org/c/openstack/kolla-ansible/+/909971 | 10:44 |
| frickler | we can "borrow" from devstack I guess https://opendev.org/openstack/devstack/src/branch/master/functions-common#L207-L237 | 10:46 |
| SvenKieske | the doc comment there makes me very nervous, "if var is unset use the value of the second argument as a default"? | 10:48 |
| SvenKieske | also you would have to audit all the bash code before introducing it to match this function | 10:48 |
| SvenKieske | not sure what is easier, teach people bash or do this :D | 10:48 |
| SvenKieske | that code also does some weird stuff, like that grep for xtrace..?is that copied from some posix stuff? | 10:49 |
| SvenKieske | that's quite a rabbit hole if you follow the commits (don't! I warned you! :D ) | 10:52 |
| SvenKieske | frickler: this also does not address the fact people want to do "if $var; then" in bash, this is just broken usage of bash and can't be fixed. you need to either change the language (use python?) or change the people. both are hard :) | 10:58 |
| opendevreview | Merged openstack/kolla master: mariadb-clustercheck: Add a script handling signals https://review.opendev.org/c/openstack/kolla/+/904981 | 11:40 |
| opendevreview | Merged openstack/kolla-ansible stable/2023.1: rabbitmq: bump wait timeout to 60 seconds https://review.opendev.org/c/openstack/kolla-ansible/+/912080 | 11:43 |
| opendevreview | Merged openstack/kolla-ansible stable/2023.2: rabbitmq: bump wait timeout to 60 seconds https://review.opendev.org/c/openstack/kolla-ansible/+/912079 | 11:43 |
| opendevreview | Merged openstack/kolla-ansible stable/zed: rabbitmq: bump wait timeout to 60 seconds https://review.opendev.org/c/openstack/kolla-ansible/+/912741 | 11:43 |
| mnasiadka | SvenKieske: I'm not a fan of bash usage in kolla-ansible, we should think of moving it more to ansible and splitting out tests/run.yml into something more sensible | 11:47 |
| opendevreview | Marius proposed openstack/kolla master: Updated letsencrypt-lego to version v4.15.0. https://review.opendev.org/c/openstack/kolla/+/913316 | 11:50 |
| opendevreview | Michal Arbet proposed openstack/kolla master: Bump proxysql to 2.6.x https://review.opendev.org/c/openstack/kolla/+/912731 | 11:53 |
| SvenKieske | mnasiadka: yeah I'm generally no fan of bash anywhere :) | 11:54 |
| opendevreview | Alex Welsh proposed openstack/kolla-ansible master: WIP: Automate prometheus blackbox endpoint config https://review.opendev.org/c/openstack/kolla-ansible/+/912420 | 12:17 |
| halomiva | hey, is something missing from this patchset or it can be merged? https://review.opendev.org/c/openstack/ansible-collection-kolla/+/910751 | 12:53 |
| halomiva | hey, is something missing from this patchset or it can be merged? https://review.opendev.org/c/openstack/ansible-collection-kolla/+/910751 | 12:54 |
| kevko | halomiva: done | 12:59 |
| halomiva | kevko: thanks | 13:00 |
| opendevreview | Will Szumski proposed openstack/kayobe master: Add podman support https://review.opendev.org/c/openstack/kayobe/+/909686 | 13:06 |
| opendevreview | Martin Hiner proposed openstack/kolla-ansible master: Add action for getting container names list https://review.opendev.org/c/openstack/kolla-ansible/+/905837 | 13:22 |
| opendevreview | Martin Hiner proposed openstack/kolla-ansible master: Add container engine migration scenario https://review.opendev.org/c/openstack/kolla-ansible/+/836941 | 13:22 |
| dawudm | Would appreciate if we could get this workflow +1 please https://review.opendev.org/c/openstack/kolla-ansible/+/913320 | 13:49 |
| opendevreview | Merged openstack/kolla-ansible stable/2023.2: octavia: add missing bool to the enable_octavia_jobboard parameter https://review.opendev.org/c/openstack/kolla-ansible/+/912759 | 13:51 |
| SvenKieske | dawudm: all I could do was -1 review :/ | 13:59 |
| opendevreview | Merged openstack/ansible-collection-kolla master: Bump min. version of docker-py https://review.opendev.org/c/openstack/ansible-collection-kolla/+/910751 | 14:09 |
| opendevreview | Sven Kieske proposed openstack/kolla-ansible master: Run shellcheck on scripts https://review.opendev.org/c/openstack/kolla-ansible/+/882497 | 14:14 |
| opendevreview | Mark Goddard proposed openstack/kolla-ansible master: Revert "podman: install "rich" dependency" https://review.opendev.org/c/openstack/kolla-ansible/+/902230 | 14:17 |
| opendevreview | Mark Goddard proposed openstack/kolla-ansible master: Remove "distutils" library https://review.opendev.org/c/openstack/kolla-ansible/+/857003 | 14:19 |
| opendevreview | Matt Crees proposed openstack/kayobe master: Bump KA Ansible versions to match new defaults https://review.opendev.org/c/openstack/kayobe/+/913571 | 14:22 |
| opendevreview | Matt Crees proposed openstack/kayobe master: Bump up Ansible supported versions to 8.x/9.x https://review.opendev.org/c/openstack/kayobe/+/910513 | 14:23 |
| opendevreview | Matt Crees proposed openstack/kayobe master: Support installing latest requirements on upgrade https://review.opendev.org/c/openstack/kayobe/+/790921 | 14:25 |
| Lockesmith | When I deploy a cluster using kolla-ansible, if network_interface is on a vlan, then none of my VMs are able to access my vip. They can access other machines within that vlan, just not my VIP. Is there something in my globals I need to configure to allow this? | 14:40 |
| jovial | Lockesmith, do you need the host interface to be on a bridge? Something like: br0.<api vlan> and br0 gets plugged into OVS. | 14:43 |
| opendevreview | Will Szumski proposed openstack/kayobe-config-dev master: Replace docker group with container-engine https://review.opendev.org/c/openstack/kayobe-config-dev/+/910017 | 14:49 |
| Lockesmith | jovial: I'm not sure what I need, that's what I'm trying to get to the bottom of. Currently I just have the network_interface and neutron_external_interface. Both are pointed directly at physical interfaces currently (not doing any tagging) and all traffic appears to work, except that my VMs are unable to reach my VIP. But they can reach any other address on that network. | 14:52 |
| Lockesmith | And that's what's got me stumped. | 14:52 |
| jovial | Another thing to watch out for is asymmetric routing if they get there via a router | 14:55 |
| Lockesmith | That may very well be what's happenning. They are getting there via a router. I'll have to start looking that direction. | 14:59 |
| jovial | Is it a router defined in Openstack and runs on the control nodes? I think that is when you need the bridge. | 15:02 |
| Lockesmith | There is a router defined in Openstack for the project, but the traffic also leaves that network and has my physical router to deal with when reaching anything on my physical network. | 15:04 |
| Lockesmith | And let me just apologize now if I'm not following or any of my answers are poorly structured. My networking is weak. | 15:05 |
| opendevreview | Will Szumski proposed openstack/ansible-collection-kolla master: Add feature to use binary version of crun https://review.opendev.org/c/openstack/ansible-collection-kolla/+/911422 | 15:47 |
| opendevreview | Sven Kieske proposed openstack/kolla-ansible master: Run shellcheck on scripts https://review.opendev.org/c/openstack/kolla-ansible/+/882497 | 16:04 |
| mnasiadka | SvenKieske: did we agree anywhere to move to using shellcheck? | 16:21 |
| mnasiadka | that mount of changes in that patchset is a drama | 16:22 |
| mnasiadka | *amount | 16:22 |
| opendevreview | Will Szumski proposed openstack/ansible-collection-kolla master: Document known issue with podman on Ubuntu https://review.opendev.org/c/openstack/ansible-collection-kolla/+/911422 | 16:23 |
| SvenKieske | well I asked during PTG and no one said "no". to be honest this change is not an MVP but also fixes quite some bugs/streamlines the usage | 16:25 |
| SvenKieske | why is it "a drama"? what does that even mean? | 16:25 |
| mnasiadka | a drama to review, it changes so many files - that it will create merge conflicts on all other patches changing anything in these files | 16:27 |
| mnasiadka | so all other changes will need to be rebased, reworked, you name it | 16:27 |
| mnasiadka | if we don't plan to backport that, it will make other backports difficult | 16:27 |
| SvenKieske | mnasiadka: most of the changes are really just variable quoting and most of the rest is resolving issues that arose due to people relying on accidental correct splitting of their stuff. | 16:27 |
| SvenKieske | how where other linters introduced? can I look maybe at something to make it less invasive? | 16:28 |
| mnasiadka | While I might understand the changes in scripts used on real deployments - the ones in tests/* are not helping anybody | 16:28 |
| SvenKieske | well if you don't lint your tests, people will write buggy tests :) imho tests should have the same code quality, if not higher, than "real deployments" because they are used to test the very code that does the real deployments? | 16:29 |
| SvenKieske | so the tests must really work, no? if they don't work and silently pass because they are broken we really don't need the test. one small way to catch buggy tests is use linters | 16:30 |
| SvenKieske | and I agree, it would be better to just port all the bash testing stuff to something like tempest, but I guess that is even more work? | 16:30 |
| opendevreview | Will Szumski proposed openstack/ansible-collection-kolla master: Document known issue with podman on Ubuntu https://review.opendev.org/c/openstack/ansible-collection-kolla/+/911422 | 16:31 |
| SvenKieske | so if we have a clear plan with a deadline to remove 99% of bash usage I'm fine with no bash linting. the status quo is really not maintainable, only by ignoring the problem imho. | 16:33 |
| SvenKieske | unless there is such a plan I want to improve the bash code, making it more reliable and resilient. and quoting your bash vars is stuff that is 101 that is in every bash best practice guide. | 16:34 |
| SvenKieske | mnasiadka: for stable/zed there are 7 changesets that touch "*.sh" files in total: https://review.opendev.org/q/project:openstack/kolla-ansible+branch:stable/zed+file:%22%5E.*%5C.sh%24%22+-status:abandoned | 16:38 |
| mnasiadka | Personally I would prefer getting rid of those bash scripts or minimising their amount | 16:39 |
| SvenKieske | yeah, we already agreed upon that multiple times. I believe someone voluntered for introducing tempest during C cycle, if my memory serves right? | 16:40 |
| mnasiadka | And I don't care if they get linted or not - as long as they work | 16:40 |
| mnasiadka | Nobody else than CI uses those, introducing more rules on scripts that just work doesn't seem like a good use of anybody's time | 16:41 |
| mnasiadka | but if other cores feel that it's needed - sure, they will review your patch and get that merged | 16:41 |
| SvenKieske | if they all "work" we would have no flaky CI, ever :) could we please stop viewing these problems in such a minimalist way? There's a spectrum between "works" and "doesn't work" which most of the time is "has bugs in edge cases like networked services and timeouts" | 16:42 |
| opendevreview | Alex Welsh proposed openstack/kolla-ansible master: WIP: Automate prometheus blackbox endpoint config https://review.opendev.org/c/openstack/kolla-ansible/+/912420 | 16:43 |
| SvenKieske | I can also try to work on https://review.opendev.org/c/openstack/kolla-ansible/+/878826 instead, if that's more future proof | 16:43 |
| SvenKieske | revisiting the caracal document, there are many people in favor of tempest but it seems nobody's doing the work | 16:44 |
| SvenKieske | I guess I fill both as a topic for D cycle then | 16:44 |
| opendevreview | Will Szumski proposed openstack/kayobe-config-dev master: Replace docker group with container-engine https://review.opendev.org/c/openstack/kayobe-config-dev/+/910017 | 16:46 |
| opendevreview | Will Szumski proposed openstack/kayobe master: Add podman support https://review.opendev.org/c/openstack/kayobe/+/909686 | 16:47 |
| opendevreview | Will Szumski proposed openstack/kayobe master: Add podman support https://review.opendev.org/c/openstack/kayobe/+/909686 | 16:49 |
| mnasiadka | SvenKieske: CI flakiness is not because of quotes and other ,,rules'' shellcheck is checking for, show me one occurence and I'll change my mind ;) | 16:50 |
| SvenKieske | I can surely built something in that breaks CI and you won't find during Code Review (this reminds me of the "underhanded C contest" http://www.underhanded-c.org/) :-) | 16:51 |
| SvenKieske | thing is: I write bash since over 12 years, and I don't trust myself writing good bash, because the "language" is, well shit, for writing stuff longer than a paragraph. And I know for a fact I'm not the most inexperienced bash programmer on this project. so I would really like to provide guardrails for all of us. | 16:52 |
| SvenKieske | mnasiadka: for reference, the change from today/yesterday: https://review.opendev.org/c/openstack/kolla-ansible/+/909971 the bash if condition was totally broken, had even a plus 1, and I was initially tempted to also +1 it, until I took a closer look | 16:54 |
| SvenKieske | then there was even wrong advice to "re-fix" it back into the broken patch, from before, after it was fixed. | 16:54 |
| mnasiadka | kolla-ansible cmd is written in bash, while we're at it ;) | 16:54 |
| SvenKieske | I really don't want to pick upon people here, that could've happened to myself I guess. | 16:54 |
| SvenKieske | it just shows people don't have much knowledge about bash and do basic mistakes, that's why I always triple check my code locally with examples and tests. | 16:55 |
| SvenKieske | so yeah, rewrite it all the way to python, I'm actually a fan of rewrites, if they are well reasoned. | 16:56 |
| SvenKieske | not using bash is a good reason if most devs are more comfortable with python | 16:56 |
| SvenKieske | but we also need to be realistic: will we rewrite 200 lines of bash in python? probably? will someone convert everything in one release? probably not. so if we don't have some kind of shared community goal and end up adding more bash then we are removing..well | 16:57 |
| SvenKieske | still, we even lint python, and that has way more builtin checks because it's more of a compiled language these days. and it has _data structures_ *gasp* :D | 16:58 |
| mnasiadka | well, I'm not saying we do everything at once | 16:59 |
| mnasiadka | but the CLI would benefit from being rewritten in a normal language | 16:59 |
| SvenKieske | big +1 :) | 16:59 |
| SvenKieske | mhm, I'm really tempted to do that, but I should first finish this ovn-exporter stuff | 17:00 |
| opendevreview | Will Szumski proposed openstack/kayobe master: Fix issue removing docker volumes https://review.opendev.org/c/openstack/kayobe/+/909594 | 17:14 |
| opendevreview | Will Szumski proposed openstack/kayobe master: Support configuring persistent systemd journal https://review.opendev.org/c/openstack/kayobe/+/909685 | 17:14 |
| opendevreview | Will Szumski proposed openstack/kayobe master: Add podman support https://review.opendev.org/c/openstack/kayobe/+/909686 | 17:14 |
| opendevreview | Doug Szumski proposed openstack/kolla-ansible master: Support custom Nova Compute Ironic host names https://review.opendev.org/c/openstack/kolla-ansible/+/912483 | 17:20 |
| opendevreview | Merged openstack/kayobe master: docs: Add note on SLURP upgrades https://review.opendev.org/c/openstack/kayobe/+/913192 | 17:33 |
| opendevreview | Will Szumski proposed openstack/kayobe master: Register baremetal compute nodes in Ironic https://review.opendev.org/c/openstack/kayobe/+/909671 | 18:03 |
| opendevreview | Dawud proposed openstack/kolla master: Change logic for plugins in grafana image build https://review.opendev.org/c/openstack/kolla/+/913184 | 18:04 |
| opendevreview | Michal Arbet proposed openstack/kolla master: Bump proxysql to 2.6.x https://review.opendev.org/c/openstack/kolla/+/912731 | 19:32 |
| opendevreview | Dawud proposed openstack/kolla-ansible master: Configure log level field for the Grafana OpenSearch datasource https://review.opendev.org/c/openstack/kolla-ansible/+/913597 | 21:10 |
| opendevreview | Dawud proposed openstack/kolla-ansible master: Configure log level field for the Grafana OpenSearch datasource https://review.opendev.org/c/openstack/kolla-ansible/+/913597 | 21:11 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!