| tis_at_teuto | kevko: Good morning. Hi, is there anything needed to get https://review.opendev.org/c/openstack/kolla-ansible/+/922391 merged, or will it take some time? I see, there was a failed pipeline-run. Not sure but i think that was not related to my change. | 06:07 |
|---|---|---|
| mnasiadka | tis_at_teuto: replied in the patch | 06:55 |
| mnasiadka | morning | 06:56 |
| SvenKieske | morning | 07:30 |
| SvenKieske | that's why I voted -1 for backporting :D | 07:30 |
| stan_ | SvenKieske: I just deployed a fresh antelope cluster with OVN, I'm trying to list router with ovn-nbctl commands, but it is unable to connect to DB sock. ovn-nbctl: unix:/var/run/ovn/ovnnb_db.sock: database connection failed () | 07:35 |
| stan_ | am I missing something? | 07:35 |
| SvenKieske | are you on a host/container where that socket is available? you need to check that you're actually somewhere where ovn-northdb runs | 07:39 |
| stan_ | SvenKieske: I checked on container - ovn_nb_db on controller, is that incorrect? I'm not network guy, just figuring things out | 07:40 |
| mnasiadka | stan_: you need to run this command on the active node in the cluster - or add --no-leader-only option to the CLI | 07:41 |
| stan_ | Where should I add this when deploying in kolla | 07:42 |
| stan_ | ovn-nbctl list Logical_Router --no-leader-only ovn-nbctl: unix:/var/run/ovn/ovnnb_db.sock: database connection failed () ? | 07:43 |
| mnasiadka | when deploying? | 07:43 |
| mnasiadka | ovn-nbctl --no-leader-only list ? | 07:43 |
| stan_ | one minute let me try this command | 07:44 |
| SvenKieske | so that command works for me - without the no-leader part - in the ovn-northd container on a controller. | 07:48 |
| SvenKieske | ah nvm, I had a typo | 07:48 |
| mhiner | Please review the migration patch, thanks: https://review.opendev.org/c/openstack/kolla-ansible/+/836941 | 07:53 |
| SvenKieske | stan_ good question, actually, there's the --db flag with which you can alter the socket path | 07:55 |
| SvenKieske | mhm weird, I see a dangling symlink - or the user has insufficient permissions, between /etc/openvswitch/conf.db and /var/lib/openvswitch/conf.db | 07:57 |
| opendevreview | Michal Nasiadka proposed openstack/kayobe master: Add internal vip address to no_proxy https://review.opendev.org/c/openstack/kayobe/+/922787 | 07:58 |
| SvenKieske | you can specify a different db/socket (the terminology is weird there in ovn land) via "--db=unix:/run/ovn/ovn-northd.ctl", but I also get "database connection failed ()". never used ovn-nbctl though | 08:03 |
| stan_ | SvenKieske: mnasiadka --no-leader-only listed the routers | 08:03 |
| SvenKieske | so you say it works? can you post your complete cli command? because it doesn't work for me | 08:06 |
| stan_ | give me one second, | 08:06 |
| stan_ | docker container exec -t -u root ovn_nb_db ovn-nbctl --no-leader-only ls-list > This works on controller.. | 08:09 |
| stan_ | docker container exec -t -u root ovn_nb_db ovn-nbctl ls-list >> Same controller, but did not work | 08:10 |
| SvenKieske | stan_ that works for me as well, notice, you don't need "-u root" afaik the container already runs as root | 08:11 |
| stan_ | Where should I check traffic flow.. | 08:11 |
| SvenKieske | so I tested it without root and it works | 08:11 |
| SvenKieske | you mean logical flow or physical flow? :) | 08:12 |
| stan_ | Logical flow | 08:15 |
| stan_ | sudo docker container exec -t -u root ovn_controller ovn-nbctl --no-leader-only ls-list >> This did not work on network node ovn-nbctl: unix:/var/run/ovn/ovnnb_db.sock: database connection failed (No such file or directory) | 08:15 |
| stan_ | This is command working for you in network node? | 08:21 |
| PrzemekK | for me docker container exec -t -u root ovn_nb_db ovn-nbctl --no-leader-only ls-list > This works on all controllers / docker container exec -t -u root ovn_nb_db ovn-nbctl ls-list This works on 1 controller | 08:22 |
| PrzemekK | docker container exec -t -u root ovn_controller ovn-nbctl --no-leader-only ls-list / this is not working on network nodes (gw) | 08:25 |
| opendevreview | Piotr Parczewski proposed openstack/kolla master: Add Let's Encrypt custom arguments support https://review.opendev.org/c/openstack/kolla/+/916617 | 08:25 |
| stan_ | PrzemekK Is this nbctl command supposed to work only on controllers? | 08:26 |
| stan_ | I'm having trouble accessing my openstack instances from outside, so we are trying to figure out where exactly is the traffic dropped. We are used to OVS, but not sure how to check this | 08:29 |
| jheikkin | I have a question about kolla-ansible upgrade process. Let's say a service DB schema needs to be upgraded as part of openstack version update. Is kolla-ansible doing this upgrade with some customised container image? Is there somewhere a Dockerfile which shows what kolla-ansible actually changes within the image? | 08:30 |
| PrzemekK | stan_ from what i know on GW (network node) there is only local copy of database https://docs.openstack.org/networking-ovn/latest/admin/refarch/refarch.html Service layout | 08:32 |
| SvenKieske | jheikkin: yes kolla does this, let me show you an example... | 08:38 |
| SvenKieske | these are usually just tasks, e.g. here for keystone: https://github.com/openstack/kolla-ansible/blob/master/ansible/roles/keystone/tasks/upgrade.yml#L23 they trigger some handlers | 08:39 |
| SvenKieske | this just starts, e.g. for keystone the container with a magic env variable "KOLLA_UPGRADE": https://github.com/openstack/kolla-ansible/blob/master/ansible/roles/keystone/handlers/main.yml#L15 | 08:41 |
| SvenKieske | you can find then how the images handle this in our kolla repo: https://github.com/search?q=repo%3Aopenstack%2Fkolla%20KOLLA_UPGRADE&type=code | 08:42 |
| SvenKieske | jheikkin: It's even documented: https://github.com/openstack/kolla/blob/428f37124986019bedc897469f9b1f3470b62082/doc/source/admin/kolla_api.rst#environment-variables | 08:43 |
| SvenKieske | HTH :) | 08:43 |
| PrzemekK | stan_ it looks like ovn-nbctl show command etc are on "You run OVN commands, such as ovn-nbctl show, in the ovn_controller container. The container runs on the Controller node and Compute nodes." https://docs.redhat.com/en/documentation/red_hat_openstack_platform/17.0/html/networking_guide/neutron-troubleshoot_rhosp-network#ovn-db-aliases-creating_neutron-troubleshoot | 08:53 |
| jheikkin | SvenKieske: Thank you. I didn't realise to check under kolla, found the dockerfiles too. | 09:15 |
| SvenKieske | jheikkin: you're welcome, no worries :) "kolla" is for building the images, and k-a for configuration/running, restarting, upgrading etc. | 09:17 |
| SvenKieske | so we find the process to trigger the upgrade in kolla-ansible and the necessary tooling is of course already installed in kolla and defined there. | 09:25 |
| *** mgoddard- is now known as mgoddard | 10:12 | |
| opendevreview | Sylvère Kanapa proposed openstack/kayobe master: Add missing "cloud-init" as "cloud-init-datasource" don't have dependency on it. https://review.opendev.org/c/openstack/kayobe/+/922795 | 11:23 |
| sylvr | Hello #openstack-kolla ! | 11:23 |
| sylvr | jovial: I submitted the patch you asked for ! | 11:24 |
| jovial | slyvr: Amazing, thanks. I'll take a look shortly :) | 11:25 |
| opendevreview | Sylvère Kanapa proposed openstack/kayobe master: Add missing "cloud-init" as "cloud-init-datasource" don't have dependency on it. https://review.opendev.org/c/openstack/kayobe/+/922795 | 11:35 |
| SvenKieske | is anybody in here (still) using vagrant for anything? I got curious as I found out it even supports docker. | 12:30 |
| SvenKieske | and I wanted to test if our vagrant docs still work at all :D | 12:31 |
| opendevreview | Pedro Henrique Pereira Martins proposed openstack/kolla-ansible master: Fix the docker container dimensions comparison for short notation https://review.opendev.org/c/openstack/kolla-ansible/+/886500 | 12:40 |
| mnasiadka | mgoddard mnasiadka bbezak frickler kevko SvenKieske mmalchuk gkoper jangutter jsuazo jovial osmanlicilegi mattcrees dougszu - weekly meeting in 9 minutes | 12:51 |
| darmach | \o | 13:01 |
| kevko | \o/ | 13:01 |
| SvenKieske | you're to early ;) | 13:01 |
| r-krcek | o/ | 13:01 |
| SvenKieske | \o/ | 13:01 |
| mnasiadka | #startmeeting kolla | 13:01 |
| opendevmeet | Meeting started Wed Jun 26 13:01:37 2024 UTC and is due to finish in 60 minutes. The chair is mnasiadka. Information about MeetBot at http://wiki.debian.org/MeetBot. | 13:01 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 13:01 |
| opendevmeet | The meeting name has been set to 'kolla' | 13:01 |
| mnasiadka | #topic rollcall | 13:01 |
| mnasiadka | now | 13:01 |
| mnasiadka | o/ | 13:01 |
| bbezak | o/ | 13:01 |
| SvenKieske | o/ | 13:01 |
| darmach | o/ | 13:01 |
| r-krcek | o/ | 13:01 |
| mhiner | o/ | 13:01 |
| mnasiadka | #topic agenda | 13:02 |
| mnasiadka | * Announcements | 13:02 |
| mnasiadka | * CI status | 13:02 |
| mnasiadka | * Release tasks | 13:02 |
| mnasiadka | * Regular stable releases (first meeting in a month) | 13:02 |
| mnasiadka | * Current cycle planning | 13:02 |
| mnasiadka | * Additional agenda (from whiteboard) | 13:02 |
| mnasiadka | * Open discussion | 13:02 |
| mnasiadka | #topic Announcements | 13:02 |
| mnasiadka | next OpenInfra PTG[1] which will take place October 21-25, 2024 virtually! | 13:03 |
| mnasiadka | [1] PTG Site: https://openinfra.dev/ptg/ | 13:03 |
| mnasiadka | [2] PTG Registration: http://ptg.openinfra.dev/ | 13:03 |
| mnasiadka | #topic CI status | 13:03 |
| mnasiadka | So, Koalas - how is the CI? | 13:03 |
| mnasiadka | I assume green that nobody is complaining :) | 13:04 |
| mnasiadka | #topic Release tasks | 13:04 |
| SvenKieske | I personally didn't notice any strange failures, but I didn't look at overall CI status, just the on the reviews I did. | 13:04 |
| mnasiadka | Today is R-14 week of the Dalmatian (2024.2) cycle - nothing for us | 13:05 |
| mnasiadka | #topic Current cycle planning | 13:05 |
| mnasiadka | So, what features do we need to get it - probably Ubuntu Noble (24.04) | 13:06 |
| mnasiadka | darmach: you're taking that together with some advice from bbezak and me, right? | 13:06 |
| darmach | Yes, that's right | 13:06 |
| bbezak | and some tasks from last cycle | 13:07 |
| bbezak | like OVN BGP agent | 13:07 |
| jovial | I would like to get kayobe podman supported landed sooner rather than later. I added some dependent patches to the meeting agenda. Would it be possible to prioritize these? | 13:07 |
| mnasiadka | any other features that we can't release without? | 13:07 |
| kevko | i have older patch for CI and disk preparation for swift -> https://review.opendev.org/c/openstack/kolla-ansible/+/913942 << it's non-voting and probably we used to don't try to resolve ...but from time to time the old version of disk preparation is just not working ... this is working always ... | 13:08 |
| SvenKieske | do we need to get the rmq/slurp stuff in before? | 13:09 |
| mnasiadka | mattcrees: how is it going with newer rmq releases in antelope and bobcat? | 13:09 |
| bbezak | yeah, this is needs to land in C | 13:09 |
| SvenKieske | and what happened to the systemd logging stuff? I can't remember any patches in that regard, but I guess it's only "nice to have"? | 13:10 |
| mnasiadka | yeah, rather yes | 13:10 |
| mnasiadka | I'm focusing on top priority and must have for now ;) | 13:10 |
| SvenKieske | ok | 13:10 |
| mnasiadka | kevko: SvenKieske has -1 on your patch, can you guys resolve it first? ;-) | 13:11 |
| mattcrees | For rmq it's getting close, I'm looking for reviews for two backport chains right now. https://review.opendev.org/c/openstack/kolla/+/922605 https://review.opendev.org/c/openstack/kolla-ansible/+/922607 | 13:12 |
| mattcrees | Then the main patch should be good to go https://review.opendev.org/c/openstack/kolla/+/918974 | 13:12 |
| mnasiadka | bbezak: https://review.opendev.org/c/openstack/kolla/+/916617 - can you take RP from that change? | 13:12 |
| kevko | mnasiadka: well, actually ..i am not sure if I can ..because -1 was because of usage bash_result ; if $? -eq 0 ...... but i am using that on several places in script ... | 13:13 |
| bbezak | done mnasiadka | 13:14 |
| SvenKieske | kevko: there are six unaddressed comments in that changeset, maybe write something. even if you disagree with all of them, just write that down at least? I certainly won't lift a -1 from a patchset when I see no communication :) | 13:14 |
| kevko | SvenKieske: ack | 13:14 |
| mnasiadka | Ok, from Kolla RP+1 patches I see https://review.opendev.org/c/openstack/kolla/+/915440 - bbezak, frickler - should we try to prioritise reviews of that please? | 13:14 |
| kevko | SvenKieske: i will ..but now working on something better ..but ack | 13:14 |
| mnasiadka | oh boy, kolla-ansible RP+1 list is huge | 13:15 |
| SvenKieske | that last thing goes to all patch authors btw, I often see comments just..ignored? at least write that I'm wrong or whatever :P | 13:15 |
| SvenKieske | ;) | 13:15 |
| kevko | SvenKieske: ack | 13:16 |
| SvenKieske | or not just me, I also see that for other comments as well | 13:16 |
| SvenKieske | I mean that's maybe 1-5% of comments, I guess it happens to everyone :) no big deal, most of the time. | 13:17 |
| mnasiadka | I would very much like https://review.opendev.org/c/openstack/kolla-ansible/+/913908 to go forward, but I think we've been merging patches to this important functionality without any CI coverage | 13:19 |
| mnasiadka | Shouldn't there be some mock OIDC server code we could use in a CI job? | 13:19 |
| SvenKieske | I can only support that, but my - albeit short - search didn't reveal anything in that regard, now that I'm thinking about it we _might_ have something in our downstream CI | 13:20 |
| SvenKieske | and as we also use zuul maybe that could be adapted somehow | 13:21 |
| mnasiadka | Something like this https://github.com/appvia/mock-oidc-user-server ? | 13:21 |
| SvenKieske | nah, that's the problem, we deploy keycloak via k3s on the control plane: https://github.com/osism/ansible-collection-services/blob/main/roles/keycloak/tasks/deploy.yml | 13:23 |
| kevko | keycloak ? | 13:23 |
| mnasiadka | well, I don't know if keycloak is small and fast ;) | 13:23 |
| SvenKieske | so I think there aren't any tests we can leverage inside kolla from our side, at least not if you don't want to deploy k3s ;) | 13:23 |
| SvenKieske | that mock server has no commits since 3 years ago :-/ | 13:24 |
| mnasiadka | maybe it's feature complete ;) | 13:24 |
| SvenKieske | could be :) | 13:24 |
| mnasiadka | anyway, if there's a volunteer that wants to have a look into that - I'm happy to help :) | 13:24 |
| SvenKieske | that's 4 month old: https://github.com/oauth2-proxy/mockoidc | 13:25 |
| SvenKieske | but only a go package :( | 13:25 |
| kevko | hmm, if we have proxies and caches and registries in opendev/zuul ...can we have also some permanent deployment of keycloak ? | 13:26 |
| SvenKieske | if I ever finish wrangling with ovn-exporter libs I can take a look, but that will be probably not this summer, as vacation time draws near | 13:26 |
| SvenKieske | at least it's basically greenfield testing, so I don't need to fit it in with existing other tests I guess..but if anybody has more time, don't wait for me | 13:27 |
| mnasiadka | ok, let's move on | 13:28 |
| opendevreview | Matúš Jenča proposed openstack/kolla-ansible master: Add support for RabbitMQ internode tls https://review.opendev.org/c/openstack/kolla-ansible/+/921381 | 13:29 |
| SvenKieske | back to the original topic: I still think we should accept bugfixes for not tested code, it naturally will have the most need for bugfixes ;) | 13:29 |
| mnasiadka | well, I'm just worried the bugfix for some people will cause issues for other people :) | 13:29 |
| mnasiadka | so we'll need another bugfix | 13:29 |
| frickler | but that how community support works, isn't it? | 13:30 |
| frickler | +'s | 13:30 |
| mnasiadka | in theory yes ;) | 13:30 |
| kevko | yeah, and it's master ... bugs expected .. | 13:30 |
| mnasiadka | well, then we backport the fix :) | 13:31 |
| kevko | exactly | 13:31 |
| mnasiadka | anyway - if there are cores more knowledgeable in haproxy area - https://review.opendev.org/c/openstack/kolla-ansible/+/913908 | 13:31 |
| mnasiadka | but I have no clue how haproxy would even start with bad configuration right now :) | 13:31 |
| SvenKieske | I _think_ it rejects bad config, but I can test that part I guess. | 13:32 |
| mnasiadka | #topic Additional agenda (from whiteboard) | 13:32 |
| mnasiadka | (SvenKieske: please review patch for broken keystone federation backend https://bugs.launchpad.net/kolla-ansible/+bug/2058656 | 13:32 |
| mnasiadka | trivial patch at https://review.opendev.org/c/openstack/kolla-ansible/+/913908 needs Cores <--still needs input | 13:32 |
| mnasiadka | https://review.opendev.org/c/openstack/kolla-ansible/+/921381/comment/56b11f74_f98925e8/ (rmq startuptime high, reason found, please share your opinion) | 13:32 |
| mnasiadka | ok, so the patch has been discussed | 13:32 |
| mnasiadka | I think the rmq startuptime with inter node TLS has also been discussed | 13:33 |
| mnasiadka | right SvenKieske ? | 13:33 |
| kevko | i will check | 13:33 |
| SvenKieske | yes | 13:34 |
| SvenKieske | sorry, I need to clean that up on the whiteboard :) | 13:34 |
| bbezak | very trivial for second core - https://review.opendev.org/c/openstack/kolla/+/922593 | 13:34 |
| bbezak | pls | 13:34 |
| SvenKieske | bbezak: mhm, why not backport that docs change? if people read old docs they end up at the wrong bugtracker? | 13:35 |
| SvenKieske | and google most of the time spits out old doc links :( | 13:36 |
| mnasiadka | bbezak: why is that not gating? :D | 13:36 |
| mnasiadka | oh, it started now | 13:36 |
| mnasiadka | phew | 13:36 |
| bbezak | :) | 13:37 |
| mnasiadka | (r-krcek): [26th June] anything more needed? https://review.opendev.org/c/openstack/kolla-ansible/+/914997 | 13:37 |
| SvenKieske | time for the review I guess, it's a large patch ;) | 13:38 |
| r-krcek | Ok, thanks for the info. | 13:38 |
| mnasiadka | I replied - the patch is LGTM, but we need the improvement numbers in the commit message | 13:38 |
| SvenKieske | yes, it would actually be a shame to sweep the rather good analysis under the rug of a gerrit comment :) | 13:39 |
| mnasiadka | (jovial): [26th June] Podman support patches (please review) | 13:39 |
| mnasiadka | https://review.opendev.org/c/openstack/kolla-ansible/+/910016 | 13:39 |
| mnasiadka | https://review.opendev.org/c/openstack/kolla-ansible/+/910501 | 13:39 |
| mnasiadka | Could I possibly get an RP+1. Hoping to land these soon, so podman support can be added to Kayobe. | 13:39 |
| mnasiadka | I reviewed the first one | 13:40 |
| mnasiadka | the second one seems needs some reno love and resolving the comment ;) | 13:40 |
| mnasiadka | jovial: can you have a look? | 13:40 |
| mnasiadka | next one | 13:41 |
| jovial | Yeah, have just seen that comment from Sven. Seems reasonable, I will update the patch. | 13:41 |
| mnasiadka | (jovial): [26th June] Additional node exporter targets (please review) | 13:41 |
| mnasiadka | https://review.opendev.org/c/openstack/kolla-ansible/+/915975 | 13:41 |
| mnasiadka | and jovial - please resolve the comments when you adapt code to fix them :) | 13:42 |
| mnasiadka | reviewed | 13:42 |
| mnasiadka | needs second core outside of SHPC | 13:42 |
| mnasiadka | last one | 13:43 |
| mnasiadka | (mhiner): Please review: https://review.opendev.org/c/openstack/kolla-ansible/+/836941 | 13:43 |
| jovial | Thanks - good point. I was never sure if I should be resolving the comments or if the person who made the comment should resolve them if they are happy with the change | 13:43 |
| mnasiadka | let's assume they will reply on your resolution if they are not fine | 13:43 |
| mnasiadka | :) | 13:43 |
| mnasiadka | mhiner: can you make zuul happy before I start reviewing? :) | 13:43 |
| mnasiadka | #topic Open discussion | 13:44 |
| mnasiadka | Time for everything else | 13:44 |
| SvenKieske | yes, I usually reply on every reply I get. you can ping me here if I did miss something. Sometimes it just takes 1-3 days, due to the amount of reviews. | 13:44 |
| opendevreview | Martin Hiner proposed openstack/kolla-ansible master: Add container engine migration scenario https://review.opendev.org/c/openstack/kolla-ansible/+/836941 | 13:46 |
| mnasiadka | Anybody anything? | 13:46 |
| SvenKieske | nope :) | 13:47 |
| mnasiadka | then let's finish :) | 13:47 |
| mnasiadka | Thanks for coming | 13:47 |
| mnasiadka | #endmeeting | 13:47 |
| opendevmeet | Meeting ended Wed Jun 26 13:47:48 2024 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 13:47 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/kolla/2024/kolla.2024-06-26-13.01.html | 13:47 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/kolla/2024/kolla.2024-06-26-13.01.txt | 13:47 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/kolla/2024/kolla.2024-06-26-13.01.log.html | 13:47 |
| kevko | i just read SvenKieske comment here | 13:48 |
| kevko | https://review.opendev.org/c/openstack/oslo.messaging/+/922609 | 13:48 |
| kevko | just wanted to point that this is nice example why we should have patch feature for kolla :) | 13:48 |
| kevko | oh, i forgot ...did anyone notice that openstack-base is built every time ? even if i have it built already ? i mean ..why cache is not used ? | 13:49 |
| kevko | (without dockerfile changes ..without cwd changes ) | 13:50 |
| opendevreview | Rafael Weingartner proposed openstack/kolla-ansible master: Customize the authentication error timeout page in modOIDC https://review.opendev.org/c/openstack/kolla-ansible/+/832806 | 13:51 |
| opendevreview | Matúš Jenča proposed openstack/kolla-ansible master: Add support for RabbitMQ internode tls https://review.opendev.org/c/openstack/kolla-ansible/+/921381 | 13:51 |
| SvenKieske | kevko: yeah I really didn't get that -1; I mean I personally don't need zed, but still, what's the point if there is a volunteered patch? | 13:51 |
| kevko | SvenKieske: that just means ..sorry we are not going to do anything with unmaintained/* branches | 13:51 |
| mnasiadka | If a person doesn't want to maintain the unmaintained (TM) branch - he should just ignore that Gerrit review ;-) | 13:51 |
| mnasiadka | but another thing is - they are not going to release a new tag | 13:52 |
| kevko | SvenKieske: point is that unmaintained changes can be approved from unmaintainedcore group in gerrit ...and for example noone from oslo team is inside ... :D | 13:52 |
| kevko | mnasiadka: yeah, new tag which basically means new pip package will not be released anymore for unmaintained branches ... | 13:53 |
| kevko | because of new policy | 13:53 |
| mnasiadka | well, upgrade to antelope ;-) | 13:53 |
| kevko | mnasiadka: haha :D ... dear customer, we've found critical bug in oslo.messaging but we can't fix it in zed ...we need to upgrade all 150 servers to the antelope | 13:54 |
| kevko | mnasiadka: even if it is two lines :D | 13:55 |
| kevko | instead of ...dear customer, we fixed bug and we are prepared to reconfigure your zed platform with fixed images | 13:55 |
| mnasiadka | well, dear customer - you're using unmaintained version of OpenStack - just plan the damn upgrade | 13:56 |
| kevko | it's planned in october ...and there are another processes to be done before action will start ...what to do until october ? | 13:57 |
| mnasiadka | in october they should be on C ;-) | 14:03 |
| opendevreview | Pierre Riteau proposed openstack/kayobe master: Allow operators to customise selinux_reboot_timeout https://review.opendev.org/c/openstack/kayobe/+/922826 | 14:04 |
| opendevreview | Pierre Riteau proposed openstack/kayobe master: Support customising selinux_reboot_timeout https://review.opendev.org/c/openstack/kayobe/+/922826 | 14:06 |
| opendevreview | Pierre Riteau proposed openstack/kayobe master: Support customising selinux_reboot_timeout https://review.opendev.org/c/openstack/kayobe/+/922826 | 14:08 |
| kevko | mnasiadka: A is slurp ..so we will upgrade to C at the end of the year and then upgrade will be more easy because of slurp | 14:22 |
| kevko | it's corporate ...you know ..processes | 14:22 |
| SvenKieske | don | 14:23 |
| SvenKieske | 't we all love corporate upgrade cycles? | 14:23 |
| opendevreview | Ivan Halomi proposed openstack/kolla-ansible master: Refactor of docker worker https://review.opendev.org/c/openstack/kolla-ansible/+/908295 | 14:25 |
| opendevreview | Ivan Halomi proposed openstack/kolla-ansible master: Documentation update: Enhanced explanation of volume handling with podman https://review.opendev.org/c/openstack/kolla-ansible/+/922717 | 14:30 |
| kevko | SvenKieske: yep :D | 14:38 |
| kevko | oh, i forgot ...did anyone notice that openstack-base is built every time ? even if i have it built already ? i mean ..why cache is not used ? | 14:48 |
| kevko | (cwd not changed ) | 14:49 |
| opendevreview | Merged openstack/kolla stable/2023.2: Move apt_prefs pins to respective images https://review.opendev.org/c/openstack/kolla/+/922605 | 14:51 |
| opendevreview | Merged openstack/kolla master: add kayobe to release management doc's Launchpad admin section https://review.opendev.org/c/openstack/kolla/+/922593 | 14:51 |
| SvenKieske | I did comment on https://review.opendev.org/c/openstack/kolla-ansible/+/920368 again. I think we should discuss someday if it's a real requirement for kolla(-ansible) to not break kayobe. If it is we should add kayobe tests into kolla CI because I will never know if any patch break kayobe. | 15:03 |
| jovial | SvenKieske, Agree with your point around adding a Kayobe job. Really we are just users of the documented interface i.e post-deploy.yml, so It wouldn't just be use you'd be breaking. Generally I think that change looks nice, but does seem like we should keep current behaviour of post-deploy.yml. | 15:33 |
| jovial | Just thinking I should probably have replied in the thread. Will also post the comment there. | 15:35 |
| kevko | jovial: thank you | 15:39 |
| opendevreview | Sven Kieske proposed openstack/kolla-ansible master: add a config validation precheck for haproxy https://review.opendev.org/c/openstack/kolla-ansible/+/922840 | 15:41 |
| opendevreview | Sven Kieske proposed openstack/kolla-ansible master: add a config validation precheck for haproxy https://review.opendev.org/c/openstack/kolla-ansible/+/922840 | 15:44 |
| SvenKieske | so, now we have hopefully at least some minimal detection if we break haproxy config :) | 15:44 |
| opendevreview | Sven Kieske proposed openstack/kolla stable/2024.1: add kayobe to release management doc's Launchpad admin section https://review.opendev.org/c/openstack/kolla/+/922844 | 15:56 |
| mnasiadka | kevko: I know too many corporate processes | 16:31 |
| kevko | mnasiadka: i just have a feeling that your are argumenting against me ..i don't know why :D | 16:33 |
| kevko | regarding that feature about patching the code | 16:33 |
| mnasiadka | kevko: naah, just saying that in most cases people that need patching that code are the ones staying on exotic retro openstack releases - but I'm certain we'll merge it one day ;) | 16:33 |
| kevko | mnasiadka: did you notice that example i've gave is about critical oslo.messaging bugfix which is in all releases ? including caracal ? :D | 16:35 |
| kevko | mnasiadka: https://review.opendev.org/c/openstack/oslo.messaging/+/922601 | 16:35 |
| mnasiadka | well, the wording unmaintained is for a reason | 16:35 |
| kevko | mnasiadka: it's merged already...so downstream user can be sure that it is approved by an upstream community ..but you know how it is with pip release ...it will take another days/weeks to land pypi :D | 16:36 |
| mnasiadka | it depends when somebody will tag a release | 16:36 |
| kevko | mnasiadka: unmaintained users never get oslo.messaging update ... because unmaintained ..other users can fix their images until new oslo.messaging pypi release will be available | 16:37 |
| mnasiadka | I would actually start thinking about using stable branches for those critical libraries like oslo | 16:37 |
| kevko | they can just be proactive | 16:37 |
| kevko | mnasiadka: i was thinking about this also ! | 16:37 |
| mnasiadka | because that's basically OpenStack deliverable, so we should treat it the same way as Nova and Neutron code | 16:37 |
| kevko | mnasiadka: well, from my perspective it would be the best ...but on the other side ...you want to copy devstack setup ... and they are also using pip + stable services (if i am correct) | 16:38 |
| kevko | mnasiadka: i tried this in a past ... you of course need to remove entry in u-c | 16:39 |
| kevko | mnasiadka: more packages ..more removal .. can be fatal for depency resolve ... | 16:40 |
| kevko | mnasiadka: for example in debian packaging you have version 3.0.0 and 3.0.0-1 and 3.0.0-1+bpo1 for example ... compare versioning works fine ... if i remember ... pbr has broken generating version and don't support +version | 16:41 |
| kevko | like ..i was trying several ways ... patch feature is the simplest one and very handy for several releases ...for me .. | 16:42 |
| opendevreview | Rafael Weingartner proposed openstack/kolla-ansible master: Customize the authentication error timeout page in modOIDC https://review.opendev.org/c/openstack/kolla-ansible/+/832806 | 16:51 |
| opendevreview | Sylvère Kanapa proposed openstack/kayobe master: Add missing "cloud-init" as "cloud-init-datasource" don't have dependency on it. https://review.opendev.org/c/openstack/kayobe/+/922795 | 16:52 |
| opendevreview | Sylvère Kanapa proposed openstack/kayobe master: Add missing "cloud-init" as "cloud-init-datasource" don't have dependency on it. https://review.opendev.org/c/openstack/kayobe/+/922795 | 16:53 |
| opendevreview | Sylvère Kanapa proposed openstack/kayobe master: Add missing "cloud-init" as "cloud-init-datasource" doesn't have dependency on it. https://review.opendev.org/c/openstack/kayobe/+/922795 | 16:58 |
| sylvr | sorry for the spam, made some mistakes but learnt some lessons ^^ | 17:10 |
| mnasiadka | no worries sylvr | 17:13 |
| opendevreview | Merged openstack/kolla-ansible master: Support mode in named volumes when using podman https://review.opendev.org/c/openstack/kolla-ansible/+/910016 | 17:37 |
| opendevreview | Pierre Riteau proposed openstack/kolla-ansible master: Adds option to add extra scrape targets for node exporter https://review.opendev.org/c/openstack/kolla-ansible/+/915975 | 17:43 |
| opendevreview | Pierre Riteau proposed openstack/kolla-ansible master: Adds option to add extra scrape targets for node exporter https://review.opendev.org/c/openstack/kolla-ansible/+/915975 | 17:44 |
| opendevreview | Merged openstack/kolla-ansible master: Fix the docker container dimensions comparison for short notation https://review.opendev.org/c/openstack/kolla-ansible/+/886500 | 19:00 |
| opendevreview | Merged openstack/kolla-ansible master: Masakari: Fix incorrect reference to dictionary key. https://review.opendev.org/c/openstack/kolla-ansible/+/915760 | 19:03 |
| opendevreview | Pierre Riteau proposed openstack/kolla-ansible master: heat: set backups_enabled based on cinder-backup https://review.opendev.org/c/openstack/kolla-ansible/+/918455 | 20:21 |
| opendevreview | Merged openstack/kolla-ansible master: Adds option to add extra scrape targets for node exporter https://review.opendev.org/c/openstack/kolla-ansible/+/915975 | 21:05 |
| opendevreview | Michal Arbet proposed openstack/kolla master: Add rally images for testing https://review.opendev.org/c/openstack/kolla/+/922892 | 21:59 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!