Monday, 2023-09-18

kevkoMorning 07:47
SvenKieskeo/07:49
kevkoSven are u ok now with ^^^07:57
kevkohttps://review.opendev.org/c/openstack/kolla-ansible/+/89554607:58
kevko>> 07:58
Viiihi i would like to start vitrage. But I have an error. I am using the latest code 2023.108:22
Viiimsg": "Could not find or access '/etc/kolla/config/vitrage/prometheus_conf.yaml' on the Ansible Controller.\nIf you are using a module and expect the file to exist on the remote, see the remote_src option08:22
Viiiwhat should be in this configuration file? any example. I can't find anything in the documentation08:23
mnasiadkait seems you have set enable_vitrage_prometheus_datasource to yes, I guess you should read Vitrage documentation to understand what needs to be in this file08:24
mnasiadkasorry, I doubt anybody here uses vitrage ;)08:24
Viiicat /etc/kolla/globals.yml |grep vitrage   ||| #enable_horizon_vitrage: "{{ enable_vitrage | bool }}".  ||| enable_vitrage: "yes"08:26
Viiiok, out of curiosity I check out various services08:27
mnasiadkaseems it's enabled by default when you use prometheus - https://github.com/openstack/kolla-ansible/blob/f6d0ffd61c950416f4e521064772291e65de9623/ansible/group_vars/all.yml#L126908:27
ViiiOk, thx08:28
kevkoSvenKieske replied 08:28
kevkoMnasiadka: letsencrypt Kolla path comments addressed if you have a time ... 08:29
Viiiand what do you most often use to monitor openstation?08:29
Viiiopensatck*08:29
ViiiI guess it depends :) but such a good proven way?08:31
SvenKieskeViii: per default kolla-ansible deploys prometheus & grafana08:31
opendevreviewChristian Berendt proposed openstack/kolla-ansible master: keystone: ensure bool for two parameters  https://review.opendev.org/c/openstack/kolla-ansible/+/89554608:31
kevkoI would like to ask again SvenKieske, mnasiadka - how to deal with https://review.opendev.org/c/openstack/kolla-ansible/+/895546 for example ? It's fixing bug ....but reno missing with 'fixes' block inside ... mnasiadka gave +2 SvenKieske gave +1 08:49
mnasiadkakevko: we agreed we don't require bugfix renos08:49
kevkook, nice to hear :) 08:49
mnasiadkathey are nice to have and probably needed on some complicated bugfixes08:51
opendevreviewChristian Berendt proposed openstack/kolla-ansible master: keystone: ensure bool for two parameters  https://review.opendev.org/c/openstack/kolla-ansible/+/89554608:52
mnasiadkawell, you've got your reno now :)08:53
SvenKieskewe're being watched ;)08:57
SvenKieskekevko: about the let's encrypt stuff: I realize some other part writes the cert via rsync to tmp.. I lost track: is that part of the same changeset? if yes, we should alter this path, because hardcoded /tmp/ paths are the bane of my existence and the root of all evil.08:59
SvenKieskeeither way we should change that, I'll do it myself if I find the corresponding code.08:59
kevkoSvenKieske: expand-all -> ctrl + f -> write rsync09:01
mnasiadkayeah, we should handle tmp properly - that's usually the first place security bugs appear ;)09:03
SvenKieskemhm09:03
SvenKieskecan't docker do the same stuff, systemd does? private tmp? that would mitigate a log, still we should not write this way to tmp09:04
SvenKieskekevko: thx for the expand all tip, I just was going through all files manually..my coffeine storage is depleted :(09:04
opendevreviewJake Hutchinson proposed openstack/kayobe master: Add NTP parameter configuration  https://review.opendev.org/c/openstack/kayobe/+/89519909:06
kevkoSvenKieske: no problem, i will move /tmp/ to /var/lib/haproxy home dir ...so give me few minute s09:07
kevkoansible/roles/mariadb/tasks/recover_cluster.yml:      command: "cp {{ docker_runtime_directory or '/var/lib/docker' }}/volumes/kolla_logs/_data/mariadb/mariadb.log /tmp/mariadb_tmp.log" 09:08
kevko:D 09:08
opendevreviewDr. Jens Harbott proposed openstack/kolla-ansible master: keystone: ensure bool for two parameters  https://review.opendev.org/c/openstack/kolla-ansible/+/89554609:10
SvenKieskemaybe I can refactor some of the tmp stuff in other scripts, guilty of this cardinal sin.09:13
opendevreviewMichal Arbet proposed openstack/kolla master: Rework letsencrypt  https://review.opendev.org/c/openstack/kolla/+/88734709:27
kevkoSvenKieske: done 09:27
SvenKieskety09:33
opendevreviewMerged openstack/kolla-ansible stable/2023.1: senlin: add missing cafile parameter  https://review.opendev.org/c/openstack/kolla-ansible/+/89443209:48
kevkoSvenKieske: anything else ? 09:55
SvenKieskeI wanted to check if the httpd binary is really always located in /usr/sbin/ on all distros - I don't know - but haven't got the time just yet :)10:04
SvenKieskechances are, this is correct, but I wanted to check either way, just paranoia10:04
opendevreviewMerged openstack/kolla-ansible master: Configure Nova libvirt.num_pcie_ports to 16 by default  https://review.opendev.org/c/openstack/kolla-ansible/+/76830110:08
opendevreviewMichal Arbet proposed openstack/kolla-ansible master: Add support for LetsEncrypt-managed certs  https://review.opendev.org/c/openstack/kolla-ansible/+/74134010:08
opendevreviewBartosz Bezak proposed openstack/kolla-ansible stable/2023.1: Configure Nova libvirt.num_pcie_ports to 16 by default  https://review.opendev.org/c/openstack/kolla-ansible/+/89537110:13
opendevreviewBartosz Bezak proposed openstack/kolla-ansible stable/zed: Configure Nova libvirt.num_pcie_ports to 16 by default  https://review.opendev.org/c/openstack/kolla-ansible/+/89537210:13
opendevreviewBartosz Bezak proposed openstack/kolla-ansible stable/yoga: Configure Nova libvirt.num_pcie_ports to 16 by default  https://review.opendev.org/c/openstack/kolla-ansible/+/89537310:13
opendevreviewMerged openstack/kolla-ansible master: ceilometer: process custom event_pipeline.yaml with merge_yaml  https://review.opendev.org/c/openstack/kolla-ansible/+/89418210:18
kevkoSvenKieske: haha, instead of leave a comment you can just check another apache service in k-a ? :D 10:18
SvenKieskekevko: right, this might have already been done elsewhere10:18
kevkoSvenKieske: for example placement ? https://github.com/openstack/kolla-ansible/blob/master/ansible/roles/placement/templates/placement-api.json.j210:19
SvenKieskestill not really awoken I guess, it was a long weekend.10:19
kevko:D :D :D 10:19
kevkoi fell asleep at 5 :D and at 9 i was already working :D 10:19
kevkosleep is overrated 10:20
kevko:D 10:20
SvenKieskeit seems to be10:22
mnasiadkaand then after a week of that, you go to sleep at 9pm and wake up at 9pm ;-)10:30
kevkohaha :D sometimes :D 10:41
mnasiadkaI have a feeling apache_binary and apache_conf_dir should be moved to group_vars/all.yml instead of multiplying that code everywhere11:05
SvenKieskethat seems also like a good point12:17
opendevreviewMerged openstack/kolla-ansible master: keystone: ensure bool for two parameters  https://review.opendev.org/c/openstack/kolla-ansible/+/89554612:26
opendevreviewBartosz Bezak proposed openstack/kolla-ansible master: Add option for extra kernel modules in neutron role  https://review.opendev.org/c/openstack/kolla-ansible/+/89473512:32
opendevreviewBartosz Bezak proposed openstack/kolla-ansible master: Add option for extra kernel modules in neutron role  https://review.opendev.org/c/openstack/kolla-ansible/+/89473512:32
opendevreviewSven Kieske proposed openstack/kolla-ansible stable/2023.1: keystone: ensure bool for two parameters  https://review.opendev.org/c/openstack/kolla-ansible/+/89537912:54
opendevreviewSven Kieske proposed openstack/kolla-ansible stable/zed: keystone: ensure bool for two parameters  https://review.opendev.org/c/openstack/kolla-ansible/+/89538012:55
opendevreviewSven Kieske proposed openstack/kolla-ansible stable/yoga: keystone: ensure bool for two parameters  https://review.opendev.org/c/openstack/kolla-ansible/+/89566112:55
kevkomnasiadka: SvenKieske: no problem, but i'm not going to store apache binary  value to groups vars in LE patch ...if it is another blocker for LE, ok just let me know ... give me 5 minutes and i will propose another patch for this and rebase onto it 13:07
kevkoi just wanted to say, focus on real changes which LE patch is proposing, or on functionality , let's not run into other corners which are not relevant and IF we want to change we NEED to change in another patchsets 13:10
PerduI'm trying to figure out what I need to do to get my (very small) patch finally merged (https://review.opendev.org/c/openstack/kolla-ansible/+/894930). Documentation seems to say I need to get attention from a core reviewer to approve merge13:21
kevkoPerdu: two +2 and one +w13:29
mnasiadkakevko: no no, just saying about a possible refactor of everything, not doing that now :)13:34
jangutterkevko: want to save a small amount of CI resources (job unlikely to be fixed until C-series)? https://review.opendev.org/c/openstack/kolla-ansible/+/89516313:34
mnasiadkaPerdu: thanks for doing that, completely forgot it when we moved from wiki - it's merging now13:35
fricklerPerdu: generally you should not need to do anything, core reviewer will find open patches by themselves and act upon them, but asking here after waiting a bit is fine, too13:36
SvenKieskeinteresting bug report from the scs lean operator call: https://bugs.launchpad.net/kolla-ansible/+bug/203398013:40
SvenKieskesounds like we should relocate that one to neutron though, no?13:40
SvenKieskekevko: no problem regarding LE patch, was just wondering, path is fine for me and you don't need to do anything regarding it :)13:41
SvenKieskePerdu: just pester frickler and mnasiadka here, that's what we do as well ;)13:42
fricklerSvenKieske: added neutron to that bug so they can have a look13:43
mnasiadkait's rather radvd in the neutron agent container13:44
opendevreviewCélestin Matte proposed openstack/kolla-ansible master: Fix ansible-lint warnings related to jinja spacing  https://review.opendev.org/c/openstack/kolla-ansible/+/89568413:46
SvenKieskethe condition under which this happens is funny, they provided 5 dns servers to the neutron ipv6 network, then it died like described in this bug13:49
SvenKieskeadded that info to the bug13:49
SvenKieskebut I have not reproduced myself13:49
SvenKieskefrickler: thanks13:51
opendevreviewMerged openstack/kolla-ansible master: Support customising Nova config at the individual service level  https://review.opendev.org/c/openstack/kolla-ansible/+/89429814:29
opendevreviewMerged openstack/kolla-ansible master: Add genconfig to the Operating Kolla guide  https://review.opendev.org/c/openstack/kolla-ansible/+/89039714:29
opendevreviewMerged openstack/kolla-ansible master: README: link to the meetings page directly  https://review.opendev.org/c/openstack/kolla-ansible/+/89493014:29
SvenKieskemnasiadka: should we maybe add the doodle links to the etherpad?14:53
SvenKieskedo we have any special pid file handling code in kolla? still wondering about this radvd issue, so far haven't found much.15:00
SvenKieskenvm, it's all over the place :)15:11
SvenKieskeokay, neutron invented it's own pid file handling, it's somewhere in this maze: https://github.com/openstack/neutron/blob/c33805919eb89f5ea6d8b54d4142d4c829d9a2a0/neutron/agent/linux/ra.py15:36
mnasiadkaSvenKieske: good idea, will add tomorrow15:43
SvenKieskeyeah I'm also out for today, need to get some sleep.. o/16:06
fricklermnasiadka: maybe use https://framadate.org/ instead of doodle as a privacy friendly alternative. also I like the idea of ignoring the prescribed timeslots16:12
opendevreviewMerged openstack/kolla master: toolbox: Add sleep script handling signals  https://review.opendev.org/c/openstack/kolla/+/89177718:36
opendevreviewMerged openstack/kolla-ansible master: toolbox: use script handling signals  https://review.opendev.org/c/openstack/kolla-ansible/+/89177918:44
opendevreviewCélestin Matte proposed openstack/kolla-ansible master: kolla_address: check correct error return value  https://review.opendev.org/c/openstack/kolla-ansible/+/89573619:24
opendevreviewRocky proposed openstack/kolla-ansible master: Set config files to root/horizon  https://review.opendev.org/c/openstack/kolla-ansible/+/89330723:22

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!