opendevreview | Roman Krček proposed openstack/kolla-ansible master: Rewrite kolla-ansible bash script to python https://review.opendev.org/c/openstack/kolla-ansible/+/923110 | 06:47 |
---|---|---|
opendevreview | Jörn Kaster proposed openstack/kolla-ansible master: Enable object storage metrics also with ceph rgw https://review.opendev.org/c/openstack/kolla-ansible/+/922920 | 07:12 |
sylvr | Hello ! I think I made a mistake while trying to reset my deployment environment after provisioning, and now some of my nodes aren't enrolled in Bifrost. (forgot to delete them) | 07:30 |
sylvr | I inspected the nodes, and they boot on their local disk even when I used ipmitool to set the next boot/default boot to PXEboot.. | 07:30 |
sylvr | sorry, when I said "inspected" I meant investigated on theses nodes* | 07:37 |
opendevreview | Gaël THEROND proposed openstack/kolla-ansible master: Add new hardware prometheus exporter: https://review.opendev.org/c/openstack/kolla-ansible/+/922211 | 08:00 |
opendevreview | Gaël THEROND proposed openstack/kolla-ansible master: Add new hardware prometheus exporter: https://review.opendev.org/c/openstack/kolla-ansible/+/922211 | 08:01 |
kevko | morning \o/ | 09:29 |
kevko | Has anyone noticed that there's something wrong with the apt mirror in CI? Openstack-base is not building :( | 09:30 |
opendevreview | Michal Arbet proposed openstack/kolla master: [CI] [DNM] Test openstack-base CI build https://review.opendev.org/c/openstack/kolla/+/923142 | 09:33 |
frickler | "libssl-dev : Depends: libssl3 (= 3.0.11-1~deb12u2) but 3.0.13-1~deb12u1 is to be installed" that's usually a sign that mirrors are outdated compared to images, let me check | 09:38 |
frickler | hmm, or maybe it is the other way around in this case? the updated version seems to be > 3 months old. debian mirrors were updated just fine over the last couple of days afaict | 09:42 |
kevko | frickler: my local build is fine ..it's definitely something with mirror | 09:49 |
kevko | frickler: hmmmmm, okay this is bookworm | 09:50 |
kevko | https://packages.debian.org/bookworm/libssl3 | 09:50 |
kevko | frickler: https://packages.debian.org/bookworm/libssl-dev << bookworm is 3.0.13-1~deb12u1 ...so i really don't know from which source the libssl-dev = 3.0.11-1~deb12u2 is "synced" | 09:51 |
opendevreview | Michal Arbet proposed openstack/kolla master: [CI] [DNM] Test openstack-base CI build https://review.opendev.org/c/openstack/kolla/+/923142 | 09:54 |
frickler | kevko: the 3.0.11 version is what is in bookwork-security afaict. not sure why apt seems to prefer that version, I'll hold a node to look at it | 09:59 |
kevko | frickler: because of -security ? :) | 10:00 |
kevko | frickler: it will be something in mirror configuration i would say .. | 10:03 |
kevko | frickler: if i am building locally ..fresh master branch ...it's resolved ok ...so mirror is configured to download libssl-dev from security ...but didn't download dependencies | 10:05 |
kevko | i would say | 10:10 |
frickler | security is a dedicated repo and mirrored separately. also libssl3 and libssl-dev should be treated the same | 10:11 |
kevko | frickler: btw, just question ..is it possible to sometimes get the access to those infra ci nodes ? | 10:13 |
opendevreview | Michal Arbet proposed openstack/kolla master: [CI] [DNM] Test openstack-base CI build https://review.opendev.org/c/openstack/kolla/+/923142 | 10:14 |
kevko | frickler: this is non-mirror build https://paste.openstack.org/show/b8mZW7maBR1RzKCEhkKt/ | 10:15 |
kevko | frickler: security has older version than bookworm stable ... | 10:15 |
kevko | frickler: maybe there is some pin for libssl3 | 10:17 |
kevko | frickler: this is CI build -> | 10:53 |
kevko | https://paste.openstack.org/show/buh7fIcz87FqGlt9YMPj/ | 10:53 |
kevko | frickler: mirror is outdated ... there is no 3.0.13-1~deb12u1 in mirror ... (libssl3 has no remote apt ...so it looks like libssl3 is installed in debian base image ..and mirror is outdated ) | 11:00 |
frickler | kevko: ok, so there indeed seems to be an issue with our mirror. this isn't noticed with our CI images, as they are also built from the mirror, but only with a newer docker image. cc fungi | 11:00 |
frickler | will continue this discussion in #opendev | 11:01 |
opendevreview | Gaël THEROND proposed openstack/kolla-ansible master: Add new hardware prometheus exporter: https://review.opendev.org/c/openstack/kolla-ansible/+/922211 | 11:03 |
Fl1nt | ok, I think I've finished the whole BMC patch, anyone willing to review it? | 11:05 |
kevko | frickler: but it's still strange little bit ..as if I download fresh debian:bookworm there is no preintalled libssl3 | 11:19 |
frickler | kevko: seems to do the initial updates in the base image before switching to the mirror https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_d29/922892/6/check/kolla-build-debian/d29746e/kolla/build/base.log | 11:20 |
frickler | which could be considered a bug in itself | 11:20 |
frickler | s/to/we/ | 11:20 |
kevko | frickler: yeah, it seems to be | 11:24 |
kevko | frickler: are u sure ? it is ok to switch to the mirror ..if mirror is really mirrored and nout outdated ...so still sounds like outdated mirror ... | 11:28 |
kevko | frickler: there shouldn't be an issue if mirror is non-outdated | 11:29 |
frickler | the mirror certainly is outdated (just by two days, but bad enough). using inconsistent mirrors during kolla builds is still not a good idea | 11:36 |
kevko | frickler: we can disscuss this , but for now - to unblock all CI jobs ..is it possible to do the sync manually ? | 11:37 |
frickler | sure, we're on it, but it will take some time | 11:45 |
opendevreview | Michal Arbet proposed openstack/kolla master: [CI] [DNM] Test openstack-base CI build https://review.opendev.org/c/openstack/kolla/+/923142 | 11:53 |
kevko | frickler: okay, in meantime i will probably fix this ... | 11:54 |
ftarasenko | Hello team! Can you suggest me, how to correctly upstream powerdns support to designate deployment? Should I start with spec/blueprint, or reporting bug should be enough? | 12:09 |
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 | 12:17 |
opendevreview | Michal Arbet proposed openstack/kolla master: [CI] [DNM] Test openstack-base CI build https://review.opendev.org/c/openstack/kolla/+/923142 | 12:22 |
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 | 12:22 |
ihalomi | hey, what is the problem with this https://review.opendev.org/c/openstack/ansible-collection-kolla/+/916258?tab=comments ? Its stucked on +2 for a month now | 12:25 |
opendevreview | Michal Arbet proposed openstack/kolla master: Fix build debian packages https://review.opendev.org/c/openstack/kolla/+/923162 | 12:37 |
opendevreview | Michal Arbet proposed openstack/kolla master: [CI] Fix build debian packages https://review.opendev.org/c/openstack/kolla/+/923162 | 12:37 |
opendevreview | Michal Arbet proposed openstack/kolla master: Add rally images for testing https://review.opendev.org/c/openstack/kolla/+/922892 | 12:38 |
frickler | ftarasenko: likely a bug would be enough, also a short discussion during the weekly meeting might be helpful. is there anything in particular why you would prefer pdns over bind? | 13:19 |
ftarasenko | frickler: 1. legacy/historical reasons) 2. I have demand from our customers, they asking for pdns just for the same reason - previous installation or they just used to. Designate has complete support for powerdns, so I do not see any reason why we cannot support powerdns in k-a) | 13:21 |
ftarasenko | For now, I see only one problem - upgrade path for powerdns is not complete out of the box, I need to add powerdns database migrations. I have kolla containers and kolla-ansible role ready for upstream. | 13:23 |
kevko | frickler: mnasiadka: bbezak: can we merge this to unblock kolla CI ? https://review.opendev.org/c/openstack/kolla/+/923162 | 14:35 |
kevko | zuul wil verify it in minutes | 14:36 |
frickler | kevko: the mirror update has just finished. tbh I'd rather see us make the mirror usage consistent than move pkgs around | 14:53 |
fungi | kevko: try again, mirrors should be up to date now. it was stalled because the weekend's stable point release took longer than our configured timeout to complete | 14:56 |
fungi | ah, i see that was already reported | 14:56 |
*** darmach7 is now known as darmach | 15:04 | |
kevko | frickler: check the patch ...even if it is no mirror issue ..it makes sense to move that package from images to base ... | 15:14 |
kevko | frickler: kolla-toolbox has in dockerfile FROM base ...and openstack-base has in dockerfile from base ....but both of them are installing libssl-dev in their layers ...it makes sense to install in base only ... | 15:23 |
opendevreview | Michal Arbet proposed openstack/kolla master: Add rally images for testing https://review.opendev.org/c/openstack/kolla/+/922892 | 17:02 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Add REQUESTS_CA_BUNDLE to kolla-toolbox container https://review.opendev.org/c/openstack/kolla-ansible/+/923105 | 17:02 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Add rally role https://review.opendev.org/c/openstack/kolla-ansible/+/922900 | 17:02 |
opendevreview | Michal Arbet proposed openstack/kolla-ansible master: Try to test via rally https://review.opendev.org/c/openstack/kolla-ansible/+/922901 | 17:02 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!