opendevreview | Ronelle Landy proposed openstack/tripleo-ci master: WIP: Exclude fwupd-1.7.1-1 from tripleo-repos https://review.opendev.org/c/openstack/tripleo-ci/+/823737 | 00:01 |
---|---|---|
opendevreview | Ronelle Landy proposed openstack/tripleo-ci master: WIP: Exclude fwupd-1.7.1-1 from tripleo-repos https://review.opendev.org/c/openstack/tripleo-ci/+/823737 | 00:03 |
opendevreview | Ronelle Landy proposed openstack/tripleo-quickstart master: WIP: Exclude fwupd-1.7.1-1 from baseos repo https://review.opendev.org/c/openstack/tripleo-quickstart/+/823738 | 00:04 |
opendevreview | Brendan Shephard proposed openstack/python-tripleoclient master: Move baremetal provide commands from Ansible https://review.opendev.org/c/openstack/python-tripleoclient/+/823618 | 00:07 |
opendevreview | Brendan Shephard proposed openstack/python-tripleoclient master: Move baremetal clean commands from Ansible https://review.opendev.org/c/openstack/python-tripleoclient/+/823625 | 00:25 |
opendevreview | Brendan Shephard proposed openstack/python-tripleoclient master: Move baremetal clean commands from Ansible https://review.opendev.org/c/openstack/python-tripleoclient/+/823625 | 00:26 |
opendevreview | Brendan Shephard proposed openstack/python-tripleoclient master: Move baremetal configure commands from Ansible https://review.opendev.org/c/openstack/python-tripleoclient/+/823685 | 00:27 |
opendevreview | David Hill proposed openstack/tripleo-ansible master: Added ControllerStorageNfs hosts https://review.opendev.org/c/openstack/tripleo-ansible/+/823740 | 01:00 |
opendevreview | David Hill proposed openstack/tripleo-ansible master: Added all Controller* hosts to prepare-overcloud-backup.yaml https://review.opendev.org/c/openstack/tripleo-ansible/+/823740 | 01:06 |
opendevreview | David Hill proposed openstack/tripleo-ansible master: Added all Controller* hosts to prepare-overcloud-backup.yaml https://review.opendev.org/c/openstack/tripleo-ansible/+/823740 | 01:11 |
opendevreview | Merged openstack/tripleo-heat-templates stable/wallaby: Add chrony waitsync back in https://review.opendev.org/c/openstack/tripleo-heat-templates/+/822428 | 01:33 |
opendevreview | Merged openstack/tripleo-heat-templates master: Parameters for API policies to show a host status https://review.opendev.org/c/openstack/tripleo-heat-templates/+/816319 | 01:50 |
opendevreview | Merged openstack/validations-libs master: Add backward compatibility for python pathlib module https://review.opendev.org/c/openstack/validations-libs/+/823729 | 01:50 |
opendevreview | Rabi Mishra proposed openstack/tripleo-heat-templates master: Move *NetworkConfigTemplate to parameter_defaults https://review.opendev.org/c/openstack/tripleo-heat-templates/+/823741 | 02:05 |
*** ysandeep|out is now known as ysandeep | 02:15 | |
*** rcastillo|rover is now known as rcastillo | 02:15 | |
*** pmannidi is now known as pmannidi|brb | 02:55 | |
opendevreview | Merged openstack/tripleo-heat-templates master: neutron: Stop setting the allow_overlapping_ips parameter https://review.opendev.org/c/openstack/tripleo-heat-templates/+/817347 | 02:58 |
opendevreview | Ade Lee proposed openstack/tripleo-validations master: Make sure fact is available to find hostname https://review.opendev.org/c/openstack/tripleo-validations/+/823746 | 04:32 |
opendevreview | Ade Lee proposed openstack/tripleo-validations master: Remove check for certmonger user service https://review.opendev.org/c/openstack/tripleo-validations/+/823747 | 04:32 |
opendevreview | Ade Lee proposed openstack/tripleo-validations master: Remove references to novajoin-server and novajoin-notifier https://review.opendev.org/c/openstack/tripleo-validations/+/823748 | 04:32 |
opendevreview | Ade Lee proposed openstack/tripleo-validations master: Remove domain equivalence check https://review.opendev.org/c/openstack/tripleo-validations/+/823749 | 04:32 |
opendevreview | Ade Lee proposed openstack/tripleo-validations master: Simplify and consolidate files https://review.opendev.org/c/openstack/tripleo-validations/+/823770 | 04:32 |
*** rlandy|ruck is now known as rlandy|out | 05:39 | |
*** ysandeep is now known as ysandeep|afk | 06:40 | |
*** amoralej|off is now known as amoralej | 08:09 | |
*** ysandeep|afk is now known as ysandeep | 09:38 | |
hjensas | mwhahaha: on backporting https://review.opendev.org/q/topic:%22overcloud_deploy_clean%22+(status:open%20OR%20status:merged) | 10:16 |
hjensas | mwhahaha: I did not plan to backport that, do we have so many large changes that need backporting to justify backporting them? (I expect those patches won't backport cleanly.) | 10:19 |
*** rlandy|out is now known as rlandy|ruck | 11:15 | |
*** dviroel|afk is now known as dviroel | 11:25 | |
opendevreview | Ronelle Landy proposed openstack/tripleo-ci master: WIP: Remove fwupd-redfish.conf file from overcloud images https://review.opendev.org/c/openstack/tripleo-ci/+/823736 | 11:59 |
opendevreview | Arx Cruz proposed openstack/tripleo-ci master: WIP: Remove fwupd-redfish.conf file from overcloud images https://review.opendev.org/c/openstack/tripleo-ci/+/823736 | 12:30 |
Tengu | fultonj: hello there! It seems the docker images for prometheus, as shipped from quay.ceph.io, is built using the old manifest format, resulting in an error when we try to push it to some local docker registry: err.detail="manifest schema v1 unsupported" - this is specifically for quay.ceph.io/prometheus/prometheus:v2.7.2 . Do you happen to know the owner of this image? | 12:34 |
Tengu | fultonj: and, by extension, would it be possible to get the newer schema format so that things are easier to store in a local registry? | 12:34 |
Tengu | it's probably missing an option to either `podman build' or `buildah' in order to set the format to "docker" instead of OCI - one may export BUILDAH_FORMAT env var as well. | 12:36 |
opendevreview | Douglas Viroel proposed openstack/tripleo-quickstart-extras master: DNM - Reverting DLRN config filename https://review.opendev.org/c/openstack/tripleo-quickstart-extras/+/823793 | 12:42 |
*** jcapitao is now known as jcapitao_lunch | 12:53 | |
opendevreview | mbu proposed openstack/python-tripleoclient stable/train: Passes playbook to the validation action as a list https://review.opendev.org/c/openstack/python-tripleoclient/+/823760 | 13:22 |
opendevreview | mbu proposed openstack/python-tripleoclient stable/train: Passes playbook to the validation action as a list https://review.opendev.org/c/openstack/python-tripleoclient/+/823760 | 13:22 |
*** rcastillo is now known as rcastillo|rover | 13:28 | |
opendevreview | chandan kumar proposed openstack/tripleo-ansible master: Blacklist ipmi_si for CentOS 8 distros https://review.opendev.org/c/openstack/tripleo-ansible/+/823798 | 13:37 |
*** amoralej is now known as amoralej|lunch | 13:40 | |
*** jcapitao_lunch is now known as jcapitao | 14:05 | |
*** akahat_ is now known as akahat | 14:07 | |
opendevreview | David Hill proposed openstack/tripleo-ansible master: Added all Controller* hosts to prepare-overcloud-backup.yaml https://review.opendev.org/c/openstack/tripleo-ansible/+/823740 | 14:11 |
fultonj | hi Tengu, how can I reproduce the error with quay.ceph.io/prometheus/prometheus:v2.7.2 ? | 14:19 |
fultonj | e.g. can I run container image prepare from a recent undercloud? | 14:20 |
fultonj | Tengu: i'd like to reproduce it in a non-openstack context and then get help from ceph | 14:21 |
*** amoralej|lunch is now known as amoralej | 14:22 | |
opendevreview | David Hill proposed openstack/tripleo-ansible master: Added all Controller* hosts to prepare-overcloud-backup.yaml https://review.opendev.org/c/openstack/tripleo-ansible/+/823740 | 14:31 |
mwhahaha | hjensas_: unfortunately i need to backport something that's in the validate_args bits and without all those patches it's very conflicty. I guess i'll just rework my cherry pick to be wallaby specific | 14:45 |
*** ysandeep is now known as ysandeep|dinner | 14:45 | |
mwhahaha | Tengu, fultonj: i ran into that issue previously with container image prepare. I thought I patched it | 14:46 |
fultonj | mwhahaha: do you have your patch from that handy? | 14:46 |
mwhahaha | https://review.opendev.org/c/openstack/tripleo-common/+/812963 | 14:47 |
mwhahaha | was just looking for it | 14:47 |
* fultonj running openstack tripleo container image prepare right now to try to reproduce | 14:47 | |
mwhahaha | however the prometheus container is built is using an old format | 14:48 |
mwhahaha | it should work with triple now | 14:48 |
mwhahaha | with that patch, but it's using old metadata format | 14:48 |
fultonj | so we want them to move to the new format either way | 14:48 |
Tengu | fultonj: manual tagging and push to a local docker registr:2 is enough | 14:48 |
mwhahaha | it'd be nice | 14:48 |
mwhahaha | ah so you he's pushing to docker-distribution | 14:49 |
mwhahaha | it should work with image-serve | 14:49 |
Tengu | fultonj: am about to leave, will update with command later | 14:49 |
mwhahaha | (never thought i'd say that) | 14:49 |
Tengu | mwhahaha: hehe | 14:49 |
Tengu | miracles exist! | 14:49 |
Tengu | anyway, am out now | 14:49 |
* fultonj cannot reproduce `openstack tripleo container image prepare` probably because of mwhahaha's patch | 14:49 | |
mwhahaha | maybe | 14:50 |
mwhahaha | anwyay you can use skopeo to inspect the metadata to compare | 14:50 |
mwhahaha | skopeo inspect --raw docker://quay.ceph.io/prometheus/prometheus:v2.7.2 | 14:50 |
fultonj | skopeo inspect --raw docker://quay.ceph.io/prometheus/prometheus:v2.7.2 | jq .schemaVersion | 14:52 |
fultonj | = 1 | 14:52 |
fultonj | i'll ask cephers if they know who maintains this thing | 14:53 |
opendevreview | Tom Barron proposed openstack/tripleo-ansible stable/train: Use service_net_map for ganesha_vip https://review.opendev.org/c/openstack/tripleo-ansible/+/822044 | 14:53 |
mwhahaha | skopeo inspect --raw docker://quay.io/centos/centos for comparison which uses v2 | 14:53 |
fultonj | thanks mwhahaha | 14:54 |
opendevreview | Damien Ciabrini proposed openstack/puppet-tripleo master: WIP mariadb auth_ed25519 with FIPS crypto https://review.opendev.org/c/openstack/puppet-tripleo/+/822189 | 14:55 |
fultonj | mwhahaha Tengu guits is looking into it | 14:59 |
opendevreview | Arx Cruz proposed openstack/tripleo-ci master: Remove fwupd-redfish.conf file from overcloud images https://review.opendev.org/c/openstack/tripleo-ci/+/823736 | 15:03 |
opendevreview | Arx Cruz proposed openstack/tripleo-ci master: Remove fwupd-redfish.conf file from overcloud images https://review.opendev.org/c/openstack/tripleo-ci/+/823736 | 15:05 |
*** ysandeep|dinner is now known as ysandeep | 15:10 | |
*** dviroel is now known as dviroel|lunch | 15:37 | |
opendevreview | Sandeep Yadav proposed openstack/tripleo-common master: Don't enable module stream in rhel9 https://review.opendev.org/c/openstack/tripleo-common/+/823820 | 15:43 |
*** ysandeep is now known as ysandeep|out | 15:49 | |
opendevreview | Julia Kreger proposed openstack/python-tripleoclient master: Fix ironic command references https://review.opendev.org/c/openstack/python-tripleoclient/+/823821 | 15:59 |
opendevreview | chandan kumar proposed openstack/tripleo-ansible master: Blacklist ipmi_si for CentOS 8 distros https://review.opendev.org/c/openstack/tripleo-ansible/+/823798 | 16:24 |
*** dviroel|lunch is now known as dviroel | 16:33 | |
Tengu | fultonj: ah, nice! | 16:34 |
fultonj | Tengu: yeah, he said he'd update it to v2 | 16:35 |
Tengu | fultonj: \o/ | 16:38 |
Tengu | fultonj: any ETA? just wanting to know if it's still worth the effort to tweak some image-serve in the meanwhile :) | 16:50 |
fultonj | no ETA | 16:50 |
* fultonj is going to re-ping him next week for an update | 16:50 | |
* fultonj is expecting sometime next week | 16:51 | |
Tengu | fultonj: wokay. well. Won't lose more time trying to get that image-serve hack working then | 16:52 |
fultonj | Tengu: perhaps deploy without ceph dashboard until it's fixed | 16:53 |
Tengu | that way I'll just use my registry pod (with redis) | 16:53 |
Tengu | fultonj: it's not deploying :). it's mirroring | 16:53 |
Tengu | fultonj: basically, I'm working on a podman container that will just get python3-tripleoclient and its deps, a small override setting "push_destination: my-registry" as well as the actual "tag", and run it once a day. | 16:54 |
Tengu | that will allow me to get local mirror of *all* containers, making any deploy really faster :) | 16:54 |
Tengu | I also have a container allowing to *build* containers, but I'm hitting an issue with cs9 container (cgroups, apparently - didn't debug more yet) | 16:55 |
fultonj | sounds like a good idea | 16:55 |
fultonj | your fast deploy could exclude that contianer if it's not mirrored | 16:56 |
Tengu | and the registry is actually a pod running registry:2 + redis, exposing the 5000 port on my internal interface, and so on | 16:56 |
fultonj | your mirroring could exclude that container too | 16:56 |
fultonj | until it's available | 16:56 |
Tengu | right. iirc there's an "exclude" key | 16:56 |
fultonj | yes | 16:56 |
Tengu | I forgot about that one :D | 16:56 |
fultonj | https://docs.openstack.org/project-deploy-guide/tripleo-docs/latest/deployment/container_image_prepare.html#layering-image-preparation-entries | 16:58 |
Tengu | yup | 16:58 |
Tengu | excludes: [prometheus] | 16:59 |
Tengu | should be enough. Will check for grafana.. | 16:59 |
Tengu | let's see. thanks for the reminder about that "excludes" thing, fultonj ! | 17:00 |
fultonj | glad i could help in some way | 17:00 |
Tengu | :) | 17:00 |
Tengu | well, you just helped with a workaround and a long-term solution :) | 17:01 |
fultonj | Tengu: on another topic... are you using a cent9 deployed tripleo-lab? | 17:01 |
fultonj | cent9 hypervisor? | 17:01 |
fultonj | cent9 cloud images? | 17:01 |
* fultonj has gotten away with cent8 so far and hasn't checked the lastest on 9 for tripleo-lab | 17:05 | |
opendevreview | Douglas Viroel proposed openstack/tripleo-quickstart-extras master: DNM - Testing DLRN config filename change https://review.opendev.org/c/openstack/tripleo-quickstart-extras/+/823793 | 17:08 |
Tengu | fultonj: sooo. there's a centos9 branch on tripleo-lab that SHOULD allow deploying cs9 VMs on a cs8 hypervisor | 17:09 |
Tengu | I don't remember if I already merged it on master though... | 17:10 |
fultonj | This branch is 3 commits ahead of master. | 17:10 |
Tengu | nah, not merged - no wonder, I didn't test it that much yet | 17:10 |
fultonj | https://github.com/cjeanner/tripleo-lab/tree/centos-9 | 17:10 |
Tengu | ah, there's a PR | 17:11 |
fultonj | Tengu: i'll have to test that sometime | 17:11 |
Tengu | fultonj: please, any feedback would be nice :) | 17:11 |
Tengu | I worked with the maintainer of libguestfs in order to get an updated version of the appliance, and that one works - this is already a good point. | 17:11 |
fultonj | no ETA on that either :) | 17:12 |
Tengu | heh | 17:12 |
Tengu | Once I get my mirroring square, I should be able to test. | 17:12 |
fultonj | but i do intend to use it | 17:12 |
Tengu | probably next week. | 17:12 |
Tengu | fultonj: ah, apparently node-exporter has the same issue | 17:14 |
opendevreview | Julia Kreger proposed openstack/python-tripleoclient master: Abort root device detection if an MD device is found https://review.opendev.org/c/openstack/python-tripleoclient/+/823825 | 17:21 |
fultonj | Tengu: right so exclude that too | 17:21 |
Tengu | fultonj: yep, I'm running after failing containers :) | 17:23 |
Tengu | ah, dang. I've set the excludes at the wrong level. it won't help, right. | 17:25 |
fultonj | skopeo inspect --raw docker://quay.ceph.io/prometheus/node-exporter:v0.17.0| jq .schemaVersionn | 17:25 |
fultonj | Tengu: i'll ask if guits could please fix ^ too | 17:25 |
Tengu | ++ | 17:28 |
Tengu | w00t! | 17:30 |
opendevreview | Merged openstack/tripleo-heat-templates master: Move *NetworkConfigTemplate to parameter_defaults https://review.opendev.org/c/openstack/tripleo-heat-templates/+/823741 | 17:31 |
*** amoralej is now known as amoralej|off | 17:50 | |
mwhahaha | sshnaidm: i got the reproducer working with libvirt, have you got it working with openstack? i'm currently hitting configure-swap problems | 17:53 |
sshnaidm | mwhahaha, yeah, same problem got yesterday, it's because cloud-init disk is /dev/vdb and the job is trying to create there swap | 17:54 |
mwhahaha | ok so it's not just me :D | 17:55 |
sshnaidm | need to look if swap device is configurable, or why cloud-init is not removed | 17:55 |
sshnaidm | maybe to try a different flavor | 17:55 |
sshnaidm | because it worked before | 17:55 |
mwhahaha | it's cause it's a 5gb partition | 17:59 |
mwhahaha | ci.m1.large.ephemeral | 8192 | 35 | 5 | 8 | True | 18:00 |
mwhahaha | and it wants to create an 8gb partition | 18:00 |
mwhahaha | so yea | 18:00 |
mwhahaha | sdb on that one is not cloud-init, it's a 5gb ephemeral thing | 18:03 |
mwhahaha | i don't think there's a flavor like what the ci nodes get | 18:04 |
sshnaidm | mwhahaha, try use the usual m1.* | 18:14 |
sshnaidm | mwhahaha, you are using psi, right? | 18:14 |
opendevreview | mbu proposed openstack/python-tripleoclient stable/train: Passes playbook to the validation action as a list https://review.opendev.org/c/openstack/python-tripleoclient/+/823760 | 18:24 |
opendevreview | Douglas Viroel proposed openstack/tripleo-quickstart-extras master: DNM - Testing DLRN config filename change https://review.opendev.org/c/openstack/tripleo-quickstart-extras/+/823793 | 18:24 |
*** rlandy|ruck is now known as rlandy|ruck|biab | 18:34 | |
Tengu | fultonj: fyi, I can confirm only those two containers have the wrong format. Which is really cool :). | 18:39 |
mwhahaha | sshnaidm: yea it's psi but it's using that flavor by default | 18:40 |
fultonj | that's good | 18:41 |
Tengu | and I get the tripleomastercentos9 mirror as well. | 18:53 |
Tengu | that's a great thing. Next week, i'll test tripleoab with cs9. | 18:53 |
Tengu | and now - weekend for real :D | 18:54 |
opendevreview | Merged openstack/tripleo-validations master: fix(molecule): Add ability to run molecule on Fedora https://review.opendev.org/c/openstack/tripleo-validations/+/822336 | 18:57 |
*** rlandy|ruck|biab is now known as rlandy|ruck | 19:13 | |
mwhahaha | sshnaidm: i think zuul is picking ci.m1.large.ephemeral instead of m1.large for some reason | 19:16 |
opendevreview | Julia Kreger proposed openstack/python-tripleoclient master: Fix ironic command references https://review.opendev.org/c/openstack/python-tripleoclient/+/823821 | 19:35 |
*** dviroel is now known as dviroel|afk | 19:37 | |
opendevreview | Alex Schultz proposed openstack/tripleo-ci master: [WIP] Allow swap size to be configurable on a job basis https://review.opendev.org/c/openstack/tripleo-ci/+/823864 | 20:14 |
opendevreview | Alex Schultz proposed openstack/tripleo-ci master: [WIP] Allow swap size to be configurable on a job basis https://review.opendev.org/c/openstack/tripleo-ci/+/823864 | 20:43 |
opendevreview | Alex Schultz proposed openstack/tripleo-ci master: [WIP] Allow swap size to be configurable on a job basis https://review.opendev.org/c/openstack/tripleo-ci/+/823864 | 21:08 |
opendevreview | Merged openstack/tripleo-upgrade stable/train: Fast and furious upgrade https://review.opendev.org/c/openstack/tripleo-upgrade/+/817053 | 21:13 |
opendevreview | Tom Barron proposed openstack/tripleo-heat-templates stable/train: Remove ganesha_vip extra config workaround https://review.opendev.org/c/openstack/tripleo-heat-templates/+/823750 | 21:33 |
opendevreview | Merged openstack/tripleo-ci master: Remove fwupd-redfish.conf file from overcloud images https://review.opendev.org/c/openstack/tripleo-ci/+/823736 | 21:43 |
opendevreview | Harald Jensås proposed openstack/python-tripleoclient stable/wallaby: overcloud node extract provisioned - fixed_ips https://review.opendev.org/c/openstack/python-tripleoclient/+/822416 | 22:47 |
opendevreview | David Hill proposed openstack/tripleo-heat-templates master: Add OVNContainerCpusetCpus https://review.opendev.org/c/openstack/tripleo-heat-templates/+/820535 | 23:09 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!