Friday, 2023-09-08

mnasiadkafrickler: it seems validate-all-file.py was not validating block tasks - can you have a look in https://review.opendev.org/c/openstack/kolla-ansible/+/894155 ?05:25
mnasiadkaI'll rebase it on the fixed code now, so it doesn't fail - but you check the historical run for logs (https://zuul.opendev.org/t/openstack/build/3656d5fa57c34c60aebcfdb8a849dd02)05:26
opendevreviewMichal Nasiadka proposed openstack/kolla-ansible master: CI: add block support to validate-all-file.py  https://review.opendev.org/c/openstack/kolla-ansible/+/89415505:26
opendevreviewMarcin Juszkiewicz proposed openstack/kolla master: build: distutils.StrictVersion() is deprecated  https://review.opendev.org/c/openstack/kolla/+/88674105:45
opendevreviewMarcin Juszkiewicz proposed openstack/kolla master: base: switch to new Grafana key  https://review.opendev.org/c/openstack/kolla/+/88608605:46
opendevreviewMarcin Juszkiewicz proposed openstack/kolla-ansible master: init-runonce: update CirrOS to 0.6.2  https://review.opendev.org/c/openstack/kolla-ansible/+/88484905:46
hrwrebased05:46
fricklermnasiadka: I don't understand this. since when is "become: true" on the block not working? and if it doesn't work, why isn't it dropped?05:48
opendevreviewMerged openstack/kolla-ansible stable/yoga: [yoga only] followup ovn clustering fix  https://review.opendev.org/c/openstack/kolla-ansible/+/89419406:37
mnasiadkafrickler: right, missed that - need to update validate-all-file, argh06:40
mnasiadkafrickler: need more coffee ;)06:59
opendevreviewDr. Jens Harbott proposed openstack/kolla-ansible master: DNM: Test octavia amp subnet  https://review.opendev.org/c/openstack/kolla-ansible/+/89421907:05
fricklermnasiadka: I wasn't sure if it was me not seeing something important07:06
frickleralso fyi ^^ is likely a regression in a-o-c07:07
mnasiadkaNo breakage no fun ;)07:07
fricklerI hope https://review.opendev.org/c/openstack/ansible-collections-openstack/+/894254 is a reproducer07:13
fricklermnasiadka: the fun thing is that this seems to be breaking things only with OVN, while octavia in our CI still runs on OVS07:14
Viiihello, how can I overwrite a newer version of the plugin when creating an image?07:19
Viii"kolla-build.conf" "[neutron-server-plugin-vpnaas-agent]" "type = local" "location=path/neutron-vpnaas-22.1.0.dev3.tar.gz07:20
ViiiI have an error while building image07:20
Viii"ERROR: Cannot install neutron-vpnaas 22.0.0.0rc2.dev2 (from /plugins/neutron-vpnaas-22.0.0.0rc2.dev2) and neutron-vpnaas 22.1.0.dev3 (from /plugins/neutron-vpnaas-22.1.0.dev3) because these package versions have conflicting dependencies."07:20
mnasiadkafrickler: so you're saying we should have an ovn flavor of the octavia job?07:35
mnasiadkaor maybe we need to reorganise the CI jobs and include octavia in the standard ones07:36
mmalchukViii use git07:38
mmalchuktype = git07:39
mmalchuklocation = http://....07:39
mmalchukreference = your_branch07:39
opendevreviewMichal Nasiadka proposed openstack/kolla-ansible master: CI: add block support to validate-all-file.py  https://review.opendev.org/c/openstack/kolla-ansible/+/89415507:40
fricklermnasiadka: maybe we should just switch to OVN everywhere. or maybe next cycle, not sure yet07:41
mnasiadkafrickler: first we need a precheck to not break existing ML2/OVS environments, but we have someone working on that in SHPC - once that's done, we could change the default.07:42
fricklerare you thinking about migrating OVS to OVN? otherwise the precheck would be easy. and I'm not convinced migration during normal kolla operations is possible, I think it might need to be a dediccated tools07:44
ViiiI need to use VPNaaS for OVN07:47
fricklerViii: I don't think that is working yet07:48
Viiiversion neutron-vpnaas-22.1.0.dev3.tar.gz supports this07:48
fricklerwith https://review.opendev.org/c/openstack/neutron-vpnaas/+/765353 still open I very much doubt this. anyway there's no support for it in kolla yet I guess07:51
Viiihttps://pastebin.com/Jd0YjeKn07:51
ViiiI know it doesn't support it, so I want to build an image with the new version of the plugin07:53
fricklerthen what mmalchuk says. with "your_branch=master" I guess07:55
opendevreviewMichal Nasiadka proposed openstack/kolla-ansible master: CI: add block support to validate-all-file.py  https://review.opendev.org/c/openstack/kolla-ansible/+/89415508:03
mnasiadkafrickler: fixed ^^08:03
mnasiadkaargh, pep808:07
fricklerlinting the linter failed? ;)08:09
mnasiadkapep8: OK (14.92=setup[12.01]+cmd[2.91] seconds)08:12
mnasiadkashould be good now ;)08:12
opendevreviewMichal Nasiadka proposed openstack/kolla-ansible master: CI: add block support to validate-all-file.py  https://review.opendev.org/c/openstack/kolla-ansible/+/89415508:12
mnasiadkaI want to backport that all the way to stable/yoga and add a check there for not doing kolla_container_engine - because we backported 5th patch with that and I had to do followup again :D08:12
SvenKieskecurrently going through the version bumps for bobcat cycle, do we still need CentOS 7 Repositories? https://opendev.org/openstack/kolla/src/branch/master/docker/base/crmsh.repo09:27
SvenKieskethey are disabled, can't we delete this from master?09:28
mnasiadkahuh09:30
mnasiadkasurely we can, question what was installed from that repo09:30
SvenKieskedon't know, I don't use EL since...5-8 years?09:31
SvenKieskeI'll provide a changeset and we can discuss there with the experts09:32
SvenKieskeseemingly was used for corsync/pacemaker ha stuff09:32
opendevreviewSven Kieske proposed openstack/kolla master: delete seemingly unused EL7 repo  https://review.opendev.org/c/openstack/kolla/+/89429009:35
SvenKieskedo we run EL7 anywhere still? some old yoga stuff maybe?09:36
SvenKieskeokay, this is referenced by the hacluster role, but that role seems to not have been touched since 4 years, and it's still using EL7 packages? does this even work?09:40
mnasiadkano09:41
mnasiadkaEL7 was Wallaby I think?09:41
SvenKieskethe referenced URL is broken; I doubt this "hacluster" role works at all http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/CentOS_CentOS-7/ is wrong09:41
SvenKieskecorrect URL: http://download.opensuse.org/repositories/network:/ha-clustering:/Stable/CentOS_7/09:42
mnasiadkaIs that repo file used anywhere? Like do we add it in base image?09:42
SvenKieskeshould I extend the changeset to delete that role? :D09:42
SvenKieskethe base url of that repo is a 404, so I hope we don't use it..investigating09:42
mnasiadkawe used that for masakari, but might be crmsh is now in some standard CentOS/Rocky repos09:43
SvenKieskethis got introduced here: https://review.opendev.org/c/openstack/kolla/+/66876509:43
mnasiadkaif hacluster containers are buildable, then I guess everything is all right and it's just a dangling file09:44
SvenKieske # FIXME(mgoddard): Not available for CentOS 8 yet.09:44
SvenKieske    #   'crmsh.repo',09:44
SvenKieskethat file is nowhere else referenced, so I guess it's save to delete09:45
SvenKieskeat least not by direct name09:45
SvenKieskeokay, nice, moving on, changeset is up for review if anybody cares :) now upgrade those versions..09:46
SvenKieskemhm, I guess we don't want to switch from mariadb 10.11 to 11.1?09:54
SvenKieskeI'll include it, just for the sake of it09:54
mnasiadkammalchuk: I thought we agreed we could EOL Xena beginning of September? :)09:59
SvenKieskedoing this, I'm wondering: why do we mix packages from distributions with packages from upstream on another distribution? wouldn't it be more secure if we - in cases where $dist doesn't offer $version we switch all affected repos to upstream, to get the same compilation flags etc?10:00
mnasiadkaSvenKieske: we use mariadb LTS, 11.1 is not LTS, right?10:00
mmalchukmnasiadka as I said will support Xena until the end10:00
SvenKieskeah only LTS, wasn't aware, makes sense10:00
mnasiadkahttps://mariadb.com/kb/en/mariadb-server-release-dates/10:00
mnasiadkayeah, 10.11 is long term stable10:00
mmalchukmnasiadka agreed, but lets merge fixes until tagged10:01
mmalchukmnasiadka ok?10:01
mnasiadkasure10:01
mnasiadkabbezak: want to raise xena EOL patch? :)10:01
mmalchukbbezak want review and merge last Xena patches? please10:02
mmalchukkolla: https://review.opendev.org/c/openstack/kolla/+/890541 and https://review.opendev.org/c/openstack/kolla/+/89385410:03
mmalchukkolla-ansible: https://review.opendev.org/c/openstack/kolla-ansible/+/89342210:03
mmalchukhttps://review.opendev.org/c/openstack/kolla-ansible/+/893954 and https://review.opendev.org/c/openstack/kolla-ansible/+/89344110:04
mmalchukthats all10:04
mmalchukkayobe still lack of review even in master ;(10:05
bbezaksure mnasiadka10:08
SvenKieskeoh no, td-agent upstream got renamed to "fluent-package" new LTS Version is 5 instead of 4..10:08
mnasiadkanice, good bump :)10:10
fricklerrocky9 seems to be acting up, if someone wants to take a look https://zuul.opendev.org/t/openstack/build/5efa57a4ffde47518a5a5d37f159d9ca10:14
mnasiadkafrickler: that's weird, but it might happen when there are no Nova services registered, I've seen this in the past - maybe we need to have a look10:20
opendevreviewMerged openstack/kolla master: Add python 3.10 to setup.cfg metadata  https://review.opendev.org/c/openstack/kolla/+/88654710:35
opendevreviewMerged openstack/kolla stable/yoga: Add comprehensive checks for container restarts  https://review.opendev.org/c/openstack/kolla/+/89287810:35
opendevreviewMerged openstack/kolla stable/2023.1: Revert "Disable build of Gnocchi containers"  https://review.opendev.org/c/openstack/kolla/+/89053811:17
opendevreviewMichal Nasiadka proposed openstack/kolla-ansible master: CI: bump up ansible-lint  https://review.opendev.org/c/openstack/kolla-ansible/+/89429511:20
opendevreviewMerged openstack/kolla-ansible master: Remove duplicate whitespace  https://review.opendev.org/c/openstack/kolla-ansible/+/89304511:27
opendevreviewMerged openstack/kolla stable/zed: Revert "Disable build of Gnocchi containers"  https://review.opendev.org/c/openstack/kolla/+/89053911:27
opendevreviewMerged openstack/kolla-ansible stable/xena: Use better default bind address for ironic-tftp  https://review.opendev.org/c/openstack/kolla-ansible/+/89342211:38
opendevreviewMerged openstack/kolla-ansible stable/xena: Fix designate-producers not running properly  https://review.opendev.org/c/openstack/kolla-ansible/+/89344111:38
mmalchukmnasiadka you miss the Yoga change https://review.opendev.org/c/openstack/kolla/+/89054011:46
mmalchukcores, please +W last two kolla change: https://review.opendev.org/q/project:openstack/kolla+branch:stable/xena+status:open11:48
mmalchukand the last one in kolla-ansible: https://review.opendev.org/q/project:openstack/kolla-ansible+branch:stable/xena+status:open11:50
opendevreviewJohn Garbutt proposed openstack/kolla-ansible master: WIP: Overide of [default]host in nova-compute-ironic  https://review.opendev.org/c/openstack/kolla-ansible/+/89429813:24
opendevreviewMerged openstack/kolla-ansible stable/zed: Use better default bind address for ironic-tftp  https://review.opendev.org/c/openstack/kolla-ansible/+/89338613:56
opendevreviewMerged openstack/kolla-ansible stable/2023.1: Use better default bind address for ironic-tftp  https://review.opendev.org/c/openstack/kolla-ansible/+/89338513:58
opendevreviewMerged openstack/kolla-ansible stable/xena: Fix typo in endpoint influxdb_internal_endpoint variable  https://review.opendev.org/c/openstack/kolla-ansible/+/89395414:00
opendevreviewDoug Szumski proposed openstack/kolla-ansible master: Support customising Nova config at the individual service level  https://review.opendev.org/c/openstack/kolla-ansible/+/89429814:15
opendevreviewBartosz Bezak proposed openstack/kolla-ansible master: CI: add q35 hardware machine type to tests  https://review.opendev.org/c/openstack/kolla-ansible/+/89419914:28
opendevreviewBartosz Bezak proposed openstack/kolla-ansible master: Configure Nova libvirt.num_pcie_ports to 16 by default  https://review.opendev.org/c/openstack/kolla-ansible/+/76830114:30
opendevreviewMerged openstack/kolla stable/xena: Revert "Disable build of Gnocchi containers"  https://review.opendev.org/c/openstack/kolla/+/89054114:41
opendevreviewWill Szumski proposed openstack/kolla-ansible master: Support exposing prometheus_server externally  https://review.opendev.org/c/openstack/kolla-ansible/+/82869514:50
opendevreviewMerged openstack/kolla stable/xena: Improve kolla-toolbox offline support.  https://review.opendev.org/c/openstack/kolla/+/89385415:00
opendevreviewWill Szumski proposed openstack/kolla-ansible master: Support exposing prometheus_server externally  https://review.opendev.org/c/openstack/kolla-ansible/+/82869515:18
opendevreviewJake Hutchinson proposed openstack/kolla-ansible master: Ironic parameter rework and default NTP server  https://review.opendev.org/c/openstack/kolla-ansible/+/89303115:49
opendevreviewMerged openstack/kolla-ansible stable/yoga: Fix typo in endpoint influxdb_internal_endpoint variable  https://review.opendev.org/c/openstack/kolla-ansible/+/89395315:56
opendevreviewBartosz Bezak proposed openstack/kolla-ansible master: CI: add q35 hardware machine type to tests  https://review.opendev.org/c/openstack/kolla-ansible/+/89419916:03
opendevreviewBartosz Bezak proposed openstack/kolla-ansible master: Configure Nova libvirt.num_pcie_ports to 16 by default  https://review.opendev.org/c/openstack/kolla-ansible/+/76830116:05
spatelis barbican-ui is supported in kolla-ansible?16:13
spatelmmalchuk ^16:13
spatelI have seen its not so very curious https://review.opendev.org/c/openstack/kolla-ansible/+/89128416:13
opendevreviewChristian Berendt proposed openstack/kolla-ansible master: senlin: add missing cafile parameter  https://review.opendev.org/c/openstack/kolla-ansible/+/89438316:14
opendevreviewChristian Berendt proposed openstack/kolla-ansible master: senlin: add missing cafile parameter  https://review.opendev.org/c/openstack/kolla-ansible/+/89438316:24
opendevreviewWill Szumski proposed openstack/kolla-ansible master: Support exposing prometheus_server externally  https://review.opendev.org/c/openstack/kolla-ansible/+/82869516:31
opendevreviewJuan Pablo Suazo proposed openstack/kolla-ansible master: Deploy Glance with S3 backend support  https://review.opendev.org/c/openstack/kolla-ansible/+/84461416:34
opendevreviewWill Szumski proposed openstack/kolla-ansible master: Support exposing prometheus_server externally  https://review.opendev.org/c/openstack/kolla-ansible/+/82869516:45
opendevreviewWill Szumski proposed openstack/kolla-ansible master: Support exposing prometheus_server externally  https://review.opendev.org/c/openstack/kolla-ansible/+/82869516:55
opendevreviewDoug Szumski proposed openstack/kolla-ansible master: Support customising Nova config at the individual service level  https://review.opendev.org/c/openstack/kolla-ansible/+/89429816:57
atmarkhello, I'm trying to get designate playbook to have designate-mdns service listen on `dns_interface` and `kolla_external_vip_interface` but I'm undefined network. 18:16
atmarkI'm getting undefined networK*18:17
atmarkI have this in my designate.conf 18:17
atmarklisten = {{ 'dns' | kolla_address | put_address_in_context('url') }}:{{ designate_mdns_port }}, {{ 'kolla_external_vip' | kolla_address | put_address_in_context('url') }}:{{ designate_mdns_port }}18:17
atmarkdns_interface defaults to network_interface while kolla_external_vip_interface is mapped to an actual interface 18:18
atmarkerror I'm getting ` FAILED! => {"msg": "Address family undefined for network 'kolla_external_vip' (set 'kolla_external_vip_address_family')"} `18:19
atmarkI figured it out18:44
atmarklisten = {{ 'dns' | kolla_address | put_address_in_context('url') }}:{{ designate_mdns_port }}, {{ kolla_external_vip_address }}:{{ designate_mdns_port }}18:44

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