akahat|rover | o/ | 04:32 |
---|---|---|
marios | o/ | 06:09 |
*** ysandeep|out is now known as ysandeep | 06:20 | |
ysandeep | akahat|rover, good morning o/ | 06:25 |
chandankumar | ysandeep: marios akahat|rover Good morning and Happy new year o/ | 06:26 |
ysandeep | akahat|rover, You need to update promotion-testing-hash-* files for periodic job to pick it up. https://review.rdoproject.org/zuul/status#46525 will not work | 06:26 |
ysandeep | chandankumar, happy new year | 06:27 |
marios | \o chandankumar happy new year | 06:27 |
*** soniya is now known as soniya29 | 06:32 | |
chandankumar | thank you ysandeep: akahat|rover rlandy|out for taking care of growvols bug :-) | 06:32 |
ysandeep | That's a weird bug, we are now saying newer lvm2/device-mapper is culprit, but I had bad results with older lvm2/device-mapper as well. | 06:37 |
ysandeep | https://logserver.rdoproject.org/37/28537/66/check/periodic-tripleo-ci-centos-9-ovb-1ctlr_2comp-featureset020-zed/2f30180/logs/undercloud/home/zuul/overcloud_node_provision.log.txt.gz | 06:37 |
ysandeep | overcloud package list: https://logserver.rdoproject.org/37/28537/66/check/periodic-tripleo-ci-centos-9-ovb-1ctlr_2comp-featureset020-zed/2f30180/logs/overcloud-controller-0/var/log/extra/package-list-installed.txt.gz | 06:38 |
marios | :/ so we are still chasing that one then | 06:38 |
chandankumar | ysandeep: may be we can first get passing job package exclude list patch merged and then start removing packages packages one by one or combination? | 06:41 |
chandankumar | to find the actual culprit | 06:42 |
marios | akahat|rover: pojadhav|ruck: sent invite for sync call at usual time please check inbox | 06:44 |
ysandeep | chandankumar, with https://review.opendev.org/c/openstack/tripleo-quickstart/+/868605/13/config/release/tripleo-ci/CentOS-9/master.yml last 8/8 job pass node provisioning | 06:45 |
ysandeep | let me reduce the diff to just device-mapper/lvm2 to see if that enough(as per steve's comment).. but last week i hit issues with older device-mapper/lvm2 too. | 06:45 |
marios | could it be something really nasty like combination of kernel and lvm for example | 06:46 |
ysandeep | yes could be | 06:46 |
marios | ysandeep: arent we already testing that per akahat https://review.opendev.org/c/openstack/tripleo-quickstart/+/869099/3/config/release/tripleo-ci/CentOS-9/master.yml | 06:46 |
chandankumar | not sure selinux-policy package | 06:46 |
marios | ysandeep: i mean 08:45 < ysandeep> let me reduce the diff to just device-mapper/lvm2 to see if that enough(as per steve's comment).. but last week i hit issues with older device-mapper/lvm2 too. | 06:46 |
ysandeep | marios, yes I mean to update akahat|rover patch, his patch won't work with periodic job | 06:47 |
ysandeep | >> <ysandeep> akahat|rover, You need to update promotion-testing-hash-* files for periodic job to pick it up. https://review.rdoproject.org/zuul/status#46525 will not work | 06:47 |
marios | ysandeep: ah ok | 06:47 |
pojadhav|ruck | marios, accepted ! | 06:49 |
marios | \o | 06:51 |
marios | akahat|rover: ysandeep: chandankumar: can you check https://review.opendev.org/c/openstack/tripleo-quickstart/+/869121 rabi posted revert and content providers look good there | 06:53 |
marios | (when you have time please thanks) | 06:53 |
chandankumar | done! | 06:54 |
ysandeep | double +wed :D | 06:55 |
marios | thanks :) | 06:56 |
akahat|rover | ysandeep, o/ hey.. thanks i'll update the patch.. | 07:01 |
ysandeep | akahat|rover, nw, I already posted it | 07:02 |
akahat|rover | marios, accepted invite. thanks! | 07:02 |
marios | o/ | 07:02 |
* akahat|rover lunch | 07:30 | |
*** ysandeep is now known as ysandeep|afk | 07:42 | |
Tengu | hello there | 07:57 |
Tengu | chandankumar: heya! fyi, I'll work a bit on tripleo-lab in order to support the standalone + external compute. That should help me understanding why the modifications I'm trying to get in are apparently failing. I think I may have an idea, but it's really weird. | 07:58 |
chandankumar | Tengu: hello o/ Happy new year :-) | 08:27 |
Tengu | chandankumar: HNY :) | 08:27 |
chandankumar | Tengu: I was checking the current logs(not fully), I think few tasks are getting skipped that's why compute is not getting deployed | 08:28 |
Tengu | chandankumar: apparently the "target_host" points to "undercloud" instead of "subnode-1" | 08:29 |
Tengu | guess it's not properly passed. | 08:29 |
Tengu | that's why | 08:29 |
Tengu | so it's apparently pure ansible being dumb (with my help) | 08:29 |
Tengu | it *should* work. | 08:29 |
chandankumar | ok | 08:30 |
chandankumar | ah https://review.opendev.org/c/openstack/tripleo-quickstart-extras/+/866864/25/playbooks/deploy-standalone-compute.yml#44 | 08:30 |
chandankumar | that's why you are defining target host here | 08:30 |
Tengu | yeah | 08:31 |
chandankumar | by the way the what is the idea | 08:31 |
Tengu | trying to. last log showed "undercloud" as the value instead of "subnode-1". | 08:31 |
Tengu | chandankumar: which idea? | 08:31 |
chandankumar | Tengu: sorry on this one "I think I may have an idea, but it's really weird." | 08:32 |
Tengu | ah, that. well, the param not being passed from the play level.... sounds weird. | 08:32 |
Tengu | basically, everything is launched from the undercloud, ON the undercloud :/. | 08:32 |
Tengu | the only ansible trace we see on the subnode-1 is the first play, when it gathers facts; and when collect-logs role is launched. | 08:33 |
chandankumar | Tengu: https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_1fb/866864/24/check/tripleo-ci-centos-9-standalone-external-compute-target-host/1fbb44c/logs/undercloud/usr/share/ansible/tripleo-inventory/02-computes | 08:37 |
chandankumar | can we change this localhost: to subnode-1 ip or hostname | 08:37 |
Tengu | chandankumar: it's not supposed to use that inventory | 08:38 |
Tengu | chandankumar: https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_1fb/866864/24/check/tripleo-ci-centos-9-standalone-external-compute-target-host/1fbb44c/logs/undercloud/home/zuul/tripleo-deploy/edpm-inventory/02-computes | 08:38 |
chandankumar | yes correct as per the chnage | 08:39 |
pojadhav|ruck | marios, https://sf.hosted.upshift.rdu2.redhat.com/zuul/t/tripleo-ci-internal/status | 08:40 |
Tengu | chandankumar: I must say, I'm a bit surprised by the overally behavior.... | 08:43 |
chandankumar | there is nothing on subnode-12 | 08:44 |
Tengu | chandankumar: even with the --limit, it seems to not do the intended thing. | 08:44 |
chandankumar | Tengu: Do you want to hold a node on rdo side via tp and debug that? | 08:46 |
chandankumar | what if we put the ip of subnode-1 here https://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_1fb/866864/24/check/tripleo-ci-centos-9-standalone-external-compute-target-host/1fbb44c/logs/undercloud/home/zuul/tripleo-deploy/edpm-inventory/02-computes | 08:47 |
chandankumar | and before running the playbook we can run some adhoc command | 08:47 |
chandankumar | and then run the playbook | 08:47 |
Tengu | chandankumar: that's the enxt step yeah - injecting the ansible_host in the subnode inventory entry. | 08:48 |
* chandankumar running a tp and holding the node | 08:49 | |
Tengu | for now I want to see if we use the correct node. | 08:49 |
akahat|rover | ysandeep|afk, marios excluding lvm2 and libblockdev didn't worked: https://review.rdoproject.org/zuul/status#46525 I think we need to add more packages. | 08:52 |
akahat|rover | lvm2 and device-mapper * | 08:53 |
chandankumar | Tengu: ok | 08:55 |
Tengu | chandankumar: and, as said, I'm working in order to add that support in tripleo-lab - that will help me better understanding the whole thing. | 08:55 |
Tengu | chandankumar: also, we may add some more tests in the validate-compute, for instance by ensuring we DO have a hypervisor not named "standalone.localdomain" or the like. | 08:56 |
chandankumar | Tengu: yup, sounds good . | 08:56 |
marios | thanks akahat|rover so lets see with ysandeep|afk ongoing tests too then | 08:56 |
chandankumar | Let me restart the tempest work there also so that we have proper validation in place there | 08:57 |
marios | pojadhav|ruck: dasm: fyi added task into backlog https://issues.redhat.com/browse/TRIPLEOCI-1342 (via chat rhos-ops just now) | 08:57 |
Tengu | sure | 08:57 |
marios | not sure who or when but just captured it for now pojadhav|ruck dasm | 08:57 |
pojadhav|ruck | marios, ack | 09:09 |
* pojadhav|ruck lunch | 09:09 | |
*** soniya29 is now known as soniya29|afk | 09:20 | |
marios | chandankumar: bhagyashris: ysandeep|afk: did we already move jobs to rhos1? i didn't think we did? (d/stream node_failures may be due to missing rhos-1 flavor) | 09:39 |
marios | dpawlik: ^^ | 09:40 |
chandankumar | marios: nope, we have not moved any jobs to rhos-01 | 09:57 |
chandankumar | marios: I tried to run jobs on rhos-01 last week but got hit by node failure | 09:57 |
*** ysandeep|afk is now known as ysandeep | 09:59 | |
marios | k thanks for confirm chandankumar | 10:04 |
dpawlik | chandankumar: ack | 10:08 |
dpawlik | chandankumar: you can try now | 10:08 |
dpawlik | we changed the flavor, job should start now | 10:08 |
chandankumar | dpawlik: ok | 10:17 |
*** soniya29|afk is now known as soniya29 | 10:56 | |
* akahat|rover coffee | 10:59 | |
*** ysandeep is now known as ysandeep|afk | 11:07 | |
*** rlandy|out is now known as rlandy | 11:14 | |
marios | akahat|rover: o/ around can you join us please https://meet.google.com/vzm-nrah-qqf | 11:31 |
chandankumar | Tengu: I think switching to ip will work | 11:32 |
*** dviroe|out is now known as dviroe | 11:32 | |
*** dviroe is now known as dviroel | 11:32 | |
chandankumar | Tengu: https://paste.openstack.org/raw/bmXceV5YmRZYcdvJyJD7/ | 11:33 |
chandankumar | --limit does not work | 11:33 |
chandankumar | rlandy: dviroel good morning & Happy new year o/ | 11:33 |
rlandy | chandankumar: sec - rr syncing - will talk to you in a few | 11:34 |
dviroel | chandankumar: hi o/ | 11:34 |
rlandy | akahat|rover: if you are around - pls join | 11:37 |
akahat|rover | rlandy, yes. joining | 11:37 |
bhagyashris | marios, hey kindly +w on https://code.engineering.redhat.com/gerrit/c/openstack/tripleo-ci-internal-config/+/438309 | 11:43 |
chandankumar | Tengu: left comment on the patch | 11:45 |
pojadhav|ruck | rlandy, https://code.engineering.redhat.com/gerrit/c/openstack/sf-config/+/438313 | 11:49 |
pojadhav|ruck | rlandy, https://code.engineering.redhat.com/gerrit/c/openstack/sf-config/+/438310 | 11:49 |
marios | rlandy: https://code.engineering.redhat.com/gerrit/c/openstack/sf-config/+/438310 | 11:52 |
pojadhav|ruck | rlandy, marios : https://code.engineering.redhat.com/gerrit/c/testproject/+/436232/11#message-ec2b3a0c582010d3867c8fc912fca84ba66675ad | 12:04 |
rlandy | akahat|rover: https://72bac32aa1e395a959c2-fe1bae3c47bea9de602eec2d1e53cbdb.ssl.cf1.rackcdn.com/869020/4/check/tripleo-ci-centos-9-content-provider/59909ec/logs/undercloud/var/log/extra/package-list-installed.txt | 12:15 |
rlandy | https://b2b9d08c5b85917c3495-18cd7f889ae1379763e7ec0a1ec70a8d.ssl.cf2.rackcdn.com/869121/1/gate/tripleo-ci-centos-9-content-provider/ae8d9f4/logs/undercloud/var/log/extra/package-list-installed.txt | 12:15 |
rlandy | qemu-img.x86_64 17:7.2.0-2.el9 @quickstart-centos-appstreams | 12:15 |
rlandy | qemu-img.x86_64 17:7.1.0-6.el9 @quickstart-centos-appstreams | 12:15 |
rlandy | https://b2b9d08c5b85917c3495-18cd7f889ae1379763e7ec0a1ec70a8d.ssl.cf2.rackcdn.com/869121/1/gate/tripleo-ci-centos-9-content-provider/ae8d9f4/logs/container-builds/86794729-b591-48ae-b45a-050377a223ac/base/os/nova-base/nova-libvirt/nova-libvirt-build.log | 12:17 |
rlandy | jq: error: Could not open file /usr/share/qemu/firmware/50-edk2-ovmf-cc.json: No such file or directory | 12:17 |
rlandy | jq: error: syntax error, unexpected end of file (Unix shell quoting issues?) at <top-level>, line 1: | 12:17 |
rlandy | .mapping["nvram-template"] = | 12:17 |
rlandy | jq: 1 compile error | 12:17 |
* pojadhav|ruck needing tea break.. back in few mins.. | 12:18 | |
chandankumar | dviroel: I got the kubeconfig for rhos-1 cluster and I am able to install ansiblee operator | 12:18 |
chandankumar | and run sample playbook | 12:18 |
* marios food | 12:18 | |
chandankumar | next thing I need to deploy a full control plane on that operator | 12:18 |
chandankumar | for getting creds https://code.engineering.redhat.com/gerrit/c/testproject/+/438178/2/playbooks/cluster_creds.sh#4 | 12:18 |
chandankumar | I passed the rhos-01 namespace it gave me the creds | 12:19 |
Tengu | chandankumar: oh, you mean the inventory file is borked, and "all" doesn't work? | 12:22 |
chandankumar | yes all is not working | 12:22 |
Tengu | pfrtt. | 12:22 |
*** ysandeep|afk is now known as ysandeep | 12:27 | |
bhagyashris | rlandy, marios ysandeep|afk fyi, added all the required reviews link to start the rhos18 on rhel9 integration line in the review hackmd https://hackmd.io/FGMoCiRfSNa8puA1BpTQ-Q#Jan-4-2023 as their dependency | 12:27 |
reviewbot | Do you want me to add your patch to the Review list? Please type something like add to review list <your_patch> so that I can understand. Thanks. | 12:27 |
dviroel | chandankumar: looking | 12:28 |
dviroel | chandankumar: hive says that there is no claim for rhos-1, you might be using another clusterclaim | 12:28 |
chandankumar | dviroel: I have created the claim, it is still in initialized state | 12:30 |
chandankumar | dviroel: without claim, I can get the kubeconfig for a namespace | 12:30 |
dviroel | chandankumar: "stage-1-claim" | 12:30 |
dviroel | on your change | 12:30 |
dviroel | "bash /home/zuul/src/code.engineering.redhat.com/testproject/playbooks/cluster_creds.sh stage-1-claim > ~/.kube/stage-1-claim-kubeconfig" | 12:31 |
dviroel | chandankumar: only rhos-01-01-bk9fs-master-0 is on rhos-1 | 12:32 |
rlandy | akahat|rover: can you try exclude that qemu rpm so it uses the old one as a workaround to unblock gate? | 12:32 |
rlandy | marios: ^^ | 12:33 |
rlandy | in release files | 12:33 |
rlandy | chandankumar: dviroel: sorry - reading back - was on rr issues | 12:33 |
chandankumar | dviroel: http://pastebin.test.redhat.com/1086861 | 12:34 |
chandankumar | rlandy: dviroel if you can come on a call, then I can show the same | 12:35 |
chandankumar | let me know when free | 12:39 |
soniya29 | arxcruz, rlandy, please add/edit today's agenda for tempest meeting - https://hackmd.io/xWJLuBHGSI2YGX7UlkwRbg?both | 12:40 |
ysandeep | marios, hey o/ I saw your comment on https://review.opendev.org/c/openstack/tripleo-quickstart/+/868605/13/config/release/tripleo-ci/CentOS-9/master.yml#177 to only try with kernel/device-mapper/lvm2.. akahat|rover already tried that here: https://review.opendev.org/c/openstack/tripleo-quickstart/+/869099/6/config/release/tripleo-ci/CentOS-9/promotion-testing-hash-master.yml , didn't help | 12:40 |
arxcruz | soniya29: i don't have anything | 12:40 |
arxcruz | should we cancel it? | 12:40 |
rlandy | soniya29: arxcruz: ok with canceling | 12:40 |
soniya29 | yeah, we can, i also dont have anything new and martin is also on pto | 12:41 |
ysandeep | akahat|rover, looks like only wallaby test where we excluded systemd is procedding further | 12:41 |
akahat|rover | ysandeep, yes. https://review.opendev.org/c/openstack/tripleo-quickstart/+/869099/6/config/release/tripleo-ci/CentOS-9/promotion-testing-hash-wallaby.yml#144 | 12:41 |
soniya29 | alright, i will cancel the meeting :) | 12:41 |
akahat|rover | this is looks promissing | 12:41 |
rlandy | bhagyashris: thanks - will review later | 12:41 |
akahat|rover | ysandeep, i'll exclude same packages in all release | 12:41 |
akahat|rover | and then retry | 12:41 |
akahat|rover | after job passes | 12:41 |
ysandeep | akahat|rover, ack, for 1 release please only exclude systemd | 12:42 |
dviroel | chandankumar: your clusterclaim is in the wrong namespace, thats why it is still PENDING -> Initialized. Must be on "oooci" namespace. | 12:42 |
akahat|rover | ysandeep, sure.. got your point. :) | 12:42 |
dviroel | chandankumar: you still can reach credentials of that cluster, in the way you did with "jsonpath='{.spec.clusterMetadata.adminKubeconfigSecretRef.name}')" | 12:43 |
ysandeep | dviroel, hey o/ merged your flavor change PR few hours ago. | 12:43 |
dviroel | chandankumar: but you don't have the control of who is claiming the cluster | 12:44 |
dviroel | ysandeep: thanks man, hive is already creating new clusters on the new flavor | 12:44 |
chandankumar | dviroel: thanks | 12:44 |
chandankumar | dviroel: it worked now | 12:44 |
dviroel | chandankumar: \o/ | 12:45 |
ysandeep | dviroel, I have a query - what was the reason to delete some files there like install-config-sno_secret_osp_03.yaml | 12:45 |
chandankumar | dviroel: the job is running in rhos-1 cloud | 12:45 |
dviroel | ysandeep: we were not using them anymore. I merged those files when Hive Inventory wasn't working. If you check those deleted files, you will see FIP info there. | 12:46 |
dviroel | ysandeep: this info lives on ClusterDeploymentCustomization now: https://github.com/openshift/release/blob/master/clusters/hive/pools/openstack-k8s-operators/openstack-k8s-operators-cdc-vexx-01.yaml | 12:47 |
dviroel | ysandeep: and are consumed by the ClusterPool here: https://github.com/openshift/release/blob/9cd4cc1fa636452d52f63ae62d9c50663e8114d6/clusters/hive/pools/openstack-k8s-operators/openstack-k8s-operators-4-10-vexxhost-sno_clusterpool.yaml#L23 | 12:47 |
dviroel | chandankumar: i see a node there | 12:47 |
ysandeep | dviroel, ahh okay, thanks! | 12:47 |
rlandy | chandankumar: can meet now if you want | 12:49 |
chandankumar | rlandy: sure | 12:49 |
chandankumar | rlandy: dviroel https://meet.google.com/gkq-weqq-xzo?authuser=0&hl=en | 12:50 |
chandankumar | rlandy: dviroel just joined | 12:57 |
*** bhagyashris is now known as bhagyashris|afk | 13:17 | |
akahat|rover | ysandeep, marios rlandy updated patch with requested package exclude by ysandeep. https://review.opendev.org/c/openstack/tripleo-quickstart/+/869099, https://review.rdoproject.org/r/c/testproject/+/46525 | 13:32 |
akahat|rover | i've rekicked tp.. | 13:33 |
akahat|rover | marios, rlandy https://bugs.launchpad.net/tripleo/+bug/2001626, Bogdan working on the issue, he has some patches up for it. | 13:34 |
dasm | o/ | 13:40 |
marios | thanks akahat|rover can you please also notify mailing list since its new gate blocker | 13:40 |
akahat|rover | marios, rlandy i'm leaving to home.. i'll be back online after some time.. I've updated the CIX cards with the latest info | 13:40 |
akahat|rover | marios, yes. i've already done. | 13:40 |
dviroel | rlandy: while I was writing the 3rd solution: it ends up in the same situation as the 1st. Even if we have a isolated tenant, we plan to run nodepool nodes there, so we need to care about leaking FIPS there. In the end, where nodepool runs, we need to have a FIP cleanup solution. | 13:40 |
dasm | 08:57 <marios> pojadhav|ruck: dasm: fyi added task into backlog | 13:40 |
marios | akahat|rover: can you please update the xfs card with latest? | 13:41 |
marios | looking | 13:41 |
dasm | marios: thanks. not need to announce though :) | 13:41 |
marios | ack thanks akahat|rover looks like you did | 13:41 |
marios | dasm: o/ | 13:41 |
akahat|rover | marios, yeah.. see you in a bit. | 13:41 |
*** ysandeep is now known as ysandeep|away | 13:49 | |
*** pojadhav- is now known as pojadhav|ruck | 13:58 | |
pojadhav | rlandy, marios : i can rhos17.1 line running smoothly atleast build image/containers and standalone jobs... | 14:06 |
rlandy | dviroel: correct - but we would not impact all jobs | 14:25 |
rlandy | just the ones that require have | 14:25 |
rlandy | ie: if FIPs caused an issue, other jobs would keep running ie: less impact | 14:25 |
rlandy | chandankumar: hi - I thought I added the EDPM team meeting to my calendar but I don;t see it | 14:27 |
rlandy | can you remind me of the time/day? | 14:27 |
dviroel | rlandy: new vexx flavor worked, job is passing again | 14:27 |
rlandy | dviroel: \o/ | 14:28 |
chandankumar | rlandy: OSP NG: External Data Plane Weekly Status: Timing Monday, 2:30 PM UTC | 14:29 |
rlandy | chandankumar: ah 0 I have it - it was just not scheduled this week | 14:30 |
chandankumar | rlandy: yup | 14:30 |
marios | rlandy: firefox finally gave up... restarting it and rejoining | 14:40 |
*** dviroel is now known as dviroel|lunch | 14:54 | |
* pojadhav lunch | 15:08 | |
* pojadhav sorry... dinner | 15:08 | |
Tengu | marios: that's why I use a dedicated chromium instance for meet/calendar/gchat - the rest is running in firefox :). | 15:11 |
marios | Tengu: :) yeah i mean i also have chrome just for google stuff (mail/calendar/chat etc also slack and also radio(s)/sounds) but for some reason sharing doesn't work for me with chrome & meet | 15:12 |
marios | Tengu: so i use ffox for it | 15:12 |
marios | Tengu: maybe should try chromium instead | 15:12 |
rlandy | I have the same setup as Tengu | 15:15 |
rlandy | chrome runs G stuff | 15:15 |
rlandy | firefox runs the rest | 15:15 |
marios | me too :) | 15:19 |
chandankumar | see ya people tomorrow! | 15:19 |
rlandy | see you tomorroew | 15:19 |
marios | except i cant get meet to work properly on chrome so have ff for it | 15:19 |
marios | o/ ch | 15:19 |
marios | chandankumar: o/ | 15:19 |
Tengu | marios: chromium + sharing works fine here at least. | 15:23 |
Tengu | see you tomorrow chandankumar | 15:23 |
*** dviroel|lunch is now known as dviroel | 16:06 | |
rlandy | lunch brb | 16:30 |
* marios off in few have a good rest of day oooq o/ | 16:34 | |
*** marios is now known as marios|out | 16:39 | |
dasm | marios|out: o/ | 16:41 |
Tengu | ok, bogdando patch is correcting the content-provider issue. | 16:46 |
Tengu | could validate it with my EDPM work. | 16:46 |
Tengu | heh, we may need to get it in Zed before master | 16:46 |
Tengu | https://review.opendev.org/c/openstack/tripleo-common/+/869206 | 16:46 |
Tengu | rlandy, marios|out wdyt? zed before master, or content-provider-zed on master as non-voting until we get it in master + zed? | 16:47 |
rlandy | Tengu: master first | 16:48 |
rlandy | has more jobs | 16:48 |
rlandy | then zed | 16:48 |
Tengu | rlandy: so we must switch content-provider-zed to non-voting, as well as the dependents... | 16:49 |
Tengu | since content-provider-zed takes its code from the stable/zed branch. which is unpatched. | 16:49 |
Tengu | :) | 16:49 |
rlandy | if you did zed first? | 16:50 |
Tengu | that's a nice Ouroboros situation. | 16:50 |
rlandy | you would need to make master non-voting | 16:50 |
rlandy | on a job that changes master .... | 16:50 |
rlandy | only zed and master run | 16:50 |
rlandy | oh and wallaby | 16:51 |
Tengu | wallaby doesn't seem affected? | 16:51 |
rlandy | if you look at the dependencies ... | 16:51 |
Tengu | so, if we're to get master in first, we must set content-provider-zed and undercloud-upgrade to non-voting | 16:52 |
rlandy | yeah | 16:52 |
rlandy | smaller list than everything that depends on master | 16:52 |
Tengu | ok | 16:52 |
Tengu | soooo it's tripleo-upgrades-master-pipeline apparently? | 16:53 |
rlandy | temp comment that out? | 16:55 |
Tengu | probably easier/better than hacking through tripleo-ci | 16:55 |
marios|out | really got to go Tengu will check in morning... but if we have content provider broken then what is providing the container (regardless if it is non voting) | 16:55 |
Tengu | as a dedicated patch so that it's easier to revert? | 16:55 |
Tengu | marios|out: np - will figure out a way | 16:56 |
marios|out | oh i see now .. you mean content provider zed non voting to merge the master fix | 16:57 |
Tengu | marios|out: or https://review.opendev.org/c/openstack/tripleo-common/+/869240 (cc rlandy ) | 16:58 |
Tengu | let's see if it does the intended thing, then I can rebase bogdan's fix onto it. | 16:58 |
Tengu | and we should get out of that chicken'n'egg issue. | 16:58 |
marios|out | Tengu: yeah maybe this one is better... but can be combined into bogdan patch and followed by partial revert which is rebased onto that | 16:59 |
Tengu | marios|out: I hage partial reverts.. | 16:59 |
marios|out | :) | 16:59 |
marios|out | anyway got to go now... will pickup in ~13 hours ;) | 16:59 |
marios|out | have a good one o/ | 16:59 |
Tengu | ok - repushed as: comment / bogdan's / revert-comment | 17:01 |
Tengu | that should do it. | 17:01 |
akahat|rover | What if we non-vote all zed jobs? | 17:01 |
rlandy | Tengu: I'd vote for https://review.opendev.org/c/openstack/tripleo-common/+/869240 if it works | 17:02 |
rlandy | simpler | 17:02 |
Tengu | errr. | 17:03 |
Tengu | that one won't work | 17:03 |
Tengu | it will fail due to the very issue bogdan is correcting in the patch that's following this comment-out. | 17:03 |
Tengu | so I'll need to amend his patch and do the partial revert.... | 17:03 |
Tengu | I hate that. | 17:03 |
Tengu | ok, signing off. Hopefully the CI will be green tomorrow so that Marios or anyone else may be able to merge it. | 17:19 |
akahat|rover | rlandy, https://review.rdoproject.org/r/c/testproject/+/46525 -> we got one master job passing.. other failed on undercloud install.. rerunning | 17:23 |
rlandy | akahat|rover: thanks - let's try settle on the list of exclude and merge those up tomorrow morning | 17:38 |
rlandy | Tengu: sorry - meeting reading back | 17:38 |
rlandy | tomorrow is fine | 17:39 |
akahat|rover | rlandy, may be we need some more green runs.. | 17:50 |
akahat|rover | in my morning i'll keep poking it.. and will get merge tomorrow. | 17:51 |
rlandy | thank you | 17:53 |
* dviroel out | 21:09 | |
*** dviroel is now known as dviroel|ourt | 21:09 | |
dasm | dviroel|ourt: o/ | 21:09 |
* dasm => offline | 22:14 | |
dasm | o/ | 22:14 |
*** dasm is now known as dasm|off | 22:14 | |
*** rlandy is now known as rlandy|out | 23:19 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!