Closed jsturtevant closed 10 months ago
the last run failed because I had not replicated the images across the regions we test against. running again
/test pull-e2e-capz-windows-2022-extension
failed this time due to older containerd version, https://github.com/kubernetes/test-infra/pull/31400 fixes that then will re-run
/test pull-e2e-capz-windows-2022-extension
I guess there was some eventual consistency in prow, that run still has 1.7.0
/test pull-e2e-capz-windows-2022-extension
Nodes came online
Fri, 08 Dec 2023 01:21:59 +0000: Nodes Ready
NAME STATUS ROLES AGE VERSION INTERNAL-IP EXTERNAL-IP OS-IMAGE KERNEL-VERSION CONTAINER-RUNTIME
capz-conf-2jugc7-control-plane-5jlxb Ready control-plane 6m2s v1.30.0-alpha.0.21+89dfbebe2e823b 10.0.0.4 <none> Ubuntu 22.04.3 LTS 6.2.0-1017-azure containerd://1.7.6
capz-conf-65l6x Ready <none> 52s v1.30.0-alpha.0.21+89dfbebe2e823b <none> <none> Windows Server 2022 Datacenter 10.0.25398.469 containerd://1.7.10
capz-conf-d4mdm Ready <none> 59s v1.30.0-alpha.0.21+89dfbebe2e823b <none> <none> Windows Server 2022 Datacenter 10.0.25398.469 containerd://1.7.10
but doesn't looks like cluster initialized properly with cloudprovider:
Dec 8 01:52:06.712: INFO: ==== node wait: 1 out of 3 nodes are ready, max notReady allowed 0. Need 2 more before starting.
Dec 8 01:52:06.756: INFO: Unschedulable nodes= 2, maximum value for starting tests= 0
Dec 8 01:52:06.756: INFO: -> Node capz-conf-65l6x [[[ Ready=true, Network(available)=true, Taints=[{node.cluster.x-k8s.io/uninitialized NoSchedule <nil>} {node.cloudprovider.kubernetes.io/uninitialized true NoSchedule <nil>}], NonblockingTaints=node-role.kubernetes.io/control-plane ]]]
Dec 8 01:52:06.756: INFO: -> Node capz-conf-d4mdm [[[ Ready=true, Network(available)=true, Taints=[{node.cluster.x-k8s.io/uninitialized NoSchedule <nil>} {node.cloudprovider.kubernetes.io/uninitialized true NoSchedule <nil>}], NonblockingTaints=node-role.kubernetes.io/control-plane ]]]
cloudnode manager failed to pull:
Pulling image "capzci.azurecr.io/azure-cloud-node-manager:30e6f4f"
Warning Failed 41m (x4 over 43m) kubelet Error: ErrImagePull
Normal BackOff 3m32s (x172 over 43m) kubelet Back-off pulling image "capzci.azurecr.io/azure-cloud-node-manager:30e6f4f"
other pods looks like the worked but there is not logs for any of the containers running
/test pull-e2e-capz-windows-2022-extension
Overall looks good. I think we should rename the new template away from sig-ci
tho. Maybe shared-image-gallery-ci
?
The previous test run passed with annual channel https://prow.k8s.io/view/gs/kubernetes-jenkins/pr-logs/pull/kubernetes-sigs_windows-testing/417/pull-e2e-capz-windows-2022-extension/1733186423831400448
I've renamed the template and switched the PR back to using the standard ci template by default. Once this merged can open a Prow config with the shared-image-gallery-ci.yaml
as the template for annual channel tests.
/hold cancel
[APPROVALNOTIFIER] This PR is APPROVED
This pull-request has been approved by: jsturtevant, marosset
The full list of commands accepted by this bot can be found here.
The pull request process is described here
This sets up testing scripts for the Windows Annual channel https://techcommunity.microsoft.com/t5/windows-server-news-and-best/windows-server-annual-channel-for-containers/ba-p/3866248
The image is not in the azure market place, so currently need to download and convert it to VHD and upload it to an Azure SIG. Image builder can then use this to build a CAPI ready image
/sig windows /hold I've changed the base template here to force the CI in this PR to test this as some of the CI infra is new