Closed lb4368 closed 3 years ago
I can work on this, please assign it to me
Comment from @sshiba
Also make sure CAPI manifests are pinned to v1apha3. See error from Zuul airship-airshipctl-gate-script-runner-docker below:
Using configuration File="/workdir/.cluster-api/clusterctl.yaml" Installing the clusterctl inventory CRD Creating CustomResourceDefinition="providers.clusterctl.cluster.x-k8s.io" Fetching providers Using Override="core-components.yaml" Provider="cluster-api" Version="v0.4.2" Using Override="bootstrap-components.yaml" Provider="bootstrap-kubeadm" Version="v0.4.2" Using Override="control-plane-components.yaml" Provider="control-plane-kubeadm" Version="v0.4.2" Using Override="infrastructure-components.yaml" Provider="infrastructure-docker" Version="v0.3.11" Using Override="metadata.yaml" Provider="cluster-api" Version="v0.4.2" Using Override="metadata.yaml" Provider="bootstrap-kubeadm" Version="v0.4.2" Using Override="metadata.yaml" Provider="control-plane-kubeadm" Version="v0.4.2" Using Override="metadata.yaml" Provider="infrastructure-docker" Version="v0.3.11" Error: current version of clusterctl is only compatible with v1alpha4 providers, detected v1alpha3 for provider infrastructure-docker
Problem description (if applicable) In order to fix issue when uplifting to CAPI v0.4.2 (#518), we need the Docker provider to use the krm supporting clusterctl: 0.3.22 version instead of latest. The reason is when uplifting to CAPI v0.4.2, the clusterctl krm image is build using clusterctl:0.4.2 binary, which may be causing Zuul to fail during docker provider test.
Proposed change The proposed approach (but not limited to) is to create a strategic merge on the docker-test-site level for the clusterctl executor.