Fix kubectl patch command for v2.5.x

- An agent is no longer deployed to the local cluster in Rancher v2.5.x, so the local cluster can be excluded
- There are two cluster CRDs in v2.5.x  clusters.management.cattle.io and clusters.rancher.cattle.io, so this needs to be fully-qualified
This commit is contained in:
Alex Seymour
2021-05-12 17:13:46 +02:00
committed by GitHub
parent 791a252aac
commit 7099b82bb1
@@ -110,10 +110,10 @@ Method 3 can be used as a fallback if method 1 and 2 are unfeasible.
### Method 1: Kubectl command
For each cluster under Rancher management (including `local`) run the following command using the Kubeconfig file of the Rancher management cluster (RKE or K3S).
For each cluster under Rancher management (except the `local` Rancher management cluster) run the following command using the Kubeconfig file of the Rancher management cluster (RKE or K3S).
```
kubectl patch clusters <REPLACE_WITH_CLUSTERID> -p '{"status":{"agentImage":"dummy"}}' --type merge
kubectl patch clusters.management.cattle.io <REPLACE_WITH_CLUSTERID> -p '{"status":{"agentImage":"dummy"}}' --type merge
```
This command will cause all Agent Kubernetes resources to be reconfigured with the checksum of the new certificate.
@@ -142,4 +142,4 @@ With this method you are recreating the Rancher agents by running a set of comma
First, generate the agent definitions as described here: https://gist.github.com/superseb/076f20146e012f1d4e289f5bd1bd4971
Then, connect to a controlplane node of the downstream cluster via SSH, create a Kubeconfig and apply the definitions:
https://gist.github.com/superseb/b14ed3b5535f621ad3d2aa6a4cd6443b
https://gist.github.com/superseb/b14ed3b5535f621ad3d2aa6a4cd6443b