Search code examples
dockerkuberneteskubernetes-helmrancherrancheros

Cluster not shown when Rancher machine restarts


Whenever I restart my virtual machine ( Rancher OS) containing a single node cluster (Rancher server and Rancher agent exists on the same machine) I could not find my previously created cluster listed on Rancher Dashboard.

Steps to reproduce:

  • Start a virtual machine (Rancher OS)
  • Set up Rancher Server on the Rancher OS ( to get Rancher running )
  • Create a single node cluster. (Node exists on the same Rancher OS )
  • Shutdown the machine and again start the machine.

Expectation - The created cluster must be shown.

Result: - Rancher dashboard shown no cluster.

Other details that may be helpful: It is a single node cluster with etcd,worker and control plane. The node exists on the same machine where the Rancher Server exists.

Environment information -Rancher OS : v1.5.1 -Rancher Server : Rancher Stable release (v2.2.3) (rancher:stable image used) - Installation option (single install/HA): Single Install

Cluster information - Cluster type (Hosted/Infrastructure Provider/Custom/Imported): Custom - Machine type (cloud/VM/metal) and specifications (CPU/memory): VM (Rancher OS)

I tried to investigate the issue and found that when the Rancher OS machine restarted Rancher Agent is showing restarting status.

image

I tried to check the logs of Rancher agent and found the issue with checksum. image

The issue : https://github.com/rancher/rancher/issues/15012 seems to be linked but I cant find the solution as provided by him.


Solution

  • I have solved the issue by removing the checksum from the command to create the cluster. If we remove the checksum the certificate match does not happen and the issue gets resolved. I believe this solution works fine on a development environment but it is a security concern in a production environment. In my case there is not much security concern.