Konvoy node is unreachable after restarting AWS EC2 instance

Scenario

After stopping and starting an AWS EC2 instance, Konvoy is unable to communicate with one of the cluster nodes:

% konvoy check nodes

TASK [Gathering Facts] ********************************
ok: [10.0.194.251]
ok: [10.0.128.241]
fatal: [10.0.129.190]: UNREACHABLE! => {
    "changed": false,
    "unreachable": true
}

MSG: Connection timed out.

This is due to the fact that restarting an AWS EC2 instance will cause that instance to be assigned a new public IP address. Konvoy communicates with cluster nodes on AWS via their public IP address so now that it has changed, Konvoy throws this error.

 

Solution

Run `konvoy up` to update the cluster's inventory.yaml with the new public IP of the node. Konvoy will then be able to connect to the node successfully.