SOLO-1009

RemoteConfigMissingForDeployError

Description

Thrown by network deploy when the solo-remote-config ConfigMap that backs a deployment cannot be found. network destroy leaves the namespace and the recorded deployment in place unless it is asked to remove PVCs and secrets as well, so the deployment itself is not gone — only the ConfigMap tracking it. deployment cluster attach only (re)creates that ConfigMap when the deployment has no clusters recorded yet; once a cluster-ref is already attached it instead expects the ConfigMap to exist and fails the same way. The only way back to a working state is therefore to delete and recreate the deployment’s local config so the cluster-ref attach is treated as new, which recreates the ConfigMap fresh. The consensus keys have to be regenerated alongside it, since a successful network deploy removes the cached copies from disk once they are uploaded to cluster secrets.

Troubleshooting Steps

1. Inspect the remote config ConfigMap: kubectl get configmap solo-remote-config -n
2. The deployment record still exists locally, but the remote config backing it is gone and cannot be recreated in place. Recreate the local deployment config so cluster attach treats it as new, then reattach the cluster and regenerate the consensus keys:
3. solo deployment config delete –deployment
4. solo deployment config create -d -n -q
5. solo deployment cluster attach -d -c -q –num-consensus-nodes
6. solo keys consensus generate -d -q –gossip-keys –tls-keys
7. Set –num-consensus-nodes to the node count this deployment had before, then re-run network deploy