Skip to content

harvester.md

Latest commit

 

History

History
41 lines (30 loc) · 1.43 KB

harvester.md

File metadata and controls

41 lines (30 loc) · 1.43 KB

HARVESTER

Fetch Harvester Kubeconfig 
ssh rancher@192.168.1.50 "sudo cat /etc/rancher/rke2/rke2.yaml" > ./harvester.yaml
yq -i '.clusters[].cluster.server = "https://192.168.1.50:6443"' ./harvester.yaml
yq '.clusters[].cluster.server' ./harvester.yaml
kubectl get nodes --kubeconfig ./harvester.yaml
ReplicaSchedulingFailure Single-Node Harvester Cluster

You have a single-node Harvester cluster (sthings) Longhorn is configured for 3 replicas per volume Only 1 replica can run because Longhorn's anti-affinity prevents multiple replicas on the same node Status: ReplicaSchedulingFailure: precheck new replica failed

Why this happens: Longhorn's default configuration assumes multi-node clusters for data redundancy. With one node, it can't satisfy the "3 replicas on different nodes" requirement. Solutions:

For existing volumes
kubectl --kubeconfig ~/.kube/harvester patch volume pvc-2ae20462-92b3-4ce2-93f5-8d90812b03bb \
  -n longhorn-system --type=merge -p '{"spec":{"numberOfReplicas":1}}'

kubectl --kubeconfig ~/.kube/harvester patch volume pvc-5626539c-5b15-4b9c-8bf9-cda3625887a4 \
  -n longhorn-system --type=merge -p '{"spec":{"numberOfReplicas":1}}'

# Set global default for future volumes
kubectl --kubeconfig ~/.kube/harvester patch settings.longhorn.io default-replica-count \
  -n longhorn-system --type=merge -p '{"value":"1"}'