Restoring a cluster configuration from an existing configuration

To restore a cluster configuration from an existing configuration after a cluster failure, you re-create the cluster using the cluster configuration that you chose and made available to the recovering node, and then rejoin each additional node to the new cluster.

1. Disable storage failover for each HA pair: storage failover modify -node node_name -enabled false

   You only need to disable storage failover once for each HA pair. When you disable storage failover for a node, storage failover is also disabled on the node's partner.
2. Halt each node except for the recovering node: system node halt -node node_name -reason "text"

   Example

   cluster1::*> system node halt -node node0 -reason "recovering cluster"
   
   Warning: Are you sure you want to halt the node? {y|n}: y
   
   
3. Set the privilege level to advanced: set -privilege advanced
4. On the recovering node, use the system configuration recovery cluster recreate command to re-create the cluster.

   Example

   This example re-creates the cluster using the configuration information stored on the recovering node:

   cluster1::*> configuration recovery cluster recreate -from node
   
   Warning: This command will destroy your existing cluster. It will
            rebuild a new single-node cluster consisting of this node
            and its current configuration. This feature should only be
            used to recover from a disaster. Do not perform any other
            recovery operations while this operation is in progress.
   Do you want to continue? {y|n}: y
   
   

   Result

   A new cluster is created on the recovering node.

5. If you are re-creating the cluster from a configuration backup file, verify that the cluster recovery is still in progress: system configuration recovery cluster show

   You do not need to verify the cluster recovery state if you are re-creating the cluster from a healthy node.

   Example

   cluster1::*> system configuration recovery cluster show 
    Recovery Status: in-progress
    Is Recovery Status Persisted: false
   
6. Boot each node that needs to be rejoined to the re-created cluster.

   You must reboot the nodes one at a time.
7. For each node that needs to be joined to the re-created cluster, do the following:
   1. From a healthy node on the re-created cluster, rejoin the target node: system configuration recovery cluster rejoin -node node_name

      Example

      This example rejoins the node2 target node to the re-created cluster:

      cluster1::*> system configuration recovery cluster rejoin -node node2
      
      Warning: This command will rejoin node "node2" into the local
               cluster, potentially overwriting critical cluster
               configuration files. This command should only be used
               to recover from a disaster. Do not perform any other
               recovery operations while this operation is in progress.
               This command will cause node "node2" to reboot.
      Do you want to continue? {y|n}: y
      
      

      Result

      The target node reboots and then joins the cluster.

   2. Verify that the target node is healthy and has formed quorum with the rest of the nodes in the cluster: cluster show -eligibility true

      The target node must rejoin the re-created cluster before you can rejoin another node.

      Example

      cluster1::*> cluster show -eligibility true
      Node                 Health  Eligibility   Epsilon
      -------------------- ------- ------------  ------------
      node0                true    true          false
      node1                true    true          false
      2 entries were displayed.
      
      
8. If you re-created the cluster from a configuration backup file, set the recovery status to be complete: system configuration recovery cluster modify -recovery-status complete
9. Return to the admin privilege level: set -privilege admin
10. If the cluster consists of only two nodes, use the cluster ha modify command to reenable cluster HA.
11. Use the storage failover modify command to reenable storage failover for each HA pair.