Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

Navigation  Back up to About Overview 
  
[+] Expand All
[-] Collapse All

Proper Order for Stopping the Entire Cluster

If you intend to shut down the entire cluster, use the following procedure.

Caution: Before proceeding with this procedure, be sure that you want to shut down all nodes in the cluster.

When you stop a cluster, the system prompts you with the following warning:

WARNING: This function is capable of stopping multiple nodes. Do not use this function if you intend to stop only one node. Do you intend to stop the entire cluster? (y,n): y Are you sure? (y,n): y Really? (y,n): y

Caution: Stopping multiple systems and processes removes all redundancy, can create multiple faults, and may damage the cluster. If you do need to stop the cluster, be sure to restart the cluster properly. See Starting the Cluster.

Follow this procedure in precise order to properly shut down the cluster. Log in as root on each machine, in the order specified, and execute the specified command.

  1. On each SBR node (s), one at a time, execute:
    sbrd stop cluster
    sbrd status
  2. On each SBR/management node (sm), one at a time, execute:
    sbrd stop cluster
    sbrd status
  3. On each management node (m), one at a time, execute:
    sbrd stop cluster
    sbrd status
  4. On each data node (d), one at a time, execute:
    sbrd stop cluster
    sbrd status

 

Modified: 2017-03-07