Procedures to restart Cloud data Protection Cassandra Database

Products

• CDP for ServiceNow
• CDP for Salesforce
• CDP for Xactly
• CDP Communication Server
• CDP for Oracle Sales Cloud
• CDP for Oracle CRM On Demand
• CDP Policy Builder
• CDP for Yammer
• CDP Integration Server

Issue/Introduction

Beginning in Cloud Data Protection (CDP) 4.13.x or newer, Cassandra database is introduced and tightly integrated with the CDP.

In general, the Cassandra DB should not be restarted even if the CDP is restarted.

There are situations when a restart of the Cassandra DB is unavoidable (example: OS or Java patching).

Environment

• CDP 4.13.x or newer

Resolution

If a restart of the Cassandra DB becomes necessary as part of the CDP restart, please follow these procedures

1. NEVER use kill -9 against Cassandra process
2. Shutdown all CDP services
3. Go to the cassandra/bin folder (default path /usr/local/perspecsys/container/cassandra/bin), run "./nodetool flush" on ALL CDP container servers
4. Run command "service Cassandra stop" on ALL CDP container servers
5. Wait till Cassandra processes are shutdown
6. Run command "service Cassandra start" on ALL CDP container servers
7. Monitor the Cassandra status by going to the cassandra/bin folder (default path /usr/local/perspecsys/container/cassandra/bin), run "./nodetool status" on ALL container servers
8. After running the last step, eventually, all Cassandra nodes should become "UN" (up-node)
9. Start container1 and wait till it gets into a ready state before starting container2, container3, etc.
10. Start the rest of the CDP services