Performance Management CopyCluster requirements for Vertica

Products

• CA Performance Management
• Network Observability

Issue/Introduction

What are the requirements for copycluster to work between 2 separate environments

DX NetOps Performance Management Data Repository (DR) Vertica copycluster requirements.

Environment

All supported DX NetOps Performance Management releases

Resolution

Configuring the Target Cluster

Configure the target to allow the source database to connect to it and restore the database. The target cluster must:

• Have the same number of nodes the source cluster.
• Have a database with the same name as the source database.
  • The target database can be completely empty.
• Have the same node names as the source cluster.
  • The nodes names listed in the NODES system tables on both clusters must match.
  • Ex: Node 1 on source cluster is named v_drdata_node0001, Node 1 on target cluster must be the same.
• The catalog directory, and data directory must be the same as the original source data repository.
• Have the same database administrator account name.
• All nodes must be configured for passwordless SSH in both directions.
  • Allow database administrator user passwordless SSH access from all source nodes to all target nodes
  • Allow database administrator user passwordless SSH access from all target nodes to all source nodes
  • Ex: Node 1 on source must be able to SSH to all other nodes in it's cluster as well as all nodes in the target cluster. The same must be configured from target to source.
• SELinux must be in Permissive or Disabled mode
• Database administrator user must be able to create directories in, read and write files in, the temp directory specified in the configuration *.ini file in the tempDir variable value
• The sshd services on all nodes in source and target system must be configured to allow TCP forwarding. Ensure "AllowTcpForwarding" is set to Yes and uncommented/enabled in the /etc/ssh/sshd_config file. Restart sshd services if sshd_config file is changed.
• Have an available and accessible /tmp directory.
  • Used to write copycluster related log files.
  • Used to write copycluster related working files during its operation. The files are added/removed as it works on them. 
  • 4 GB or more of available disk space should be sufficient. We've never had reports of copycluster failures due to low /tmp disk space availability.
  • If concerned with existing /tmp locations, or if the /tmp location is very small with no options to enlarge it, we can direct the copycluster configuration to use a different directory than /tmp.
    • To configure an alternative directory set the target alternative directory against the tempDir variable in the *.ini file configured for copycluster runs.

Additional Information

See the following URLs for more information.

Vertica Documentation: Copying the database to another cluster

DX NetOps Performance Management Documentation: Migrate the Data Repository