Vertica copycluster returns error about paths that do not match

Products

• CA Performance Management - Usage and Administration
• DX NetOps

Issue/Introduction

When we run copycluster as follows we see these errors.

-bash-4.1$ ./vbr.py --task copycluster --config-file /opt/CA/db/CopyClusterConfig.ini
Starting copy of database drdata.
Participating nodes: v_drdata_node0001.
Error: Database path does not match on source (v_drdata_node0001: /opt/CA/db/catalog/drdata/v_drdata_node0001_catalog) and destination (dc2sdeldpfm002: /opt/CA/db/data/drdata/v_drdata_node0001_catalog) nodes.
Copycluster FAILED

Why does the Vertica Data Repository copycluster command fail in a DX NetOps Performance Management environment?

Environment

All supported DX NetOps Performance Management releases

Cause

The new destination system was installed using different DbDataDir and DbCatalogDir values in drinstall.properties file than used for the source system

Resolution

To resolve this the following steps should be followed. It will result in matched paths allowing the copycluster to function properly.

1. On the new destination DR DB system open adminTools.
   1. Enter option 6: Configuration Menu
   2. Choose option 7: Drop Database
   3. Follow the prompts to drop the DB with wrong catalog dir
   4. Check the existing catalog and data directories to make sure there is no drdata dir remaining or present.
      • Delete drdata recursively if present.
   5. Ensure you've correctly mounted the correct catalog disk on the same catalog path as the source system.
2. Edit the drinstall.properties file on the new host so it matches the old host file values. Make sure data and catalog dir values match.
3. Re-run the dr_validate using the corrected drinstall.properties file
4. Re-run the dr_install using the corrected drinstall.properties file
5. Run the copycluster command that failed previously.

Additional Information

Performance Management CopyCluster requirements for Vertica