Aria Operations upgrade pre-upgrade assessment check failure with error "Issue #2794269 - Potential issue with Analytics node."

Products

• VMware vRealize Operations 8.x
• VMware Aria Operations 8.x

Issue/Introduction

During the pre-upgrade assessment tool check, you may encounter the following error message.

 

Environment

Aria Operations 8.1.1 to 8.6.4

8.10 and later 

Resolution

For versions Aria Operations 8.1.1 to 8.6.4 please follow below steps to fix the pgpass permission issue:

1. Take the cluster offline and take snapshots per How to take a Snapshot of VMware Aria Operations 
   
   
2. SSH to the primary node as root, and run the following commands to correct the permissions on .pgpass
   

   # rm /storage/db/vcops/recovery.conf.bootstrap
   
   # chown postgres:users /var/vmware/vpostgres/current/.pgpass

3. SSH to the replica node as root, and run the following command to correct the permissions on .pgpass
   

   # chown postgres:users /var/vmware/vpostgres/current/.pgpass

4. From SSH to the primary node, manually start repl db service
   

   # /etc/init.d/vpostgres-repl start

5. From SSH to the replica node, manually start repl db service
   

   # /etc/init.d/vpostgres-repl start

6. Bring the cluster back online
7. Re-run the assessment tool.

Note: For versions Aria Operations 8.1.1 to 8.6.4, the "recovery.conf.bootstrap" should be present only on the replica node. Do not remove the file from replica node. 

 

For versions Aria Operations 8.10 and later please follow below steps to fix pgpass permission issue

1. Take the cluster offline and take snapshots per How to take a Snapshot of VMware Aria Operations 
2. SSH to the primary and replica node as root, and run the following commands

3. # rm /storage/db/vcops/recovery.conf.bootstrap
   # chown postgres:users /var/vmware/vpostgres/current/.pgpass

4. From SSH to the primary node, manually start repl db service
   

   # /etc/init.d/vpostgres-repl start

5. From SSH to the replica node, manually start repl db service
   

   # /etc/init.d/vpostgres-repl start

6. Bring the cluster back online
7. Re-run the assessment tool.

Note: For versions 8.10 and later the "recovery.conf.bootstrap" should not be present on the primary or replica node. 

Additional Information

VMware Aria Operations Pre-Upgrade Readiness Assessment Tools: https://knowledge.broadcom.com/external/article/324352/vmware-aria-operations-preupgrade-readin.html