Upgrade fails due to broken replication between the Primary and Replica node in VMware Aria Operations 8.x.

Products

• VCF Operations/Automation (formerly VMware Aria Suite)

Issue/Introduction

• The upgrade fails due to error:
  

  Failed"bootstrap_file_existence.

• The last step on "Aria Operations 8.x upgrade hangs/fails due to incorrect .pgpass permissions." points to broken replication.
• When checking the status of replication on the Replica node with the /etc/init.d/vpostgres-repl status command, you receive output similar to:
  

  pg_ctl: directory "/storage/db/vcops/vpostgres/repl" is not a database cluster directory
  vRealize Operations vPostgres Replication Database (vpostgres-repl) is not running.

• When trying to start the replication service with the /etc/init.d/vpostgres-repl start command, you receive output similar to:
  

  Reset vRealize Operations vPostgres Replication Database (vpostgres-repl)...
  Test connection to (masterIP)
  Base backup from (masterIP)
  Failed creating base backup from (masterIP)
  Starting vRealize Operations vPostgres Replication Database (vpostgres-repl)...
  vRealize Operations vPostgres Replication Database (vpostgres-repl) is already running.

Environment

VMware Aria Operations 8.x

Cause

The issue occurs due to incorrect information in the /storage/db/vcops/vpostgres/repl/pg_hba.conf file in VMware Aria Operations (formerly known as vRealize Operations) 8.x.

Resolution

This is a known issue in Aria Operations, and there is currently no resolution available. If you believe that you have encountered this issue, please raise a case with Broadcom support.