VCF Automation Orchestrator tab loads slowly due to stale cluster members after an upgrade

Products

• VCF Automation

Issue/Introduction

• In VCF Automation 9.0.2, navigating to or switching back to the System tab under Orchestrator in the VMware Aria Automation UI takes 3 to 4 minutes to load.

Environment

 

• VCF Automation 9.x

• Embedded VCF Automation Orchestrator 9.x

 

Cause

• Following an upgrade from version 8.x to 9.0.x, a stale node record from the previous deployment remains in the database. As a result, the system incorrectly registers four cluster members in the vmo_clustermember table instead of the expected three.

Resolution

To resolve this issue, clear out stale cluster data from the embedded Orchestrator database:

1. Ensure you have a valid backup of VCF Automation. 

2. Log into the Orchestrator database (vco_db) and execute the following SQL command:

   Delete FROM vmo_clustermember;
   

3. Restart the Orchestrator application pods using kubectl:

   kubectl -n prelude delete pods -l app=vco-app
   

4. Wait until the pods are running and operational, then reload the Orchestrator UI.