LCMVRAVACONFIG590070 and identity-service-app pods stuck when trying to start Aria Automation after experiencing VMware Identity Manager Health Issues

Products

• VCF Operations/Automation (formerly VMware Aria Suite)

Issue/Introduction

Following remediation of VMware Identity Manager (vIDM) critical health via reboot, VMware Aria Automation (vRA) fails to start.

Users encounter the following symptoms or errors:

• Aria Automation web UI: “No healthy upstream”
• VMware Aria Suite Lifecycle Request UI shows errors LCMVRAVACONFIG590093 during inventory sync and LCMVRAVACONFIG590070 on service restart of Aria Automation.
• identity-service-app pods remain at 0/1 Running

Environment

VMware Aria Automation 8.18.x

VMware Identity Manager 3.3.7

VMware Aria Suite Lifecycle 8.18.x

Cause

A degraded vIDM cluster prevented the identity-service-app from reaching a ready state:

• The lead opensearch node (node03) failed to start horizon-workspace and opensearch
• Remaining nodes (node01 and node02) lost quorum, resulting in failed opensearch daemons
• Identity-service-app, which depends on vIDM, could not complete startup

Resolution

1. Diagnose and Stabilize vIDM Cluster:

   1. Access: https://vIDM_FQDN:8443

   2. Review System Diagnostics and verify opensearch health reports as unknown

   3. Identify the Primary vIDM node.

2. Fix quorum and daemon failures:

   1. Check the status of key services:

      1. /etc/init.d/vpostgres status

      2. /etc/init.d/horizon-workspace status

      3. /etc/init.d/opensearch status

   2. Check the expiration of the vIDM "postgres" user account:

      1. chage -l postgres

      2. To change this run:
         

         chage -M number_of_days_here postgres

         e.g.:

         chage -M 9999 postgres

   3.  Stop and start all services in the following order (Starting with the Primary viDM node, then the Secondary nodes):

      /etc/init.d/opensearch stop
      /etc/init.d/horizon-workspace stop
      /etc/init.d/vpostgres stop
      
      /etc/init.d/vpostgres start
      /etc/init.d/horizon-workspace start
      /etc/init.d/opensearch start

3. Monitor opensearch health:

   1. Expect unassigned shards to resolve over time (may take up to 24 hours)

   2. Once green, proceed to Aria Automation service restart

4. Restart Aria Automation services:

   1. SSH to an Aria Automation appliance

   2. Run:

       

      /opt/scripts/deploy.sh

5. Verify:

   1. Confirm that identity-service-app pods now show 1/1 Running

   2. UI and inventory sync should resume normal operation.

Additional Information

Opensearch requires quorum to function when clustered. If a lead node is inaccessible, dependent services will fail. Unassigned shards will normally self-recover post-restart, depending on shard volume and node availability. Always verify vIDM stability before restarting Aria Automation services.