Bad gateway errors are returned by Aria Automation when trying to login after redirected to vIDM when auto recovery is disabled
search cancel

Bad gateway errors are returned by Aria Automation when trying to login after redirected to vIDM when auto recovery is disabled

book

Article ID: 417563

calendar_today

Updated On:

Products

VCF Operations/Automation (formerly VMware Aria Suite)

Issue/Introduction

  • When attempting to log in to the Aria Automation 8.x (vRA) UI, users are unable to access the portal and receive a "Bad Gateway" error in the browser. This issue indicates a problem with the underlying VMware Identity Manager cluster responsible for authentication.
  • Pinging deletgateIP from any node fails.
  • When reviewing VMTools for each node, you cannot find any node with two IPs for one of the nodes which should be the floating IP outside of the cluster of IPs that is the delegateIP indicating the delegateIP is not assigned to any node in the cluster.

Environment

VMware Identity Manager 3.3.7

VMware Aria Automation 8.x

Cause

The delegateIP (virtual IP) for the vIDM cluster is not assigned to any active node. This failure is caused by the presence of the following file on the vIDM appliance nodes:

/usr/local/etc/LCM_DISABLE_AUTO_RECOVERY

This file explicitly prevents the Aria Suite Lifecycle (vRSLCM) auto-recovery feature from managing and assigning the delegateIP, even if a failure is detected.

Resolution

To resolve this issue, you must remove the blocking file from all vIDM nodes to allow the cluster to self-heal.

  1. Log in to each vIDM appliance node via SSH.

  2. Check for and remove the auto-recovery disable file using the following command:

    rm /usr/local/etc/LCM_DISABLE_AUTO_RECOVERY
    
  3. Repeat this step on all nodes in the cluster.

  4. After the file is removed from all nodes, restart the horizon-workspace service to trigger the auto-recovery process.

    /etc/init.d/horizon-workspace restart
    
  5. After the service restarts, the delegateIP will be automatically reassigned to a healthy node in the vIDM cluster, and the Aria Automation login page will become accessible.

Additional Information

The troubleshooting script verify-cluster-prerequisites-v5.sh can be used to confirm the presence of this file as a prerequisite failure. The check will return as a warning that auto-recovery is disabled.

Attachments

verify-cluster-prerequisites-v5.sh get_app