SSP Upgrade Stalls at 93% due to vCenter Connection Timeout

Products

• VMware vDefend Firewall
• VMware vDefend Firewall with Advanced Threat Prevention

Issue/Introduction

SSP Upgrade stalls at 93% with app-discovery and redis pods in CrashLoopBackOff due to disconnected vCenter.

Symptoms:

During the upgrade process to version 5.1.1.0, the progress bar becomes stuck at 93% and eventually fails to proceed.

Observed Symptoms:

• In the SSP-Installer UI -> Instance Management -> vCenter Parameters, an error banner appears stating: "Failed to establish a connection to vCenter... Unable to connect to the vCenter."

• Running k get pods -A on the SSP-Installer CLI, reveals that app-discovery, csi-controller, and redis pods are in a CrashLoopBackOff or Error state.

• Upgrade failed message when expanded, show errors related to storage class health checks:

  Unexpected error while upgrading upgrade unit: Fabric8StorageClass... health check is not supported

Environment

Security Services Platform (SSP) Version Upgrade from 5.1.0.0 to 5.1.1.0

Cause

The upgrade process requires active communication with the vCenter API to validate resources and manage storage classes (specifically the csi.vsphere.vmware.com provisioner).

If the vCenter credentials have expired, the network path is interrupted, or the session has timed out, the CSI (Container Storage Interface) controller cannot mount or validate the volumes required by core services like Redis and App-Discovery.

This prevents those pods from starting, which in turn halts the upgrade orchestration at the final stages (93%).

Resolution

To resolve this issue, you must refresh the connection between the SSP Installer and your vCenter environment:

1. Access the Installer UI: Log in to the SSP-Installer web interface.

2. Navigate to Parameters: Go to the Instance Management -> vCenter Parameters section.

3. Verify Status: Look for a red error icon or a "Connection Failed" message next to the vCenter configuration.

4. Edit Connection: * Click on EDIT CONNECTION.

   • Re-enter the vCenter credentials (username/password) if necessary.

5. Monitor Pod Recovery: * Once the connection is successfully re-established, the csi-controller will be able to communicate with vCenter.

   • The redis and app-discovery pods should automatically transition from CrashLoopBackOff to Running as their storage volumes become available.

6. Resume Upgrade: Return to the upgrade dashboard and click Retry or Refresh. The upgrade should now proceed past 93% and complete successfully.

7. Final Step: Proceed with the SSP-Installer component upgrade as prompted to ensure all management tools are aligned with version 5.1.1.0.