VCF Fleet lifecycle upgrade fails at vmsp_upgrade with "Backup workflow failed for component: vcf-fleet-lcm"

Products

• VMware SDDC Manager / VCF Installer

Issue/Introduction

During a VMware Cloud Foundation (VCF) upgrade to version 9.1.x, the Fleet lifecycle component sub-task Perform Fleet lifecycle upgrade fails. The upgrade process stops during the vmsp_upgrade step when a backup is initiated.

Symptoms:

• Upgrade fails at the vmsp_upgrade stage.
• Error code: Backup workflow failed for component: vcf-fleet-lcm.
• Reference Codes: E6F5E4F3, A8A3CFC6, or 2DFCE4EF.
• Logs show: stderr: jq: parse error: Invalid escape at line 4, column 230.

• Error Message: 

  Failed with error code "Backup workflow failed for component: vcf-fleet-lcm"
  Reference Code: E6F5E4F3
  Step vmsp_upgrade failed. Reference Code: A8A3CFC6.
  stderr: jq: parse error: Invalid escape at line 4, column 230
  stderr: Error: exit status 5 

Environment

VMware Cloud Foundation 9.1.0.x

Cause

The failure is caused by a bug in the VMSP backup-upload script (upload_to_sftp.sh). When the configured SFTP backup account uses a domain-style username containing a backslash (e.g., domain\username), the script generates an invalid JSON manifest during the post-upload step. Specifically, the backslash is not properly escaped before being passed to jq, resulting in a parse error: Invalid escape and causing the upgrade task to fail.

Resolution

To resolve this issue, reconfigure the backup credentials to use the User Principal Name (UPN) format, which avoids the use of a literal backslash.

1. Log in to the SDDC Manager / Ops interface.
2. Navigate to Build > VCF Management > Backup & Restore > Backup.
3. Locate the existing SFTP configuration.
4. Update the username from the DOMAIN\user format to the user@domain (UPN) format.
5. If the update does not apply correctly, select Delete Location from the menu next to the configuration and re-add the backup location using the user@domain format.
6. Retry the Fleet lifecycle upgrade task.