"Upgrade failed" in VAMI Cluster tab for IaaS component
search cancel

"Upgrade failed" in VAMI Cluster tab for IaaS component

book

Article ID: 325975

calendar_today

Updated On:

Products

VMware Aria Suite

Issue/Introduction

Symptoms:
  • After a failed automatic IaaS upgrade attempt IaaS is upgraded successfully with the manual IaaS installer.
  • For one or more IaaS component you see:
Upgrade failed
or
Default Web Site Upgrade to 7.6.0.16195 pending


Environment

VMware vRealize Automation 7.6.x

Cause

The Upgrade failed warning indicates automatic IaaS upgrade has failed for the respective IaaS component and the node it resides on has not been reverted. 

This status would prevent subsequent automatic upgrades as the environment might be in a corrupted state.

In general, it is always recommended to revert the node where the IaaS upgrade has failed before retrying the upgrade. This is especially valid in case of failed primary web node upgrade. In that case the IaaS MS SQL database must also be reverted to its pre-upgrade snapshot or back-up.

Resolution

Before clearing the Failed status it is important to understand how and why the automatic upgrade failed and to make sure that the manual upgrade was indeed successful and environment is in a healthy state.

The logs from the automatic upgrade are found on the primary appliance at /opt/vmware/var/log/vami/upgrade-iaas.log and then check the Management Agent All.log file on the node where it failed.

The Upgrade failed state for IaaS components can be cleared in the registry. 
  1. Locate the respective IaaS component key under one of the following registry locations and expand its entries.
"HKEY_LOCAL_MACHINE\SOFTWARE\Wow6432Node\VMware, Inc.\VMware vCloud Automation Center Server"
"HKEY_LOCAL_MACHINE\SOFTWARE\Wow6432Node\VMware, Inc.\VMware vCloud Automation Center WAPI"
"HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\VMware, Inc.\VMware vCloud Automation Center DEM"
"HKEY_LOCAL_MACHINE\SOFTWARE\WOW6432Node\VMware, Inc.\VMware vCloud Automation Center Agents"
  1. Locate the ComponentCondition registry entry and delete it, or modify its value to Completed.

The change will be visible in VAMI Cluster tab immediately after the next Management Agent ping report (by default every 30 seconds).