When attempting to upgrade VMware Site Recovery Manager (e.g., from 9.0.2.x to 9.0.4.x or 9.0.5.x), the operation fails immediately.
The UI displays the error message: unable to upgrade the SRM to the version.
DNS server ordering may differ between the Source (Old) and Destination (New) appliances.
The Destination appliance fails to resolve the Source appliance's FQDN during the initial handshake.
VMware Site Recovery Manager (SRM) 9.x
VMware vSphere Replication (if applicable)
This issue is caused by environmental prerequisites not being met during the upgrade handshake:
DNS Resolution Failure: The new (destination) appliance cannot resolve the FQDN of the source (old) appliance, or the DNS response is too slow, causing a handshake timeout.
Account Expiration: The admin account on the source appliance may be expired.
Time Skew: Significant time drift exists between the appliances.
To resolve this issue, verify the appliance health and enforce name resolution locally.
SSH into the Source (Old) SRM appliance as root.
Verify the admin account is not expired:
chage -l admin
If expired, reset the password or expiry date.
Verify time synchronization:
date
Ensure the time matches the Destination appliance and NTP is active.
SSH into the Destination (New) SRM appliance as root.
Backup the current hosts file:
cp /etc/hosts /etc/hosts.bak
Edit the hosts file:
nano /etc/hosts
Add a static entry for the Source appliance at the end of the file in the following format:
<Source_IP_Address> <Source_FQDN> <Source_Shortname>
Example: 192.168.10.20 srm-source.example.com srm-source
Save the file.
Retry the upgrade procedure from the SRM Appliance Management Interface.
Ensure that port 443 and port 9086 are open between the Source and Destination appliances.
Once the upgrade is complete, the static entry in /etc/hosts can be removed if DNS is functioning correctly for the new instance.