Migration from Windows vCenter 5.5 to vCenter Server Appliance 6.2 - Part 2 - Upgrade vCSA 5.5 to vCSA 6.0 U2

Interrestingly, the process of vCSA upgrade is similar to vCenter Server to vCSA migration. 

First new appliance is deployed and then data is copied over from the old appliance. That is actually a significantly safer of option of upgrade as at any point of the upgrade process you can roll back by simply powering back on your existing vCenter.

Before you start the upgrade procedure you need to complete the following:

1. Install Integration Plug-In from the vCSA ISO file

2. Create temp Ephermal PortGroup. 

This will be used as a temp network by new vCSA 6 to connect to the current vCSA to copy data across. Once the copy is over the new vCSA will take network settings of the old vCSA and this network connection will be dropped. Make sure new vCSA can talk to existing vCSA from this temp network.

Once you complete these prerequisites download and mount vCSA ISO file and run open vcsa-setup.html


1. Press Upgrade button

2. Confirm your current vCenter version allows to upgrade to vCenter 6.0


3. Accept EULA


4. Provide FQDN and credentials for the ESXi server you want to deploy new vCSA to




5. Specify name for new vCSA - that's only the VM's name.
Optionally you can enable ssh on the appliance.




6. Enter details of vCSA 5.5 and the host where it resides




7. And I bumped into the first problem




8. Luckily I quicky found the solution in VMware Communities. 

The problem was in extra sequence in Postgres schema which had to be removed.

Run the following command in the vCSA shell and restart the upgrade procedure.

/opt/vmware/vpostgres/current/bin/psql -U postgres -d VCDB -c "drop sequence if exists vpx_host_cnx_seq cascade"


9. Select Appliance Size

10. Select datastore where you want to place new vCSA 6 U2

11. Configure temp network setting. 

I created a Temp portgroup on the same vlan where vCSA 5.5 resides and took advantage of my dhcp server

12. Confirm upgrade wizard settings

13. And the upgrade process failed with message 'Internal error occurs during export'.

I read vSphere Release notes and found similar error that suggested to use static IP address instead of DHCP config for temp network in step 11. Once I re-ran the upgarde wizard again with adjusted settings it all worked just fine

14. Then I spotted there was something wrong with vFRC config.

ESXi was reporting no disk space for vFRC.



But when I tried to remove the SSD and add it again the task failed with error that there are VMs using vFRC.

I checked VMs' settings, but it looked like these were not configured for vFRC even though they were.

So I had to shutdown VMs and re-add SSD disk as VFRC capacity which fixed the issue.