The TKGi Management Console (or former EPMC) is quite handy, when it comes to automatically deploying TKGi. You can also deploy and manage your Kubernetes cluster from there (instead of using the pkscli / tkgicli).
But lately, I faced an issue where I couldn’t reach the Console anymore more and more often. This never happened during runtime, only after (re)booting the appliance.
I faced that issue mostly in my HomeLab, where my shutdown-procedure is …. well … highly professional … =D
But I had also some occurrences at customer production environments.
If you can’t reach the WebGUI anymore, check the VM. You will find that the DNS Name is “errorEOF” and the actual Managment IP address is not listed under IP Addresses anymore.
If you log into the VM’s Console, you will notice, that the prompt also says errorEOF. Additionally, the eth0 Interface does not have an IP (this is where the Management IP should be)
The fix is actually quite easy – we simply enable the Installation boot in the vApp Options.
To do so, you need to select the (1) TKGi Management Console VM –> (2) Configure –> (3) vApp Options –> (4) Edit
In the newly opened Wizard, go to (1) OVF Details and Check (2) Enable next to Installation Boot – then click (3) OK
Now, you have to shutdown the VM and Power On again.
(Note: Restart will not work).
This will now re-apply all settings, you’ve set during the initial OVA Deployment. Since the the important services in this VMs are running in Containers anyway (thus getting “redeployed” after each reboot), this isn’t a bad thing, like you would assume with “usual” appliances.
Now, your TKGi Management Console should be working fine again =)