I am having an odd issue with only 1 of 5 hosts connected to a cluster inside of vcenter. One host randomly disconnects from vcenter.
I right click the host and select connect. I get the default warning that "reconnecting a host will override any resource management changes that were made directly on the host....". I select yes.
I then get an error
Authenticity of the host's SSL certifcate is not verified. Only option is to close.
Then I am taken to the add host wizard. where I can enter the IP and root password to add the host back to the cluster.
I now get a security alert:
"Unable to verify the authenticity of the host: the SHA1 thumbprint of the certificate is: "
The shown thumbprint is different everytime I add this host.
I have worked through the following vmware KB's with no resolution.
We have tried, replacing the network card that the management network is on, replacing cables, switch the port on the switch this machine is plugged into, completely reinstalling esxi. Nothing has solved this issue.
Does anyone have any other things to try?
Turns out the problem was that our nimble was taking up ip's that it wasn't supposed to and it was a duplicate IP of that esxi host. Very odd behavior. But this is the fix.
The shown thumbprint is different everytime I add this host.
That sounds weird. Did you check if the certificate on the host in /etc/vmware/ssl/rui.crt is really changing constantly? You can check the thumbprint of the base64'd certificate file on the local console with openssl x509 -text -in /etc/vmware/ssl/rui.crt
You can also connect with your browser to your host and check the presented SSL certificate.
Did you restart the management agents and completely remove the host from vCenter and re-add it in?
Thanks for the reply.
No I have not checked that this has changed. I will take note what is currently there and check it when it disconnects again.
I just checked it and the host's certificate at /etc/vmware/ssl/rui.crt is actually staying the same.
Also, I have attempted removing the host from vcenter and adding it along with restarting the management agent, to no avail.
Turns out the problem was that our nimble was taking up ip's that it wasn't supposed to and it was a duplicate IP of that esxi host. Very odd behavior. But this is the fix.