In my previous blog post, I described how to update ESXi nodes in a cluster from vSphere 6.0 U3 to 6.5 U1 with VUM. This process went smoothly for all nodes in the cluster, except two.

I noticed, during the upgrade procedure, that after quite some time the ESXi nodes where still not accessible.

Console Session

After I logged into the console session, I saw the following error: “Expecting 2 bootbanks, found 0

Schermafbeelding 2018-05-01 om 19.53.15

I could only press Enter and the ESXi node would reboot and revert to the same error message. In other words, I couldn’t do anything from here.

New Installation

I decided to directly attach the ESXi 6.5 U1 ISO file to the console session and start a new installation process.

When the installation process started, it stopped in about 9% with a new error message: “Unable to successfully execute ‘tryFormatDevice’ after 3 tries

Schermafbeelding 2018-05-01 om 19.48.15

When I googled this error I found the exact same issue on the VMware Communities forum. I too had an 8Gb boot-LUN attached to the ESXi nodes that were facing this issue.

Just like the forum posts mentioned, I increased the boot-LUN from 8 to 16Gb in the NetApp OnCommand System Manager.

Schermafbeelding 2018-05-01 om 20.19.52

After this step, I rebooted the ESXi node(s) and re-started the installation. I noticed the boot-LUN was indeed increased to 16Gb.

Schermafbeelding 2018-05-01 om 20.15.02

Succes!

When the installation continued I immediately noticed the process went beyond 9%, and after a while, I saw the installation was completed successfully!

Schermafbeelding 2018-05-01 om 20.15.26

I hope this blog post will help someone when encountering the same error messages.