Getting an error at end of install to iSCSI lun. "Expecting 2 bootbanks, found 0"

[marcomosa]

I'm in trouble installing ESXi 5.0 booting from iSCSI LUN. The installation wizard ends with an error: "Expecting 2 bootbanks, found 0".

There's someone who observed the same issue and could help me solving this problem?

Thanks

[LYuan]

Have you figured out a solution for this problem?  I'm having the exact same issue.  Server will detect the iSCSI LUN, and install until 90%, and then fail.  Has anyone else seen this?

Larry

[rlund]

What lun id are you using?

[LYuan]

Lun ID is '0', it's the only lun I have attached to the Intel NIC that I am booting up with.  Should I try a different LUN ID?

[rlund]

I am assuming you have booted iSCSI with esxi 4.x with this server hardware and san? Not all of either support iSCSI boot.

If you do alt f12 , from memory here, might be 11, it shows the kernel message, what's that show for a error?

Roger

[LYuan]

Sorry Roger,

I'm having trouble following - Where, or at what point would you like me to press ALT-F11/12?  Actually, I'm attempting to boot off a Synology DS1511.  I've actually done this before.  I installed ESX 5.0 on the LUN the first time, and worked flawlessly.  After that, I proceeded to delete the LUN and the entire setup to do it again for documentation.  However, the install now continuously fails.  I haven't changed my process, or I don't think I have.

Thanks,

Larry

[AnthonyChow]

When you delete the LUN and re-create it does any of the parameter changed not matching on your ESXi host?

[LYuan]

I'm pretty sure everything is exactly the same - I do notice on the synology, that at that point during the install, it looks like the system disconnects from the LUN before giving the error.  It stops at 90% for 1 minute before showing the error.  What does the message usually mean?

[jl3128]

Same thing here.  This is a deal breaker.  ESXi 4.1 works fine on bootable iSCSI luns with the same exact setup.  Please help figure out why 5.0 doesn't install properly.  Using NetApp luns (-t vmware) for the record although it doesn't seem like it matters.  Dropping to alt+F12 shows the following (more or less):

The {} UI encountered an error...

Traceback (most recent call last):

     File "weasel/main.py", line 108, in main

          userInterface._execute()

     File "/usr/lib/vmware/weasel/console_install.py", line 51, in _execute

ERROR      Caught an exception: Error (see log for more info):

Expecting 2 bootbanks, found 0.

[jl3128]

also changing lun id from 0 to 1 doesn't help...

[fairmont50]

I came across the same error booting from an iSCSI LUN connected to a NetApp controller. Just for kicks I rebooted the box and ESXi 5.0 booted right up..

[BradACS]

Had this same problem yesterday trying to install ESXi 5.0 on a iSCSI LUN on Cisco UCS B230 blade.   Reading through the logs in the F12 screen, it appears that at some point in the install, ESXi drops the static IP assigned to the iSCSI vNIC from UCS Manager and attempts to get a DHCP address. In our case, that NIC is an access port in layer 2 VLAN for storage only, with no DHCP server.  As you can imagine this causes the iSCSI LUN to become unavailable.  While it did reboot OK and "seems" to come up as expected, the one thing I did notice is that the root password was blank, when I know I set a password during the install.

It seems to be a problem with the networking portion of the installation scripts around 90%.  This possibly would not happen with more than one vNIC assigned in UCS manager, but we were just testing the iSCSI boot in UCS with a Cisco rep and only had the one vNIC assigned for iSCSI boot.

[BradACS]

We just redeployed this UCS blade with 6 vNICS, the last two for primary and secondary iSCSI.  The problem went away and everything appears to have worked as expected.