Upgrade from ESXi 5.1 to 5.5 (free) and stuck at "iodm loaded successfully"

[fmoreira86]

Hi everyone!

So I run a little whitebox HP Compaq 8100 SFF with ESXi for my own lab tests.

Today I upgraded to 5.5 and when the ESXi was loading it just stuck at "iodm loaded successfully".

The odd thing is, I tried to ping it... and it "replied". So I thought "let me check if my VM's are up"... and it was!! The console just kept showing "iodm loaded successfully"

So I tried a fresh install and no problems... but as soon as I enabled my graphics and sound passthrough, rebooted and.... boom! Stuck at "iodm loaded successfully"!!

Any ideas?

I'm going back to 5.1 til I figure out this issue.

Thanks!

[nkrishnan]

Please provide additional details and screenshot if possible,

Does ESXI booted completely ?.

Did you get the user intervention screens Is your hardware is part of HCL (especially the pass through the device)

Did you able to connect the ESXi with vSphere client.

Did you have any 3rd party driver or software (management software) installed on ESXI5.1 . Or it was a plain VMware ESXI image.

[zXi_Gamer]

Hi guyz. I have seen this thread multiple times and I have tried on my setup and following is the observation I had noted:

When the onboard or the primary Graphics unit is given as passthrough and the server is rebooted, during reboot, after the loading of the cnic_register driver / iodm, it would be the time of the graphics card to get register or the display to be handed over to vmkernel. But we have made the graphics unit as passthrough, hence the display driver is NOT taken by the vmkernel, but Left For Alive Dead.

Meaning that the server will continue to perform only that the user will not be able to see anything at all in the DCUI. I was able to confirm this by having a serial port attached and observed that the ESX server was still booting and the ip was accessible and the server manageable.

To revert the settings when you have made the graphics unit passthrough, do a ping on the server during reboot, as usual after 90 secs or so, the server will start replying to ping. Now, disable the graphic unit from passthrough and you are good to go.

Hope that helps,

Hardware passthrough causing crash...Cnic_register...

_zXi_

[fmoreira86]

So it basically means that my whitebox can't passthrough my Graphics, right?

[ramkrishna1]

Hi

Welcome to the communities.

To make sure graphics problem please go through HCL .

"I take a decision and make it right."

[zXi_Gamer]

Not exactly, you cannot passthrough the primary display device which ESX is using. But then again, you have to use the HCL card, else, it might end in other probs

[eRJe]

I know this is a bit of an old post but I just like to add my experience. Perhaps it is useful for someone else.

I enabled (1394 (firewire) pass-through on the ESXi 5.5 host. Rebooting to make the change active, resulted in a what seemed to be a frozen ESXi host at boot.Just like described by the OP. It turned out that the 1394 had a dependency with the gfx chip, putting the main (only) gfx also on pass-through. The ESXi host continues loading, but the display freezes when the pass-through is being activated.

I was just being impatience waiting for the host to continue to boot. (I have to wait approximately 10 minutes longer because the host tries to find the remote maintenance controller which is currently not installed.) And because I also made some other changes that were BIOS related, I was looking into an other direction.