SOLVED BIOS hangs on B4

[Rilo Ravestein]

Well, after finally figuring out why my USB sticks kept dying (Ultra Fit 32GB issue in FreeBSD), i now have what appears to be a dead mobo . (X10SLH-F)
I updated the BIOS, everything seemed to have succeeded. Rebooted into some bootable freedos usb stick again to be sure, which worked fine. Rebooted, entered BIOS settings to change the boot option to the other USB and saved and rebooted again.....

And now the BIOS hangs at the Supermicro logo (with the 'hit DEL to enter BIOS settings' and other options screen) with code B4 in the lower right corner.

Anny suggestions or just RMA?

 

[Rilo Ravestein]

Weird. I just connected with the IPMI view utility and reflashed the IPMI firmware et volá.... everything is working again. I have no idea what happened here

I flashed the BIOS again, i just did not trust the previous flash anymore, and everything works like a charm now.

 

[Ericloewe]

Just the IPMI firmware? That's odd.

The BMC does have some pretty deep hooks, though. Deep enough to make you question the wisdom of maintaining compatibility all the way back to the IBM PC and the Intel 8086.

 

[Rilo Ravestein]

Yep, only the IMPI firmware. Don't even know why i tried that, but i thought i'd just give it a try.

 

[jgreco]

Oh hell yeah, if the IPMI/BMC is hosed up in the wrong way, the BIOS won't POST. Which gets you into that fun situation of how do you get in via the BIOS to bonk the IPMI back into shape.

 

[Rilo Ravestein]

Well, I wonder how the IPMI/BMC got hosed up by a BIOS upgrade in the first place. Getting IPMI back was surprisingly easy, using Supermicro IPMIview (V2.10.2 on Windows) to flash the IPMI/BMC firmware. So somewhere it seems like IPMI was not completely hosed down. Well, i could connect but the iKVM viewer didn't work, but you don't need that to flash the fimware fortunately.

 

[Z300M]

Just the IPMI firmware? That's odd.

The BMC does have some pretty deep hooks, though. Deep enough to make you question the wisdom of maintaining compatibility all the way back to the IBM PC and the Intel 8086.

When I emailed Supermicro tech support about long boot times and restarts immediately after shutdown, they told me that BIOS and IPMI should be updated together: I had updated IPMI but not the BIOS.

 

[Rilo Ravestein]

If that is true, than big shame on Supermicro. Nowhere in any documentation/readme files/website/... is that mentioned.

 

[Z300M]

If that is true, than big shame on Supermicro. Nowhere in any documentation/readme files/website/... is that mentioned.

Supermicro may have covered themselves by their general warning not to update BIOS or IPMI without good reason, which they might interpret as meaning "unless we tell you to" -- and if one were to contact them about a problem, they might well advise updating both IPMI and BIOS.

Or maybe it was merely my particular combination of updated IPMI and earlier BIOS that caused the problem.

 

[Rilo Ravestein]

I think the advice you got was from some intern/trainee talking through his hat.

 

[Ericloewe]

I think the advice you got was from some intern/trainee talking through his hat.

It does make some sense though. The BIOS update includes the Management Engine, which interfaces with the BMC. Inter-version compatibility is probably not a high priority.

 

[Rilo Ravestein]

If that is true, they should mentioned it in TFM, TF readme, or TF website. Sorry... i'm (a bit) frustrated today

 

[jgreco]

Sounds about right. And, yes, it's frustrating, there's a lot of "wait, what, that wasn't obvious to you?!?" in this biz. :-/

 

[Rilo Ravestein]

...there's a lot of "wait, what, that wasn't obvious to you?!?"

How is this obvious? It isn't even true/possible. The BIOS updates do not come out at the same dates as the IPMI updates. The latest BIOS update for my mobo is from 2015-04-24. The latest IPMI update is from 2014-05-21. In my case, I already had the latest IPMI update.

 

[jgreco]

How is this obvious?

It's not obvious if you're not in the loop. "Obviously!" (everyone groans.)