H12SSL-I Stuck at "bmc initiating"

[ocfguy]

Ugh - *just* got hit by this. I installed a NIC card and lo and behold - system would not boot. The LE1 LED flashes and turns off. It stays lit when power is applied. But there's no sign of life from the LEDM1 BMC LED. Booting produces nothing, system turns 'on' in the sense the fans power up but no output from the onboard VGA.

Looks like U4 might have had some scrapes and a damaged edge. For those who sent this motherboard back to Supermicro - how much did it cost to repair?

View attachment 28212


Keep in mind I've known of this thread and I have been careful when I put the new NIC card in - it is indeed amazing how fragile this board is !

They replaced my board (out of warranty) for free, but obviously YMMV.

 

[codgician]

can someone designe a 3D model of a U4-6 protector ? mounting with the MH1 hole shown on the picture and one or both of the heatsink holes of the Broadcom chip.
some grid holes in it for air circulation.

I protected it by cutting an SSD thermal pad and pasting to the area of U4/U5/U6. It works good so far but not sure about possible side-effects.

 

[zolti]

Hi, here is a photo of my broken motherboard.



Most likely, damage to the elements occurs during the removal of the PCI-E peripherals.

Repair in a local service (Russia) cost about $250.
I have no chance of a warranty repair.

The fastest and easiest way to protect yourself from such situations in this photo.

 

[RolloZ170]

The fastest and easiest way to protect yourself from such situations in this photo.

but remove after install/remove PCIe Cards. some parts(U4 U5 U6) are VERY hot.

 

[joet]

but remove after install/remove PCIe Cards. some parts(U4 U5 U6) are VERY hot.

I was wondering whether something like a squidgy thick strip of thermal pad might not be better - it's designed for heat flow and would spread any accidental impact forces from clumsy pcie card insertion/removal. Not sure I'm curious enough to buy another H12-SSL to find out though!

 

[hmw]

Supermicro replaced my motherboard for free - they noted 'damage to U4' and did not charge, the only cost was shipping to them for RMA. I guess this is a known design issue. It's enough to stop me from using the H12SSL, will probably sell it off - I am sure the U4 is of no consequence in the usual server environment where the servers are configured and then never touched for years and years. But for a home lab this motherboard is very nice but very fragile

 

[abirkill]

I just bought an H12SSL-I on eBay. I'd been doing some research to find out more about the board before it arrived, and learn about any issues, and found this thread. This issue was a concern as IPMI was pretty high on my list of priorities, so when the board arrived I studied it to learn where the danger area was, so I could avoid it.

However, it appears that this area of the board has a different design than that shown elsewhere in this thread, and these components aren't populated (in fact, there aren't even any pads in this area):



I've had a look in the surrounding area for any similar bare-silicon chips, but wasn't able to find any (there are a couple near the CPU socket, but well out of the way of inserting/removing cards).

So it looks like my board isn't as susceptible to damage in this area. My board revision is 1.02, I don't know if this is newer or older than the other boards pictured (it was sold as a new motherboard, but that doesn't mean it wasn't sitting on a shelf for a year or more).



Thought this might be useful to others who might be buying these boards and spotted this issue.

 

[RolloZ170]

So it looks like my board isn't as susceptible to damage in this area. My board revision is 1.02, I don't know if this is newer or older than the other boards pictured (it was sold as a new motherboard, but that doesn't mean it wasn't sitting on a shelf for a year or more).

the problematic rev. is 1.00(rare) and 1.01

 

[Tim_E]

Does anyone happen to know the resistance of R52 on an H12SSL? (exact board is a -CT, but I suspect they're all the same)

I had some fun with the hot air rework station and took a run at replacing a visibly damaged U5 voltage regulator. I don't want to jinx myself, but it think I managed to do it! (so small!!) Unfortunately... it looks like I knocked R52 off somewhere in the process. Does anyone happen to have a board handy that they can measure a resistance off of? Pad-to-pad is open circuit, so an in-circuit measurement would be golden.
I'll be crawling around on the floor searching in the meantime

Ref:

 

[oneplane]

I wouldn't be surprised if that is just the same as R51, seems like current limiting resistors for data lines.

 

[Tim_E]

I wouldn't be surprised if that is just the same as R51, seems like current limiting resistors for data lines.

good intuition. I had the radical idea to just ask supermicro, and I’m kind of surprised they answered! R52 is 2.2k ohm, just like its neighbor. I’m glad to know for sure because there are also 10k and 4.7k resistors in the area

 

[joet]

I'd imagine what we all want to know now is, 'Does it boot?'

My broken H12SSL-NT is hovering over an E-waste bin, these last few updates might just save it!

 

[Tim_E]

I'd imagine what we all want to know now is, 'Does it boot?'

My broken H12SSL-NT is hovering over an E-waste bin, these last few updates might just save it!

hah, well not yet... The board powers up. Thankfully no fireworks, but still not getting voltage out of the regulator. It's really hard to get the motherboard up to lead-free solder temps, so I'm not too surprised that there's a bad joint under there somewhere. I ordered an IR preheater that will hopefully help. I'll take another run at it when that arrives. Smart money would give up at this point, but I'm in it for the challenge

 

[RolloZ170]

I'll take another run at it when that arrives. Smart money would give up at this point, but I'm in it for the challenge

too late now. SM would have swapped it because this is a known issue.

 

[Tim_E]

too late now. SM would have swapped it because this is a known issue.

This board unfortunately has a pedigree such that supermicro wouldn't give me an RMA. (ebay shenanigans...)

 

[bitbckt]

This is more fun than complaining about the board on the internet, for sure.

 

[joet]

hah, well not yet... The board powers up. Thankfully no fireworks, but still not getting voltage out of the regulator. It's really hard to get the motherboard up to lead-free solder temps, so I'm not too surprised that there's a bad joint under there somewhere. I ordered an IR preheater that will hopefully help. I'll take another run at it when that arrives. Smart money would give up at this point, but I'm in it for the challenge

Well, I do have access to a preheater big enough, and a hot-air gun, but I also have previous experience of tombstoning parts near the one I was replacing, so am only half tempted to give it a go. Then again, it's scrap unless I try...

 

[RolloZ170]

but I also have previous experience of tombstoning parts near the one I was replacing

i use self adhesive copper foil.

 

[principino1984]

This board unfortunately has a pedigree such that supermicro wouldn't give me an RMA. (ebay shenanigans...)

I bought from ebay and SM accepted the RMA .. it's not a crime to buy in china since in europe when I bought it was not available at anywhere.
For sure you have to "cry" a little more than expected and you have to ship to them by yourself...but at the end they will open an RMA number/case for you

Marco

 

[RolloZ170]

it's not a crime to buy in china

sure. but the warranty rules may different, china only parts do not have 3 years warranty.