Not, sure. Went into bios to change some things and now its fine. Not fond of the temps though. The bios on this board is amazing. Allot of adjustments can be done. At first, had it set to run all cores at max turbo (3.9) but the temps were 65c to 70c on air. Readjusted back to normal and temps are under 50c. Bad tim on these, just like the ivy bridge quads. Was told the lids are soldered on so no taking these off. Need more time to play with it to see what I can do. Going to look for a better cooler though. Never water cooled before but and thinking about it now. Just need the new AMD GPU to go with this rig. Overall, you would like this chip.
Not sure what caused the BSOD but think it was my fan controls I played with. Looking at the log, it pointed towards this. My case has 3 pin and need 4 pin for bios to control the fans (forgot about the pins).

You should take a look periodically at the jobs running. With Linux there is some kind of bug which will cause the autodock jobs to hang. They will look like they are running, but if you watch the percentage complete, you will notice it does not change. When the job is running properly the percentage done will change very quickly, every 2 seconds or so. The time will continue to increment and the time remaining will grow. If I suspend the WU and then resume it, it will take off and run correctly when its turn comes again. This only happens with the autodock units and never with the Vina units. It is hit and miss and does not happen with all the autodock units. When I look in task manager it will show one core is doing nothing while the other cores are cooking along at 100%. As soon as another unit fires up, all the cores are at 100%. When the stuck unit does resume, it will back off on the time running to what ever time had accrued when it initially got stuck. Once resumed, they all appear to run normally. It does cause some lost time on my Linux rigs.
Hope this helps.
Cheers

I only have 4gb of ram in the 8 core servers and currently task manager is only showing 20 to 25% of memory being used (FAAH only). With 3 times as many jobs running he may be bumping up to 75%+ or more depending on what else he is running. If he is running CEP2 then he could be swapping like mad. At any rate 4gb of ram with 24 cores is probably pushing the limits pretty hard. Going to 16gb will certainly give him some headroom.
Cheers

I've upgraded the RAM, and disabled BOINC use of the page swap file. System reports that RAM is only 15% utilized.

At the moment it's doing 24 Autodock units and there doesn't appear to be any hung units.

Another weird thing I found is when going to view the system information in Linux Mint, it says Opteron 6176 X 18. But when viewing the BOINC message log at the beginning, it says Genuine AMD Opteron 6176 X 24.

System monitor reports 24 cpu's running at 99-100% so it certainly seems to be doing 24 cores worth of work.

It normally does mostly VINA, just a rare occurence I guess, probably because it has several days of work downloaded and it seems to kind of group things together.

Another thing I didn't mention before was it's using an old wireless usb stick for network connnectivity. I'm a bit wary of wireless, maybe better to get it closer to the router so it can do a LAN connection.