PowerEdge T410 Memory Configuration Error

Yes, you can populate all four slots on each CPU and have more than 32GB of memory. The memory configuration guidelines/population rules should be in the manual.

http://www.dell.com/support/

The memory mode, memory specifications, or a faulty slot or module may prevent all memory from being available.

I have followed the instructions in the manual, populating all banks with 8GB Dell A3078601 DIMMs. However no matter what I do, the BIOS always disables A1/A4 and B1/B4 if all 8 slots are filled. There is nothing wrong with the motherboard, memory or sockets as Windows Server 2016 reports 64GB of memory but with only 32GB available for use. BIOS has also locked the memory mode option and I cannot choose anything other than Optimizer (which appears to be the right mode when all DIMM sockets are full. I have also tried using three DIMMs for each processor (A1/A2/A3 and B1/B2/B3) which results only in 16GB of available memory. 

Is there some sort of other BIOS settings that I have missed? Are the DIMMs I'm using perhaps somehow not compatible with this system? It seems that no matter what size or amount of DIMMs I populate, the BIOS never allows all system memory to be available. 

Having same problem!

After a lot of faffing, have upgraded the system BIOS [not to be confused with the DRAC/iDRAC BIOD] from my original system's 1.5.3 to what seems to be the most recent BIOS v1.14.0

Still have the MEMBIST failure on slot B3.

Weirdly, the Dell T410 system inventory report lists all 8 memory lots as being populated correctly and active - no failures!

hi, you can try running diagnostics.

 

https://dell.to/2YMlTIw

Many thanks for at least replying. As an IT engineer of over 25 years experience, and personal experience of home computers dating back to the Commodore 64, I'm aware of various testing procedures. I have tried various Dell and non-Dell diagnostic utilities. Running the Dell system inventory tool does return all 8 ram sticks being installed and active. However, BIOS seems to actually test most of the significant memory addresses and found a fault which then cased the BIOS to disable that slot. Using the non-Dell diagnostics, and advice from elsewhere, leads me to believe there is damage to at least one of the conductors somewhere along of the address and/or data traces between the second cpu and the B3 ram slots. Possibly just some dirt or oxidization of the copper but enough to cause an intermittent fault. As these traces are so tiny and there are about 300 of them, it isn't practical to test them with a multi meter! Fortunately, I've found a replacement motherboard for £26 so I'll try that when it is delivered. Otherwise, as this is such a clunky and overly proprietary machine, I'll sell the whole thing for parts.