MP 1,1-5,1 Mojave upgrade kills a DIMM.

[mattspace]

So I finally ugraded my machine to Mojave, and it seems to have coincided with one of my DIMMs dying in the process - I'm only showing 80GB, rather than 96. It's no longer showing up as a filled socket, and the red LED is lit. Problem persists on older macOS volumes, and follows the DIMM if I move it to other sockets.

Reset PRAM and left the machine unplugged from power overnight, problem persists.

Any thoughts on what part of the install process could do that?

 

[tsialex]

So I finally ugraded my machine to Mojave, and it seems to have coincided with one of my DIMMs dying in the process - I'm only showing 80GB, rather than 96. It's no longer showing up as a filled socket, and the red LED is lit. Problem persists on older macOS volumes, and follows the DIMM if I move it to other sockets.

Reset PRAM and left the machine unplugged from power overnight, problem persists.

Any thoughts on what part of the install process could do that?

Mojave has nothing to do with your DIMM problem, you have a bad DIMM or a bad contact within the CPU socket and processor or the DIMM and the slot - since the DIMM problem persists when you move to other slots, it's fairly to say that the DIMM itself has a problem.

 

[mattspace]

Mojave has nothing to do with your DIMM problem, you have a bad DIMM or a bad contact within the CPU socket and processor or the DIMM and the slot - since the DIMM problem persists when you move to other slots, it's fairly to say that the DIMM itself has a problem.

Yeah, it's just interesting the problem didn't seem to manifest until the first boot following the upgrade (which didn't involve opening the machine up or messing around with the hardware). I was wondering if there was some "problem identified, blacklist this DIMM" firmware thing that might have been going on. Or, if part of the upgrade process would have stressed an already fragile DIMM in a way that any other task wouldn't.

 

[tsialex]

Yeah, it's just interesting the problem didn't seem to manifest until the first boot following the upgrade (which didn't involve opening the machine up or messing around with the hardware). I was wondering if there was some "problem identified, blacklist this DIMM" firmware thing that might have been going on. Or, if part of the upgrade process would have stressed an already fragile DIMM in a way that any other task wouldn't.

Probably stressed enough the already flaky DIMM module that the DIMM ECC POST test marked it as bad in the first cold start.

 

[mattspace]

Probably stressed enough the already flaky DIMM module that the DIMM ECC POST test marked it as bad in the first cold start.

So the problem manifesting now in multiple different boot disks, is that because marking is persistent, or is it because each boot on each system is failing it for that session, afresh?

 

[Krevnik]

is it because each boot on each system is failing it for that session, afresh?

Yes.

As tsialex says, the DIMM was already flaky. The cold boot was the straw that broke the camel's back, as it were.

Usually electronics will fail when power is applied. That's the "shock" to the system that causes the sort of physical wear that will trigger failures.

 

[tsialex]

So the problem manifesting now in multiple different boot disks, is that because marking is persistent, or is it because each boot on each system is failing it for that session, afresh?

macOS has nothing to do with the bad DIMM detection, it's the POST process of the BootROM that detects a bad DIMM and this happens at every cold boot.

You could try to clean the gold contacts "fingers" of the DIMM with an eraser then brushing it with IPA and try the DIMM one more time before trashing it.