Hi, On 2016-05-12 21:03, Steven Hartland wrote:
I wouldn't rule out a bad cpu as we had a very similar issue and that's what it was.Quick way to confirm is to move all the dram from the disabled CPU to one of the other CPUs and see if the issue stays away with the current CPU still disabled.
One core is still running seemingly without problems it is only one core I disabled not the entire cpu. APIC 1 and 2 I believe are on the same chip. I am not a super CPU design expert, but if the two cores are on the same cpu chip do they not share the same memory bus with this model of the AMD cpu?
If that's the case it's likely the on chip memory controller has developed a fault
Or you could just move around two cpu cards and se if the error jumps from apic 1+2(err) to apic 3+4(err). If these are issued in order by FreeBSD? Or is the ordering random?
I suppose I could move all of the boards one step to the right and test it that way regardless.
If it does it is probably a DIMM or, as you say, the memory bus if not it is probably the cpuboard slot on the mainboard itself.
I will try this and post my findings. Offtopic:I cannot belive how poor the onboard bios diagnostics are on this server compared to my old IBM netfinity 5000.
rgrds Nikolaj Hansen
smime.p7s
Description: S/MIME Cryptographic Signature
