[j-nsp] Multibit ECC errors on M7i CFEB

[Andrew Hoyos]

Hi j-nsp folks,

I've got a an M7i, and a crashing CFEB, which appears due to multibit ECC memory errors. 

At this point, we've replaced both the CFEB card itself, and the memory a few times, figuring it was just a bad card or bad memory, but within a day, same thing.

Any other thoughts as to what might be causing this? Beginning to consider chassis replacement, etc. 

Thanks in advance,
Andrew

relevant log:

Jun  5 06:06:09  splt-mdpb-cr01 cfeb NMI detected from Watchdog (nmi_stat 0x4)
Jun  5 06:06:09  splt-mdpb-cr01 cfeb mpc106 machine check caused by error on the Processor Bus
Jun  5 06:06:09  splt-mdpb-cr01 cfeb mpc106 PCI status register: 0x0020, error detect register 1: 0x00, 2: 0x08
Jun  5 06:06:09  splt-mdpb-cr01 cfeb mpc106 error ack count = 0
Jun  5 06:06:09  splt-mdpb-cr01 cfeb mpc106 error address: 0x0f4023c0
Jun  5 06:06:09  splt-mdpb-cr01 cfeb mpc106 Processor bus error status register: 0x72
Jun  5 06:06:09  splt-mdpb-cr01 cfeb transfer type 0b01110, transfer size 2
Jun  5 06:06:09  splt-mdpb-cr01 cfeb mpc106 error detection reg2: ECC multibit
Jun  5 06:06:09  splt-mdpb-cr01 cfeb ^B
Jun  5 06:06:09  splt-mdpb-cr01 cfeb last message repeated 7 times
Jun  5 06:06:09  splt-mdpb-cr01 cfeb Registers:
Jun  5 06:06:09  splt-mdpb-cr01 cfeb R00: 0x0000b030 R01: 0x0079ace0 R02: 0x0079a55c R03: 0x00000001
Jun  5 06:06:09  splt-mdpb-cr01 cfeb R04: 0x00008000 R05: 0x00670000 R06: 0x0066c018 R07: 0x0079af44
Jun  5 06:06:09  splt-mdpb-cr01 cfeb R08: 0x0079adb4 R09: 0x0066b920 R10: 0x0079adb4 R11: 0x00000000
Jun  5 06:06:09  splt-mdpb-cr01 cfeb R12: 0x28002024 R13: 0xb0d42200 R14: 0xe1f2c2bc R15: 0xccb59042
Jun  5 06:06:09  splt-mdpb-cr01 cfeb R16: 0x66b05803 R17: 0x296904ca R18: 0x00670000 R19: 0x00670000
Jun  5 06:06:09  splt-mdpb-cr01 cfeb R20: 0x00670000 R21: 0x00670000 R22: 0x00000000 R23: 0x002513f9
Jun  5 06:06:09  splt-mdpb-cr01 cfeb R24: 0x00000000 R25: 0x0028d903 R26: 0x00000000 R27: 0x0028c57b
Jun  5 06:06:09  splt-mdpb-cr01 cfeb R28: 0x00000000 R29: 0x0028d48a R30: 0x00000000 R31: 0x0028d903
Jun  5 06:06:09  splt-mdpb-cr01 cfeb MSR: 0x0008b030 CTR: 0x00000000 Link:0x00039850 SP:  0x0079ace0
Jun  5 06:06:09  splt-mdpb-cr01 cfeb CCR: 0x28002022 XER: 0x00000000 PC:  0x000361e4
Jun  5 06:06:09  splt-mdpb-cr01 cfeb DSISR: 0x00000000 DAR: 0x00000000 K_MSR: 0x00000030
Jun  5 06:06:09  splt-mdpb-cr01 cfeb Stack Traceback:
Jun  5 06:06:09  splt-mdpb-cr01 cfeb Frame 01: sp = 0x0079ace0, pc = 0x0002674c
Jun  5 06:06:09  splt-mdpb-cr01 cfeb Frame 02: sp = 0x0079acf8, pc = 0x00464c90
Jun  5 06:06:09  splt-mdpb-cr01 cfeb Frame 03: sp = 0x0079ad10, pc = 0x0002c6a8
Jun  5 06:06:09  splt-mdpb-cr01 cfeb Frame 04: sp = 0x0079ad58, pc = 0x00027468