kern.log errors

[dandart]

Hi all,

My server keeps dying with this in the last lines of the kern.log for that boot cycle. Here are the errors:


Sep 12 16:25:35 pollux kernel: [ 1801.010028] [Hardware Error]: MC0_STATUS[Over|CE|-|-|AddrV|CECC]: 0xd420400000000833
Sep 12 16:25:35 pollux kernel: [ 1801.012883] [Hardware Error]: Data Cache Error: during system linefill.
Sep 12 16:25:35 pollux kernel: [ 1801.015282] [Hardware Error]: cache level: L3/GEN, mem/io: MEM, mem-tx: DRD, part-proc: SRC (no timeout)
Sep 12 16:25:35 pollux kernel: [ 1801.018821] Disabling lock debugging due to kernel taint
Sep 12 16:25:35 pollux kernel: [ 1801.018825] [Hardware Error]: MC1_STATUS[Over|CE|-|-|AddrV|CECC]: 0xd400400000000853
Sep 12 16:25:35 pollux kernel: [ 1801.021693] [Hardware Error]: Instruction Cache Error: during system linefill.
Sep 12 16:25:35 pollux kernel: [ 1801.024312] [Hardware Error]: cache level: L3/GEN, mem/io: MEM, mem-tx: IRD, part-proc: SRC (no timeout)
Sep 12 16:25:35 pollux kernel: [ 1801.027847] [Hardware Error]: MC2_STATUS[Over|CE|-|-|-|CECC]: 0xd000400000000863
Sep 12 16:25:35 pollux kernel: [ 1801.030594] [Hardware Error]: Bus Unit Error: PRF/ECC error in data read from NB: SRC.
Sep 12 16:25:35 pollux kernel: [ 1801.033464] [Hardware Error]: cache level: L3/GEN, mem/io: MEM, mem-tx: PRF, part-proc: SRC (no timeout)
Sep 12 16:25:35 pollux kernel: [ 1801.037000] [Hardware Error]: MC4_STATUS[Over|CE|-|-|AddrV|CECC]: 0xd420400100000813
Sep 12 16:25:35 pollux kernel: [ 1801.039839] [Hardware Error]: Northbridge Error (node 1, core 0): DRAM ECC error detected on the NB.
Sep 12 16:25:35 pollux kernel: [ 1801.043528] EDAC amd64 MC1: CE ERROR_ADDRESS= 0x425aba90
Sep 12 16:25:35 pollux kernel: [ 1801.045424] EDAC MC1: CE page 0x425ab, offset 0xa90, grain 0, syndrome 0x40, row 0, channel 0, label "": amd64_edac
Sep 12 16:25:35 pollux kernel: [ 1801.045427] [Hardware Error]: cache level: L3/GEN, mem/io: MEM, mem-tx: RD, part-proc: SRC (no timeout)
Sep 12 16:25:35 pollux kernel: [ 1801.106945] [Hardware Error]: Machine check events logged


Is this just a RAM error - and I need to buy more RAM, or is it something deeper?

Thanks

[zackwasa]

try running memtest to see if everything is fine with your RAM chips. If you have more then one try removing one of them and see if it fixes, swap them until you see the problem again to find out what exact chip went bad