I found mah ZFS baby with its teeth clenched, refusing to run. The mcelog analysis yielded this:
The questions:
1. How can I figure out from the motherboard layout which one is the bank #8? See the attached PDF, presumably this is the correct manual for my motherboard - a X8DT3-LN4F.
2. Will a subsequent scrub confirm that there are no ZFS block errors that have persisted on the existing datasets? There are no known disk issues - the 5 disks I have in this pool are almost new, and there have been no errors reported by the previous scrubs, automated or manually-launched.
3. I have attached also the latest dmidecode. What would I need to look for in there?
Thank you much.
root@nas02[/var/log]# mcelog --ascii --file /var/log/messages
Hardware event. This is not a software error.
CPU 0 BANK 8
MISC 30aed90400014200
MCG status:
Memory ECC error occurred during scrub
Memory corrected error count (CORE_ERR_CNT): 1
Memory transaction Tracker ID (RTId): 0
Memory DIMM ID of error: 1
Memory channel ID of error: 0
Memory ECC syndrome: 30aed904
STATUS 88000040000200cf MCGSTATUS 0
MCGCAP 1c09 APICID 0 SOCKETID 0
CPUID Vendor Intel Family 6 Model 44 Step 2
The questions:
1. How can I figure out from the motherboard layout which one is the bank #8? See the attached PDF, presumably this is the correct manual for my motherboard - a X8DT3-LN4F.
2. Will a subsequent scrub confirm that there are no ZFS block errors that have persisted on the existing datasets? There are no known disk issues - the 5 disks I have in this pool are almost new, and there have been no errors reported by the previous scrubs, automated or manually-launched.
3. I have attached also the latest dmidecode. What would I need to look for in there?
Thank you much.