20130828 Storage is amazing, and nobody's happy
August 28, 2013•136 words
Customer wants to know why their 5 TB filesystem with a gazillion files went offline for a while.
Well, because the filesystem software found some unexpected data and it flagged the issue for repair.
But why?
Because it's smart.
But why did it do that?
Because it knew it should sweep up before you actually start losing something important.
We demand a Root Cause Analysis.
The root cause is that something unexpected happened.
But what and why?
I don't know: maybe a subtle filesystem bug, maybe an array controller hiccup, maybe coincidental bad sectors on multiple disks in the RAID, maybe cosmic rays, I DON'T KNOW and BTW do you have any idea how much your data is scrambled every day and the computer MAGICALLY FIXES 99.99999% OF THOSE ERRORS WITHOUT YOU EVER KNOWING ABOUT IT?