Use smarttools to dump the statistics for the drive and see what has been going on.
Some "green" drives spin down the disk rather quickly upon a lack of use. It should come back automatically however, but perhaps Linux is timing out the drive before it is up to speed. Have a look at the output of "dmesg" after the fault has been experienced. Here's a nice summary of the situation with Western Digital
drives. Some other drives can use hdparm to alter the spindown time.