Interpreting SMART output

[Alex <mysqlstudent@xxxxxxxxx>]

Hi all,

I was using rsync to copy a very large file (95GB kvm Linux host
image) on the same filesystem to make a backup, and it was taking a
much longer time than it should have. The system was otherwise
completely idle. The time, as shown from --progress, would fluctuate
from 50 minutes to three hours, and never really settle on a steady
time. It would copy between 8MB/s and 30MB/s during the copy. The kvm
image was on a RAID5 filesystem with three disks, all the same.

iostat would show it was reading about 20MB/s then would go to nearly
0, then back to 20MB/s, and also would show that the transfer speed
would fluctuate.

Is this to be expected? It got me thinking that the disks are a few
years old now, and perhaps are on the verge of failure? I'm not sure I
can properly interpret the SMART output, though. It passed a short
test, but the disks have 18439 hours on them!

Here is the SMART output for one of the disks:
http://pastebin.com/vnnAvbrD

It appears to show quite a few errors, but none of them have exceeded
the threshold:
  7 Seek_Error_Rate         0x000f   085   060   030    Pre-fail
Always       -       354385893

Does hardware ECC affect performance?
195 Hardware_ECC_Recovered  0x001a   071   053   000    Old_age
Always       -       209030577

Thanks for any ideas.
Best,
Alex
--
To unsubscribe from this list: send the line "unsubscribe linux-admin" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html