Hi,

While running a backup, I noticed one of the files on my server could not be copied. (Server is running Linux.)

When I try to cat or copy the file to /dev/null, I get the same error, for ex:

ls -l hal.log.061210.gz
-rw-r----- 1 81199433 Dec 10 20:10 hal.log.061210.gz

cat hal.log.061210.gz > /dev/null
cat: hal.log.061210.gz: Input/output error

I presume this indicates I have some bad blocks on the hard drive.

My question is, how should I proceed? Should I create a support ticket on Orbit and ask for a new drive to be installed? (I have two drives in the system, the 2nd drive is a backup of the 1st drive, and I can boot off the 2nd drive.)

If I had physical access to the system, I guess I would:
- boot off backup drive
- replace drive with I/O error
- copy contents of backup drive back to new drive
- boot off new drive

Anyway, figured I'd ask for suggestions in case there are other possibilities...

Thanks,

Bill
tastyspleen.net

Try running smartctl (from the smartmontools package) and hdparm to test drive diagnostics/speed to see if there's anything of concern. It might be aa corruption from a hard reboot or nothing at all.

A good old fashion FSCK wouldn't hurt either.

If the issue persists you should definitely open a ticket about testing and possible drive replacement.

smartctl is nice but can sometimes be wrong. I would say just run a badblocks, which works below the partition level, to see whats going on and if it is having problems.