HDD GURU FORUMS

They mean the drive is failing time to get the data backed up and replace the drive. Not something you can "fix" and "they don't make them like they used to" drives seem to fail sooner due to the size increase. Backups always have good backups !

So basically, if anything doesn't have a check, it's time to get a new hard drive?
Thanks!

@Fimuxu77,

As networks says, always have backups. In answer to your question:

Fimuxu77 wrote:So basically, if anything doesn't have a check, it's time to get a new hard drive?

In my experience, it is not that black & white - it is a judgement call for you to make, depending on your attitude to risk, your budget, the rate of change of some of those parameters etc. Of course the lowest risk option is to replace a disk as soon as there is any indication of any problem - but that's also the highest cost option. Your data, your risk, your choice.

My experience with disk manufacturers is also that simply having small, non-zero values for things like pending or reallocated sectors, is typically not "bad enough" for them to agree to RMA a drive which is in-warranty. You didn't supply the full raw, SMART data, so it's not worth discussing too much hypothetically.

Just to add a couple more points:

- You cannot rely on getting a clear problem indication from the SMART data, before a disk fails catastrophically - hence why you need backups.

- You can read more about the meaning of the SMART attributes in many places on the web, here is one place:

http://en.wikipedia.org/wiki/S.M.A.R.T.

- As you will see, the last of the attributes that you mentioned (UltraDMA CRC errors, attribute 199) is related to the interface, and the cause of that being non-zero is usually external to the drive, in my experience.