Chkdsk /r - How often?

[mtp]

Hi,

I recently had a hard disk go (WD Green 2TB) after about 3 years of 24/7 use. The disk that failed was used to store music and videos - no apps. There were no symptoms apart from my http software slowing down - which I attributed to network speeds and an ever increasing media library size. Then one day, I noticed that copying a large file to the drive was going very very slowly. The disk was only 60% full so that should not be an issue ... I did a bunch of network trouble shooting... to no avail... Finally I decided to run a chkdsk /r on the disk.

It found lots of bad sectors.... (fortunately everything was backed up) I replaced the disk and everything speeded up dramatically...

My question is... how often should I run chkdsk to discover disks that are failing? Especially considering chkdsk takes a long time to run (8 hrs for a 2 TB disk) and stresses the disk.

 

[nommy the first]

Hello mtp,

In my opinion there really isn't a "fixed time" for when to run it. It's just like the SFC /SCANNOW command, or even an anti virus scan - you run it either when you feel the need or when you feel that something is wrong (as in the situation that you described).

I do however recommend running it at least once a month so that you are on the safe side and won't get any surprises.

More info:
How to run Disk Check in Windows 7
How to read Event Viewer Logs for Disk Check


Good luck,
Nommy

 

[kado897]

Alternatively, and much faster, display the s.m.a.r.t. data using Crystaldiskinfo.

 

[Golden]

Agree. CrystalDiskInfo is quicker....just watch out for unwanted software during the installation. ...check those check boxes carefully

 

[kado897]

I use the PortableApps version which is free from nasties.

 

[mtp]

Alternatively, and much faster, display the s.m.a.r.t. data using Crystaldiskinfo.

Thanks for the suggestion - I actually used crystal disk info to look at the smart data before i figured out what the problem was - the smart data showed the drive as healthy. The drive also passed WDs quick test.

It was not until I ran the chkdsk /r that the bad sectors were identified and the bad sector count in smart were updated. Perhaps I should have said this in the original post, but i had been looking at the smart data on a fairly regular basis to monitor the health of my disks.

It seems that regular reads and writes that fail are not enough to trigger a sector becoming marked as bad. On an empirical basis, only format, chkdsk /r and the specialised utilities do it...

Thanks

Mark

edit: After running the chkdsk, smart correctly showed the problem

 

[kado897]

That's good to know.