Typical scenario: I made a raid 1 array out of 2 disks, that contains data I wouldn't like to lose.

I found that average life-time of HDD (rotating drive) is 3 - 5 years.

Does it mean that if the array contained disks that are older than 3 years I should, if I wanted to keep the data on reliable storage, add 1 new HDD into array just to make sure? Or should I rather get 2 new disks and completely decommission previous drives? Even if they still work?

I was always thinking that having raid 1 consisting of 2 drives could work until one of the 2 drives fail, and then I would just replace the failed one, but considering that old disk may not be reliable anymore, I am afraid there is a possibility that if one of disks fail, upcoming sync with new drive, which would put heavy load on last remaining disk in array, could kill it?

Also, from my personal experience, I had few disks that worked fine even for 10 years, and saw brand new disks that failed after few months. So at some point, I still believe that having old disks shouldn't be so much of a big deal. But maybe I am wrong?

migrated from unix.stackexchange.com Apr 5 '15 at 19:10

This question came from our site for users of Linux, FreeBSD and other Un*x-like operating systems.

  • Why to SU and not to SF? – Manuel Faux Apr 5 '15 at 19:18
  • It depends on the value of the data, but I would rather put the new disk as a backup than a raid member if I had to buy one disk – user2813274 Apr 5 '15 at 20:36
  • @ManuelFaux SF is what? – Petr Apr 7 '15 at 7:12
  • @Petr: see Server Fault/Super User... – Manuel Faux Apr 7 '15 at 9:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.