• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

RAID 1 and bad sectors

S

Sean Gagnon

n00b
Joined
Dec 5, 2013
Messages
2
Hello,

got something i am not sure about, and looking for an answer.

the reason i scanned it, was because of weird "going-ons" and i am trying to rule out what is causing it. i was kind of leaning to possible MOBO starting to flake out.

I have a raid 1 mirror on a system. it appears to have bad sectors on one of two drives according to the dell diagnostic. The Raid controller never showed the raid as degraded. so my question is:
will the Raid, while mirroring, just mark the sectors bad on on both drives and continue to work around it, or should it have shown me a possible degraded status?

feel like i knew this answer before, but i am blank, and Google ain't helping me today :(
 
In good raid software, in my case Linux md, it works like this:

one drive can't read a block and gives up, reports an error up to the OS. The raid subsystem reconstructs the contents of the block from the good drive(s) and overwrites the faulty position on the bad disk with the intended contents. This will then cause the drive to reallocate that block and turn it into a good block. The raid never goes into degraded mode from this, you will only have the error in the kernel syslog and in SMART.

Really bad things happen if the RAID is already degraded by the time this occurs on one of the drives still in the active array. Because now the RAID software cannot reconstruct the intended contents of the block. In that case md will take the array offline (but it can be force-reassembled later after you did something about the bad-read block manually, since nothing was written) and ZFS will report errors further upstream, .e.g. to the application or whatever else sits on top of the affected file (Linux md can't do that since it can't signal through the layers above it, e.g. the filesystem).

The latter is why it is so important to scrub the array while it is not degraded.
 
What uOpt said.

Simplifying some of it:
As long as there are free sectors available to copy data to when sectors go bad, the drive should function, all drives have reserved space for bad sectors.
When the drive runs out of reserved sectors to replace the bad ones, you can lose data/drive integrity.

An increasing amount of bad sectors is not a good sign.
It might be a one off, but it could be signalling imminent failure/problems in the not too distant future.
 
You must log in or register to reply here.
Back
Top