Topic: Raid disk replacement problem

[mark]

a while ago I built a sme 7.2 box with 4 500gb sata disks using the sme nospare option (1.5tb total size).

A disk failed on Sunday and hung the server on reboot. With the dead disk disconnected the server booted ok and I (fortunately) backed up the 1.2 TB of data before shutting down and putting in a 750gb drive to replace the dead one and rebooting. Server came up fine with all data running on 3 disks so I went into admin and added the new disk into the array. I checked /proc/mdstat a couple of time and it seemed to be syncing fine so I left the customer to it.

Later that day the server died with a kernel panic - it seems to have decided one of the existing working disks was a spare and could only find 2 drives in the array. looks like its toasted

what did I do wrong?

[Stefano]

what did I do wrong?

well, you wrote here and not in bugzilla

open a bug please, thank you
Stefano

[mark]

this forum is how you work towards determining weather a problem is a configuration issue or a bug.

As it turns out there may be intermittent problems on another of the disks in the server causing the problem

I don't think that's a bug - do you?

[Stefano]

as Charlie Brady sometime says, you don't have to think if it's a bug or not.. it's not up to you to decide.. so, please, raise a bug.. you will steered in the right direction to solve/understand your problem

Ciao
Stefano