> Hi!
>
> raid5 oopses at me at time of mdrun. Oops is in do_md_run:
could you try the newest version of the RAID driver? It's at
http://www.kernel.org/pub/linux/daemons/raid/alpha/. (it's against
2.1.125 currently, a 2.1.131 patch will be released soon) I have not had a
RAID related oops for a long time.
> (this is ckraid.c)
ckraid is obsolete, try the newest raidtools.
> Looks to me like you don't check error code from read, which is bad
the reconstruction code in the kernel looks for errors.
> idea in itself. It is even worse idea with nbd, as nbd returns -EIO on
> end of device. Is it for ckraid to read only area it needs to?
does nbd report correct blocksize and device size? If yes then it should
not happen that we try to read past the end of the device.
> PS: I'm sorry for not posting patch for ckraid, but I kind of do not
> dare to touch threaded code I do not understand.
it's gone, so never mind. Kernel based reconstruction is far more
superior.
-- mingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/