Neil Brown <[EMAIL PROTECTED]> writes:
> On January 14, [EMAIL PROTECTED] wrote:
> > How I have to start a RAID-10 array after disk failure?
> >
> > [ I try to use raidstart without success ]
> >
> > Why?
>
> Because "raidstart" is overly simplistic. [...]
OK, I've learned that I have to use mkraid in such situations.
I have my root fs on a RAID-10 array. I have a linuxrc script on
initrd that will do the work for the RAID-10 arrays. All work fine.
But in the above situation of failure I have to use mkraid. Now I have
written a new version of the script that detect the failure situation
(it looks in /proc/mdstat). If one RAID-0 array is not up (eg. it is
not auto detected and started by kernel) the script makes a branch,
writes a new raidtab (with marked the missing RAID-0 array as failure)
and call mkraid.
linuxrc call
mkraid -f /dev/md010
mkraid say's:
disk 0: /dev/md100, failed
disk 1: /dev/md101, ...kB, raid superblock at ...kB
No error message, but it doesn't work! After that there is no md010
reported in /proc/mdstat.
If I boot the machine from an other disk and call the same procedure,
all is right and md010 exist after that.
Why it doesn't work from linuxrc?
Peter
If it does matter: I'm using 2.4.0-ac11.
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to [EMAIL PROTECTED]
