> I'm running 2.2.12 and raidtools/patches from 19990824.
> I'm trying to create a raid-1 partition with the "failed-disk" feature
> to transfer existing data from one of the partitions.
>
> raiddev /dev/md6
> raid-level 1
> nr-raid-disks 2
> nr-spare-disks 0
> chunk-size 64
> persistent-superblock 1
> device /dev/sdb6
> raid-disk 0
> device /dev/sda6
> raid-disk 1
> failed-disk 1
A lot of people trying to use the failed-disk method, are using these last two
lines in their raidtab. I thought the failed-disk line was instead of the
raid-disk line, not as well as it. But Martin's next message implies that this
works. Can someone clarify whether a raidtab should include both lines, just the
failed-disk line or whether either way is acceptable.
Cheers,
Bruno Prior [EMAIL PROTECTED]
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Martin Lichtin
> Sent: 29 September 1999 17:40
> To: [EMAIL PROTECTED]
> Subject: Problem using "failed-disk" feature
>
>
> I'm running 2.2.12 and raidtools/patches from 19990824.
> I'm trying to create a raid-1 partition with the "failed-disk" feature
> to transfer existing data from one of the partitions.
>
> raiddev /dev/md6
> raid-level 1
> nr-raid-disks 2
> nr-spare-disks 0
> chunk-size 64
> persistent-superblock 1
> device /dev/sdb6
> raid-disk 0
> device /dev/sda6
> raid-disk 1
> failed-disk 1
>
> So I have sda6 mounted (shouldn't be a problem) and try to
> create the raid partition:
>
> % mkraid /dev/md6
> handling MD device /dev/md6
> analyzing super-block
> disk 0: /dev/sdb6, 6964146kB, raid superblock at 6964032kB
> disk 1: /dev/sda6, failed
> mkraid: aborted, see the syslog and /proc/mdstat for potential clues.
>
> Hmm, syslog messages are:
>
> Sep 29 05:10:08 carve kernel: bind<sdb6,1>
> Sep 29 05:10:08 carve kernel: blkdev_open() failed: -19
> Sep 29 05:10:08 carve kernel: md: [dev 00:00] has zero size, marking faulty!
> Sep 29 05:10:08 carve kernel: md: error, md_import_device() returned -22
>
> % cat /proc/mdstat
> Personalities :
> read_ahead not set
> md6 : inactive sdb6[0] 0 blocks
> unused devices: <none>
>
> Any ideas what's wrong here?
