Hello everyone,
I have a Intel RST RAID 1 array composed of 2 physical drives and then
split into a 2-volume Intel Matrix. One physical drive appears to have
gone bad. I am trying to recover the data on another computer without the
RST controller. I am working on copies of the original drives but the bad
drive copy doesn't look usable (its --examine output is very different).
Using only the copy of the good drive, mdadm seems to assemble and run it
without having to force it with --run which looks promising.
# IMSM_NO_PLATFORM=1 mdadm --assemble -e imsm -v /dev/md0 /dev/sdc
mdadm: looking for devices for /dev/md0
mdadm: /dev/sdc is identified as a member of /dev/md0, slot -1.
mdadm: added /dev/sdc to /dev/md0 as -1
mdadm: Container /dev/md0 has been assembled with 1 drive
After this I get stuck trying to assemble the Matrix volumes. When I
--examine the physical drive I see a UUID at the top, and this matches the
UUID of the currently assembled device (according to --detail). There are
two volumes listed below that, each with their own different UUID. I think
I'm supposed to use the --incremental mode to continue adding the volumes,
but this fails:
# IMSM_NO_PLATFORM=1 mdadm --incremental -e imsm -v /dev/md0
mdadm: /dev/md0 is not part of an md array.
I have also tried using --assemble using just those UUID separately in a
number of ways, all unsuccessful.
With only the parent container assembled I have this:
# cat /proc/mdstat
Personalities :
md0 : inactive sdc[0](S)
5201 blocks super external:imsm
Is "inactive" a problem? Adding --run to the --assemble makes no
difference. Is the small block count only for the superblocks?
How can I assemble and access those Intel Matrix volumes?
