On Nov 8, 2005, at 02:55 , Neil Brown wrote:
<snip>
Can you find that ramdisk (probably in /boot/initrd* somewhere), mount
it, (-t cramfs -o loop) and look inside it.
Look particularly for /etc/mdadm.conf. but look around for anything
else that might be related to md or raid.
... actually, you said which distribution you were using (Debian) so I
should be able to figure it out from that... Hm.. mkinitrd in pretty
broken in debian wrt mdadm :-(
You should find some script called 'md-1script' or something like that
which has something like
mdadm -A /dev/md1 -R -u 4b22b17d:06048bd3:ecec156c:31fabbaf /
dev/hda3 /dev..
This is probably broken.
Ah. In the file etc/script I found the line
mdadm -A /dev/md1 -R -u 4b22b17d:06048bd3:ecec156c:31fabbaf /dev/
hda3 /dev/hdg2
-wouldn't the best thing just be to add /dev/hdc3 there?
I'm terribly sorry, but it seems I've been wasting your time on
a) a Debian-specific question and
b) a problem I've caused myself.
You see, originally I was using a 2.4.17 kernel when the old drive
crashed, but when I tried resyncing, the system froze.
Therefore I upgraded to the 2.6.12 kernel which had no problems
syncing...but of course only hda3 and hdg2 were active at the time of
installation, so the last drive didn't make it into the initrd image.
However, you've been incredibly helpful. You've solved my problem,
thanks!
Troels
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
