me again, sorry.
I have a raidframe raid5 array, I had it originally installed, had a kernel
panic of some sort, don't remember, and turned off the machine. I have had
to clean the disklabels on the raw drives, newfs them, raidctl -C the array,
-I the array, -vP the array, and had a new array. I rebooted and -s to make
sure everything was optimum and clean.
I create a database on the raid with postgresql and then attempt to upload
an .sql file to 'restore' (that is move a copy of my system from a different
box) my database onto this Sun box, and I get a kernel panic that says
'biodone already.'
three questions:
1) is there a command I can put at the ddb> prompt so that I don't lose my
raids?
when I try boot, it says that root wasn't umounted properly and I have to
rebuild raid0 and my SAN array (raid1 - where the database is supposed to
sit). When I try 'continue',
it responds with syncing disk and then reboots, and lose my raids again.
2) is there a command where I can entirely erase my drives so I can start
from scratch? This is a sparc64 build and I can't use fdisk - is newfs
enough?
3) is there any way to prevent this biodone panic? like turning off a
feature or recompiling the kernel?
It takes a whole day to rebuild the SAN array because it's so big, if this
were to happen in a production machine I suspect I'd lose everything.
thanks
