Ok, after tinkering somewhat, the problem was that each system could only
see one boot disk.
Once I'd relabeled the disks (disklabel -z) they could see them once booted,
and I simply installed TruCluster on top, and then booted the other node,
which thought it was the first one, then added the member. I did have to
remove the old file domains which were stale, though.
If anyone has any questions about strange installations, I could probably
answer them all.
Thanks for all the help.
Chewy
I've got a cluster that's basically totally broken. I've decided to
re-install, after trying to patch it up, and remove and add members, and
re-configure services, and things like. It's just not happening.
So, the disks are off an HSG, and when I install Tru645.0A (I know I should
upgrade) on a disk, and then try to configure the other disks, it starts to
complain. It says 'The requested disk is either not read access enabled, or
busy. This is all the information that is available'.
So, I need to know an easy way to make the disks available to the new
machine. I can still boot up an old cluster member, and mark all the disks
as unavailable, but I don't know if this will wreck the system, or simply
not work - I suspect I wouldn't be able to do it to the root partitions, for
example.
My question is this:
How do I free up the disks again?
Received on Thu Apr 05 2001 - 14:45:02 NZST