From: Jenny Butler (jbutler@utmem.edu)
Date: Thu Sep 12 2002 - 15:42:12 EDT
I have an TruCluster v5.1A configuration, which consists of two nodes
and the
associated disk configuration. In the beginning, at version 5.1 I installed
software
on a single Tru64 Unix disk, with patches and layered software. Once I did
the
cluster creation and got things into a production mode, the original system
disk
has not been touched even though it is still configured. I have since then
upgraded
the cluster (worked pretty well) and changed up the layered software some.
My big question is whether to keep that original system disk in my
storage
rack, to pull it out (doing the appropriate steps to preserve the contents),
or just
reuse it. If I had to recreate the cluster from scratch, I would have to
rebuild
that disk or upgrade it. Has anyone else come to this point and figured out
what
benefit keeping the original would give versus just starting over if it got
that bad.
Some documentation indicated being able to go back and boot from it if
the
cluster got hammered. If it's that bad, I'd probably boot a distribution
CDROM.
Opinions and experiences welcome. Email to me and I will summarize.
Thanks, Jenny Butler
jbutler@utmem.edu
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:48:53 EDT