Hi
we have a Trucluster 1.6 availability server running
on two AlphaServer4100 servers connected via the
memory channel interconnect. The Alpha servers are
running Tru64 UNIX 4.0f.
ASE is used for availability of nfs services to
various clients over the LAN. This config was being in
use (non-stop) for the last two years and we've had no
problem.
But today, both servers crashed with the following
error
"panic cpu(0) alloc_mcell, read page N1 instead of N2
N1=974032, N2 = 279"
Both the systems kept crashing on repeated attempts of
booting at a certain stage.
By trial and error (renaming of S* files in
/sbin/rc3.d), we found, without S00aseam and
S14asemember, the system comes up. But this wouldn't
bring the cluster up and hence the nfs services
wouldn't be available.
Has anyone faced such a situation.
Please clarify as early as possible...as the situation
is very critical. .
thanks in advance
____________________________________________________________
Do You Yahoo!?
Get your free _at_yahoo.co.uk address at
http://mail.yahoo.co.uk
or your free _at_yahoo.ie address at
http://mail.yahoo.ie
Received on Wed Apr 18 2001 - 01:59:28 NZST