Many thanks for the following who replied immediately:
Adel K. F. Tadros
Boris Berkovich
Dedi Buditriyanto
Corinne Haesaerts
Mohd Nayle
They all agreed on using only one service to re-locate the disks and start
oracle on the second machine.
This can be done by adding start/stop commands to the start/stop action
scripts of the disk service.
In this case there will be no need for the oracle user defined service to do
this job.
This method worked very fine and oracle fails over correctly a long with the
disk service. Thanks a lot again.
----------------------------------------------------------------------------
----------------------------------------------------------------------------
-----
Original message starts here:
----------------------------------------------------------------------------
----------------------------------------------------------------------------
-----
Hello Managers,
I recently installed two ES40 systems as an Available Server Cluster
members, sharing disks
on an RA3000 storage works. Oracle Database was installed on the
shared disks, and it runs
fine on any of the two systems. I defined two services on the
cluster. One for the disks, and the
other is a user defined service to fail over the application. I
created simple action scripts to start
and stop oracle when it fails over. The two services are named disk1
and ora_serv. When one
system goes down, the services fail over to the second system
correctly, but when we check
oracle processes on that system, it's not there. In other words,
oracle did not start up on that
system. If I manually restart ora_serv from the asemgr utility,
oracle starts up fine. Only automatic
fail over doesn't work.!
Does anybody have a clue?! Your help is kindly appreciated.
Best Regards
Yousef Nour
Tel : +966 1 2180909 Ext 3829
Fax : +966 1 2180800
E-mail : [mailto:yousef.nour_at_digital-natcom.com]
Received on Wed Sep 27 2000 - 17:51:54 NZST