If you are setting up High Availability (HA) for DB2, the
hadb2eee.setup program completed this step for you.
See Appendix C. "Implementing Failover Support in the Solaris Operating Environment" for more information.
To make DB2 Extended Enterprise Edition accessible to users, you must ensure that the instance owner ID can execute any command using the rsh command on all nodes defined in the db2nodes.cfg file. To do this, you can either create an INSTHOME/.rhosts file or an /etc/hosts.equiv file.
If you want to use the autoloader command, you must also include the netname values for each machine. These values are specified in the fourth column of the db2nodes.cfg file.
For example, the INSTHOME/.rhosts file may look like:
machine1.torolab.ibm.com db2inst1 machine2.torolab.ibm.com db2inst1 machine3.torolab.ibm.com db2inst1 switch01.torolab.ibm.com db2inst1 switch02.torolab.ibm.com db2inst1 switch03.torolab.ibm.com db2inst1
For more information about the INSTHOME/.rhosts or /etc/hosts.equiv files, refer to your operating system's reference manual.
You should also ensure that the IP address of every machine can be resolved
for the names in the /etc/hosts and the
/etc/resolv.conf files. These files are also
described in your operating system's reference manual.
The system database directory contains information about all defined
databases. You must ensure that all nodes always access the same
system database directory file,
INSTHOME/sqllib/sqldbdir/sqldbdir. Unpredictable errors can
occur if either the system database directory or the system intention
file (INSTHOME/sqllib/sqldbdir/sqldbins) is a symbolic link
to another file. All database partition servers need to access the same
copy of these files.
[ DB2 List of Books | Search the DB2 Books ]