HP Serviceguard Enterprise Cluster Master Toolkit User Guide, December 2012 (5900-2145)

MIG29
master tcp ether 192.168.10.1 5000
query tcp ether 192.168.10.1 5000
If you require a re-locatable hostname to be associated with this re-locatable IP-address, do the
followingf:
1. Use the re-locatable hostname in the srvbuild.adaptive_server.rs and similar files for all the
other 'server' processes and use srvbuildres utility to configure the servers.
2. Make the (relocatable_,P relocatable_hostname) pairs entry in the /etc/hosts (and also in the
nameserver, if necessary). The /etc/hosts, must contain the <IP-address, hostname> associated
with each ASE instance along with the physical IP-address and corresponding physical
hostname entries.
- /etc/hosts
Ensure that the entries mentioned above are present in the /etc/hosts file before an attempt
is made to bring up the ASE processes.
For example, consider that the IP 192.168.10.1 is the associated IP and sybase0 is the
associated hostname with the ASE instance SYBASE0. The entry required in /etc/hosts is:
192.168.10.1 sybase0
In the adaptive server configuration file, the re-locatable hostname sybase0 must be used in
place of the hostname:
sqlsrv.network_hostname_list: PUT_YOUR_HOSTNAME_HERE
as -
sqlsrv.network_hostname_list: sybase0
NOTE: If you don not want to use a re-locatable hostname with the ASE instance, the /etc/
hosts file on each node must contain the re-locatable IP-address associated with the physical
hostname of that node. For example, if node1_hostname and node2_hostname are the physical
hostnames of two nodes in the cluster, and if 192.168.10.1 is the re-locatable IP-address for
the ASE instance, the /etc/hosts of node1 must have an entry like: 192.168.10.1
node1_hostname
And the /etc/hosts file of node2 have an entry like: 192.168.10.1 node2_hostname.
Database Maintenance
There might be situations, when the Sybase ASE database must to be taken down for maintenance
purposes like changing configuration, without migrating the instance to standby node. To take
down the database:
NOTE: The example assumes that the package name is SYBASE0, the package directory is
/opt/cmcluster/pkg/SYBASE0, and the SYBASE_HOME is configured as /SYBASE0.
1. To disable the failover of the package, run the following command:
$ cmmodpkg -d SYBASE0
2. Pause the monitor script.
Create an empty file <TKIT_DIR>/sybase.debug
$ touch <TKIT_DIR>/sybase.debug
With this command, the toolkit monitors script, that continuously checks the Sybase ASE
process, stops monitoring this process. A message, "Sybase ASE toolkit pausing monitoring
and entering maintenance mode" appears in the Serviceguard Package Control script log.
3. To stop the Sybase ASE database instance, if required, run the following commands::
72 Using the Sybase ASE Toolkit in a Serviceguard Cluster on HP-UX