Admin ServiceOptimizer System Configuration Distributed System Configuration
- Matthew Wren
On This Page:
Database Configuration
It is recommended that your ServiceOptimizer instance be run in non-distributed mode initially, to confirm that the basic server installation and configuration is correct.
Admin ServiceOptimizer System Configuration Example Configurations has a worked example showing how to migrate from a non-distributed ServiceOptimizer instance to a distributed ServiceOptimizer instance. It is recommended that initially a single ServiceOptimizer process is configured to run on a different node: once this has been verified then other processes can be reconfigured as required.
Tolerating Staggered Startup
If all the computers in a distributed ServiceOptimizer instance are rebooted, they are likely to become available at different times.
The System Monitors can be configured to wait for the ServiceScheduling database server to be available. System monitors should be started with “-r” and “-t” parameters set to suitable values (see Admin ServiceOptimizer Further Detail Core Processes System Monitor).
The Master System Monitor can be configured to wait for the Slave System Monitors to become available using the Database sp083_system_parameters values mon_slave_conn_attempt_limit and mon_slave_connect_retry_delay (see Admin ServiceOptimizer Further Detail Setting up Auto-Recovery of Components).
Application Logfiles
It is recommended that a central archive location be configured for the ServiceOptimizer application log files. This makes it easier to analyse and support a distributed ServiceOptimizer instance. See section Admin ServiceOptimizer System Configuration Application Logfiles for further details.