|
Configuring Monitoring Studio to monitor your application strongly depends on the application itself. Nevertheless, certain basic guidelines can be followed:
1. Specify the resources used by your application:
| • | Directories (folders) where your application runs (where transactional data is stored, where queries are queued, etc.) |
| • | Files that are critical to your application (to check their presence, size, growth, etc.) as well as their security settings |
| • | Other middleware information provided by other KMs if your application relies on some middleware software that can be monitored by specific KMs (Oracle, Tuxedo, MSMQ, etc.) |
| • | Other middleware information provided by Windows performance counters if your application relies on middleware software under Windows and provides perfmon data |
2. Specify the information sources (where or how can you get information about the application):
| • | Flat files that contain useful information about your application |
| • | LOG files in which your application writes data (traces of the operations) |
| • | Web pages (your application provides a web front-end that should be tested) |
| • | Database queries (your application relies on a database server and some tables contain useful data or procedures that should be tested) |
| • | SNMP agents (your application provides information through an SNMP agent that can be polled or that sends SNMP traps) |
| • | JMX-enabled servers that you want to poll and monitor |
3. Specify what should be searched for in these information sources:
| • | Strings (regular expressions) that are to be found or not to be found |
| • | Numeric values that will be used to build a graph in the console (with alert thresholds) |
4. Specify what Monitoring Studio should do when an application failure has been detected:
| • | Decide what kind of Alert Actions are to be taken when an error is detected. |
|
