Monitoring is collecting historic data points that you then use to generate alerts (for any anomalies), to optimize databases and lastly to be proactive in case something goes wrong (for example, a failing database).
There are five key reasons to monitor FEP database.
It is a very simple equation that if you do not have a database in running, your application will not work. If the application is critical, it directly effects on users and the organization.
Monitoring helps to identify the system bottlenecks and according to the user can make changes to your system to see if it resolves the problem or not. To put this into perspective, there may be a situation where users see a very high load on the system. And figured out that there is a host parameter that can be set to a better value.
Proactive monitoring can help you to identify future performance problems. From the database side, it could be related to bloating, slow running queries, table and index statistics, or the vacuum being unable to catch up.
Every database user has a different need and priority. Knowing the system (load, user activity, etc.) helps you to prioritize customer tasks, reporting, or downtime. Monitoring helps to make business process improvement.
More user or application growth means more system resources. It leads to key questions: Do you need more disk space? Do you need a new read replica? Do you need to scale your database system vertically? Monitoring helps you to understand your current system utilization—and if you have data, points spread over a few weeks or months, it helps to forecast system scaling needs.
This article describes monitoring and alerting operations using OpenShift's standard Pod alive monitoring, resource monitoring and database statistics provided by the FEP Exporter.
