We are currently receiving a number of false positive alerts associated to application monitoring which we can't seem to isolate the root cause. The alerts are configured to send out a notification if a component status is down. However, we are also receiving alerts when the component status are up as well. In fact, on the notification email, we've added the following fields in the email body and they are listed as such:
- Components with problems: (empty)
- All monitored components: component1 (Up), component2 (Up)
These components can be port monitors and/or windows services statuses.
Initially, we suspected that this may be due to the polling interval and alert evaluation and escalation intervals; in which, the polling interval is set for every 3 minutes while the alert evaluation is done every 1 minute and the escalation is done every 10 minutes. We ended up changing the escalation interval to 9 minutes to line it up with the polling interval. However, we are still receiving these false positive alerts.
Any suggestions?