Application monitors allow the user to set thresholds for the statistic (number of connections, size of the database in MB, etc.). In addition to the thresholds, the user can specify how many times the threshold must be breached before changing the monitor's status to warning or critical. For instance, the user can specify that warning threshold has to be breached on 3 of the last 4 tests, or has to be breached for at least 5 consecutive tests. This is incredible useful because it means that short-term breaches don't have to trigger an alert.
Solarwinds should add this logic for determining when to change the monitor's status to down. We have many monitors that, for whatever reason, sometimes don't respond fast enough or return an unexpected value. Currently, Solarwinds changes the monitor's status to down on the first failure. With transient errors due to network traffic, etc., the situation has often "corrected" itself by the time someone responds. We'd like to enforce "X out of Y"/"X consecutive" logic so we know that the monitor is really down.
