Filtering Event Reports
Feeling the need to control the flow of events? Dealing with too many I/O transitions? Flooding monitoring stations? Then the techniques shown here will bring some relief.
To establish a baseline, lets look at an arbitrary sequence of I/O events. As can be seen, all recognized transitions are reported 1:1 to the monitoring station(s). No suppression or event filtering takes place. This places a large burden on the central monitoring station, as it needs to cope with unfiltered events streaming in from a potentially large number of remote sites.
Sometimes it may become necessary to simply ignore the transitions of a particular I/O. Temporary manual suppression of this type is supported using the IGNORE flag. The user controls the point in time when the ignore function takes place and how long it will last. Most often used during maintenance periods or when site alterations take place.
Ignored transitions are shown in the I/O dashboard, but will not be reported to the central monitoring station(s).
Squelching an I/O is one step above using manual ignore. Here the user determines when the I/O reporting should be suppressed and for how long. RIO xR3 will start a SQUELCH timer, and once the timer runs out automatically re-enables the I/O reporting. This relieves the user from having to remember to re-enable the I/O reporting at the end of the maintenance/upgrade procedure and ensures an orderly return to regular I/O monitoring.
Squelched transitions are shown in the I/O dashboard, but will not be reported to the central monitoring station(s).
Some I/O events only last for very brief periods of time, and may thus be difficult to catch or trace. Use the LATCH flag to help visualize even short I/O events. Once an I/O transitions into an ALARM state (no longer NORMAL or INFO), RIO xR3 will automatically latch and hold the I/O at the specified level (any level above INFO like WARNING, MINORx, MAJORx, CRITICALx, ERROR or FAIL) until manually cleared by a user.
I/O transitions occuring during a latch phase are shown in the I/O dashboard, but will not be reported to the central monitoring station(s).
Adaptive Alarm Suppression (AAS)
Enable AAS to guard against excessive I/O event reports during normal operations. An I/O is automatically placed in alarm suppression mode if it exceeds a preset number of transitions within a guard period.
The guard period is implemented as a moving window which activates with the first alarm raised by the corresponding I/O. Once activated, the RIO xR3 units start counting the number of I/O transitions detected during the guard period. The following actions will be taken should the count exceeds the preset number of allowed transitions:
- the guard period ends,
- the I/O will be placed in alarm suppression mode,
- an alarm suppression notification is sent, and
- a suppression period timer will be started.
Notifications are issued for each I/O placed in alarm suppression mode. RIO xR3 units keep a tally of all suppressed transition notifications until the end of the suppression period. At the end of the suppression period, the RIO xR3 units release the I/O from the alarm suppression mode and issue an alarm suppression release notification.
Safeguards have been implemented to improve the resiliency of the I/Os in noisy environments and to guard against contact degradation. See the user guide for more details.
In addition to the methods shown above, RIO xR3 units offer the ability to throttle event notifications. THROTTLING will only be applied if the total number of event notifications to be sent in one reporting interval (1 minute) exceeds a user defined maximum. Throttled notifications are instead held over until the next reporting interval.
As can be seen in the diagram, events report normally until a user defined maximum number of event notifications per reporting interval has been reached. This enforces a strict upper limit to notifications generated by each RIO xR3 during each interval, regardless of the source or urgency of the notification.
RIO xR3 units keep all pending event notifications in a sorted list, with the highest severity notifications receiving the highest priority. This prevents low priority events from using up the available notifications budget.