Knowledge Base

Monitoring System Latency

OneRain 2019-11-25 21:35:59
Popular1 Views97

We typically recommend two different types of rules for system latency. The first uses the time_since_last_report_received function to Alarm if an entire system ceases to receive data. Although we recommend this for ALERT2 systems as well, this is especially useful for data that is processed from a single port at a time, such as ALERT, USGS, METAR, and more. An example of this type of alarm would be time_since_last_report_received_system_key(SYSTEM_KEY) >= MINUTES, where SYSTEM_KEY is the System Key, and MINUTES is the number of minutes since last the time of the last Reading.

The second type of rule uses the time_since_process_status function to alarm if a specific data feed is interrupted. We recommend this grouping for ALERT2 systems that use two or more ports to send data to Contrail. There should be a rule written for each active port associated with the system. An example of this type of alarm would be time_since_process_status(CATEGORY, ID) >= MINUTES, where MINUTES is the number of minutes since the last process status update. Process status infomration is available to licensed Contrail owners, under Configuration > Settings > Status. Examples of Service processes include ALERT2 ASCII, USGS, and Alarms. For these examples, the CATEGORY is the Category specified for each given process (i.e. alert2-port-60012, USGS, CheckAlarms). The ID is the Service key specified for each defined Service. This can be found in Configuration > Settings > Services, then select a Service of interest (i.e. ALERT2 ASCII) to find the key information.

Was this article helpful?

Contrail® Help © 2004 - 2020