I am thinking about making a meta-script that will 'watch' a number of points, and monitor them for conditions or sets of conditions, and then write these to a 'log file' or 'messages' data point.
We have a number of Mango servers with "a few hundred" data points. We can set alarms for specific conditions, but to quickly-fault find, we need to see the chronological series of events across multiple points.
Having messages logged to a single 'log file' data point has a few advantages that I can see. One is that it is easy to publish this single point (via persistent TCP, etc.) or put in a table on a DGLux dashboard.
Another is that the logic can be more finely tuned; alarms can be raised only if another condition is true too, improving the signal to noise ratio.
One really major benefit would be the ability to chart this 'log' on a DGLux graph. By disabling fill and line, and enabling a marker, a point can be shown and the hover tooltip set to display the log message (the point value). This would enable us or the person investigating to very quickly identify the series of events that led up to a failure.
Our installations use many different pieces of equipment with many different messages, conditions and fault statuses. So consolidating them into a meta script, which would allow us to additionally build in logic and natural language, could be a really powerful thing.
Would be interested to hear others' thoughts on this approach ..