This is a bit esoteric but for those who use the DSC toolkit to read trace data from a Citadel DB you will find that you cannot always obtain the most current value
of a trace. Here is the text describing what a ghost point is:
Citadel uses ghost points to maintain real-time awareness of traces that do not change frequently.
For example, suppose you configure a thermocouple measurement for
logging with a deadband of 1°F. The thermocouple is in an environment
that remains at a constant temperature most of the time, but you need
to know when any major changes occur. If the temperature remains within
1°F for one hour, Citadel does not log redundant points, even if the
temperature is sampled once every five seconds. However, if the logging
system experiences a failure at some point during the sampling, you
want to know the temperature at the time of the system failure. You
also want to know the approximate time of the failure. Citadel handles
this situation by logging a single ghost point for each trace. The
ghost point in each trace updates constantly at a rate based on the
time group specified for the trace. The ghost point always reflects the
most recent time and value of a trace.
Citadel stores ghost points in a .tgpffile that updates once every
ten seconds. This file ensures that Citadel has an accurate record of
the value for every trace at the time of a system failure.
It would be very useful to be able to specify in the trace read vi that you would like it to return the ghost data points. Otherwise, it is very cumbersome to try to detect if you are missing this data point and then have to programmatically read it yourself from the originating data system.
