The pump activity report gives insane results if there is any gap in the historian. Such as the gap that happens when the power is lost to the server.

The pump activity report actually goes into the recorded historical data and adds up the number of starts and the time the tag was high to calculate the run time. This means that it relies on the communication to the remote to be accurate.

For example, if you are polling a device once an hour and,. on that particular poll, the pump was on, then VTScada will record the pump as running for the hour regardless of what happened between polls. If the pump goes on and off three times in that hour, then VTScada will not record those.

A similar thing will occur if VTScada is actually off line or it cannot communicate with the device.

This all changes if you are using a device that logs historical data such as DNP3. Then the historical data should be collected on the next good connection and the activity report should be valid.

Having said that, I am not sure what you mean by insane. Are the values not there or just not correct? What type of device are you using? The pump activity report only works with pump status or digital status tags so what tags are you using it against?

It might be better to send us an email with the report and maybe a changeset so we can have a look?

Just to clear this one up it was the daily total report on the runtime accumulator tags in the remote [cellular] PLC's.
Problem was solved with the HOLD DATA toggle in the communication driver.

I'm considering using a midnight reset on the accumulators and wonder how that will respond if the VTscada and PLC's clock don't exactly match.