Unreliable results when combining Solar and Eagle smart meter data

Problems
#1

I’ve been trying to get a nice combination graph to show my generation and consumption using a rainforest Eagle device (and also tried fetching the data via Wattvision also), but I’m finding that the data often doesn’t make it to the graph.

So what I’ve done is create a couple of different ‘systems’ under my profile.

The first is only the solar generation, fed by the Oxley solar Android app, which is quite reliable.
https://pvoutput.org/intraday.jsp?id=42066&sid=38406

Second graph is just my smart meter data, which is fetched from Rainforest. I restart my Eagle device nightly, and it now seems generally reliable.
https://pvoutput.org/intraday.jsp?id=42066&sid=41738

The third graph is a combination of the two, where I also receive a duplicate stream of data from both sources. Very often I find that I’ll lose half a day, or bunches of data from either the Solar generation or smart meter consumption data.
https://pvoutput.org/intraday.jsp?id=42066&sid=42424

I’ve tried all sorts of combinations, like trying to set one graph as a parent of another and aggregating the data that way, but I get a similar result of missing data.

Any help would be greatly appreciated here. Thanks!

#2

There are requests to add data directly to the parent 42424 which will overwrite the results of the aggregation.

This should be disabled and only have children data from 38406 and 41738 aggregated.

#3

Thanks. I’ve created a new aggregation system and set individual systems for the eagle data and the solar generation to to use the new one as a parent. Will see how that goes.

Thanks

#4

@bankstownbloke So I’ve reconfigured my systems. I’m now using
https://pvoutput.org/intraday.jsp?id=42066&sid=42525
which has both the Solar generation data fed in, and the Eagle configured. Earlier today it appeared to be doing the right thing, but I’ve just checked now, and the smart meter data has now disappeared. Any suggestions? I just can’t seem to get the right magic.