Quirky Temperature during Daylight Saving End

[Super-T]

Over the period of the change to normal time from Summer time at 2am last night, the Temperature data became very erratic.
I recovered back to a couple of days before and the data was OK suggesting that the problem was not the 1081 Console but something in Cumulus.
Pics are before and after data recovery.
Not really a problem as easily fixed.
Interesting that the file sizes are very different. I must have a closer look later to see what is different.

[steve]

It's because there are two sets of data with timestamps between 0200 and 0300. Cumulus doesn't do anything special for daylight savings; I'm not sure what it could do, anyway. I thought everyone understood and accepted this?

By 'rewinding', you've forced Cumulus to timestamp everything according to the current time setting, so although it looks 'better', all of your data from the point you rewound to up to when the clock changed is now timestamped incorrectly.

The new file is smaller because the data is at the logger interval instead of your 1-minute Cumulus interval.

[Super-T]

Old age mate. Makes you forget what happened last year.
I may do a rewind back to show that original data....
I guess I could just erase 1 set of data by recovering back to 2am Sunday but it may be just a waste of time anyway.