1 week with MX and fine offset.

[xayide]

So just wanted to give som thoughts after first week. I must say it seems to be stable for me. Run it as a service using NSSM in Windows server 2012R2 X64. It has yet to crash or do anything out of the ordinary. Although it seems to behave differently than 1.9.4 when it comes to "lost transmission" it seems to max out the datatype. For instance when console lost solar value it spiked to Max of 3 bytes (i.e. 24bit value). Something like it happened when I had to remove the raingauge and it said 8890 mm last hour or if it was day. The rain didn't glue to the records while the value for solar rad stays on the yesterday record.

Sorry I do not have anything more specific to bugs or odd behaviours to report yet. It just seems pretty stable, isn't that a good thing after all?

[steve]

Although it seems to behave differently than 1.9.4 when it comes to "lost transmission" it seems to max out the datatype. For instance when console lost solar value it spiked to Max of 3 bytes (i.e. 24bit value).

MX does the same as Cumulus 1 when the 'lost contact' flag is set - it ignores all of the outside data. It logs a message "Sensor contact lost; ignoring outdoor data" when it does that.

C1 also has a check on 'rogue' Lux values (in the case where the lost contact flag is not set) which MX currently does not have; I'll add that to the next build.