deviation data

[Gerrit]

Hello.

I have quite a difference in data representation, and do not know where it is to

Monthly rainfall, this is true, correct and gives the rainfall for February weather
https://www.meteoplug.com/cgi-bin/meteo ... 58425d4b6a

Spring rain, the sum of January and February is far too low
https://www.meteoplug.com/cgi-bin/meteo ... f3ebeaf2d1


Monthly temperature, this is true.
https://www.meteoplug.com/cgi-bin/meteo ... bfa7a6b695

Annual temperature shows in February to 5.4 degrees, which should be -0.1
https://www.meteoplug.com/cgi-bin/meteo ... 041e190f2e

Regards, Gerrit

[oliverb]

I also have error. Meteoplug is showing 12mm of rain for 1/04/2010. https://www.meteoplug.com/cgi-bin/meteo ... 3b2b353735
my Meteohub is showing 0. There is also no rain for 29/03 when my Meteohub recorded 10mm. http://home.btconnect.com/brettoliver/rainweek.png

[admin]

Gerrit, I just initiated a recomputation of your aggregated data. When this fixes the issue I have to find out, why incremental aggregation has not worked as expected. If deviation remains, I will have to check computation functions inside aggregation routines. Please stay tuned

[Gerrit]

Hello

There are always exceptions, though temperatures in both rain beat many months.
At 01-04 gives meteohub 6mm rainfall and meteoplug 2.4mm , it should be 6.0.

Regards, Gerrit

[admin]

I found a well hidden bug in generation of aggregated data. Hopefully, this will fix deviations you have discovered.

I started a recomputation of data of all activated users. This will for sure take a day or two until finished.

I will drop a note here, when done, so that you can check if inconsistencies are gone by that, or if I do have to invest a second look.

Thanks for your patience.

[oliverb]

I have done a quick check with my charts from Meteohub and Meteoplug (see my previous comments in this thread) and they now match.

I understand that in may take a couple of days to complete the recomputation of data but it is looking good so far.

Thanks.
Brett.

[admin]

jobs are queued one after the other... looks our one has already run through. 2/3 are done by now.

[Gerrit]

Hello.

All looks good, I'm still 1 little mistake in the annual temperature shows only 9 days in January to 4 days 5 days max mac 0.0 and 5.0, the monthly temperature is okay -2.1
I can not find more different.

https://www.meteoplug.com/cgi-bin/meteo ... 041e190f2e

Regards, Gerrit

[admin]

Sensor "th0" does have a large gap in data for Jan 2010:

2010-01-01 00:56:32 th0 -0.7°C 86% -2.7°C
2010-01-01 00:57:34 th0 -0.7°C 86% -2.7°C
2010-01-01 00:58:27 th0 -0.8°C 86% -2.8°C
2010-01-01 00:59:28 th0 -0.8°C 86% -2.8°C
2010-01-24 16:41:11 th0 -1.2°C 93% -2.2°C
2010-01-24 16:42:13 th0 -1.2°C 93% -2.2°C
2010-01-24 16:43:15 th0 -1.2°C 93% -2.2°C
2010-01-24 16:44:17 th0 -1.2°C 93% -2.2°C
2010-01-24 16:45:19 th0 -1.2°C 93% -2.2°C

Maybe your upload of weather data did fail during that period? When you have Meteohub data for January, feel free to upload.
While recomputation is scheduled or running for your data you get a message in web interface indicating that. When message has gone, you might try to draw the graph again, which will hopefully come up correctly this time.

[Gerrit]

Hello.

Yet there remain errors.
I got up again in April, with rain on April 3 this gave 9.6 mm was 10.2 now correct it.
January, continues to give errors, even after the New upload, so you see the rainy day on January 24 are 27.8 mm, which is not true, there's the fallout of weeks in there, see the NOAA report.

http://www.natuurrondleidingen.nl/noaaview.wmh.html
https://www.meteoplug.com/cgi-bin/meteo ... 58425d4b6a

Regards, Gerrit