February 2024 errors with NOAA and rainfall generation

[brookr]

It's not been a good month for Cumulus MX as something is corrupting my data files and I'm unable to fix. I have reproduced my NOAA report for this month that isn't updating at all. I've also reproduced my daily rainfall tables that has a recurring error that is giving spurious large rainfall on almost a daily basis. my Weatherlink total rainfall to date is 74.8mm.

[freddie]

If you zip up and post your MXdiags log files then somebody can take a look at what's occurring.

[brookr]

Thanks Freddie. I'll do that!

[brookr]

Here's my latest MXdiag: Hope someone can see my problems. For me it's all a bit confusing to say the least!

[SamiS]

Your dayfile.txt has an error that might be the reason for report generation to fail.

Code: Select all

 Error at line 1256 of data\dayfile.txt : Input string was not in a correct format.

If that particular line seems to be ok, check if there are some corrupted lines before that. If you can’t see anything obvious, you could post the file here also.

[brookr]

Thanks. It's always difficult to see what's right and wrong for me!
Here's the file in question . Hope it's obvious.

dayfile.txt

[yamen]

I'm not sure if this is the problem and how it may happened but you can try to edit this entry or use this attached already modified dayfile

Code: Select all

01/02/24,38,247,23:20,4.7,00:50,7.9,00:00,1025.36,19:30,1031.80,10:17,0.0,00:00,0.6

error.png

[brookr]

Thanks Sami
I deleted the file and replaced it with your corrected file and hey it works! However, not the end of the story since the gigantic rainfall totals are still amiss and continuing, but at least I can see the blatant errors in the two location of "daily rain" and on the "Noaa report" I like the way you identified the issue. How did you find the error in the dayfile?
Much appreciated.

https://leedsweathercentre.co.uk/noaareports.php

[brookr]

Sorry I see It's not Sami that identified. Thanks for the file regardless!

[SamiS]

I deleted the file and replaced it with your corrected file and hey it works! However, not the end of the story since the gigantic rainfall totals are still amiss and continuing, but at least I can see the blatant errors in the two location of "daily rain" and on the "Noaa report" I like the way you identified the issue. How did you find the error in the dayfile?

Not to take credit of @yamen’s work… But when using a proper editor (Notepad++ on the screenshot) you can see the line numbers. Then you have just to look at the error specified line to see visually what is wrong. The data in your case is comma separated, and with decimal dots, so it is very obvious that a value should not have two dots. It usually helps to spot the differences when comparing with the line above and below. This was very strange error itself, usually the whole line is just garbage if there is corruption.

If you are still having problems with improper rainfall figures, you might want to first check your station settings that the rain correction factor and measuring unit is correctly set. If necessary, you may want to open a new thread for that problem. It may need some debug logging by Mark that developes Cumulus code.

[brookr]

Thanks
No errors with Weatherlink data, or correction factors. It's definitely Cumulus software

[brookr]

Hi
I've now managed to correct the DAy rainfall editing the dayfile log, but once again NOAA rainfall is being fed from another file that is causing the program to produce huge rainfall totals on a daily basis. How can I correct this ongoing problem without causing any further anomalies. This is my only criticism of CumulusMX. Error correcting is a nightmare!

[brookr]

Dayfile corrections work for this report

[philpugh]

Have you tried to regenerate the NOAA report? (Button at top of report page)

[mcrossley]

The report is generated from data in the dayfile.

If you have regenerated the report and still have the problem; how did you edit the dayfile? Using the built-in editor, or manually?