v3.10.0 - NOAA Year report does not generate

[mcrossley]

I have found the problem, the name of the data item plural vs singular

If you want to try it, then grab the cumuluscharts.js from my test site page: https://weather.wilmslowastro.com/test/newMX/trends.htm

[mdbaines1]

I have found the problem, the name of the data item plural vs singular

If you want to try it, then grab the cumuluscharts.js from my test site page: https://weather.wilmslowastro.com/test/newMX/trends.htm

After I worked out how to download the .js file all works now, the Daily Temperatures button has returned on my site.

[mdbaines1]

Going back to the NOAA Year report issue,

I was just going through the MXDiag logs, and when you click the button no errors are logged, but it did catch this error at the Midnight autogeneration time.

2021-02-27 00:00:00.887 Writing yesterday.ini
2021-02-27 00:00:00.940 Written yesterday.ini
2021-02-27 00:00:00.941 Creating NOAA monthly report for 26 February 2021
2021-02-27 00:00:01.657 Saving monthly report as Reports\NOAAMO0221.txt
2021-02-27 00:00:01.809 Creating NOAA yearly report
2021-02-27 00:00:01.899 SendLoopCommand: Starting - LPS 2 1
2021-02-27 00:00:01.974 SendLoopCommand: Starting - LOOP 50
2021-02-27 00:00:02.131 Error creating NOAA reports: Input string was not in a correct format.
2021-02-27 00:00:02.131 NOAA reports will be uploaded at next web update
2021-02-27 00:00:02.606 If required the daily graph data files will be uploaded at next web update
2021-02-27 00:00:02.606 Executing daily program: c:\cumulus\upload-dayfile.bat params:
2021-02-27 00:00:04.144 === Day reset complete

That appears to be the only Error recorded.

[mcrossley]

I'll create a debug build with some extra error messages if you'd be willing to try running that?

[mdbaines1]

Yes - no problem, I expect it's something simple causing it.

[AndyKF650]

Just to confirm I saw the same error message in MXdiags at the midnight roll over.

[mcrossley]

OK, who is willing to run a debug version for me?

[PaulMy]

Sorry, I don't have a Pi.

Paul

[mdbaines1]

OK, who is willing to run a debug version for me?

I will - what's needed?

[AndyKF650]

count me in as well

[mcrossley]

I'll PM you both...

[AndyKF650]

Hi Mark

Just did a bit of digging on this problem.
I opened the .ini file and noted that most of the NOAA numbers were at 16dp rather than 1. I cleaned all the numbers back to 1dp and then saved the .ini file. On reopening the numbers were still 1dp.
Then restart CMX and on the desktop I tried to generate monthly NOAA report OK and annual NOAA report FAIL.
Then reviewed the .ini file and found some of the NOAA numbers had reverted to 16dp.
I am not sure if this helps you in your research on this problem

[mdbaines1]

I noticed that as well - but disregarded it due to it being like that in version 3.9.x where the NOAA reports still worked, I checked the cumulus.ini file I had in a backup made before doing the v3.10 upgrade.

I also notice we are both using David VP2s wonder if its something related to the data provided from them?

[mcrossley]

Have either of you run the debug version I PM'd you about yet? I really need to see those diags files.

The decimals in the ini file is normal. Decimals are not stored exactly by computers.

[mdbaines1]

I've not had any PM from you yet?