meteohub.de

Hello,

Since a couple of days I am missing values of some variables in the output of meteohub. It seems the variables with "day1........" and "last60......." are missing.
I am using versie 5.0d of the meteohub application. I have checked the logs and there are no messages of "bad data" or anything else which looks not ok.
I hope someone of you will have an idea what could be the cause of this.

In the attachment I have the output of the ipaddress-meteohubpc:5558 of the meteohub which is running on a Alix.1D:

actual_system_version_text 5.0d
actual_system_version_num 50
actual_system_build_num 7217
actual_system_platform_text ALIX.1D_(500_MHz)
actual_system_language_text Dutch
actual_system_temp_unit c
actual_system_hum_unit rel
actual_system_press_unit hpa
actual_system_wind_unit bft
actual_system_rain_unit mm
actual_utcdate 20150103182829
actual_utcdate2 03.01.2015 18:28:29
actual_date0_puredate_utc 03.01.2015
actual_date0_time_utc 18:28:29
actual_date0_year_utc 2015
actual_date0_month_utc 01
actual_date0_day_utc 03
actual_date0_dayofweek_utc 6
actual_date0_hour_utc 18
actual_date0_min_utc 28
actual_date0_sec_utc 29
actual_localdate 20150103192829

Thanks it advance

BUMP

Meteohub only generates actual_ values, none of the day1_ month_ and so on are changed with the new values..

This is one of the first users of my template.
So I bump this as the website is totaly broken by all the missing day1 values.
http://www.weerstation-nijbroek.nl/weat ... x.php?p=10

Is there anyone with even the faintest idea how this can happen?

Look at your own meteohub-ip:5558 output and compare with these lines; Normaly there are hundreds of lines between the last alltime and the first month line.
All the other time-slots have all sensors are present, but not for day1 - hour1- and so on.

And another clue: None of the calculated month_ alltime- and so on values are changed with the new actual_ values.
It seems Meteohub is not using the actual_ values AT ALL

Were do we have to look?

There seems to be no errors in the logs?

HELP

Wim

It took some (better much) testing and reinstalling. We discovered that removing all past data files resolved the problem.
Only after a logon by Boris the reason was found, these are his words:

Hi,
I found the problem. When you look at the first data sets, you will see that somehow data of 5th January is at the beginningof the file.
These blocks all following data from being read, as Meteohub inspects linear increasing time stamps and no jumping forth and back.
I deleted those lines and did start recomputation. When done,all should look fine again.
No idea, how this happened.

And indeed all data is now appearing as it should.

This problem started on/just after the first day of the year. Maybe a backup/restore went wrong.
As every day the data for the first few measurements was written at the beginning of the file, we will check if the problem comes back next month Feb.1 or Jan.1 next year.
And again:
Big thanks to Boris.
Wim

I remember running into this kind of issue a long time ago (data out of order in the raw files). With a simple sort the issue was addressed.

Thanks everyone for helping me out.

Everything is working again like it should.