auxhist3 output problem

I am trying to develop three days forcast for an Indian Region using WRF-4.0.3 using GFS 0.5 data. My requirements are only surface variables viz; TMAX, TMIN, Windspeed, Rainfall, Solar Radiation & RH that we use in agriculture. So I have added auxhist3 for getting these values for the day. I also added auxhist3_begin_h and auxhist3_begin_m for obtaining these values for the period in which we take observation (7:00 AM to 7:00 AM). The problem is that, the values of T2MAX and T2MIN from auxhist3 looks like Celsius but the unit indicated in header is K. In case of T2MEAN I could see abnormal values a five digit integer & decimals instead of three digit integer. However, without adding auxhist3_begin_h and auxhist3_begin_m, I found all these values are normal and the unit in K. I have not received any error mesages.

As I want these variable to be stored during our observation time I need to use auxhist3 begin hour. Is there any way by which I can correctly store these variables according to my requirement. I am using CenrOS 6.7 & Gfortran & gcc 4.4.7.

Attachments are forum1_auxhist3.tar.gz with contents
File with auxhist3_begin_h = wrfxtrm_d01_2019-03-03_01:30:00
File with auxhist3_begin_h = wrfxtrm_d01_2019-03-03_12:00:00
namelist.input

Hi,
I first would like to apologize for the delay in response. We have been preparing for the upcoming V4.1 release and have fallen a little behind with these posts. Thank you for your patience.
The auxhist3 stream is reserved for long-term climate diagnostics, which I see that you do have that turned on (output_diagnostics = 1), but it looks like you are likely asking for all of the other variables you need to also be put into that stream, possibly using the runtime I/O option? I am not sure if this may cause a problem with the overlap in streams. Just as a test, can you try to add another stream that isn't already used (7 would be good), and output everything in your varlist.txt file to that stream, instead? You can try using the auxhist7_begin_h/m parameters with that one to see if that is any different. This may not be the problem, but I'd like to see before we did some more.

Thanks,
Kelly