[Wrf-users] Segmentation faults, DX/DY, and timekeeping errors.

[Agnes Mika]

Hallo Elliot,

 I can't help you with your first question on ndown.exe, but see below
 some comments on your other points. Btw, do you have some reason for
 using ndown.exe (off-line nesting) instead of the built-in on-line
 nesting? By using the on-line version you could avoid the extra
 ndown.exe step.

Elliot wrote:
> Another problem I am having in WRFV3.0 is when trying to run the first child nest, after ndown.exe and real.exe have successfully completed.  Immediately upon the start of the first child WRF.exe, I get the following error: 
> 
> 
> WRF V3.0 MODEL
>   *************************************
>   Parent domain
>   ids,ide,jds,jde            1         151           1         151
>   ims,ime,jms,jme           -4          81          -4          44
>   ips,ipe,jps,jpe            1          75           1          38
>   *************************************
>  DYNAMICS OPTION: Eulerian Mass Coordinate
>    med_initialdata_input: calling input_model_input
>  -------------- FATAL CALLED ---------------
>  FATAL CALLED FROM FILE:  input_wrf.b  LINE:     173
>  DX and DY do not match from the namelist and the input file
>  -------------------------------------------
> [b227.uncc.edu:25854] MPI_ABORT invoked on rank 0 in communicator MPI_COMM_WORLD with errorcode 1

I would gess that the NE and SW grid resolution (DX and DY) are not
the same in your parent domain and in your namelist file. Check them
for consistency.


> Finally, I am running into problems with WRF.exe abruptly aborting in the middle of a run, ending with a segmentation fault.  Again, these segmentation faults occur in the exact same place when I attempt to repeat the same case.  The segmentation fault is given as follows:
> 
> b223:12055] *** Process received signal ***
> [b223:12055] Signal: Segmentation fault (11)
> [b223:12055] Signal code: Address not mapped (1)
> [b223:12055] Failing at address: 0xfffffffe06fcfb00
> forrtl: error (78): process killed (SIGTERM)
> Image              PC                Routine            Line        Source             
> libgcc_s.so.1      00000033180087B0  Unknown               Unknown  Unknown
> libc.so.6          00000033148E5338  Unknown               Unknown  Unknown
> libopen-pal.so.0   00002B28CAB0598E  Unknown               Unknown  Unknown
> libopen-pal.so.0   00002B28CAB044DE  Unknown               Unknown  Unknown
> libpthread.so.0    000000331540E4C0  Unknown               Unknown  Unknown
> wrf.exe            0000000001454868  Unknown               Unknown  Unknown
> wrf.exe            000000000144F6E8  Unknown               Unknown  Unknown
> wrf.exe            000000000144E6F4  Unknown               Unknown  Unknown
> wrf.exe            000000000144C528  Unknown               Unknown  Unknown
> wrf.exe            00000000011538DD  Unknown               Unknown  Unknown
> wrf.exe            0000000001238307  Unknown               Unknown  Unknown
> wrf.exe            0000000000C53DA2  Unknown               Unknown  Unknown
> wrf.exe            00000000007BEE81  Unknown               Unknown  Unknown
> wrf.exe            000000000049EAD6  Unknown               Unknown  Unknown
> wrf.exe            000000000046718B  Unknown               Unknown  Unknown
> wrf.exe            0000000000467019  Unknown               Unknown  Unknown
> wrf.exe            0000000000466FC2  Unknown               Unknown  Unknown
> libc.so.6          000000331481D974  Unknown               Unknown  Unknown
> wrf.exe            0000000000466EE9  Unknown               Unknown  Unknown


This is not enough information. You should also post to the mailing
list the section of your rsl.error file before this message.

One guess: check if you see statements like "would go off top" or "cfl
exceeded" (or similar). This would mean that your model becomes
unstable due to a violation of the CFL criterium. In this case you
would need to use a shorter timestep.

But there could be countless other reasons for such a message so
please send us more information.

Regards,
Agnes

--
Dr. A'gnes Mika
Advisor, Meteorology and Air Quality

Tel:    +31 (0)527-242299
Fax:    +31 (0)527-242016
Web:    www.bmtargoss.com

BMT ARGOSS
P.O. Box 61, 8325 ZH Vollenhove
Voorsterweg 28, 8316 PT Marknesse
The Netherlands

Confidentiality Notice & Disclaimer

The contents of this e-mail and any attachments are intended for the
use of the mail addressee(s) shown. If you are not that person, you
are not allowed to take any action based upon it or to copy it,
forward, distribute or disclose its contents and you should delete it
from your system. BMT ARGOSS does not accept liability for any errors
or omissions in the context of this e-mail or its attachments which
arise as a result of internet transmission, nor accept liability for
statements which are those of the author and clearly not made on
behalf of BMT ARGOSS.