Re: 20000107: problems reading CDC NCEP/NCAR reanalysis data

Kim,

>Date: Sat, 08 Jan 2000 01:49:43 +0900
>From: Kim Ju-Won <address@hidden>
>Organization: .
>To: address@hidden,
>To: address@hidden
>Subject: Thank you, but...
>Keywords: 200001051437.HAA03318

In the above message, you wrote:

> error 1:
> Unaligned access pid=21539 <a.out> va=0x140005f94 pc=0x12003b61c
> ra=0x12000d0e8 inst=0x8e100000

This "error" can be removed by tellin the CPU to ignore unaligned access
errors.  I don't recall how to do this.   Check your manuals regarding
this error.

> But output file was made. Still all jobs is condected at DEC Alpha.
> 
> Third, however, errors was occured again in numerical model.
> I use RAMS(Regeonal Atmospheric Modeling System, Colorado State
> University). Above output data use to input data for RAMS.
> Run was conducted at CRAYC90.
> 
> Error 2 :
>  Initial varfile read of grid-1
>  NO CONVERGENCE AT ( 1575) NUMBER ITERATIONS   50 MAXIMUM VCTM3(K) 1.000
> 
>  K       TH     THIL     TEMP   VCTM1(K)    PI      RT      RC
> RV      RICE
>  1  441.433  349.690   317.334         536.248  0.72   96.76   61.73
> 35.03        0.00
>  NO CONVERGENCE AT ( 1647) NUMBER ITERATIONS   50 MAXIMUM VCTM3(K) 1.000
> 
>  K       TH     THIL     TEMP   VCTM1(K)    PI       RT      RC
> RV      RICE
>  2  444.944  350.464   322.706         350.464  0.73  160.12    0.00
> 160.12        0.00
>  NO CONVERGENCE AT ( 1527) NUMBER ITERATIONS   50 MAXIMUM VCTM3(K) 1.000
> 
>  K       TH     THIL     TEMP   VCTM1(K)    PI      RT      RC
> RV      RICE
>  3  397.564  349.114   284.397         349.114  0.72   65.86    0.00
> 65.86        0.00
> .
> .
> .
> So, run is ended.
> I don't know what stoped run was related to above error 1.

If the CPU isn't told to ignore unaligned access errors, then the
process will abnormally terminate.

Regards,
Steve Emmerson   <http://www.unidata.ucar.edu>