[Mitgcm-support] Re: Cannot run the executable file

[mitgcm-support at dev.mitgcm.org]

Dear Dr. Adcroft,

Thanks for your message. I check all of the topog.* files.
The topog.bump file exists and is also readable in the same directory.
All of the experiments have the same problems. I guess this is due to the
Linux machine (The ieee issue when reading the binary file). I follow
the same procedure using another SunOS machine and can run the
code succefully. However, I don't know how to fix this problem in Linux.
It seems the -D_BYTESWAPIO is used to remove the binary reading problem.
I don't know if there is anyone running the code in Linux successfully.
Thanks a lot for your help!

Sincerely yours,

Yu-heng
On Sun, 3 Mar 2002, Alistair Adcroft wrote:

> Yu-heng,
>
> That error means the code isn't finding the input file "topog.bin".
> Does the file exist and is it readable? Do any other experiments
> have similar problems?  Where are you running the executable? All
> the input files all need to be in the current directory where you
> issue the execution command.
>
> Alistair Adcroft.
>
>
> > I just download release1 and test the model in our machine(Linux alpha).
> > It seems the compilation is fine however, I cannot run it due to the
> > following error after the link:
> > (PID.TID 0000.0001)  MDSREADFIELD: opening global file: topog.bump
> >
> > forrtl: severe (36): attempt to access non-existent record, unit 9, file
> > /home/directory/verification/exp4/input/topog.bump
> >
> > Anybody has this error before? We are using Compaq Fortran for alpha.
> > Thanks!
> >
> > Yu-heng
>
>