[FLASH-USERS] H5Screate_simple(): zero sized dimension for non-unlimited dimension

[Samuel Friedman]

Kunal,

The previous email is correct.  The reason why you cannot run on more than 
one processor right now is that you have one and only one block.  Your log 
says "tot blks 1".  When you add a second processor, it will crash.  Try 
increasing your maximum refinement level (lrefine_max) and that should 
increase the number of blocks you have.

Sam

On Thu, 3 Jun 2010, kunal rao wrote:

> Hi,
> I am running FLASH 3.2 on my laptop on a virtual machine running Ubuntu 9.10. I have
> configured the code using:
> $./setup Sedov -auto -makefile=gfortran
> 
> I am able to run the executable for a single process using $./flash3 . But when I try to
> run it on 2 processes using
> 
> $mpirun -np 2 ./flash3
> 
> I get the following error:
> 
> Driver_abortC called
> Error: negative return from memspace H5Screate_simple
> Calling MPI_Abort for immediate shutdown
> HDF5-DIAG: Error detected in HDF5 library version: 1.6.10 thread 0.  Back trace follows.
>   #000: H5S.c line 1595 in H5Screate_simple(): zero sized dimension for non-unlimited
> dimension
>     major(01): Function arguments
>     minor(05): Bad value
> 
> This problem was addressed in one of the previous archived emails. The response to that
> by Chris was that:
> 
> -------------------------------------------------
> 
> I think this is happening because there are zero blocks on at least
> one processor. You can check whether this is the case by looking in
> the flash log file from your run. Look for the last
> amr_refine_derefine message of the form:
> 
> [GRID amr_refine_derefine] min blks 17 max blks 18 tot blks 69
> 
> I expect there is a min blks 0 in your message. A temporary
> workaround is to run your application on fewer processors.
> 
> -------------------------
> 
> I checked in my sedov.log file and it is 0. This is what i see in the sedov.log file:
> 
> ----------------------------
> 
> WARNING : Adaptive Grid did not find any variable to refine
>  [ 06-03-2010  11:55:51.588 ] [GRID amr_refine_derefine]: initiating refinement
> 
> 
>  [GRID amr_refine_derefine] min blks 0    max blks 1    tot blks 1
>  [GRID amr_refine_derefine] min leaf blks 0    max leaf blks 1    tot leaf blks 1
>  [ 06-03-2010  11:55:51.616 ] [GRID amr_refine_derefine]: refinement complete
> 
> 
>  [ 06-03-2010  11:55:51.743 ] [GRID gr_expandDomain]: iteration=1, create level=2
> 
> -----------------------------------------------------------
> 
> The temporary workaround was to use fewer processes. But I am not able to run on even 2 
> processes. 
> 
> Can you please tell me if there is a solution for this.
> Thanks & Regards,
> Kunal
> 
>