reconstructing error with snappyHexMesh
 

Hi all,

I got an error message when I conducted reconstructPar.
What I did is:

1. run blockMesh
2. run decomposePar (I used 8 processors)
3. run snappyHexMesh -parallel (this created files named 0.01,0.02 and 0.03 in each processor)
4. mpirun --hostfile machines -np 8 mysolver -parallel > log &

The time step of my case i s 0.01 and the writeInterval=100.

I got the following error message when I input reconstructPar :

--> FOAM FATAL IO ERROR:
cannot find file
file: /home/myname/OpenFOAM/myname-2.2.0/run/tutorials/mycase/processor0/0.01/polyMesh/pointProcAddressing at line 0.

From function regIOobject::readStream()
in file db/regIOobject/regIOobjectRead.C at line 73.


Does anybody have any idea on this? Thanks.

Try snappyHexMesh -parallel -overwrite

Hi Paritosh,

Thank you for your reply.
My Allrun file is as follow:
************************************************** ********
. $WM_PROJECT_DIR/bin/tools/RunFunctions

runApplication blockMesh
runApplication decomposePar
runParallel snappyHexMesh 8
runApplication reconstrucParMesh -mergeTol 1e-6 -latestTime
************************************************** *************
Is there any problem of my Allrun script?
It creates a dir named 0.03. The mesh in this dir is ok. Then what should I do? Do I need to copy the polyMesh here to constant dir instead of the original one?
In addition, for my case, I need to run setFields. My question is that do I need to delete all the processor dir and redo decomposePar after I run setFields? Is that possible I could do setFields in parallel? Do you have any suggestion on this? Thank you!

If checkMesh is not giving you any error then it is alright.

Again I suggest you use -overwrite with snappyHexMesh command. This will automatically overwrite the mesh files and you need not worry about any thing.

Yes, you must delete the processor files first and then start again. If there is even a small modification in your mesh it will give you errors in later stage.

I am not sure about that.

Thank you very much!