[Peter_600]

Dear Foamers,

sprayEngineFoam crashes in parallel

mpirun has exited due to process rank 0 with PID 29241 on
node nusm-14 exiting improperly. There are two reasons this could occur:

1. this process did not call "init" before exiting, but others in
the job did. This can cause a job to hang indefinitely while it waits
for all processes to call "init". By rule, if one process calls "init",
then ALL processes must call "init" prior to termination.

2. this process called "init", but exited without calling "finalize".
By rule, all processes that call "init" MUST call "finalize" prior to
exiting or it will be considered an "abnormal termination"

This may have caused other processes in the application to be
terminated by signals sent by mpirun (as reported here).

I was following an bug report

http://www.openfoam.org/mantisbt/view.php?id=577

Unfortunately, I couldn't solve the problem with the proposed workaround.

Has someone of you used successfully sprayEngineFoam with OF-2.1.x ?

Cheers Peter

[ris]

Hello Peter,

I have the same problem and just want to confirm that the workaround provided on the bug report page does not work for OF2.1.x latest commit.

ris.

[kalle]

Hi,

so, there might have been changes lately in the code that renders the workaround unusable. I did not git pull since september I think, so it is about time I do an update. I also need parallel mesh motion + spray, so I'll investigate. I guess the developers have an idea to solve this later on, but for some reason they did not comment or acted upon this yet.

I'll get back here when I know more.

K

[Peter_600]

Thank you for this information!

[kalle]

Sorry for the delay. I have migrated my code to 2.2.x recently. It seems the hack still works for my case. This time I have simply commented out all calls to

meshTools::constrainToMeshCentre
meshTools::constrainDirection

in

src/lagrangian/intermediate/submodels/Kinematic/InjectionModel/InjectionModel/InjectionModel.C

src/lagrangian/intermediate/parcels/Templates/KinematicParcel/KinematicParcel.C

as 1D and 2D is not of interest for me anyway.

I run unsteady parallel with mesh motion (without topological changes) with coneNozzleInjection. Except that, the sprayCloudProperties is the same as aachenBomb. I guess for other configurations, you may fall in other traps.

Lets hope they find a more smooth solution.

K

[ris]

Hello,

I attempted the hack on OF-2.2.x for use with sprayEngineFoam and topo changes but it did not work.
I decided to use the 1st release of OF2.1.1 (http://sourceforge.net/projects/foam/files/foam/2.1.1/) and the hack (commenting out the lines) does not work. The code will run but the simulation is incorrect in the fact that some cells are not updated with the correct information.

regards,

ris