| icab |
April 18, 2017 07:04 |
foam-extend-4.0 pimpleDyMFoam stops suddenly when running in parallel
2 Attachment(s)
Dear Foamers,
I am facing a problem concerning a parallel computation in foam-extend-4.0 on openSUSE 42.1 Leap.
I want to investigate VIV of a circular cylinder using the pimpleDyMFoam solver.
The calculation works fine using 1 processor.
When I try do to a parallel computation using
Code:
mpirun -np 4 pimpleDyMFoam -parallel
the simulation stops ("hangs-up") during calculation without any error.
I already did a parallel computation with pisoFoam and this worked perfectly.
This are the last lines of the log.mpirun-file (whole log-file in attachment):
Code:
Starting time loop
Courant Number mean: 0.0159612 max: 0.361299 velocity magnitude: 1
--> FOAM Warning :
From function dlLibraryTable::open(const dictionary& dict, const word& libsEntry, const TablePtr tablePtr)
in file lnInclude/dlLibraryTableTemplates.C at line 67
library "libsampling.so" did not introduce any new entries
Searching for probe point locations
Time = 0.00025
Restraint verticalSpring spring length 0.2 force (-0 0 -0) moment (0 0 0)
Constraint fixedLine1 error (8.00292e-10 0 0) force (-0.00896327 -0 -0) moment (0 0 0) converged
sixDoFRigidBodyMotion constraints converged in 1 iterations
Constraint force: (-0.00896327 0 0)
Constraint moment: (0 0 0)
Centre of mass: (3 0 0.05)
Linear velocity: (4.80175e-07 -1.18923e-21 4.27439e-39)
Angular velocity: (8.76288e-23 -8.35888e-22 8.40257e-20)
GAMG: Solving for cellDisplacementx, Initial residual = 0, Final residual = 0, No Iterations 0
GAMG: Solving for cellDisplacementy, Initial residual = 0, Final residual = 0, No Iterations 0
volume continuity errors : volume = 2.99921, max error = 0, sum local = 6.49626e-314, global = 0
PIMPLE: iteration 1
BiCGStab: Solving for Ux, Initial residual = 1, Final residual = 9.84448e-09, No Iterations 3
BiCGStab: Solving for Uy, Initial residual = 1, Final residual = 1.37882e-07, No Iterations 2
GAMG: Solving for p, Initial residual = 1, Final residual = 0.00681398, No Iterations 24
Moving mesh time step continuity errors : sum local = 1.16265e-08, global = -4.98565e-10, cumulative = -4.98565e-10
PIMPLE: iteration 2
BiCGStab: Solving for Ux, Initial residual = 0.0346077, Final residual = 3.4695e-07, No Iterations 2
BiCGStab: Solving for Uy, Initial residual = 0.332333, Final residual = 8.04175e-08, No Iterations 2
GAMG: Solving for p, Initial residual = 0.0660635, Final residual = 8.14216e-08, No Iterations 57
Moving mesh time step continuity errors : sum local = 1.70333e-12, global = -2.45975e-14, cumulative = -4.9859e-10
ExecutionTime = 1.11 s ClockTime = 1 s
Courant Number mean: 0.0164681 max: 0.529879 velocity magnitude: 1.46659
Time = 0.0005
Restraint verticalSpring spring length 0.2 force (-0 0 -0) moment (0 0 0)
Constraint fixedLine1 error (9.35335e-08 0 0) force (-1.04758 -0 -0) moment (0 0 0) not converged
Constraint fixedLine1 error (2.806e-08 0 0) force (-0.314273 -0 -0) moment (0 0 0) not converged
Constraint fixedLine1 error (8.41801e-09 0 0) force (-0.0942818 -0 -0) moment (0 0 0) not converged
Constraint fixedLine1 error (2.5254e-09 0 0) force (-0.0282845 -0 -0) moment (0 0 0) not converged
Constraint fixedLine1 error (7.57621e-10 0 0) force (-0.00848536 -0 -0) moment (0 0 0) converged
sixDoFRigidBodyMotion constraints converged in 5 iterations
Constraint force: (-1.4929 0 0)
Constraint moment: (0 0 0)
Centre of mass: (3 -5.94617e-25 0.05)
Linear velocity: (4.54573e-07 -1.05822e-09 1.29309e-20)
Angular velocity: (8.43161e-20 -1.30344e-19 -9.97196e-12)
GAMG: Solving for cellDisplacementx, Initial residual = 1, Final residual = 7.82687e-06, No Iterations 30
GAMG: Solving for cellDisplacementy, Initial residual = 0.00334489, Final residual = 9.43799e-06, No Iterations 9
The simulation starts at the beginning using 4 processors, but immediately after the beginning 3 processors stop "working" and only 1 processor has 100% load. The simulation remains stuck in this situation.
I attached also the whole case I am using.
If want to run the case in parallel just execute
in the case-directory.
For 1 processor execute
If you need further information to solve my problem just let me know.
Kind regards,
icab |
