[kate.F]

Dear FOAMers,

I have encountered a problem while running a case on cluster and got the error(part of it):

WARNING: Error in initializing MVAPICH2 ptmalloc library.Continuing without InfiniBand registration cache support.
[7] [4] [3] [6] [10] [9] [15] [5]



[4] [7] [3]


[4] [7] [3] [6] --> FOAM FATAL IO ERROR: --> FOAM FATAL IO ERROR: --> FOAM FATAL IO ERROR:


[6]
--> FOAM FATAL IO ERROR: [4] [7] [3]
wrong token type - expected Scalar, found on line 0 the word 'nan'wrong token type - expected Scalar, found on line 0 the word 'nan'

wrong token type - expected Scalar, found on line 0 the word 'nan'

[5]
[4]

[10] [9] [15] [5]


[14] [9] --> FOAM FATAL IO ERROR: [4] --> FOAM FATAL IO ERROR:
[15] file:
[11] /scratch/03857/tg831567/SC5/processor4/system/data.solverPerformance.p[5] [9] [12] at line [8]
[11] [0] 0
.
[4] [11]

The log file:

Time = 20.558

Courant Number mean: 0.028402 max: 0.11946
smoothSolver: Solving for Ux, Initial residual = 0.758777, Final residual = 9.92998e-09, No Iterations 4
smoothSolver: Solving for Uy, Initial residual = 0.399664, Final residual = 2.49574e-08, No Iterations 4
smoothSolver: Solving for Uz, Initial residual = 0.0129346, Final residual = 3.99374e-08, No Iterations 3
GAMG: Solving for p, Initial residual = 0.00802443, Final residual = 0.000744506, No Iterations 21
GAMG: Solving for p, Initial residual = 0.00167003, Final residual = 0.000155684, No Iterations 24
time step continuity errors : sum local = 4.97087e-08, global = 8.01521e-09, cumulative = -5.85061e-07
GAMG: Solving for p, Initial residual = 0.032428, Final residual = 0.00219924, No Iterations 2
GAMG: Solving for p, Initial residual = 0.00244059, Final residual = 9.45781e-07, No Iterations 64
time step continuity errors : sum local = 2.71954e-10, global = 4.74174e-11, cumulative = -5.85013e-07
ExecutionTime = 27940.8 s ClockTime = 28738 s

Time = 20.56

Courant Number mean: 0.0282469 max: 0.118619
smoothSolver: Solving for Ux, Initial residual = 0.327472, Final residual = 4.42746e-09, No Iterations 4
smoothSolver: Solving for Uy, Initial residual = 0.180964, Final residual = 1.13125e-08, No Iterations 4
smoothSolver: Solving for Uz, Initial residual = 0.0103192, Final residual = 3.28473e-08, No Iterations 3
GAMG: Solving for p, Initial residual = 0.00895919, Final residual = 0.000816276, No Iterations 20
GAMG: Solving for p, Initial residual = 0.00165398, Final residual = 0.000153604, No Iterations 24
time step continuity errors : sum local = 2.04548e-08, global = 3.5577e-09, cumulative = -5.81455e-07
GAMG: Solving for p, Initial residual = 1, Final residual = nan, No Iterations 1000
GAMG: Solving for p, Initial residual = nan, Final residual = nan, No Iterations 1000

And I have run the identical case on the clusters of another institution, as well as on my local machine, it worked well.

It is LES case, cell number 3.75 million. subdomain: simple, (12,4,1). The OpenFOAM versions on the two clusters and my workstation are all 2.3.0. Does any one have any idea about the problem?

[alexeym]

Hi,

You problem is divergence

Time = 20.56

Courant Number mean: 0.0282469 max: 0.118619
...
GAMG: Solving for p, Initial residual = 1, Final residual = nan, No Iterations 1000
GAMG: Solving for p, Initial residual = nan, Final residual = nan, No Iterations 1000

Why GAMG is diverging? It depends on GAMG settings (and maybe IC/BC). Could you reproduce the error with PCG?

[kate.F]

Hi Alexey,
Thank you very much for you reply. I have run the identical case elsewhere( my workstation, and cluster of another institution) and it worked well, no divergence. And that is why I'm confused and don't know what is the problem.
I'll try PCG and will post the result here.

Thank you again!