Getting iter/convergence info from hung solver

General Discussion
1

I’ve setup a simple problem again ( 2 cubes in flowing air )
but with a different model equation. In setting it up, I have a few issues
with boundary conditions. In one case, the thing seems to run forever
with no progress result and no idea how to determine if it is stuck, hit
some limit cycle, or is slowly improving. Setting “verb=1” provided
nothing useful. Right now I need to just guess on formulation and BC
code. Is there some way to get more diagnostic info? Thanks.

//solve Stokes (ux, uy,uz, p,q, vx, vy,vz, solver=Crout) // oom
//solve Stokes (ux, uy,uz, p,q, vx, vy,vz,solver=sparsesolver)
solve Stokes (ux, uy,uz, p,q, vx, vy,vz,solver=CG,verb=1)

2

It looks like putting braces around the parameters helped :slight_smile:
At least now it is producing output and I can play with the model
Is there some way to get it to dump what it thinks the parameters are? lol

//solve Stokes (ux, uy,uz, p,q, vx, vy,vz,solver=sparsesolver)
solve Stokes ([ux, uy,uz, p],[vx, vy,vz,q],solver=CG,verb=1)

3

you can always use something like:

cout << "VAR1 = " << VARIABLE1 << ", VAR2 = " << VARIABLE2 << ... << endl;

to print out parameters at any point in your code.