3d lid driven problem , no velocity. I didn't change any in case file.

[ztdepztdep]

I write some code
" write(,) lsmoothing , "user_bcuser_bc "
STOP
"
in user_bc to test if this is excuted. it seemts the user_bc was not excuted at all.

------------Fluid-------------
*** WARNING: New vector not linearly indepependent! ***
*** WARNING: New vector not linearly indepependent! ***
*** WARNING: New vector not linearly indepependent! ***
*** WARNING: New vector not linearly indepependent! ***
Pressure
Iterations: Start residual: Final residual:
0 0.0000000E+00 0.1365993-318
X-Velocity
Iterations: Start residual: Final residual:
0 0.0000000E+00 0.0000000E+00
Y-Velocity
Iterations: Start residual: Final residual:
0 0.0000000E+00 0.0000000E+00
Z-Velocity
Iterations: Start residual: Final residual:
0 0.0000000E+00 0.0000000E+00
Fluid step time (s): 0.3619183E-01
Total elapsed time (s): 0.2780178E+00

--------Postprocessing--------

---------Step summary---------
Total time for step        8 (s):   0.3621853E-01
Total elapsed time (s):             0.2780445E+00

[timofeymukha]

I am not able to reproduce this. I get the expected behavior on my machine. Just to confirm, are you on the latest develop?

[ztdepztdep]

I am not able to reproduce this. I get the expected behavior on my machine. Just to confirm, are you on the latest develop?

I downloda the latest release from official website. am i right?

[ztdepztdep]

`The session file name is lid
The numer of cpu is 2
neko.......................

    _  __  ____  __ __  ____  
   / |/ / / __/ / //_/ / __ \ 
  /    / / _/  / ,<   / /_/ / 
 /_/|_/ /___/ /_/|_|  \____/  

 (version: 0.9.99)
 (build: 2025-02-18 on x86_64-unknown-linux-gnu using gnu)


 -------Job Information--------
 Start time: 14:44 / 2025-02-18
 Running on: 2 MPI ranks
 CPU type  : Intel(R) Xeon(R) CPU E5-2683 v4 @ 2.10GHz
 Bcknd type: CPU
 Real type : double precision

 -------------Case-------------
 Reading case file lid.case

   -------------Mesh-------------
   Reading a binary Neko file lid.nmsh
   gdim = 3, nelements =       108
   Reading elements
   Reading BC/zone data
   Reading deformation data
   Mesh read, setting up connectivity
   Done setting up mesh and connectivity
   Mesh and connectivity setup (excluding read) time (s):  0.003276

   ---User defined extensions----
   - Fluid initial condition
   - User check
   - Initialize modules
   - Finalize modules

   --------Gather-Scatter--------
   Comm         :          MPI
   Avg. internal:         9696
   Avg. external:         4992
   Backend      :          std

   ------------Fluid-------------
   Type       : Modular (Pn/Pn)
   Read non-dimensional material properties
   Re         : 5.000000E+03
   Poly order : 7
   GLL points : 55296
   Unique pts.: 38829
   rho        : 1.000000E+00
   mu         : 2.000000E-04
   Dealias    : T
   LES        : F
   Save bdry  : F

      -------Velocity solver--------
      Type       : (cg, jacobi)
      Abs tol    : 1.000000E-07
 *** ERROR: Invalid user defined inflow condition ***
ERROR STOP 

Error termination. Backtrace:
 *** ERROR: Invalid user defined inflow condition ***
ERROR STOP 

this is the version from git clone. cause aother error.

[timofeymukha]

Wait, did you run makeneko first to compile the user file and get a local neko executable?

[ztdepztdep]

Wait, did you run makeneko first to compile the user file and get a local neko executable?

of course. I can ray rayleigh-benard problem withthout any problem.

[timofeymukha]

OK. Can you attach your .f90 and .case file here, so I can try locally with exactly the same thing you are running?

[ztdepztdep]

OK. Can you attach your .f90 and .case file here, so I can try locally with exactly the same thing you are running?
Uploading Archive.tar.gz…

[timofeymukha]

Something happened with the upload, it is just a link to the issue's webpage.

[ztdepztdep]

Archive.tar.gz
sorry for that.
I just found that it can run with 1 cpu.
if i run it with >2 cpus, it will cause the error.

[timofeymukha]

There is something up with your mesh file. If I run your case (changing the velocity solver to cg) on the mesh in the lid example, it runs.

[timofeymukha]

But, to be honest, maybe your mesh file is fine per se. It seems this case is not very stable. There are some warnings about not reaching the tolerance even in the mesh we use. I am not sure what the reason is, I am not reall a lid expert :-).

[pschlatt1]

In principle, a lid-driven cavity should be quite robust, if properly regularised in the corners. But yeah, let's look at it in more detail.

[pschlatt1]

indeed, something is not good with the example. Looking at it.