---[precice] ERROR:  Send using sockets failed with system error: write: Broken pipe

Using preCICE
, ,
1

Dear all,

I am simulating the problem of elastic structures entering into water on the cluster, openfoam handles fluid, calculix handles solid and precice addresses the coupling. The case can run normally on the local Ubuntu 2004 system.

Fluid fsi interface
Fluid fsi interface733×540 61 KB

But when I run it on the CentOS 7 system in the cluster, it always exits the simulation at 0.001s. CalculiX provides a system error:

—[precice] e[31mERROR: e[0m Sending data to another participant (using sockets) failed with a system error: write: Broken pipe. This often means that the other participant exited with an error (look there).

I have reinstalled the softwares twice and precice’s ctest has also passed.

ctest
ctest801×882 110 KB

I am completely lost. OpenFOAM is completely convergent, and so is the coupling interface. Why does CalculiX give this error? Can someone give me some advice??
precice-config.xml (2.6 KB)
logFluid.txt (662.3 KB)
logSolid.txt (219.0 KB)
precice-Solid-convergence.log (3.7 KB)
slurm.txt (382 Bytes)

2

Hi,

The Fluid / OpenFOAM logs are either truncated, or the process was terminated by the job system.

The last preCICE message on the OpenFOAM side was:

iteration: 3 of 100, time-window: 20, time: 0.00095 of 0.04, time-window-size: 5e-05, max-timestep-length: 5e-05, ongoing: yes, time-window-complete: no, read-iteration-checkpoint

Hence, the simulation is not yet finished.

Given that you start OpenFOAM first, it may be that the job system kills it first, which leads to the observed error in the CacluliX adapter.

It is possible that the job reaches some time limits defined by the cluster admins.

I see that you define out and error files, do they contain anything useful?

#SBATCH -output=Job_Results%j.out
#SBATCH -error=Job_Results_ERROR%j.err