[MFWilliams]

Hi,


I am trying to run my first FSI tutorial on solids4foam and I have encountered an error. When trying to run the tutorial 'elastic beam in cross flow' in parallel, I get to the stage where I must decompose both the solid and fluid domains and get an error for both the solid and fluid decomposition.


upon entering the command: decomposePar -region solid


I get the error:


Decomposing mesh solid

Create time

Time = 0
Create mesh for region solid


--> FOAM FATAL ERROR:
Cannot find file "points" in directory "constant/solid/polyMesh"

From function Time::findInstance(const fileName&, const word&, const IOobject::readOption)
in file db/Time/findInstance.C at line 148.

FOAM exiting




Please could somebody advise me how to rectify this?
Thanks

[Hgholami]

Hi
When you found this error, it means the command can't find points in sub-folder constant/solid/polyMesh.
you may are in incorrect folder and run the command or you had problem in blockMesh.
first check this sube-folder and found points file, then check you folder when run the decomposePar -region solid command.

Originally Posted by MFWilliams

Hi,


I am trying to run my first FSI tutorial on solids4foam and I have encountered an error. When trying to run the tutorial 'elastic beam in cross flow' in parallel, I get to the stage where I must decompose both the solid and fluid domains and get an error for both the solid and fluid decomposition.


upon entering the command: decomposePar -region solid


I get the error:


Decomposing mesh solid

Create time

Time = 0
Create mesh for region solid


--> FOAM FATAL ERROR:
Cannot find file "points" in directory "constant/solid/polyMesh"

From function Time::findInstance(const fileName&, const word&, const IOobject::readOption)
in file db/Time/findInstance.C at line 148.

FOAM exiting




Please could somebody advise me how to rectify this?
Thanks

[MFWilliams]

Originally Posted by Hgholami

Hi
When you found this error, it means the command can't find points in sub-folder constant/solid/polyMesh.
you may are in incorrect folder and run the command or you had problem in blockMesh.
first check this sube-folder and found points file, then check you folder when run the decomposePar -region solid command.

I checked the subfolder and could not see the points file, i have not edited it though and I downloaded solids4foam straight from the official bitbucket and it has been working fine with blockmesh for the other tutorials like the solid tutorials.


I tried to run 'decomposePar -region fluid' again but whilst in the system folder instead of being in the whole case directory and recieved this error instead:





Decomposing mesh fluid

Create time



--> FOAM FATAL IO ERROR:
cannot open file

file: /mnt/c/Users/merli/simulations/s4f_crossflow_elastic_test/system/system/controlDict at line 0.

From function regIOobject::readStream()
in file db/regIOobject/regIOobjectRead.C at line 60.

FOAM exiting




Any ideas at what is wrong?
Thanks for the advice

[Hgholami]

Hi, for decomposePar, you run this command in system folder.
../system/system/controlDict
you should ensure, where run command. for run command use /mnt/c/Users/merli/simulations/s4f_crossflow_elastic_test/ folder.

Originally Posted by MFWilliams

I checked the subfolder and could not see the points file, i have not edited it though and I downloaded solids4foam straight from the official bitbucket and it has been working fine with blockmesh for the other tutorials like the solid tutorials.


I tried to run 'decomposePar -region fluid' again but whilst in the system folder instead of being in the whole case directory and recieved this error instead:





Decomposing mesh fluid

Create time



--> FOAM FATAL IO ERROR:
cannot open file

file: /mnt/c/Users/merli/simulations/s4f_crossflow_elastic_test/system/system/controlDict at line 0.

From function regIOobject::readStream()
in file db/regIOobject/regIOobjectRead.C at line 60.

FOAM exiting




Any ideas at what is wrong?
Thanks for the advice

[MFWilliams]

Originally Posted by Hgholami

Hi, for decomposePar, you run this command in system folder.
../system/system/controlDict
you should ensure, where run command. for run command use /mnt/c/Users/merli/simulations/s4f_crossflow_elastic_test/ folder.

Sorry I am not completely sure I understand what you mean. I am in the 'system' subfolder which is in the s4f_crossflow_elastic_test case folder when I input the 'decomposePar -region fluid' command and I receive the error below.


--> FOAM FATAL IO ERROR:
cannot open file

file: /mnt/c/Users/merli/simulations/s4f_crossflow_elastic_test/system/system/controlDict at line 0.

From function regIOobject::readStream()
in file db/regIOobject/regIOobjectRead.C at line 60.

FOAM exiting




I am unsure as to why in the error it says 'system' twice: '.../s4f_crossflow_elastic_test/system/system/controlDict'


I have not edited the decomposeParDict files and they are the ones that are already in the elastic beam in crossflow case when solids4foam is installed.

[Hgholami]

I am in the 'system' subfolder which is in the s4f_crossflow_elastic_test case folder when I

executing command in system folder is wrong, back to master folder "s4f_crossflow_elastic_test" folder, then run the command.
All command must run in case folder not subfolder, the case folder is where the "Allrun" file located.

[MFWilliams]

Ah ok thank you.


I have now run the 'decomposePar -region fluid' command again whilst in the whole case directory and not in the system subdirectory and I recieved the following error:


--> FOAM FATAL ERROR:
Cannot find file "points" in directory "constant/fluid/polyMesh"

From function Time::findInstance(const fileName&, const word&, const IOobject::readOption)
in file db/Time/findInstance.C at line 148.

FOAM exiting


I had a look in polyMesh and the only thing in the polyMesh directory was blockMeshDict and in blockMeshDict I could not see the file 'points'. Should it be there?
I also tried running another case 'micro beam in crossflow' as well but that generated the same error message.



Thanks in advance

[Hgholami]

first, run "blockMesh -region fluid" in the whole case directory
then send the report. If the command successes, you can see the points file in "constant/fluid/polyMesh" and can run decomposePar -region fluid.
If the command failed, you should send the report to be consider.

Originally Posted by MFWilliams

Ah ok thank you.


I have now run the 'decomposePar -region fluid' command again whilst in the whole case directory and not in the system subdirectory and I recieved the following error:


--> FOAM FATAL ERROR:
Cannot find file "points" in directory "constant/fluid/polyMesh"

From function Time::findInstance(const fileName&, const word&, const IOobject::readOption)
in file db/Time/findInstance.C at line 148.

FOAM exiting


I had a look in polyMesh and the only thing in the polyMesh directory was blockMeshDict and in blockMeshDict I could not see the file 'points'. Should it be there?
I also tried running another case 'micro beam in crossflow' as well but that generated the same error message.



Thanks in advance

[MFWilliams]

Originally Posted by Hgholami

first, run "blockMesh -region fluid" in the whole case directory
then send the report. If the command successes, you can see the points file in "constant/fluid/polyMesh" and can run decomposePar -region fluid.
If the command failed, you should send the report to be consider.

Ahhh thank you so much! That has fixed it, I did not realise that you have to also run 'blockMesh -region fluid' as it does not say that in the tutorial document. It is now running successfully.


One final question. Is the number of subdomains that you decompose it into dependant on the number of processors your device has?


Many thanks on being patient with me and helping me solve the problem.


Merlin

[Hgholami]

You're welcome.
For the number of subdomains, I should say it's not straight relation to number of processors. you can choice each number of subdomains, but it can't help you.
you should choice number of subdomains base on available thread of processors.
because the available thread of processors accelerate the simulation, not number of subdomains. so if you choice number of subdomains greater than available thread of processors, some subdomains should waits to processors be available.

[MFWilliams]

Originally Posted by Hgholami

You're welcome.
For the number of subdomains, I should say it's not straight relation to number of processors. you can choice each subdomains, but it can't help you.
you should choice number of subdomains base on available thread of processors.
because the available thread of processors accelerate the simulation, not number of subdomains. so if you choice number of subdomains greater than available thread of processors, the each subdomains should waits to processors be available.

Ah ok thank you.