[amirhosseintaran]

Hello Everyone,
When I'm trying to run a case with one of the filedFunctionObjects called interfaceHeight, I'm facing a strange problem. Imagine I have assigned the write interval to each of two-time steps; the solver freezes at the time it is trying to write. Before asking the system support to check the validity of their OpenFOAM compilation, Has anyone faced something similar to this?

I checked other function objects, and they can write to the hard drive without any problem, also, the odd thing is there is no error! it just waits without any crash till the end 😄

Please let me know your thoughts on this.

[Lorenzo210]

Hi,


Which OpenFOAM version are you using?
Can you share a test case where the problem is occurring?

Also, I am not sure what you mean by

Imagine I have assigned the write interval to each of two-time steps

I tried interfaceHeight in one of the tutorials available in OpenFOAM-v2312 and there's no issue with writing the output.


Cheers,
Lorenzo

[amirhosseintaran]

Hi Lorenzo,
First of all, I had to say that I forgot to mention it is happening during parallel processing, when I'm on a single core, everything is fine.

Which OpenFOAM version are you using?

It is 2312.

Also by

Imagine I have assigned the write interval to each of two-time steps

I meant this:

writeControl    timeStep;
writeInterval   2;

I tried interfaceHeight in one of the tutorials available in OpenFOAM-v2312, and there's no issue with writing the output.

Yes, It is working for me, too (in serial). But in this specific cluster that I am working on, at the end of each timestep, the whole system freezes when it's about to write the functionObject data. Mostly, I was looking for ideas for tackling the issue here. I tried to compile a copy of my own interfaceHeight called myInterfaceHeight (everything is the same as the original, just renamed), but that also failed to write the output.


Cheers,
Amir.

[Lorenzo210]

Hi Amir,



Thanks for the information.


I ran the tutorial even in parallel, it works with different decomposition method, number of cores, writeInterval etc.


I used both interCondensatingEvaporatingFoam and interFoam.


As an initial guess, it could be a good idea to test a coarse and fast version of your test case on a regular computer and understand if the issue is correlated with the cluster you are working with..



However, It looks weird that other functions object work, because it makes it look like interfaceHeight has something wrong here..


Unfortunately I have no experience in running interfaceHeight on clusters.




Cheers,
Lorenzo

[amirhosseintaran]

Hi Lorenzo,

Thanks for this,

I ran the tutorial even in parallel, it works with different decomposition method, number of cores, writeInterval etc.


I used both interCondensatingEvaporatingFoam and interFoam.

Yes, exactly it is working on my machine even in parallel, but fails in the cluster that I am using.

However, It looks weird that other functions object work, because it makes it look like interfaceHeight has something wrong here..


Unfortunately I have no experience in running interfaceHeight on clusters.

yes, its strange to me too, anyway, thanks for your time.

Bests,
Amir.