[LauraSumner]

Hi
Sorry in advance for the potentially banal question!!

I am trying to run my first case in converge, managed to get everything set up in the studio with no errors, and run the case. The entire case did not finish, and I am trying to see what happened using tecPlot. I have tried to post convert the output files and get this message for example:

numbers don't add up
vertex_index, face_per_cell_index, verts_per_face_index = 0 0 0
did the file finish writing?
converting file post000013_+1.20000e-01.out ..


I wonder, why does this happen?

[azandian]

Originally Posted by LauraSumner

Hi
Sorry in advance for the potentially banal question!!

I am trying to run my first case in converge, managed to get everything set up in the studio with no errors, and run the case. The entire case did not finish, and I am trying to see what happened using tecPlot. I have tried to post convert the output files and get this message for example:

numbers don't add up
vertex_index, face_per_cell_index, verts_per_face_index = 0 0 0
did the file finish writing?
converting file post000013_+1.20000e-01.out ..


I wonder, why does this happen?

Hi Laura;
The error message you are getting is from Tecplot, so my guess is that your postfile has not been written to completion and Tecplot is looking for some data that is not there. You could try converting one of your older postfiles and loading that in Tecplot to see if you get the same error message.

It is usually a better idea to check the logfile to see why a simulation has crashed. To get a more detailed logfile as the simulation progresses, turn up your screen_print_level to 3 in inputs.in. If you see the same error message when loading any of your older postfiles into Tecplot, please contact the CONVERGE support team at support@convergecfd.com for help.

[LauraSumner]

Thanks for the reply!
This was actually my first simulation so I couldn't compare it against anything. In this case the simulation ran without error, it was just that I ran it in serial but had the mpi data writing option flagged, so once I started to run using ompi it was fine!
Thanks for the heads up about screen_print_level!

[hansfoelsche]

I have a follow up question related to the output files not writing. I have tried on both Linux and Windows. The windows runs are working fine and writing proper output files, but not for the linux setup. It was writing properly before on the computer running linux, but now it is not and I am not sure what I could have changed.
Also, I am viewing the output files using Paraview, but I dont think that matters, as the output files aren't writing in the first place.
Thanks for any help!

[MCoil]

Hi-
I'm sorry you're having trouble with your output. Could you be specific about the error? Are there *no* output files, or do you think there's something wrong with them? Please also include your solver version.
You could also send an email to support@convergecfd.com and send one of the post files and the setup.

Thanks!

[hansfoelsche]

There is an output file created, but nothing is written on it. I can try emailing the Converge support to see what they can do.
Thanks!

[MCoil]

Hi -

What version of CONVERGE are you using, v2.4 or v3.0?

When you say output, do you mean the text output files (like thermo.out) or 3D output files (like postxxxxx.out)?

When you look in the log file, is the simulation running properly, or does it stop?

[hansfoelsche]

The simulation does run fine to completion, but the 3D output files (like postxxxxx.out) are not writing.

[MCoil]

Okay, so you are running a linux version of CONVERGE v2.4? Which mpi?

Have you tried both in serial and parallel?
Maybe you should set your setup in to support@convergecfd.com. Include solver version and mpi.

[Suresh Nambully]

Hello,

This is typically seen in v2.4 when mpi_write flag is on and your file system does not support it. Please try setting mpi_write_flag to 0 in inputs.in.

Best regards,

Suresh Nambully
Convergent Science