Simulation Not Running

[Daniel Romary]

Thank you, Dave. This worked great!

[David Parker]

Glad you are up and running!

I will improve the way SimVascular checks for MPI and to display the appropriate informational messages.

Cheers,
Dave

[Hannah Cebull]

Hi Dave,

Similar to Susannah, I am getting an error when I try to generate my files for simulation ("error parsing num!"). Sometimes, I can get it to run based on how I make my model, but more complex models seem to have issues. I attached the errors that I am getting from both Terminal and my Data Files window in SimVascular. I have downloaded mpiexec. I'm really at a loss and would appreciate any potential solutions.

Thanks,
Hannah

Screen Shot 2019-04-28 at 9.28.46 PM.png (496.37 KiB) Viewed 722 times

Screen Shot 2019-04-28 at 9.29.01 PM.png (67.63 KiB) Viewed 722 times

[David Parker]

Hi Hannah,

The error you are getting using mpiexec is because you need to specify the number of processors to use when executing the simulation. The mpiexec command to use 4 processors is

mpiexec -n 4 /usr/local/sv/svsolver/2019-01-19/svsolver solver.inp

The <NPROC> I used in the above message was meant to be a variable representing the number of processors, sorry for the confusion!

As for running a simulation from SimVascular, I've seen this problem before but have not been able to track down the cause. Could you please send me the .svpre file in your Simulations directory?

Cheers,
Dave

[Hannah Cebull]

Hi Dave,

Thank you for the prompt reply! I ran it from my terminal using the correct language and it gave me an error because it couldn't find my geombc.dat.1 file. This makes sense, I guess, based on my error in creating all of the files.

I tried attaching my .svpre file but it's not allowing that file type maybe because it unix executable. I've had this same error with both the 2017 and 2019 versions of the svpre files, though. I just tried redownloading everything today. It's a really confusing problem because my presolver does work some of the time. Let me know if there's another way I can try to send it to you?

Thanks for your help!
Hannah

[Hannah Cebull]

Hi Dave,

I uploaded my .svpre file to my google drive so you can download it from this link:
https://drive.google.com/file/d/1jt5ZJA ... sp=sharing

Thanks,
Hannah

[Hannah Cebull]

Hi Dave,

I just wanted to follow up and see if you were able to look at my .svpre file and If you found a potential solution? I noticed that only some of my models have this issue and if I create a simpler model with less branches I don't usually have this issue. I just can't figure out why some of my models with 3 branches run properly and others don't. Any advice on how to troubleshoot is appreciated.

Thanks,
Hannah

[David Parker]

Hi Hannah,

Sorry for the late reply, had to focus on other things for a bit.

I looked at the .svpre file you sent. The problem is that the line

set_surface_id_vtp mesh-complete/mesh-surfaces/cap_L Subclavian.vtp 5

contains a space after cap_L. You must have named a segmentation L Subclavian. You should not have spaces in the names you give to SV. SV should really check this when you enter a name, this is a bug.

Cheers,
Dave

[Hannah Cebull]

Hi Dave,

Thank you so much! I usually am careful about using underscores or no spaces, but I missed this one. After fixing it, my simulation now runs properly!!

Best,
Hannah