BAD TERMINATION ERROR_LINUX

[Danish]

Dear FDS developers/users,

I am performing a simple tunnel fire model with natural ventilation, whenever running the model using Linux machine with latest FDS version 6.7.9 getting an error. So I tested the model in different ways to understand if there is any issue in the model/ mesh itself and the summary is as follows. It is also noted that the same file is working using Windows machine with FDS version  6.7.5

a)      Case 1 - Mesh +model +fire properties = Error

b)     Case 2 - Mesh + model = Simulation Running

c)      Case 3 - Mesh =  Simulation Running

The below is the error whenever I am running with fire properties

forrtl: severe (174): SIGSEGV, segmentation fault occurred

Image              PC                Routine            Line        Source            

fds                00000000070DDA8A  Unknown               Unknown  Unknown

libpthread-2.17.s  00007FED4B18E5D0  Unknown               Unknown  Unknown

fds                0000000007437194  Unknown               Unknown  Unknown

fds                00000000071BE9FD  Unknown               Unknown  Unknown

fds                0000000000408412  Unknown               Unknown  Unknown

libc-2.17.so       00007FED48ACA3D5  __libc_start_main     Unknown  Unknown

fds                0000000000408329  Unknown               Unknown  Unknown

forrtl: severe (174): SIGSEGV, segmentation fault occurred

 

===================================================================================

=   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES

=   RANK 1 PID 6804 RUNNING AT node-001.is-1.compute.internal

=   KILLED BY SIGNAL: 9 (Killed)

 

Please advise the reason for this error, attaching a file for your reference

Regards

Wali

[Kevin McGrattan]

At what time in the simulation does the error occur?

[Kevin McGrattan]

Case 1 is running with the latest source code. I will run it with the source code of the last release.

[Kevin McGrattan]

Case 1 is running on my Linux computer, where the 7 meshes are run on seven processes, all run on a single node. I am running with the source code of 6.7.9 and with the current source code. 

[Danish]

Dear Kevin,

Thanks for your response. 

I have minor clarification, as you mentioned you tested the case with 6.7.9 and the current source. Is the current source different because when I checked the FDS - official download website I found the latest current release is 6.7.9 only?

I am attaching the snip regarding the Linux FDS version from my pc, please have a look, and can you please advise why it is not working on my pc.

Thanks & Regards

Wali

[dr_jfloyd]

FDS 6.7.9 was released on June 30, 2022. Then work began on the next release of FDS. Current source means the executable you get if you compile the current version of the source code on the master branch on the fds github site.

[Enrico Cardellino]

Dear all,

 

I’m starting with FDS using the preprocessor Pyrosim. I need to measure the HRR in different point, for this reason I used both Thermocouple with quantity equal to HRR per unit volume and HRR device  but I can’t get results.

Please, could you explane me how I can do this?

 

Thanks a lot in advance,

best regards!

[Kevin McGrattan]

What kind of processors do you have on your linux cluster? To find out, enter the command

lscpu

[Kevin McGrattan]

I do not understand what you have done. Paste the input lines here.