Problem with program stalling with wait4() in gdb

[declan]

Hey guys. I have program that generally runs for a while. I want to make sure it works with lots of different initial values, so I've created another program that just calls it in a loop, with system(). I've also implemented multiprocessing with OpenMP.

I run it in gdb, so if it crashes I can figure out why and hopefully fix it. It has been working, but occasionally stalls (I know because it outputs some stuff while it's running). I'll do Ctrl+c to kill it, so I can do backtrace and see where it's stalling.

But when I do backtrace, it doesn't say what functions of my program it was in. It says it's in wait4():

Timestep 29 out of 1000
Timestep 30 out of 1000
Timestep 31 out of 1000
^C
Program received signal SIGINT, Interrupt.
0x90ffe18e in wait4 ()
(gdb) backtrace
#0  0x90ffe18e in wait4 ()
#1  0x90ffbcd4 in system$UNIX2003 ()
#2  0x00001ab2 in main (someNum=1, argv=0xbffff71c) at tester.cpp:35 

tester.cpp is my program that calls the other program repeatedly. Because I'm calling the other program with system(), I'm guessing system$UNIX2003 is that. And I looked up wait4(), it seems to be something that waits for a child process to finish. But what child process? It must be stuck somewhere in my program, so why isn't it showing up in backtrace?

Thanks!

[jsmith]

The child process that got forked by system().

[declan]

But why doesn't that show up in backtrace? It's gotta be somewhere in my program.

[jsmith]

The child process is not part of your program. It is an independent program that got launched by your program.

[declan]

Hmmmm, ok. Best bet is to put it in a loop in my main program rather than calling it with system(), I guess.