Segmentation fault (core dumped) tzap

tonj · 03-10-2016, 04:06 AM

I'm running a centos 6.7 server (command line only) and I'm trying to auto-start a live tv stream whenever the machine boots up. I have a bash script "/root/start-stream.sh" with this command in it:

Code:

#!/bin/bash

# start tzap on pick tv
echo "starting tzap"
/usr/bin/tzap -S -r Pick
sleep 3

<snip-rest of script>

if I run the same command "/usr/bin/tzap -S -r Pick" in a shell it works perfectly, and if I manually run the script "/root/start-stream.sh" that works fine too, but if I put the script into /etc/rc.d/rc.local so it auto-starts it crashes with this:

Quote:

/root/start-stream.sh: line 6: 2922 Segmentation fault (core dumped) /usr/bin/tzap -S -r Pick

why is it crashing? and can this be solved? thanks for any help.

toothandnail · 03-18-2016, 03:19 PM

I would wonder if its starting too early - some service that it depends on has not finished initialising? I'd try putting a startup dealy in the script and see if that helps.

Paul.

rtmistler · 03-18-2016, 03:22 PM

What does the core file show? And I wasn't aware you could get a core file from BASH, if so, how do you debug it? I thought that'd come from an executable instead.

tonj · 03-19-2016, 05:09 AM

I don't know what the core file is called or where to find it.

rtmistler · 03-19-2016, 12:07 PM

Quote:

Originally Posted by tonj

I don't know what the core file is called or where to find it.

It's usually called "core" and will usually also reside in the directory where you were running the script from. Sorry I didn't look closely enough before, but based on this:

Quote:

Originally Posted by tonj

/root/start-stream.sh: line 6: 2922 Segmentation fault (core dumped) /usr/bin/tzap -S -r Pick

It is therefore having a problem at line #6 of your start-stream.sh script. And apparently that is a call to /usr/bin/tzap and my guess is that the segmentation fault is occurring in tzap. You therefore either have source for tzap, or you can possibly determine what is wrong with it by running:

Code:

$ gdb /usr/bin/tzap core

in the directory where the core file is located and see what the problems are by using the backtrace command. This also may be tzap having a problem because the input to it was bad and that particular function was poorly written where if given bad arguments it can have this problem.

tonj · 03-19-2016, 02:36 PM