10

This question already has an answer here:

I have written a loop to iterate all .out c binary file and copy their outputs into a text file (the output for every binary file is just a one line hash value, one line output for one program). Here is my code so far:

for j in {1..10}
do
    ./gcc-$j.out >> gcc-result.txt
done

Unfortunately, some binary files have some unknown issues and cannot be correctly executed (they got stuck and cannot proceed to the next program).

I am not going to fix those c code but I want my bash to automatically jump to executing the next program within a given timeout (say 10 secs), and also write "0" to the gcc-result.txt.

Thanks in advance if you have idea to solve this issue.

marked as duplicate by Anthony Geoghegan, Jeff Schaller, Timothy Martin, Jesse_b, G-Man Apr 18 '18 at 21:03

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • Does the "stuck" process need to be killed or can it just be ignored and left to run to completion? – roaima Apr 27 '15 at 10:48
  • Thank you for your reply, it should be killed, and output "0" result file (occupy one line) – Yang Xia Apr 27 '15 at 10:49
  • Isn't this an instance of the famed "Halting Problem"? – Bruce Ediger Apr 27 '15 at 12:28
  • @BruceEdiger No. The spec quite clearly says that the code should skip a file if it takes more than "say 10 secs" to terminate. You don't need to know if it's going to halt or not: just whether it did halt within a time limit. – David Richerby Apr 27 '15 at 17:54
  • @DavidRicherby - thanks for the consideration. "Say, 10 seconds" sounds like a very informal spec, and to my mind, a requirement conceived of without much thought. But, yes, that makes it not an instance of the Halting Problem. – Bruce Ediger Apr 27 '15 at 18:13
11

You could use the timeout command :

if timeout 10 ping google.fr > /dev/null
then
    echo "process successful"
else
    echo "process killed"
fi

shows process killed, and

if timeout 10 ls /usr/bin | wc -l > /dev/null
then
    echo "process successful"
else
    echo "process killed"
fi

shows process successful. Based on this, you could run each command using such an if; then; else; fi, redirect the standard output to a temporary file, and copy that temporary to the target file in the successful case, while generating the target file in the failure case.

How can I kill a process and be sure the PID hasn't been reused could be helpful in case you don't have timeout.

4

You can certainly kill the child process after some execution time and append the text file you need with '0' from within your bash-script.
You may find Bash script that kills a child process after a given timeout useful.

Not the answer you're looking for? Browse other questions tagged or ask your own question.