I'm wanting to constantly monitor syslog and perform some computations after a regex pattern appears, and use the grep output in the command, as well as continue to monitor for new matching lines. The best way I can think to accomplish this is using tail -f and piping this output to grep. The only issue is I don't know of a way to run a new command based off each new line of output and continue until termination.

Any ideas on what I could use here?


What you want is piping it to a while read. Quick example for just counting occurrences:

tail -f input.txt | grep 'pattern' | while read line ; do
    echo -e "found match in $line , occurrence no. $a"

Tried with dd if=/dev/random of=file as reference. Assumption is that you only need 'pattern' as trigger and not the previous lines.

  • 6
    Note that you need to somehow make grep not buffer its output, otherwise the script will only see anything after some 4 kB or so is matched. With GNU grep, use grep --line-buffered, with other tools, perhaps something else. See. Turn off buffering in pipe. – ilkkachu Jan 27 at 22:55
  • Note that that would stip leading/trailing blanks from each line and interpret all escape sequences so, for example, foo\tbar in the input would become foo<tab>bar in the output. Always use while IFS= read -r line unless you have a specific need not to. – Ed Morton Jan 31 at 19:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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