I wrote a rsync script which also checks for how long the process of the script has been running.
This is the relevant part:
time=$(ps -p $processid -o etime= | awk -F: '{print $2}') # number of seconds the process is running.
if [ $(ps -ef | grep $(basename $0) &>/dev/null && echo $?) -eq "0" ] && [ "$time" -gt "5" ]; then # Check if there's a running process with the script name which is running for more than 5 seconds.
echo "$message"
print_log $message
print_log "--------- END --------"
exit 1
fi
Sometimes the process gets stuck and runs for more than 5 seconds (and even days) so the above part of the script is supposed to write it to the log.
When running:
ps -p PID -o etime=
It returns how long the process has been running. Example:
43:36
But if the process has been running for more than a day, then the output looks like so:
[root@do01 ~]# ps -p 28518 -o etime=
7-22:43:36
My question is how can I get this number in seconds? because I need to make sure the process has not been running for more than 5 seconds.