logging - Create a detailed self tracing log in bash -

i know can create log of output typing in script nameoflog.txt , exit in terminal before , after running script, want write in actual script creates log automatically. there problem i'm having exec >>log_file 2>&1 line:

the code redirects output log file , user can no longer interact it. how can create log copies in output?

and, possible have automatically record process of files copied? example, if file @ /home/user/deskop/file.sh copied /home/bckup, possible have printed in log or have write manually?

is possible record amount of time took run whole process , count number of files , directories processed or going have write manually too?

my future self appreciates help!

here whole code:

#!/bin/bash      collect() { find "$directory" -name "*.sh" -print0 | xargs -0 cp -t ~/bckup #xargs handles files names spaces. gives error of "cp: not overwrite just-created" if file didn't exist } echo "starting log" exec >>log_file 2>&1 timelimit=10 echo "please enter directory collect. if no input in 10 secs, default of /home selected"  read -t $timelimit directory  if [ ! -z "$directory" ] #if directory doesn't have length of 0 echo -e "\nyou want copy $directory." #-e \n work , won't show part of string else directory=/home/ echo "time's up. backup in $directory" fi  if [ ! -d ~/bckup ] echo "directory not exist, creating now" mkdir ~/bckup fi   collect echo "finished collecting"  exit 0 

to answer "how copy output" question: use program called tee , bit of exec magic explained here: redirect copy of stdout log file within bash script itself

regarding analytics (time needed, files accessed, etc) -- bit harder. programs can time(1):

time - run programs , summarize system resource usage 

and strace(1):

   strace - trace system calls , signals 

check man pages more info. if have control on script easier logging instead of parsing strace output.