Running a simple shell script as a cronjob

The easiest way would be to use a GUI:

For Gnome use gnome-schedule (universe)

sudo apt-get install gnome-schedule 

For KDE use kde-config-cron

It should be pre installed on Kubuntu

But if you use a headless linux or don´t want GUI´s you may use:

crontab -e

If you type it into Terminal you´ll get a table.
You have to insert your cronjobs now.
Format a job like this:

*     *     *     *     *  YOURCOMMAND-     -     -     -     -|     |     |     |     ||     |     |     |     +----- Day in Week (0 to 7) (Sunday is 0 and 7)|     |     |     +------- Month (1 to 12)|     |     +--------- Day in Month (1 to 31)|     +----------- Hour (0 to 23)+------------- Minute (0 to 59)

There are some shorts, too (if you don´t want the *):

@reboot --> only once at startup@daily ---> once a day@midnight --> once a day at midnight@hourly --> once a hour@weekly --> once a week@monthly --> once a month@annually --> once a year@yearly --> once a year

If you want to use the shorts as cron (because they don´t work or so):

@daily --> 0 0 * * *@midnight --> 0 0 * * *@hourly --> 0 * * * *@weekly --> 0 0 * * 0@monthly --> 0 0 1 * *@annually --> 0 0 1 1 *@yearly --> 0 0 1 1 *

What directory is file.txt in? cron runs jobs in your home directory, so unless your script cds somewhere else, that's where it's going to look for/create file.txt.

EDIT: When you refer to a file without specifying its full path (e.g. file.txt, as opposed to the full path /home/myUser/scripts/file.txt) in shell, it's taken that you're referring to a file in your current working directory. When you run a script (whether interactively or via crontab), the script's working directory has nothing at all to do with the location of the script itself; instead, it's inherited from whatever ran the script.

Thus, if you cd (change working directory) to the directory the script's in and then run it, file.txt will refer to a file in the same directory as the script. But if you don't cd there first, file.txt will refer to a file in whatever directory you happen to be in when you ran the script. For instance, if your home directory is /home/myUser, and you open a new shell and immediately run the script (as scripts/test.sh or /home/myUser/scripts/test.sh; ./test.sh won't work), it'll touch the file /home/myUser/file.txt because /home/myUser is your current working directory (and therefore the script's).

When you run a script from cron, it does essentially the same thing: it runs it with the working directory set to your home directory. Thus all file references in the script are taken relative to your home directory, unless the script cds somewhere else or specifies an absolute path to the file.

Try,

 # cat test.sh #!/bin/bash /bin/touch file.txt

cron as:

 * * * * * /bin/sh /home/myUser/scripts/test.sh

And you can confirm this by:

 # tailf /var/log/cron