I recently started moving some personal stuff up to Github, which brought me to the issue of hard coded information that needed to be scrubbed. I am using Githubs private repositories, but still feel some kind of way about putting certain types of things in the repo. Mostly webhook urls, usernames, email addresses, etc. The general solution I came aross was using environment variables that you set on the server running the scripts. This also solves a common issue where when something hardcoded does change, you have to update multiple scripts, where as using a variable you can just update it on the system and all scripts have the new information.

The work

No problem, I update my scripts to use variables. I set those variables and test the script, no issues, so I set a Cron job. I am all set. The following day I check in to verify everything ran smoothly. Nothing. I manually kick of the script. Success. Wait on the cron. Nothing.

Come to find out, Cron jobs do not import all environment variables. Luckily they do provide a way to provide variables to Cron.

Using crontab -e you can set variables just as you would in /etc/environment

Add your variables to the top of the file.

var1=/home/user/scripts  
var2=https://webhook.url

Save the changes.
Now when you run a cron job it will have access to those variables that you set.

⤧  Next post SFTP Server with chroot ⤧  Previous post Using PSExec for a remote command prompt