Question

I run these three commands everyday:

cd /Users/xxx/code
heroku pgbackups:capture --expire
curl -o latest.dump `heroku pgbackups:url`

So I thought I’d try to move it into a cron job so I created a file called /Users/xxx/code/backup_script that has the following:

cd /Users/xxx/code
heroku pgbackups:capture --expire
curl -o latest.dump `heroku pgbackups:url`

However, when I run ./backup_script it gives this error:

./backup_script
./backup_script: line 2: heroku: command not found
./backup_script: line 3: heroku: command not found
curl: no URL specified!
curl: try 'curl --help' or 'curl --manual' for more information

I’m a unix newbie – can someone help me figure out how to fix the script above?

Answer 1

The environment variables for a cronjob are significantly reduced from your login; you’ll want to set $PATH for your crontab.

For instance:

PATH=$HOME/bin:$PATH
@daily $HOME/backup_script

You may want to run which heroku to figure out what path you need to add.