Using the Server Jobs Add-in
Overview
The Server Jobs add-in allows you to define and schedule regular server tasks (shell commands). Under the hood, these are run as cronjobs.
Adding a new server job
To add a new server job:
- Click the + button on the add-ins panel of your Application Overview
- Click the Install Now button on the Server Jobs add-in block.
- Choose a server and give your new job a name
- Specify the shell command you want to run
- Choose which (Linux) user should run the command (the options are
cloud66-userornginx) - Set a schedule for the job, or set it to run only on demand. (You can use cron syntax for more control over your scheduling)
- Clicks the green Create Job button
You will now see your new job listed in your Application Overview. You can edit it, or run it on demand by clicking on the small downward arrow. Your run results (success, failure and any output) can be seen in real-time on the job detail page.
Using parameters
When you are running a job on demand via dashboard or toolbelt, you can pass it parameters (if it is written to accept them).
Notation
Jobs use a facility in the shell called positional parameters. Positional parameters are a series of special variables ($1, $2 … $n) that contain the contents of the command line. Where n is greater than 9 using braces. For example, to refer to the 15th positional parameter, use the notation ${15}.
cp $1 $2
ls $1 $2 $3 $4 $5 $6 $7 $8 $9 ${10} ${11}
Default values
You can handle default value for the parameter with following notation: ${n:-YOUR_DEFAULT_VALUE}. This is useful when you have ha job that scheduled and also you need to run it on demand sometimes.
Example below cp file $1 to tmp directory by default
cp $1 ${2:-/tmp}
Passing parameters to job
Since job is using positional parameters pass you arguments in order, eg: if you pass arg1 arg2, $1 would contain arg1 and $2 would contain arg2
You can also quote your argument if there is a space in the value.
Example
| Job command | cp $1 ${2:-/tmp} |
| Passing arguments via dashboard | "log*.txt" tmp/logs |
| Passing arguments via Toolbelt | --arg "log*.txt" -- arg tmp/logs |
Support for whenever gem
We also support the whenever gem for managing CRON jobs. However, we recommend using our native Jobs Add-in (see above) instead. The Add-in has many benefits:
- It allows you to choose which server these tasks run on
- It gives visibility in the UI when tasks are running, and displays their outputs
- You can set tasks to run on-demand
- It alerts you if tasks fail
If you’d prefer to use the gem, simply add whenever to your Gemfile and redploy your code. We will automatically use the config/schedule.rb in your source code to build the CRON jobs on the relevant servers. You should see the message “Handling whenever schedule” on your server detail page log.
