Tasks in PE
Tasks are ad-hoc actions you can execute on a target and run from the command line or the console.
A task is a single action that you execute on target machines. With tasks, you can troubleshoot and deploy changes to individual or multiple systems in your infrastructure.
You can run tasks from your tool of choice: the console, the orchestrator command line
interface (CLI), or the orchestrator API /command/task
endpoint.
When you run a task, you can run it immediately, schedule it to run later, or schedule it to run at a recurring frequency - hourly, daily, weekly, every 2 weeks, or every four weeks. After you launch a task, you can check on the status or view the output later with the console or CLI.
If a reboot occurs or if you need to restore a backup, scheduled tasks are rescheduled based on the last execution time. If the scheduled task running in the orchestrator is what caused the reboot, the task run appears as failed.
task_concurrency
and bolt_server::concurrency
limits can accommodate your needs. To adjust these settings, go to Orchestrator and pe-orchestration-services parameters.-
Installing tasks
Puppet Enterprise comes with some pre-installed tasks, and you can install or write other tasks you want to use. -
Running tasks in PE
Use the orchestrator to set up jobs in the console or on the command line and run Bolt tasks across systems in your infrastructure. -
Writing tasks
Bolt tasks are similar to scripts, but they are kept in modules and can have metadata. This allows you to reuse and share them.