Running plans in PE
The orchestrator can run plans across systems in your infrastructure. You can set up plan jobs from the Puppet Enterprise (PE) console or the command line. Plan jobs can run once or on a recurring schedule.
Plans can't change your Puppet configuration, which defines the desired state of your infrastructure. If you run a plan that changes the state of a resource managed by Puppet (for example, upgrading a package or service), the next Puppet run changes that resource's state back to what is defined in your Puppet configuration.
If you have set up compilers and you want to use plans, you must set either
primary_uris or server_list on
your agents to point to your compilers. This setting is described in the
section on configuring compilers for orchestrator scale.
Plans in PE require Managing code with Code Manager.
Unlike Tasks in PE, you cannot stop a plan that is in progress.
-
Running plans from the console
Run ad hoc plans from the console.
-
Running plans from the command line
Run a plan using the
puppet plan runcommand. -
You can inspect plan metadata, outcomes of plan events, a list of all installed plans, and a list of plans you have permission to run.
-
Running plans alongside code deployments
The orchestrator's file sync client has a built-in locking mechanism that ensures your plans run in a consistent environment state. The locking mechanism prevents plans from starting while a code deployment is in progress, and it prevents new code deployments from synchronizing while a plan is running. You can disable this locking mechanism if you want to run plans and deploy code simultaneously. Consider the tradeoffs before deciding whether to disable the file sync locking mechanism.
