Puppet HTTP API: Index

A Puppet master server provides several services via HTTP API, and the Puppet agent application uses those services to resolve a node’s credentials, retrieve a configuration catalog, retrieve file data, and submit reports.

In general, these APIs aren’t designed for use by tools other than Puppet agent, and they’ve historically relied on a lot of shared code to work correctly. This is gradually changing, although we expect external use of these APIs to remain low for the foreseeable future.

Puppet will often send garbage URL parameters, such as fail_on_404 and ignore_cache. The server will ignore any parameters it isn’t expecting.

V1/V2 HTTP APIs (Removed)

The V1 and V2 APIs were removed in Puppet 4.0.0. The routes that were previously under / or /v2.0 can now be found under the /puppet/v3 API or /puppet-ca/v1 API.

Starting with version 2.1, the Puppet Server 2.x series provides both the current and previous API endpoints, and can serve nodes running Puppet agent 3.x and 4.x. However, Rack masters, WEBrick masters, and Puppet Server 2.0 cannot serve nodes running Puppet 3.x.

Puppet and Puppet CA APIs

Beginning with Puppet 4, Puppet’s HTTP API has been split into two APIs, which are versioned separately. There is now an API for configuration-related services and a separate one for the certificate authority (CA).

All configuration endpoints are prefixed with /puppet, while all CA endpoints are prefixed with /puppet-ca. All endpoints are explicitly versioned: the prefix is always immediately followed by a string like /v3 (a directory separator, the letter v, and the version number of the API).

Authorization

Authorization for /puppet endpoints is still controlled with Puppet’s auth.conf authorization system.

Puppet Server ignores auth.conf for /puppet-ca endpoints. Access to the certificate_status endpoint is configured in Puppet Server’s ca.conf file, and the remaining CA endpoints are always accessible. Rack Puppet master servers still use auth.conf for /puppet-ca.

When specifying authorization in auth.conf, the prefix and the version number (e.g. /puppet/v3) on the paths must be retained, since Puppet matches authorization rules against the full request path.

Puppet V3 HTTP API

The Puppet agent application uses several network services to manage systems. These services are all grouped under the /puppet API. Other tools can access these services and use the Puppet master’s data for other purposes.

The V3 API contains endpoints of two types: those that are based on dispatching to Puppet’s internal “indirector” framework, and those that are not (namely the environment endpoints).

Every HTTP endpoint that dispatches to the indirector follows the form: /puppet/v3/:indirection/:key?environment=:environment where:

  • :environment is the name of the environment that should be in effect for the request. Not all endpoints need an environment, but the query parameter must always be specified.
  • :indirection is the indirection to dispatch the request to.
  • :key is the “key” portion of the indirection call.

Using this API requires significant understanding of how Puppet’s internal services are structured, but the following documents attempt to specify what is available and how to interact with it.

Configuration Management Services

These services are all directly used by the Puppet agent application, in order to manage the configuration of a node.

These endpoints accept payload formats formatted as JSON or PSON (MIME types of application/json and text/pson, respectively) except for File Content and File Bucket File which always use application/octet-stream.

Informational Services

These services are not directly used by Puppet agent, but may be used by other tools.

Environment Endpoints

The endpoints with a different format are the /puppet/v3/environments and the /puppet/v3/environment/:environment endpoints.

These endpoints will only accept payloads formatted as JSON and respond with JSON (MIME type of application/json).

Puppet Server-specific endpoints

When using Puppet Server 2.3 or newer as a Puppet master, Puppet Server adds additional /puppet/v3/ endpoints:

Error Responses

The environments endpoint will respond to error conditions in a uniform manner and use standard HTTP response code to signify those errors.

  • When the client submits a malformed request, the API will return a 400 Bad Request response.
  • When the client is not authorized, the API will return a 403 Not Authorized response.
  • When the client attempts to use an HTTP method that is not permissible for the endpoint, the API will return a 405 Method Not Allowed response.
  • When the client asks for a response in a format other than JSON, the API will return a 406 Unacceptable response.
  • When the server encounters an unexpected error during the handling of a request, it will return a 500 Server Error response.
  • When the server is unable to find an endpoint handler for an http request, it will return a 404 Not Found response

All error responses will contain a body, except when it is a HEAD request. The error responses will uniformly be a JSON object with the following properties:

  • message: (String) A human readable message explaining the error.
  • issue_kind: (String) A unique label to identify the error class.

A JSON schema for the error objects is also available.

CA V1 HTTP API

The CA API contains all of the endpoints used in support of Puppet’s PKI system.

The CA V1 endpoints share the same basic format as the Puppet V3 API, since they are also based off of Puppet’s internal “indirector”. However, they have a different prefix and version. The endpoints thus follow the form: /puppet-ca/v1/:indirection/:key?environment=:environment where:

  • :environment is an arbitrary placeholder word, required for historical reasons. No CA endpoints actually use an environment, but the query parameter must always be specified.
  • :indirection is the indirection to dispatch the request to.
  • :key is the “key” portion of the indirection call.

As with the Puppet V3 API, using this API requires a significant amount of understanding of how Puppet’s internal services are structured. The following documents provide additional specification.

These endpoints only accept plain text payload formats. Historically, Puppet has used the MIME type s to mean text/plain. In Puppet 5, it will always use text/plain, but will continue to accept s to mean the same thing.

Serialization Formats

Puppet sends messages using several different serialization formats. Not all REST services support all of the formats.

YAML was supported in earlier versions of Puppet, but is no longer for security reasons.