Skip to content

GraphQL

Nautobot supports a Read-Only GraphQL interface that can be used to query most information present in the database. The GraphQL interface is available at the endpoint graphql/ for a human to explore and GraphQL can be queried as an API via the endpoint api/graphql/. Currently the support for GraphQL is limited to query, other operation type like mutations and subscriptions are not supported. Additionally, GraphQL variables are supported.

The GraphQL implementation is leveraging the graphene-django library and supports the standard GraphQL language.

How to use the GraphQL interface

The GraphQL interface can be used to query multiple tables at once in a single request. In GraphQL, only the information requested will be returned which can be contrasted to REST APIs. In the example below, this query will return the name of all interfaces attached to the device nyc-sw01 along with all ip_addresses attached to those interfaces.

query {
  devices(name: "nyc-sw01") {
    name
    interfaces {
      name
      ip_addresses {
        address
      }
    }
  }
}

Result

{
  "data": {
    "devices": [
      {
        "name": "nyc-sw01",
        "interfaces": [
          {
            "name": "xe-0/0/0",
            "ip_addresses": [
              {
                "address": "10.52.0.1/30"
              }
            ]
          },
          {
            "name": "xe-0/0/1",
            "ip_addresses": []
          }
        ]
      }
    ]
  }
}

It is possible to explore the Graph and create some queries in a human friendly UI at the endpoint graphql/. This interface (called graphqli) provides a great playground to build new queries as it provides full autocompletion and type validation.

Querying the GraphQL interface over the rest API

It is possible to query the GraphQL interface via the rest API as well, the endpoint is available at api/graphql/ and supports the same Token based authentication as all other Nautobot APIs.

A GraphQL Query must be encapsulated in a JSON payload with the query key and sent with a POST request. Optionally it is possible to provide a list of variables in the same payload as presented below.

{
  "query": "query ($id: Int!) { device(id: $id) { name }}",
  "variables": { "id": 3}
}

Working with Custom Fields

GraphQL custom fields data data is provided in two formats, a "greedy" and a "prefixed" format. The greedy format provides all custom field data associated with this record under a single "custom_field_data" key. This is helpful in situations where custom fields are likely to be added at a later date, the data will simply be added to the same root key and immediately accessible without the need to adjust the query.

query {
  sites {
    name
    custom_field_data
  }
}

Result

{
  "data": {
    "sites": [
      {
        "name": "nyc-site-01",
        "custom_field_data": {
          "site_type": "large"
        }
      },
      {
        "name": "nyc-site-02",
        "custom_field_data": {
          "site_type": "small"
        }
      }
    ]
  }
}

Additionally, by default, all custom fields in GraphQL will be prefixed with cf_. A custom field name site_type will appear in GraphQL as cf_site_type as an example. The prefix can be changed by setting the value of GRAPHQL_CUSTOM_FIELD_PREFIX.

query {
  sites {
    name
    cf_site_type
  }
}

Result

{
  "data": {
    "sites": [
      {
        "name": "nyc-site-01",
        "cf_site_type": "large"
      },
      {
        "name": "nyc-site-02",
        "cf_site_type": "small"
      }
    ]
  }
}

Important

Custom Fields with the prefixed cf_ are only available in GraphQL after the custom field is created and the web service is restarted.

Working with Relationships

Defined relationships are available in GraphQL as well. In most cases, the associated objects for a given relationship will be available under the key rel_<relationship_slug>. The one exception is for relationships between objects of the same type that are not defined as symmetric; for these relationships it's important to be able to distinguish between the two "sides" of the relationship, and so the associated objects will be available under rel_<relationship_slug>_source and/or rel_<relationship_slug>_destination as appropriate.

Important

Relationships are only available in GraphQL after the relationship is created and the web service is restarted.

query {
  ip_addresses {
    address
    rel_peer_address {
      address
    }
    rel_parent_child_source {
      address
    }
    rel_parent_child_destination {
      address
    }
  }
}

Result

{
  "data": {
    "ip_addresses": [
      {
        "address": "10.1.1.1/24",
        "rel_peer_address": {
          "address": "10.1.1.2/24"
        },
        "rel_parent_child_source": null,
        "rel_parent_child_destination": [
          {
            "address": "10.1.1.1/30"
          },
          {
            "address": "10.1.1.1/32"
          }
        ]
      },
      {
        "address": "10.1.1.1/30",
        "rel_peer_address": null,
        "rel_parent_child_source": {
          "address": "10.1.1.1/24"
        },
        "rel_parent_child_destination": []
      },
      {
        "address": "10.1.1.1/32",
        "rel_peer_address": null,
        "rel_parent_child_source": {
          "address": "10.1.1.1/24"
        },
        "rel_parent_child_destination": []
      },
      {
        "address": "10.1.1.2/24",
        "rel_peer_address": {
          "address": "10.1.1.1/24"
        },
        "rel_parent_child_source": null,
        "rel_parent_child_destination": []
      }
    ]
  }
}

Working with Computed Fields

By default, all custom fields in GraphQL will be prefixed with cpf_. A computed field name ip_ptr_record will appear in GraphQL as cpf_ip_ptr_record as an example. The prefix can be changed by setting the value of GRAPHQL_COMPUTED_FIELD_PREFIX.

{
  ip_addresses {
    address
    dns_name
    cpf_ip_ptr_record
  }
}

Result

{
  "data": {
    "ip_addresses": [
      {
        "address": "10.0.0.0/32",
        "dns_name": "ip-10-0-0-0.server.atl01.atc.nautobot.com",
        "cpf_ip_ptr_record": "0.0.0.10.in-addr.arpa"
      },
      {
        "address": "10.0.1.0/32",
        "dns_name": "ip-10-0-1-0.server.atl01.atc.nautobot.com",
        "cpf_ip_ptr_record": "0.1.0.10.in-addr.arpa"
      }
    ]
  }
}

Important

Computed Fields with the prefixed cpf_ are only available in GraphQL after the computed field is created and the web service is restarted.

Saved Queries

Added in version 1.1.0

Queries can now be stored inside of Nautobot, allowing the user to easily rerun previously defined queries.

Inside of Extensibility -> Data Management -> GraphQL Queries, there are views to create and manage GraphQL queries.

Saved queries can be executed from the detailed query view or via a REST API request. The queries can also be populated from the detailed query view into GraphiQL by using the "Open in GraphiQL" button. Additionally, in the GraphiQL UI, there is now a menu item, "Queries", which can be used to populate GraphiQL with any previously saved query.

To execute a stored query via the REST API, a POST request can be sent to /api/extras/graphql-queries/[slug]/run/. Any GraphQL variables required by the query can be passed in as JSON data within the request body.