The main interface to Nomad is a RESTful HTTP API. The API can query the current state of the system as well as modify the state of the system. The Nomad CLI actually invokes Nomad's HTTP for many commands.
All API routes are prefixed with
This documentation is only for the v1 API.
Backwards compatibility: At the current version, Nomad does not yet promise backwards compatibility even with the v1 prefix. We'll remove this warning when this policy changes. We expect to reach API stability by Nomad 1.0.
Nomad binds to a specific set of addresses and ports. The HTTP API is served via
http address and port. This
address:port must be accessible locally. If
you bind to
127.0.0.1:4646, the API is only available from that host. If you
bind to a private internal IP, the API will be available from within that
network. If you bind to a public IP, the API will be available from the public
Internet (not recommended).
The default port for the Nomad HTTP API is
4646. This can be overridden via
the Nomad configuration block. Here is an example curl request to query a Nomad
server with the default configuration:
$ curl http://127.0.0.1:4646/v1/agent/members
The conventions used in the API documentation do not list a port and use the
localhost:4646. Be sure to replace this with your Nomad agent URL
when using the examples.
There are five primary nouns in Nomad:
Jobs are submitted by users and represent a desired state. A job is a declarative description of tasks to run which are bounded by constraints and require resources. Jobs can also have affinities which are used to express placement preferences. Nodes are the servers in the clusters that tasks can be scheduled on. The mapping of tasks in a job to nodes is done using allocations. An allocation is used to declare that a set of tasks in a job should be run on a particular node. Scheduling is the process of determining the appropriate allocations and is done as part of an evaluation. Deployments are objects to track a rolling update of allocations between two versions of a job.
The API is modeled closely on the underlying data model. Use the links to the left for documentation about specific endpoints. There are also "Agent" APIs which interact with a specific agent and not the broader cluster used for administration.
Several endpoints in Nomad use or require ACL tokens to operate. The token are used to authenticate the request and determine if the request is allowed based on the associated authorizations. Tokens are specified per-request by using the
X-Nomad-Token request header set to the
SecretID of an ACL Token.
For more details about ACLs, please see the ACL Guide.
When ACLs are enabled, a Nomad token should be provided to API requests using the
X-Nomad-Token header. When using authentication, clients should communicate via TLS.
Here is an example using curl:
$ curl \ --header "X-Nomad-Token: aa534e09-6a07-0a45-2295-a7f77063d429" \ https://localhost:4646/v1/jobs
Nomad has support for namespaces, which allow jobs and their associated objects to be segmented from each other and other users of the cluster. When using non-default namespace, the API request must pass the target namespace as an API query parameter. Prior to Nomad 1.0 namespaces were Enterprise-only.
Here is an example using curl:
$ curl \ https://localhost:4646/v1/jobs?namespace=qa
Many endpoints in Nomad support a feature known as "blocking queries". A blocking query is used to wait for a potential change using long polling. Not all endpoints support blocking, but each endpoint uniquely documents its support for blocking queries in the documentation.
Endpoints that support blocking queries return an HTTP header named
X-Nomad-Index. This is a unique identifier representing the current state of
the requested resource. On a new Nomad cluster the value of this index starts at 1.
On subsequent requests for this resource, the client can set the
string parameter to the value of
X-Nomad-Index, indicating that the client
wishes to wait for any changes subsequent to that index.
When this is provided, the HTTP request will "hang" until a change in the system occurs, or the maximum timeout is reached. A critical note is that the return of a blocking request is no guarantee of a change. It is possible that the timeout was reached or that there was an idempotent write that does not affect the result of the query.
In addition to
index, endpoints that support blocking will also honor a
parameter specifying a maximum duration for the blocking request. This is
limited to 10 minutes. If not set, the wait time defaults to 5 minutes. This
value can be specified in the form of "10s" or "5m" (i.e., 10 seconds or 5
minutes, respectively). A small random amount of additional wait time is added
to the supplied maximum
wait time to spread out the wake up time of any
concurrent requests. This adds up to
wait / 16 additional time to the maximum
Most of the read query endpoints support multiple levels of consistency. Since no policy will suit all clients' needs, these consistency modes allow the user to have the ultimate say in how to balance the trade-offs inherent in a distributed system.
The two read modes are:
default- If not specified, the default is strongly consistent in almost all cases. However, there is a small window in which a new leader may be elected during which the old leader may service stale values. The trade-off is fast reads but potentially stale values. The condition resulting in stale reads is hard to trigger, and most clients should not need to worry about this case. Also, note that this race condition only applies to reads, not writes.
stale- This mode allows any server to service the read regardless of whether it is the leader. This means reads can be arbitrarily stale; however, results are generally consistent to within 50 milliseconds of the leader. The trade-off is very fast and scalable reads with a higher likelihood of stale values. Since this mode allows reads without a leader, a cluster that is unavailable will still be able to respond to queries.
To switch these modes, use the
stale query parameter on requests.
To support bounding the acceptable staleness of data, responses provide the
X-Nomad-LastContact header containing the time in milliseconds that a server
was last contacted by the leader node. The
X-Nomad-KnownLeader header also
indicates if there is a known leader. These can be used by clients to gauge the
staleness of a result and take appropriate action.
By default, any request to the HTTP API will default to the region on which the
machine is servicing the request. If the agent runs in "region1", the request
will query the region "region1". A target region can be explicitly request using
?region query parameter. The request will be transparently forwarded and
serviced by a server in the requested region.
The HTTP API will gzip the response if the HTTP request denotes that the client accepts gzip compression. This is achieved by passing the accept encoding:
$ curl \ --header "Accept-Encoding: gzip" \ https://localhost:4646/v1/...
By default, the output of all HTTP API requests is minimized JSON. If the client
pretty on the query string, formatted JSON will be returned.
In general, clients should prefer a client-side parser like
jq instead of
server-formatted data. Asking the server to format the data takes away
processing cycles from more important tasks.
$ curl https://localhost:4646/v1/page?pretty
Nomad's API aims to be RESTful, although there are some exceptions. The API responds to the standard HTTP verbs GET, PUT, and DELETE. Each API method will clearly document the verb(s) it responds to and the generated response. The same path with different verbs may trigger different behavior. For example:
PUT /v1/jobs GET /v1/jobs
Even though these share a path, the
PUT operation creates a new job whereas
GET operation reads all jobs.
Individual API's will contain further documentation in the case that more specific response codes are returned but all clients should handle the following:
- 200 and 204 as success codes.
- 400 indicates a validation failure and if a parameter is modified in the request, it could potentially succeed.
- 403 marks that the client isn't authenticated for the request.
- 404 indicates an unknown resource.
- 5xx means that the client should not expect the request to succeed if retried.