open-nomad/website/source/docs/jobspec/index.html.md

layout	page_title	sidebar_current	description
docs	Job Specification	docs-jobspec	Learn about the Job specification used to submit jobs to Nomad.

Job Specification

Jobs can be specified either in HCL or JSON. HCL is meant to strike a balance between human readable and editable, as well as being machine-friendly.

For machine-friendliness, Nomad can also read JSON configurations. In general, we recommend using the HCL syntax.

HCL Syntax

For a detailed description of HCL general syntax, see this guide. Here we cover the details of the Job specification for Nomad:

# Define a job called my-service
job "my-service" {
    # Job should run in the US region
    region = "us"

    # Spread tasks between us-west-1 and us-east-1
    datacenters = ["us-west-1", "us-east-1"]

    # Rolling updates should be sequential
    update {
        stagger = "30s"
        max_parallel = 1
    }

    group "webs" {
        # We want 5 web servers
        count = 5

        # Create a web front end using a docker image
        task "frontend" {
            driver = "docker"
            config {
                image = "hashicorp/web-frontend"
            }
            resources {
                cpu = 500
                memory = 128
                network {
                    mbits = 100
                    dynamic_ports = [
                      "http",
                      "https",
                    ]
                }
            }
        }
    }
}

This is a fairly simple example job, but demonstrates many of the features and syntax of the job specification. The primary "objects" are the job, task group, and task. Each job file has only a single job, however a job may have multiple task groups, and each task group may have multiple tasks. Task groups are a set of tasks that must be co-located on a machine. Groups with a single task and count of one can be declared outside of a group which is created implicitly.

Constraints can be specified at the job, task group, or task level to restrict where a task is eligible for running. An example constraint looks like:

# Restrict to only nodes running linux
constraint {
    attribute = "$attr.kernel.os"
    value = "linux"
}

Jobs can also specify additional metadata at the job, task group, or task level. This metadata is opaque to Nomad and can be used for any purpose, including defining constraints on the metadata. Metadata can be specified by:

# Setup ELB via metadata and setup foo
meta {
    foo = "bar"
    elb_mode = "tcp"
    elb_check_interval = "10s"
}

Syntax Reference

Following is a syntax reference for the possible keys that are supported and their default values if any for each type of object.

Job

The job object supports the following keys:

• all_at_once - Controls if the entire set of tasks in the job must be placed atomically or if they can be scheduled incrementally. This should only be used for special circumstances. Defaults to false.

• constraint - This can be provided multiple times to define additional constraints. See the constraint reference for more details.

• datacenters - A list of datacenters in the region which are eligible for task placement. This must be provided, and does not have a default.

• group - This can be provided multiple times to define additional task groups. See the task group reference for more details.

• meta - Annotates the job with opaque metadata.

• priority - Specifies the job priority which is used to prioritize scheduling and access to resources. Must be between 1 and 100 inclusively, and defaults to 50.

• region - The region to run the job in, defaults to "global".

• task - This can be specified multiple times to add a task as part of the job. Tasks defined directly in a job are wrapped in a task group of the same name.

• type - Specifies the job type and switches which scheduler is used. Nomad provides the service and batch schedulers, and defaults to service.

• update - Specifies the task update strategy. This requires providing max_parallel as an integer and stagger as a time duration. If stagger is provided as an integer, seconds are assumed. Otherwise the "s", "m", and "h" suffix can be used, such as "30s". Both values default to zero, which disables rolling updates.

Task Group

The group object supports the following keys:

• count - Specifies the number of the task groups that should be running. Must be positive, defaults to one.

• constraint - This can be provided multiple times to define additional constraints. See the constraint reference for more details.

• task - This can be specified multiple times, to add a task as part of the group.

• meta - Annotates the task group with opaque metadata.

Task

The task object supports the following keys:

• driver - Specifies the task driver that should be used to run the task. See the driver documentation for what is available. Examples include "docker", "qemu", "java", and "exec".

• constraint - This can be provided multiple times to define additional constraints. See the constraint reference for more details.

• config - A map of key/value configuration passed into the driver to start the task. The details of configurations are specific to each driver.

• resources - Provides the resource requirements of the task. See the resources reference for more details.

• meta - Annotates the task group with opaque metadata.

Resources

The resources object supports the following keys:

• cpu - The CPU required in MHz.

• disk - The disk required in MB.

• iops - The number of IOPS required.

• memory - The memory required in MB.

• network - The network required. Details below.

The network object supports the following keys:

• dynamic_ports - List of port labels which may contain letters, numbers and underscores (^[a-zA-Z0-9_]+$). Each label will be assigned a dynamic port when the task starts. Ports are passed to the task environment as NOMAD_PORT_{LABEL}. Drivers may infer additional semantics from the label. See the relevant driver docs for details.

• mbits - The number of MBits in bandwidth required.

• reserved_ports - This is a list of specific ports required. For applications that cannot use a dynamic port, they can request a specific port.

Constraint

The constraint object supports the following keys:

• attribute - Specifies the attribute to examine for the constraint. See the table of attributes below.

• hard - Specifies if this is a hard or soft constraint. Defaults to true. Soft constraints are not currently supported.

• operator - Specifies the comparison operator. Defaults to equality, and can be =, ==, is, !=, not.

• value - Specifies the value to compare the attribute against. This can be a literal value or another attribute.

Below is a table documenting the variables that can be interpreted:

Variable	Description
$node.id	The client node identifier
$node.datacenter	The client node datacenter
$node.name	The client node name
$attr.\	The attribute given by `key` on the client node.
$meta.\	The metadata value given by `key` on the client node.

Below is a table documenting common node attributes:

Attribute	Description
arch	CPU architecture of the client. Examples: "amd64", "386"
consul.datacenter	The Consul datacenter of the client node if Consul found
cpu.numcores	Number of CPU cores on the client
driver.\	See the [task drivers](/docs/drivers/index.html) for attribute documentation
hostname	Hostname of the client
platform.aws.ami-id	On EC2, the AMI ID of the client node
platform.aws.instance-type	On EC2, the instance type of the client node
os.name	Operating system of the client. Examples: "linux", "windows", "darwin"
os.version	Version of the client OS

JSON Syntax

Job files can also be specified in JSON. The conversion is straightforward and self-documented. The downsides of JSON are less human readability and the lack of comments. Otherwise, the two are completely interoperable.

See the API documentation for more details on the JSON structure.