Amazon ECS task definition parameters for the Fargate launch type

Type: String

Required: Yes

When you register a task definition, you give it a family, which is similar to a name for multiple versions of the task definition, specified with a revision number. The first task definition that's registered into a particular family is given a revision of 1, and any task definitions registered after that are given a sequential revision number.

Type: String array

Required: No

Valid Values: FARGATE

The launch type to validate the task definition against. This initiates a check to ensure that all of the parameters that are used in the task definition meet the requirements of the launch type.

Type: String

Required: No

When you register a task definition, you can provide a task role for an IAM role that allows the containers in the task permission to call the Amazon APIs that are specified in its associated policies on your behalf. For more information, see Amazon ECS task IAM role.

Type: String

Required: Conditional

The Amazon Resource Name (ARN) of the task execution role that grants the Amazon ECS container agent permission to make Amazon API calls on your behalf.

Type: String

Required: Yes

The Docker networking mode to use for the containers in the task. For Amazon ECS tasks hosted on Fargate, the awsvpc network mode is required.

Type: String

Required: Conditional

Default: LINUX

This parameter is required for Amazon ECS tasks that are hosted on Fargate.

When you register a task definition, you specify the operating system family.

The valid values are LINUX, WINDOWS_SERVER_2025_FULL, WINDOWS_SERVER_2025_CORE, WINDOWS_SERVER_2022_FULL, WINDOWS_SERVER_2022_CORE, WINDOWS_SERVER_2019_FULL, and WINDOWS_SERVER_2019_CORE.

All task definitions that are used in a service must have the same value for this parameter.

When a task definition is part of a service, this value must match the service platformFamily value.

Type: String

Required: Conditional

Default: X86_64

If the parameter is left as null, the default value is automatically assigned upon the initiation of a task hosted on Fargate.

When you register a task definition, you specify the CPU architecture. The valid values are X86_64 and ARM64.

All task definitions that are used in a service must have the same value for this parameter.

When you have Linux tasks, you can set the value to ARM64. For more information, see Amazon ECS task definitions for 64-bit ARM workloads.

Type: String

Required: Yes

The hard limit of CPU units to present for the task. You can specify CPU values in the JSON file as a string in CPU units or virtual CPUs (vCPUs). For example, you can specify a CPU value either as 1024 in CPU units or 1 vCPU in vCPUs. When the task definition is registered, a vCPU value is converted to an integer indicating the CPU units.

This field is required and you must use one of the following values, which determines your range of supported values for the memory parameter. The table below shows the valid combinations of task-level CPU and memory.

Type: String

Required: Yes

The hard limit of memory to present to the task. You can specify memory values in the task definition as a string in mebibytes (MiB) or gigabytes (GB). For example, you can specify a memory value either as 3072 in MiB or 3 GBin GB. When the task definition is registered, a GB value is converted to an integer indicating the MiB.

This field is required and you must use one of the following values, which determines your range of supported values for the cpu parameter:

Type: String

Required: Yes

The name of a container. Up to 255 letters (uppercase and lowercase), numbers, hyphens, and underscores are allowed. If you're linking multiple containers in a task definition, the name of one container can be entered in the links of another container. This is to connect the containers.

Type: String

Required: Yes

The image used to start a container. This string is passed directly to the Docker daemon. By default, images in the Docker Hub registry are available. You can also specify other repositories with either repository-url/image:tag or repository-url/image@digest. Up to 255 letters (uppercase and lowercase), numbers, hyphens, underscores, colons, periods, forward slashes, and number signs are allowed. This parameter maps to Image in the docker create-container command and the IMAGE parameter of the docker run command.

Type: String

Valid values: enabled|disabled

Required: No

Specifies whether Amazon ECS will resolve the container image tag provided in the container definition to an image digest. By default, this behavior is enabled. If you set the value for a container as disabled, Amazon ECS will not resolve the container image tag to a digest and will use the original image URI specified in the container definition for deployment. For more information about container image resolution, see Container image resolution.

Type: Integer

Required: No

The amount (in MiB) of memory to present to the container. If your container attempts to exceed the memory specified here, the container is killed. The total amount of memory reserved for all containers within a task must be lower than the task memory value, if one is specified. This parameter maps to Memory in the docker create-container command and the --memory option to docker run.

The Docker 20.10.0 or later daemon reserves a minimum of 6 MiB of memory for a container. So, don't specify less than 6 MiB of memory for your containers.

The Docker 19.03.13-ce or earlier daemon reserves a minimum of 4 MiB of memory for a container. So, don't specify less than 4 MiB of memory for your containers.

Type: Integer

Required: No

The soft limit (in MiB) of memory to reserve for the container. When system memory is under contention, Docker attempts to keep the container memory to this soft limit. However, your container can use more memory when needed. The container can use up to the hard limit that's specified with the memory parameter (if applicable) or all of the available memory on the container instance, whichever comes first. This parameter maps to MemoryReservation in the docker create-container command and the --memory-reservation option to docker run.

If a task-level memory value isn't specified, you must specify a non-zero integer for one or both of memory or memoryReservation in a container definition. If you specify both, memory must be greater than memoryReservation. If you specify memoryReservation, then that value is subtracted from the available memory resources for the container instance that the container is placed on. Otherwise, the value of memory is used.

For example, suppose that your container normally uses 128 MiB of memory, but occasionally bursts to 256 MiB of memory for short periods of time. You can set a memoryReservation of 128 MiB, and a memory hard limit of 300 MiB. This configuration allows the container to only reserve 128 MiB of memory from the remaining resources on the container instance. At the same time, this configuration also allows the container to use more memory resources when needed.

The Docker 20.10.0 or later daemon reserves a minimum of 6 MiB of memory for a container. So, don't specify less than 6 MiB of memory for your containers.

The Docker 19.03.13-ce or earlier daemon reserves a minimum of 4 MiB of memory for a container. So, don't specify less than 4 MiB of memory for your containers.

Type: Object array

Required: No

Port mappings expose your container's network ports to the outside world. this allows clients to access your application. It's also used for inter-container communication within the same task.

For task definitions that use the awsvpc network mode, only specify the containerPort. The hostPort is always ignored, and the container port is automatically mapped to a random high-numbered port on the host.

Port mappings on Windows use the NetNAT gateway address rather than localhost. There's no loopback for port mappings on Windows, so you can't access a container's mapped port from the host itself.

Most fields of this parameter (including containerPort, hostPort, protocol) map to PortBindings in thedocker create-container command and the --publish option to docker run. If the network mode of a task definition is set to host, host ports must either be undefined or match the container port in the port mapping.

If you're specifying a host port, use the following syntax.

"portMappings": [
    {
        "containerPort": integer,
        "hostPort": integer
    }
    ...
]

If you want an automatically assigned host port, use the following syntax.

"portMappings": [
    {
        "containerPort": integer
    }
    ...
]

Type: RepositoryCredentials object

Required: No

The repository credentials for private registry authentication.

For more information, see Using non-Amazon container images in Amazon ECS.

The container restart policy and associated configuration parameters. When you set up a restart policy for a container, Amazon ECS can restart the container without needing to replace the task. For more information, see Restart individual containers in Amazon ECS tasks with container restart policies.

The container health check command and the associated configuration parameters for the container. For more information, see Determine Amazon ECS task health using container health checks.

Type: Integer

Required: No

The number of cpu units the Amazon ECS container agent reserves for the container. On Linux, this parameter maps to CpuShares in the Create a container section.

This field is optional for tasks that use the Fargate launch type. The total amount of CPU reserved for all the containers that are within a task must be lower than the task-level cpu value.

Linux containers share unallocated CPU units with other containers on the container instance with the same ratio as their allocated amount. For example, assume that you run a single-container task on a single-core instance type with 512 CPU units specified for that container. Moreover, that task is the only task running on the container instance. In this example, the container can use the full 1,024 CPU unit share at any given time. However, assume then that you launched another copy of the same task on that container instance. Each task is guaranteed a minimum of 512 CPU units when needed. Similarly, if the other container isn't using the remaining CPU, each container can float to higher CPU usage. However, if both tasks were 100% active all of the time, they are limited to 512 CPU units.

On Linux container instances, the Docker daemon on the container instance uses the CPU value to calculate the relative CPU share ratios for running containers. The minimum valid CPU share value that the Linux kernel allows is 2, and the maximum valid CPU share value that the Linux kernel allows is 262144. However, the CPU parameter isn't required, and you can use CPU values below two and above 262144 in your container definitions. For CPU values below two (including null) and above 262144, the behavior varies based on your Amazon ECS container agent version:

On Windows container instances, the CPU quota is enforced as an absolute quota. Windows containers only have access to the specified amount of CPU that's defined in the task definition. A null or zero CPU value is passed to Docker as 0. Windows then interprets this value as 1% of one CPU.

For more examples, see How Amazon ECS manages CPU and memory resources.

This parameter isn't supported for containers that are hosted on Fargate.

Type: ResourceRequirement object

Required: No

The number of physical GPUs that the Amazon ECS container agent reserves for the container. The number of GPUs reserved for all containers in a task must not exceed the number of available GPUs on the container instance the task is launched on. For more information, see Amazon ECS task definitions for GPU workloads.

This parameter isn't supported for containers that are hosted on Fargate.

Type: ResourceRequirement object

Required: No

For the InferenceAccelerator type, the value matches the deviceName for an InferenceAccelerator specified in a task definition. For more information, see Elastic Inference accelerator name.

Type: Boolean

Required: No

Suppose that the essential parameter of a container is marked as true, and that container fails or stops for any reason. Then, all other containers that are part of the task are stopped. If the essential parameter of a container is marked as false, then its failure doesn't affect the rest of the containers in a task. If this parameter is omitted, a container is assumed to be essential.

All tasks must have at least one essential container. Suppose that you have an application that's composed of multiple containers. Then, group containers that are used for a common purpose into components, and separate the different components into multiple task definitions. For more information, see Architect your application for Amazon ECS.

"essential": true|false

Type: String array

Required: No

The entry point that's passed to the container.

"entryPoint": ["string", ...]

Type: String array

Required: No

The command that's passed to the container. This parameter maps to Cmd in the create-container command and the COMMAND parameter to docker run. If there are multiple arguments, make sure that each argument is a separated string in the array.

"command": ["string", ...]

Type: String

Required: No

The working directory to run commands inside the container in. This parameter maps to WorkingDir in the Create a container section of the Docker Remote API and the --workdir option to docker run.

"workingDirectory": "string"

This isn't available for Windows containers on Fargate.

Type: Object array

Required: No

A list of files containing the environment variables to pass to a container. This parameter maps to the --env-file option to the docker run command.

You can specify up to 10 environment files. The file must have a .env file extension. Each line in an environment file contains an environment variable in VARIABLE=VALUE format. Lines that start with # are treated as comments and are ignored.

If there are individual environment variables specified in the container definition, they take precedence over the variables contained within an environment file. If multiple environment files are specified that contain the same variable, they're processed from the top down. We recommend that you use unique variable names. For more information, see Pass an individual environment variable to an Amazon ECS container.

Type: Object array

Required: No

The environment variables to pass to a container. This parameter maps to Env in the docker create-container command and the --env option to the docker run command.

"environment" : [
    { "name" : "string", "value" : "string" },
    { "name" : "string", "value" : "string" }
]

Type: Object array

Required: No

An object that represents the secret to expose to your container. For more information, see Pass sensitive data to an Amazon ECS container.

"secrets": [
    {
        "name": "environment_variable_name",
        "valueFrom": "arn:aws:ssm:region:aws_account_id:parameter/parameter_name"
    }
]

This parameter is not supported for tasks running on Fargate.

Type: Boolean

Required: No

When this parameter is true, networking is off within the container.

The default is false.

"disableNetworking": true|false

This parameter isn't supported for tasks using the awsvpc network mode.

Type: String array

Required: No

The link parameter allows containers to communicate with each other without the need for port mappings. This parameter is only supported if the network mode of a task definition is set to bridge. The name:internalName construct is analogous to name:alias in Docker links. Up to 255 letters (uppercase and lowercase), numbers, hyphens, and underscores are allowed..

"links": ["name:internalName", ...]

Type: String

Required: No

The hostname to use for your container. This parameter maps to Hostname in the docker create-container and the --hostname option to docker run.

"hostname": "string"

This is not supported for tasks running on Fargate.

Type: String array

Required: No

A list of DNS servers that are presented to the container.

"dnsServers": ["string", ...]

This parameter isn't supported for tasks that use the awsvpc network mode.

Type: Object array

Required: No

A list of hostnames and IP address mappings to append to the /etc/hosts file on the container.

This parameter maps to ExtraHosts in the docker create-container command and the --add-host option to docker run.

"extraHosts": [
      {
        "hostname": "string",
        "ipAddress": "string"
      }
      ...
    ]

Type: Boolean

Required: No

When this parameter is true, the container is given read-only access to its root file system. This parameter maps to ReadonlyRootfs in the docker create-container command the --read-only option to docker run.

The default is false.

"readonlyRootFilesystem": true|false

Type: Object array

Required: No

The mount points for the data volumes in your container. This parameter maps to Volumes in the create-container Docker API and the --volume option to docker run.

Windows containers can mount whole directories on the same drive as $env:ProgramData. Windows containers cannot mount directories on a different drive, and mount points cannot be used across drives. You must specify mount points to attach an Amazon EBS volume directly to an Amazon ECS task.

Type: Object array

Required: No

Data volumes to mount from another container. This parameter maps to VolumesFrom in the docker create-container command and the --volumes-from option to docker run.

"volumesFrom": [
                {
                  "sourceContainer": "string",
                  "readOnly": true|false
                }
              ]

Type: LogConfiguration Object

Required: No

The log configuration specification for the container.

For example task definitions that use a log configuration, see Example Amazon ECS task definitions.

This parameter maps to LogConfig in the docker create-container command and the --log-driver option to docker run. By default, containers use the same logging driver that the Docker daemon uses. However, the container might use a different logging driver than the Docker daemon by specifying a log driver with this parameter in the container definition. To use a different logging driver for a container, the log system must be configured properly on the container instance (or on a different log server for remote logging options).

Consider the following when specifying a log configuration for your containers:

"logConfiguration": {
      "logDriver": "awslogs","fluentd","gelf","json-file","journald","splunk","syslog","awsfirelens",
      "options": {"string": "string"
        ...},
	"secretOptions": [{
		"name": "string",
		"valueFrom": "string"
	}]
}

Type: FirelensConfiguration Object

Required: No

The FireLens configuration for the container. This is used to specify and configure a log router for container logs. For more information, see Send Amazon ECS logs to an Amazon service or Amazon Partner.

{
    "firelensConfiguration": {
        "type": "fluentd",
        "options": {
            "KeyName": ""
        }
    }
}

Type: String array

Required: No

A list of ARNs in SSM or Amazon S3 to a credential spec (CredSpec) file that configures the container for Active Directory authentication. We recommend that you use this parameter instead of the dockerSecurityOptions. The maximum number of ARNs is 1.

There are two formats for each ARN.

In both formats, replace MyARN with the ARN in SSM or Amazon S3.

The credspec must provide a ARN in Secrets Manager for a secret containing the username, password, and the domain to connect to. For better security, the instance isn't joined to the domain for domainless authentication. Other applications on the instance can't use the domainless credentials. You can use this parameter to run tasks on the same instance, even it the tasks need to join different domains. For more information, see Using gMSAs for Windows Containers and Using gMSAs for Linux Containers.

Type: String

Required: No

The user to use inside the container. This parameter maps to User in the docker create-container command and the --user option to docker run.

You can specify the user using the following formats. If specifying a UID or GID, you must specify it as a positive integer.

"user": "string"

Type: Object array

Required: No

A list of ulimit values to define for a container. This value overwrites the default resource quota setting for the operating system. This parameter maps to Ulimits in the docker create-container command and the --ulimit option to docker run.

Amazon ECS tasks hosted on Fargate use the default resource limit values set by the operating system with the exception of the nofile resource limit parameter. The nofile resource limit sets a restriction on the number of open files that a container can use. On Fargate, the default nofile soft limit is 65535 and hard limit is 65535. You can set the values of both limits up to 1048576. For more information, see Task resource limits.

This parameter requires version 1.18 of the Docker Remote API or greater on your container instance.

"ulimits": [
      {
        "name": "core"|"cpu"|"data"|"fsize"|"locks"|"memlock"|"msgqueue"|"nice"|"nofile"|"nproc"|"rss"|"rtprio"|"rttime"|"sigpending"|"stack",
        "softLimit": integer,
        "hardLimit": integer
      }
      ...
    ]

Type: String to string map

Required: No

A key/value map of labels to add to the container. This parameter maps to Labels in the docker create-container command and the --label option to docker run.

This parameter requires version 1.18 of the Docker Remote API or greater on your container instance.

"dockerLabels": {"string": "string"
      ...}

Type: LinuxParameters object

Required: No

Linux-specific options that are applied to the container, such as KernelCapabilities.

"linuxParameters": {
      "capabilities": {
        "add": ["string", ...],
        "drop": ["string", ...]
        }
      }

Type: Array of ContainerDependency objects

Required: No

The dependencies defined for container startup and shutdown. A container can contain multiple dependencies. When a dependency is defined for container startup, for container shutdown it is reversed. For an example, see Container dependency.

This parameter requires that the task or service uses platform version 1.3.0 or later (Linux) or 1.0.0 (Windows).

"dependsOn": [
    {
        "containerName": "string",
        "condition": "string"
    }
]

Type: Integer

Required: No

Example values: 120

Time duration (in seconds) to wait before giving up on resolving dependencies for a container.

For example, you specify two containers in a task definition with containerA having a dependency on containerB reaching a COMPLETE, SUCCESS, or HEALTHY status. If a startTimeout value is specified for containerB and it doesn't reach the desired status within that time, then containerA doesn't start.

This parameter requires that the task or service uses platform version 1.3.0 or later (Linux). The maximum value is 120 seconds.

Type: Integer

Required: No

Example values: 120

Time duration (in seconds) to wait before the container is forcefully killed if it doesn't exit normally on its own.

This parameter requires that the task or service uses platform version 1.3.0 or later (Linux). If the parameter isn't specified, then the default value of 30 seconds is used. The maximum value is 120 seconds.

Type: SystemControl object

Required: No

A list of namespace kernel parameters to set in the container. This parameter maps to Sysctls in the docker create-container commandand the --sysctl option to docker run. For example, you can configure net.ipv4.tcp_keepalive_time setting to maintain longer lived connections.

We don't recommend that you specify network-related systemControls parameters for multiple containers in a single task that also uses either the awsvpc or host network mode. Doing this has the following disadvantages:

If you're setting an IPC resource namespace to use for the containers in the task, the following conditions apply to your system controls. For more information, see IPC mode.

"systemControls": [
    {
         "namespace":"string",
         "value":"string"
    }
]

Type: Boolean

Required: No

When this parameter is true, you can deploy containerized applications that require stdin or a tty to be allocated. This parameter maps to OpenStdin in the docker create-container command and the --interactive option to docker run.

The default is false.

Type: Boolean

Required: No

When this parameter is true, a TTY is allocated. This parameter maps to Tty in the docker create-container command and the --tty option to docker run.

The default is false.

Type: String

Required: Yes

The Elastic Inference accelerator device name. The deviceName must also be referenced in a container definition see Elastic Inference accelerator.

Type: String

Required: Yes

The Elastic Inference accelerator to use.

Type: ProxyConfiguration object

Required: No

The configuration details for the App Mesh proxy.

"proxyConfiguration": {
    "type": "APPMESH",
    "containerName": "string",
    "properties": [
        {
           "name": "string",
           "value": "string"
        }
    ]
}

Type: String

Required: No

The name of the volume. Up to 255 letters (uppercase and lowercase), numbers, hyphens (-), and underscores (_) are allowed. This name is referenced in the sourceVolume parameter of the container definition mountPoints object.

Required: No

The host parameter is used to tie the lifecycle of the bind mount to the host Amazon EC2 instance, rather than the task, and where it is stored. If the host parameter is empty, then the Docker daemon assigns a host path for your data volume, but the data is not guaranteed to persist after the containers associated with it stop running.

Windows containers can mount whole directories on the same drive as $env:ProgramData.

Type: Boolean

Required: No

Specifies whether a volume is configurable at launch. When set to true, you can configure the volume when running a standalone task, or when creating or updating a service. When set to true, you won't be able to provide another volume configuration in the task definition. This parameter must be set to true to configure an Amazon EBS volume for attachment to a task. Setting configuredAtLaunch to true and deferring volume configuration to the launch phase allows you to create task definitions that aren't constrained to a volume type or to specific volume settings. Doing this makes your task definition reusable across different execution environments. For more information, see Amazon EBS volumes.

Type: DockerVolumeConfiguration Object

Required: No

This parameter is specified when using Docker volumes. Docker volumes are supported only when running tasks on EC2 instances. Windows containers support only the use of the local driver. To use bind mounts, specify a host instead.

Type: EFSVolumeConfiguration Object

Required: No

This parameter is specified when using Amazon EFS volumes.

Type: FSxWindowsFileServerVolumeConfiguration Object

Required: Yes

This parameter is specified when you're using an Amazon FSx for Windows File Server file system for task storage.

Type: String

Required: No

One part of a key-value pair that make up a tag. A key is a general label that acts like a category for more specific tag values.

Type: String

Required: No

The optional part of a key-value pair that make up a tag. A value acts as a descriptor within a tag category (key).

Type: EphemeralStorage object

Required: No

The amount of ephemeral storage (in GB) to allocate for the task. This parameter is used to expand the total amount of ephemeral storage available, beyond the default amount, for tasks that are hosted on Amazon Fargate. For more information, see Use bind mounts with Amazon ECS.

This is not supported for tasks running on Fargate.

Type: String

Required: No

The IPC resource namespace to use for the containers in the task. The valid values are host, task, or none. If host is specified, then all the containers that are within the tasks that specified the host IPC mode on the same container instance share the same IPC resources with the host Amazon EC2 instance. If task is specified, all the containers that are within the specified task share the same IPC resources. If none is specified, then IPC resources within the containers of a task are private and not shared with other containers in a task or on the container instance. If no value is specified, then the IPC resource namespace sharing depends on the Docker daemon setting on the container instance.

If the host IPC mode is used, there's a heightened risk of undesired IPC namespace exposure.

If you're setting namespaced kernel parameters that use systemControls for the containers in the task, the following applies to your IPC resource namespace.

Type: String

Valid Values: host | task

Required: No

The process namespace to use for the containers in the task. The valid values are host or task. On Linux containers, the only valid value is task. For example, monitoring sidecars might need pidMode to access information about other containers running in the same task.

If task is specified, all containers within the specified task share the same process namespace.

If no value is specified, the default is a private namespace for each container.

Type: Boolean

Valid Values: true | false

Required: No

If this parameter is set to true, in a task's payload, Amazon ECS and Fargate accept fault injection requests from the task’s containers. By default, this parameter is set to false.