StackSets concepts

When you use StackSets, you work with stack sets, stack instances, and stacks.

Administrator and target accounts

An administrator account is the AWS account in which you create stack sets. For stack sets with service-managed permissions, the administrator account is either the organization's management account or a delegated administrator account. A stack set is managed by signing in to the AWS administrator account in which it was created.

A target account is the account into which you create, update, or delete one or more stacks in your stack set. Before you can use a stack set to create stacks in a target account, you must set up a trust relationship between the administrator and target accounts.

AWS CloudFormation StackSets

A stack set lets you create stacks in AWS accounts across regions by using a single CloudFormation template. All the resources included in each stack are defined by the stack set's CloudFormation template. As you create the stack set, you specify the template to use, in addition to any parameters and capabilities that template requires.

After you've defined a stack set, you can create, update, or delete stacks in the target accounts and AWS Regions you specify. When you create, update, or delete stacks, you can also specify operation preferences, such as the order of Regions in which you want the operation to be performed, the failure tolerance beyond which stack operations stop, and the number of accounts in which operations are performed on stacks concurrently.

A stack set is a regional resource. If you create a stack set in one AWS Region, you can't see it or change it in other Regions.

Permission models for stack sets

Stack sets can be created using either self-managed permissions or service-managed permissions.

With self-managed permissions, you create the IAM roles required by StackSets to deploy across accounts and Regions. These roles are necessary to establish a trusted relationship between the account you're administering the stack set from and the account you're deploying stack instances to. Using this permissions model, StackSets can deploy to any AWS account in which you have permissions to create an IAM role.

With service-managed permissions, you can deploy stack instances to accounts managed by AWS Organizations. Using this permissions model, you don't have to create the necessary IAM roles; StackSets creates the IAM roles on your behalf. With this model, you can also enable automatic deployments to accounts that are added to your organization in the future.

AWS Organizations integrates with CloudFormation and helps you centrally manage and govern your environment as you scale and grow your AWS resources.

• Management account – the account that you use to create the organization. For more information, see AWS Organizations terminology and concepts.

• Delegated administrator – a compatible AWS service can register an AWS member account in the organization as an administrator for the organization's accounts in that service. For more information, see AWS services that you can use with AWS Organizations.

For more information about creating and managing stack sets with service-managed permissions, see the following topics:

• Enable trusted access with AWS Organizations

• Register a delegated administrator

• Create a stack set with service-managed permissions

Stack instances

A stack instance is a reference to a stack in a target account within a Region. A stack instance can exist without a stack. For example, if the stack couldn't be created for some reason, the stack instance shows the reason for stack creation failure. A stack instance is associated with only one stack set.

The following figure shows the logical relationships between stack sets, stack operations, and stacks. When you update a stack set, all associated stack instances are updated throughout all accounts and Regions.

Stack set operations

You can perform the following operations on stack sets.

Create stack set

Creating a new stack set includes specifying a CloudFormation template that you want to use to create stacks, specifying the target accounts in which you want to create stacks, and identifying the AWS Regions in which you want to deploy stacks in your target accounts. A stack set ensures consistent deployment of the same stack resources, with the same settings, to all specified target accounts within the Regions you choose.

Update stack set

When you update a stack set, you push changes out to stacks in your stack set. You can update a stack set in one of the following ways. Your template updates always affect all stacks; you can't selectively update the template for some stacks in the stack set, but not others.

• Change existing settings in the template or add new resources, such as updating parameter settings for a specific service, or adding new Amazon EC2 instances.

• Replace the template with a different template.

• Add stacks in existing or additional target accounts, across existing or additional Regions.

Delete stacks

When you delete stacks, you are removing a stack and all its associated resources from the target accounts you specify, within the Regions you specify. You can delete stacks in the following ways.

• Delete stacks from some target accounts, while leaving other stacks in other target accounts running.

• Delete stacks from some Regions, while leaving stacks in other Regions running.

• Delete stacks from your stack set, but save them so they continue to run independently of your stack set by choosing the Retain Stacks option. Retained stacks are managed in AWS CloudFormation, outside of your stack set.

• Delete all stacks in your stack set, in preparation for deleting your entire stack set.

Delete stack set

You can delete your stack set only when there are no stack instances in it.

Stack set operation options

The options described in this section help to control the time and number of failures allowed to successfully perform stack set operations, and prevent you from losing stack resources.

Maximum concurrent accounts

This setting, available in create, update, and delete workflows, lets you specify the maximum number or percentage of target accounts in which an operation is performed at one time. A lower number or percentage means that an operation is performed in fewer target accounts at one time. Operations are performed in one Region at a time, in the order specified in the Deployment order box. For example, if you are deploying stacks to 10 target accounts within two Regions, setting Maximum concurrent accounts to 50 and By percentage will deploy stacks to five accounts in the first Region, then the second five accounts within the first Region, before moving on to the next Region and beginning deployment to the first five target accounts.

When you choose By percentage, if the specified percentage doesn't represent a whole number of your specified accounts, CloudFormation rounds down. For example, if you are deploying stacks to 10 target accounts, and you set Maximum concurrent accounts to 25 and By percentage, CloudFormation rounds down from deploying 2.5 stacks concurrently (which would not be possible) to deploying two stacks concurrently.

Note that this setting lets you specify the maximum for operations. For large deployments, under certain circumstances the actual number of accounts acted upon concurrently may be lower due to service throttling. The maximum speed of deployment is 100 concurrent stack instances per stack set.

Failure tolerance

This setting, available in create, update, and delete workflows, lets you specify the maximum number or percentage of stack operation failures that can occur, per Region, beyond which CloudFormation stops an operation automatically. A lower number or percentage means that the operation is performed on fewer stacks, but you are able to start troubleshooting failed operations faster. For example, if you are updating 10 stacks in 10 target accounts within three Regions, setting Failure tolerance to 20 and By percentage means that a maximum of two stack updates in a Region can fail for the operation to continue. If a third stack in the same Region fails, CloudFormation stops the operation. If a stack couldn't be updated in the first Region, the update operation continues in that Region, and then moves on to the next Region. If two stacks can't be updated in the second Region, the failure tolerance of 20% is reached; if a third stack in the Region fails, CloudFormation stops the update operation, and doesn't go on to subsequent Regions.

When you choose By percentage, if the specified percentage doesn't represent a whole number of your stacks within each Region, CloudFormation rounds down. For example, if you are deploying stacks to 10 target accounts in three Regions, and you set Failure tolerance to 25 and By percentage, CloudFormation rounds down from a failure tolerance of 2.5 stacks (which would not be possible) to a failure tolerance of two stacks per Region.

Retain stacks

This setting, available in delete stack workflows, lets you keep stacks and their resources running even after they have been removed from a stack set. When you retain stacks, AWS CloudFormation leaves stacks in individual accounts and Regions intact. Stacks are disassociated from the stack set, but the stack and its resources are saved. After a delete stacks operation is complete, you manage retained stacks in CloudFormation, in the target account (not the administrator account) in which they were created. Retaining stacks permanently disassociates a stack from a stack set; the stack can't be added to the stack set again, and it can't be added to a new stack set.

Region concurrency

This setting, available in create, update, and delete workflows, lets you choose how StackSets are deployed into Regions.

Sequential – Deploy StackSets operation into one Region at a time as specified by Region Deployment order box as long as a Region's deployment failures don't exceed a specified failure tolerance. Sequential deployment is the default selection.

Parallel – Deploy StackSets operations into all specified Regions in parallel as long as a Region's deployment failures don't exceed a specified failure tolerance.

Tags

You can add tags during stack set creation and update operations by specifying key and value pairs. Tags are useful for sorting and filtering stack set resources for billing and cost allocation. For more information about how tags are used in AWS, see Using cost allocation tags in the AWS Billing and Cost Management User Guide. After you specify the key-value pair, choose + to save the tag. You can delete tags that you are no longer using by choosing the red X to the right of a tag.

Tags that you apply to stack sets are applied to all stacks, and the resources that are created by your stacks. Tags can be added at the stack-only level in CloudFormation, but those tags might not show up in StackSets.

Although StackSets doesn't currently add any system-defined tags, you should not start the key names of any tags with the string aws:.

Stack set and stack instance status codes

AWS CloudFormation StackSets generates status codes for stack set operations and stack instances.

The following table describes status codes for stack set operations.

Stack set operation status	Description
RUNNING	The operation is currently in progress.
SUCCEEDED	The operation finished without exceeding the failure tolerance for the operation.
FAILED	The number of stacks on which the operation couldn't be completed exceeded the user-defined failure tolerance. The failure tolerance value you've set for an operation is applied for each Region during stack creation and update operations. If the number of failed stacks within a Region exceeds the failure tolerance, the status of the operation in the Region is set to FAILED. The status of the operation as a whole is also set to FAILED, and CloudFormation cancels the operation in any remaining Regions.
QUEUED	[Service-managed permissions] For automatic deployments that require a sequence of operations, the operation is queued to be performed. For example: Moving an account from one organizational unit (OU), OU1, to another, OU2, triggers an automatic deployment. StackSets runs a delete operation to remove the stack instance from the target OU1 account in the target Region and queues a create operation to add a stack instance to the target OU2 account in the target Region. When the delete operation is complete, StackSets will run the create operation. Adding an account AccountA to an OU triggers an automatic deployment. StackSets runs a create operation to add a stack instance to AccountA in the target Region. If you add another account AccountB to the OU while this create operation is running, StackSets queues a second create operation. When the first create operation is complete, StackSets runs the second create operation to add a stack instance to AccountB in the target Region.
STOPPING	The operation is in the process of stopping, at the user's request.
STOPPED	The operation has been stopped, at the user's request.

The following table describes status codes for stack instances within stack sets.

Stack instance status	Description
CURRENT	The stack is currently up to date with the stack set.
OUTDATED	The stack isn't currently up to date with the stack set for one of the following reasons. A CreateStackSet or UpdateStackSet operation on the associated stack failed. The stack was part of a CreateStackSet or UpdateStackSet operation that failed, or was stopped before the stack was created or updated.
INOPERABLE	A DeleteStackInstances operation has failed and left the stack in an unstable state. Stacks in this state are excluded from further UpdateStackSet operations. You might need to perform a DeleteStackInstances operation, with RetainStacks set to true, to delete the stack instance, and then delete the stack manually.
CANCELLED	The operation in the specified account and Region has been canceled. This is either because a user has stopped the stack set operation, or because the failure tolerance of the stack set operation has been exceeded.
FAILED	The operation in the specified account and Region failed. If the stack set operation fails in enough accounts within a Region, the failure tolerance for the stack set operation as a whole might be exceeded.
PENDING	The operation in the specified account and Region has yet to start.
RUNNING	The operation in the specified account and Region is currently in progress.
SUCCEEDED	The operation in the specified account and Region completed successfully.