Rain

• Documentation: https://aws-cloudformation.github.io/rain/

Rain is what happens when you have a lot of CloudFormation

Rain is also a command line tool for working with AWS CloudFormation templates and stacks.

Discord

Join us on Discord to discuss rain and all things CloudFormation! Connect and interact with CloudFormation developers and experts, find channels to discuss rain, the CloudFormation registry, StackSets, cfn-lint, Guard and more:

Key features

• Interactive deployments: With rain deploy, rain packages your CloudFormation templates, prompts you for any parameters that have not yet been defined, shows you a summary of the changes that will be made, and then displays real-time updates as your stack is being deployed. Once finished, you get a summary of the outcome along with any error messages collected along the way - including errors messages for stacks that have been rolled back and no longer exist.

• Consistent formatting of CloudFormation templates: Using rain fmt, you can format your CloudFormation templates to a consistent standard or reformat a template from JSON to YAML (or YAML to JSON if you prefer). Rain preserves your comments when using YAML and switches use of intrinsic functions to use the short syntax where possible.

• Combined logs for nested stacks with sensible filtering: When you run rain log, you will see a combined stream of logs from the stack you specified along with any nested stack associated with it. Rain also filters out uninteresting log messages by default so you just see the errors that require attention. You can also use rain log --chart to see a Gantt chart that shows you how long each operation took for a given stack.

• Build new CloudFormation templates: rain build generates new CloudFormation templates containing skeleton resources that you specify. This saves you having to look up which properties are available and which are required vs. optional. Build skeleton templates by specifying a resource name like AWS::S3::Bucket, or enable the Bedrock Claude model in your account to use generative AI with a command like rain build --prompt "A VPC with 2 subnets". (Note that Bedrock is not free, and requires some setup). NEW Use the rain build --recommend command to pick from a list of functional templates that will pass typical compliance checks by default. These templates are great starting points for infrastructure projects.

• Build policy validation files: The rain build command can now send prompts to the Bedrock Cloude3 Haiku and Sonnet models, which can write Open Policy Agent (OPA) Rego files or CloudFormation Guard files, to verify the compliance of your templates

• Manipulate CloudFormation stack sets: rain stackset deploy creates a new stackset, updates an existing one or adds a stack instance(s) to an existing stack set. You can list stack sets using rain stackset ls, review stack set details with rain stackset ls <stack set name> and delete stack set and\or its instances with rain stackset rm <stack set name>

• Predict deployment failures (EXPERIMENTAL): rain forecast analyzes a template and the target deployment account to predict things that might go wrong when you attempt to create, update, or delete a stack. This command speeds up development by giving you advanced notice for issues like missing permissions, resources that already exist, and a variety of other common resource-specific deployment blockers.

• Modules (EXPERIMENTAL): rain pkg supports client-side module development. Rain modules are partial templates that are inserted into the parent template, with some extra functionality added to enable extending existing resource types. This feature integrates with CodeArtifact to enable package publish and install.

• Content Deployment (EXPERIMENTAL): rain deploy and rain rm support metadata commands that can upload static assets to a bucket and then delete those assets when the bucket is deleted. Rain can also run build scripts before and after stack deployment to prepare content like web sites and lambda functions before uploading to S3.

Note that in order to use experimental commands, you have to add --experimental or -x as an argument.

Getting started

If you have homebrew installed, brew install rain

Or you can download the appropriate binary for your system from the releases page.

Or if you're a Gopher, you can go install github.com/aws-cloudformation/rain/cmd/rain@latest

Usage:
  rain [command]

Stack commands:
  cat         Get the CloudFormation template from a running stack
  cc          Interact with templates using Cloud Control API instead of CloudFormation
  deploy      Deploy a CloudFormation stack or changeset from a local template
  logs        Show the event log for the named stack
  ls          List running CloudFormation stacks or changesets
  rm          Delete a CloudFormation stack or changeset
  stackset    This command manipulates stack sets.
  watch       Display an updating view of a CloudFormation stack

Template commands:
  bootstrap   Creates the artifacts bucket
  build       Create CloudFormation templates
  diff        Compare CloudFormation templates
  fmt         Format CloudFormation templates
  forecast    Predict deployment failures
  merge       Merge two or more CloudFormation templates
  module      Interact with Rain modules in CodeArtifact
  pkg         Package local artifacts into a template
  tree        Find dependencies of Resources and Outputs in a local template

Other Commands:
  console     Login to the AWS console
  help        Help about any command
  info        Show your current configuration

You can find shell completion scripts in docs/bash_completion.sh and docs/zsh_completion.sh.

Contributing

Rain is written in Go and uses the AWS SDK for Go v2.

To contribute a change to Rain, fork this repository, make your changes, and submit a Pull Request.

Go Generate

The README.md, documentation in docs/, the auto completion scripts and a copy of the cloudformation specification in cft/spec/cfn.go are generated through go generate.

License

Rain is licensed under the Apache 2.0 License.

Example Usage

Packaging

The rain pkg command can be used as a replacement for the aws cloudformation package CLI command. When packaging a template, rain looks for specific directives to appear in resources.

Embed

The !Rain::Embed directive simply inserts the contents of a file into the template as a string.

The template:

Resources:
  Test:
    Type: AWS::CloudFormation::WaitConditionHandle
    Metadata:
      Comment: !Rain::Embed embed.txt

The contents of embed.txt, which is in the same directory as the template:

This is a test

The resulting packaged template:

Resources:
  Test:
    Type: AWS::CloudFormation::WaitConditionHandle
    Metadata:
      Comment: This is a test

Include

The !Rain::Include directive parses a YAML or JSON file and inserts the object into the template.

The template:

Resources:
  Test:
    !Rain::Include include-file.yaml

The file to be included:

Type: AWS::S3::Bucket
Properties:
  BucketName: test

The resulting packaged template:

Resources:
  Test:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: test

Env

The !Rain::Env directive reads environment variables and inserts them into the template as strings.

The template:

Resources:
  Test:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: !Rain::Env BUCKET_NAME

The resulting packaged template, if you have exported an environment variable named BUCKET_NAME with value abc:

Resources:
  Test:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: abc

S3Http

The !Rain::S3Http directive uploads a file or directory to S3 and inserts the HTTPS URL into the template as a string.

The template:

Resources:
  Test:
    Type: A::B::C
    Properties:
      TheS3URL: !Rain::S3Http s3http.txt

If you have a file called s3http.txt in the same directory as the template, rain will use your current default profile to upload the file to the artifact bucket that rain creates as a part of bootstrapping. If the path provided is a directory and not a file, the directory will be zipped first.

Resources:
  Test:
    Type: A::B::C
    Properties:
      TheS3URL: https://rain-artifacts-012345678912-us-east-1.s3.us-east-1.amazonaws.com/a84b588aa54068ed4b027b6e06e5e0bb283f83cf0d5a6720002d36af2225dfc3

S3

The !Rain::S3 directive is basically the same as S3Http, but it inserts the S3 URI instead of an HTTPS URL.

The template:

Resources:
  Test:
    Type: A::B::C
    Properties:
      TheS3URI: !Rain::S3 s3.txt

If you have a file called s3.txt in the same directory as the template, rain will use your current default profile to upload the file to the artifact bucket that rain creates as a part of bootstrapping. If the path provided is a directory and not a file, the directory will be zipped first.

Resources:
  Test:
    Type: A::B::C
    Properties:
      TheS3URI: s3://rain-artifacts-755952356119-us-east-1/a84b588aa54068ed4b027b6e06e5e0bb283f83cf0d5a6720002d36af2225dfc3 

If instead of providing a path to a file, you supply an object with properties, you can exercise more control over how the object is uploaded to S3. The following example is a common pattern for uploading Lambda function code. The optional Run property is a local script that you want Rain to run before uploading the content at Path.

Resources:
  MyFunction:
    Type: AWS::Lambda::Function
    Properties:
      Code: !Rain::S3 
        Path: lambda-dist
        Zip: true
        BucketProperty: S3Bucket
        KeyProperty: S3Key
        Run: buildscript.sh

The packaged template:

Resources:
  MyFunction:
    Type: AWS::Lambda::Function
    Properties:
      Code:
        S3Bucket: rain-artifacts-012345678912-us-east-1
        S3Key: 1b4844dacc843f09941c11c94f80981d3be8ae7578952c71e875ef7add37b1a7

Sometimes you require that objects uploaded to S3 have a specific extension, use the Extension property to ensure the artifact in S3 ends ..

Resources:
  Test:
    Type: A::B::C
    Properties:
      TheS3URI: !Rain::S3
        Path: test
        Extension: sh

The packaged template:

Resources:
  Test:
    Type: A::B::C
    Properties:
      TheS3URI: s3://rain-artifacts-012345678912-us-east-1/a84b588aa54068ed4b027b6e06e5e0bb283f83cf0d5a6720002d36af2225dfc3.sh

Metadata commands

You can add a metadata section to an AWS::S3::Bucket resource to take additional actions during deployment, such as running pre and post build scripts, uploading content to the bucket after stack deployment completes, and emptying the contents of the bucket when the stack is deleted.

Resources:
  Bucket:
    Type: AWS::S3::Bucket
    Metadata:
      Rain:
        EmptyOnDelete: true
        Content: site/dist
        Version: 2
        DistributionLogicalId: SiteDistribution
        RunBefore: 
          Command: buildsite.sh
          Args:
          - ALiteralArgument
        RunAfter:
          Command: buildsite.sh
          Args:
            - Rain::OutputValue AStackOutputKey

EmptyOnDelete: If true, the bucket's contents, including all versions, will be deleted so that the bucket itself can be deleted. This can be useful for development environments, but be careful about using it in production!

Version: Rain doesn't do anything with this, but incrementing the number can force the stack to deploy if there have been no infrastructure changes to the stack.

RunBefore: Rain will run this command before the stack deploys. Useful to run your website build script before bothering to deploy, to make sure it builds successfully.

RunAfter: Rain will run this command after deployment, and it is capable of doing stack output lookups to provide arguments to the script. This is useful if you deployed a resource like an API Gateway and need to know the stage URL to plug in to your website configuration. Use Rain::OutputValue OutputKey to pass one of the arguments to the script.

DistributionLogicalId: Supply the logical id of a CloudFront distribution to invalidate all files in it after the content upload completes.

See test/webapp/README.md for a complete example of using these commands with Rain modules.

Modules

You can use Rain to package templates with client-side modules, which gives CloudFormation multi-file support. This feature is compatible with (upcoming) functionality in the AWS CLI cloudformation package command. This feature is still considered experimental, so you need to use the --experimental flag to use it. The UX is fairly stable at this point, but we still reserve the right to change the behavior in a minor version release.

The rain pkg command does not actually deploy any resources if the template does not upload any objects to S3, so you always have a chance to review the packaged template. It's recommended to run linters and scanners on the packaged template, rather than a pre-processed template that makes use of these advanced directives. (We are working on native support for modules in cfn-lint and in IDE plugins)

A prior version of modules in Rain made use of the Rain::Module resource type directive, which is still available for backwards compatibility.

Modules are imported into the parent template or parent module via a new Modules section. This is a departure from registry modules and from the prior version of Rain modules, which are configured as resources.

Modules:
  Content:
    Source: ./module.yaml

Modules are very similar to CloudFormation templates. They support Parameters, Resources, Conditions, and Outputs. Parameters are configured with the Properties attribute in the parent, and act much like normal parameters, except it is possible to pass in objects and lists to a module. Any valid CloudFormation template can be used as a module.

A sample module:

Parameters:
  Name:
    Type: Scalar
Resources:
  Bucket:
    Type: AWS::S3::Bucket
    Metadata:
      OverrideMe: abc
    Properties:
      BucketName: !Ref Name
Outputs:
   BucketArn: 
    Value: !GetAtt Bucket.Arn

The Outputs of a module are reference-able in the parent by using GetAtt or Sub for scalars and Ref for objects.

An example of using the module above:

Modules:
  Content:
    Source: ./module.yaml
    Properties:
      Name: foo
    Overrides:
      Bucket:
        Metadata:
          OverrideMe: def
Outputs:
  TheArn:
    Value: !GetAtt Content.BucketArn

The Overrides attribute is a new feature that distinguishes local modules from registry modules. This allows the consumer to override content from the module that is included in the parent template, if they know the internal structure of the module. Instead of forcing a module author to anticipate every possible use case with numerous parameters, the author can focus on basic use cases and allow the consumer to get creative if they want to change certain elements of the output. Using Overrides carries a risk of breaking changes when a module author changes things, so it will generally be safer to rely on Parameters and References, but the overrides provide a flexible and easy-to-use escape hatch that can solve some tricky design challenges associated with a declarative language like YAML.

The packaged output of the above template:

Resources:
  ContentBucket:
    Type: AWS::S3::Bucket
    Metadata:
      OverrideMe: def
    Properties:
      BucketName: foo

You can also define constant values in modules. The Constants section is a simple key-value list of strings or objects that are referred to later with Fn::Sub or Ref. When referencing an object, only the entire object can be referenced, not any sub-elements of the object.

Constants:
  S3Arn: "arn:${AWS::Partition}:s3:::"
  BucketName: "${Name}-${AWS::Region}-${AWS::AccountId}"
  BucketArn: "${Const::S3Arn}${Const::BucketName}"
  AnObject:
    Foo: bar

Before any processing of the template happens, all instances of constants in !Sub strings are replaced, including in subsequent constant declarations. Constant references are prefixed by Const::. Constants are supported not just in modules, but in the parent template as well. For constants that are objects, they are referenced with !Ref Const::name.

After constants are processed, the Modules section is processed. Module source files are specified using a path that is relative to the parent template or module. The path is not relative to where the package command is being run. So, if a module is in the same directory as the parent, it is simply referred to as module.yaml or ./module.yaml. Modules can also reference an HTTPS URL as the source location.

Modules can contain other modules, with no enforced maximum limit on nesting. Modules are not allowed to refer to themselves directly or in cycles. Module A can’t import Module A, and it can’t import Module B if that module imports Module A.

Modules support a basic form of looping/foreach by adding a Map attribute to the module configuration. Special variables $MapIndex and $MapValue can be used to refer to the index and list value. Logical Ids are auto-incremented by adding an integer starting at zero. Since this is a client-side-only feature, list values must be fully resolved scalars, not values that must be resolved at deploy time. When deploy-time values are needed, Fn::ForEach is a better design option. Local modules do not process Fn::ForEach.

Parameters:
  List:
    Type: CommaDelimitedList
    Default: A,B,C

Modules:
  Content:
    Source: ./map-module.yaml
    Map: !Ref List
    Properties:
      Name: !Sub my-bucket-$MapValue

Assuming the module itself simply creates a bucket, the output would be:

Parameters:
  List: 
    Type: CommaDelimitedList  
    Default: A,B,C
Resources:  
  Content0Bucket:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: my-bucket-A
  Content1Bucket:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: my-bucket-B
  Content2Bucket:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: my-bucket-C

It’s also possible to refer to elements within a Map using something like !GetAtt Content[0].Arn for a single element, or !GetAtt Content[].Arn, which resolves to a list of all of the Arn outputs from that module.

When a module is processed, the first thing that happens is parsing of the Conditions within the module. These conditions must be fully resolvable client-side, since the package command does not have access to Parameters or deploy-time values. These conditions are converted to a dictionary of boolean values and the Conditions section is not emitted into the parent template. It is not merged into the parent. Any resources marked with a false condition are removed, and any property nodes with conditions are processed. Any values of !Ref AWS::NoValue are removed. No evidence of conditions will remain in the markup that is merged into the parent template, unless the condition is not found in the module.

Much of the value of module output is in the smart handling of Ref, Fn::GetAtt, and Fn::Sub. For the most part, we want these to “just work” in the way that an author would expect them to. Since the logical IDs in the module result in a concatenation of the module name and the ID specified inside the module, we have to modify self-references within the module. And we have to fix up any references to parameter values, so that in the final rendered output, they match up with actual IDs and property names.

From the parent, the author has two options for referring to module properties. If they know the structure of the module, they can use the predicted final name for resources, in which case we leave the strings alone. For example, in a module that creates a bucket, if the logical id within the module is Bucket and the name of the module in the parent is Content, the author could write !Ref ContentBucket or !GetAtt ContentBucket.BucketName. The second, safer way would be for the module author to specify an output reference called Name that refs the bucket name, so the parent author could write !GetAtt Content.Name. Module authors are encouraged to provide Outputs that provide access to all needed values, but there will be cases when they cannot predict everything a consumer needs. For Sub strings, if we can fully resolve the value, we get rid of the Fn::Sub and simply write the string to the output. The parent can also refer directly to the Properties defined for a module import in the Modules section.

Complex objects are merged when an override is specified, so it’s possible to do things like add statements to a policy without actually overwriting the entire thing. This is usually what the author wants, but the tradeoff is that there is no way to delete an element from a list in the module.

Intrinsics

As a part of this design, we are adding support for intrinsic functions that do not exist for normal CloudFormation templates, and can only be processed by the package command. We also augment the behavior of some intrinsics. These functions are fully resolved locally and won’t need server-side support.

Fn::Merge merges the contents of multiple objects. This is useful for things like tagging, where each module might want to add a set of tags, in addition to any tags defined by the parent module.

Fn::Select is collapsed if we can fully resolve the return value as a string. We do the same for Fn::Join.

Fn::InsertFile inserts the contents of a local file directly into the template as a string.

Fn::Invoke allows modules to be treated sort of like functions. A module can be created with only Parameters and Outputs, and by using Invoke later, you can get different outputs. An example use case is a module that creates standard resource names, accepting things like Env and App and Name as parameters, and returning a concatenated string.

Metadata

When processing modules, the package command adds metadata to the template for metrics gathering.

Metadata:
  AWSToolsMetrics:
    Rain: '{"Version":"v1.21.0","Experimental":true,"HasModules":true,"HasRainSection":false}'

Each rendered resource has an added Metadata property to indicate where it originated. This can be useful for tooling such as IDE integration, and for troubleshooting in the console.

Resources:
  ContentBucket:
    Metadata:
      SourceMap: "./modules/bucket.yaml:Bucket:35"

Packaging

In order to reference a collection of modules, you can add a Packages section to the template. Packages are zip files containing modules.

Packages:
  abc:
    Source: ./package.zip
  def:
    Source: https://example.com/packages/package.zip

Packages are referenced by using an alias to the package name that starts with $.

Modules:
  Foo:
    Source: $abc/foo.yaml
  Bar:
    Source: $def/a/b/bar.yaml

Publish modules to CodeArtifact

Rain integrates with AWS CodeArtifact to enable an experience similar to npm publish and install. A directory that includes Rain module YAML files can be packaged up with rain module publish, and then the directory can be installed by developers with rain module install.

A module package is published and released from this repository separately from the Rain binary release. This allows the package to be referenced by version numbers using tags, such as m0.1.0 as shown in the example above. The major version number will be incremented if any breaking changes are introduced to the modules. The available modules in the release package are listed below. Treat these modules as samples to be used as a proof-of-concept for building your own module packages.

simple-vpc.yaml

A VPC with just two availability zones. This module is useful for POCs and simple projects.

encrypted-bucket.yaml

A simple bucket with encryption enabled and public access blocked

compliant-bucket.yaml

A bucket, plus extra buckets for access logs and replication and a bucket policy that should pass most typical compliance checks.

bucket-policy.yaml

A bucket policy that denies requests not made with TLS.

load-balancer.yaml

An ELBv2 load balancer

static-site.yaml

An S3 bucket and a CloudFront distribution to host content for a web site

cognito.yaml

A Cognito User Pool and associated resources

rest-api.yaml

An API Gateway REST API

api-resource.yaml

A Lambda function and associated API Gateway resources

IfParam and IfNotParam

Inside a module, you can add a Metadata attribute to show or hide resources, depending on whether the parent template sets a parameter value. This is similar to the Conditional section in a template, but somewhat simpler, and it only works in modules. Modules also support normal Conditions. IfParam is basically just a shorthand for writing the complete Condition.

Resources:
  Bucket:
    Type: AWS::S3::Bucket
    Metadata:
      Rain:
        IfParam: Foo

If the parent template does not set a value for the Foo property, the module will omit the resource. The opposite is true for IfNotParam.

IfParam can be useful to make flexible modules that can optionally do things like configure permissions for related resources, like allowing access to a bucket or table.

IfNotParam is useful if you have pre-created a resource and you don't want the module to create it for you.

Gantt Chart

Output a chart to an HTML file that you can view with a browser to look at how long stack operations take for each resource.

rain log --chart CDKToolkit > ~/Desktop/chart.html

Pkl

You can now write CloudFormation templates in Apple's new configuration language, Pkl. Rain commands that accept input as JSON or YAML now also accept files with the .pkl extension. We host a Pkl package in a separate repo that is generated based on the CloudFormation registry. This package has classes that can be imported for each registry resource type, in addition to higher level patterns. This allows you to write a type-safe template and create your own client-side modules, in a way that is similar to CDK, but with declarative code.

Example Pkl template:

amends "package://github.com/aws-cloudformation/cloudformation-pkl/releases/download/[email protected]/[email protected]#/template.pkl"
import "package://github.com/aws-cloudformation/cloudformation-pkl/releases/download/[email protected]/[email protected]#/cloudformation.pkl" as cfn
import "package://github.com/aws-cloudformation/cloudformation-pkl/releases/download/[email protected]/[email protected]#/aws/s3/bucket.pkl" as bucket

Description = "Create a bucket"

Parameters {
    ["Name"] {
        Type = "String"
        Default = "baz"
    }
}

Resources {
    ["MyBucket"] = new bucket.Bucket {
        BucketName = cfn.Ref("Name")
    }
}

Constants

You can put constants into the Rain section in the template and then refer to those constants later in the template. Constants can be strings or entire objects (but only strings can be used later in Sub functions). Use the !Rain::Constant directive to refer to a constant in the template. For strings, you can add constants to !Sub strings with the ${Rain::ConstantName} pseudo-parameter. Constants can contain previosuly declared constants in Sub strings using the same format. The Rain section gets removed during packaging.

Parameters:
  Prefix:
    Type: String

Rain:
  Constants:
    Test1: ezbeard-rain-test-constants
    Test2: !Sub ${Prefix}-${Rain::Test1}-SubTest

Resources:
  Bucket:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: !Rain::Constant Test1

  Bucket2:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: !Rain::Constant Test2

  Bucket3:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: !Sub "pre-${Prefix}-${Rain::Test1}-suffix" 

Run rain pkg on that template to produce the following output:

Parameters:
  Prefix:
    Type: String

Resources:
  Bucket:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: ezbeard-rain-test-constants

  Bucket2:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: !Sub ${Prefix}-ezbeard-rain-test-constants-SubTest

  Bucket3:
    Type: AWS::S3::Bucket
    Properties:
      BucketName: !Sub pre-${Prefix}-ezbeard-rain-test-constants-suffix

Other CloudFormation tools

• cfn-lint

  Validate CloudFormation yaml/json templates against the CloudFormation spec and additional checks. Includes checking valid values for resource properties and best practices.

• cfn-guard

  Guard is a policy evaluation tool that allows you to build your own rules with a custom DSL. You can also pull rules from the guard registry to scan your templates for resources that don't comply with common best practices.

• taskcat

  taskcat is a tool that tests AWS CloudFormation templates. It deploys your AWS CloudFormation template in multiple AWS Regions and generates a report with a pass/fail grade for each region. You can specify the regions and number of Availability Zones you want to include in the test, and pass in parameter values from your AWS CloudFormation template. taskcat is implemented as a Python class that you import, instantiate, and run.

Are we missing an excellent tool? Let us know via a GitHub issue.