sbt plugin to deploy code to AWS Lambda
This plugin is available for both sbt 1.x and 0.13.x. The latest version 0.6.6 was released on 10/Oct/2018.
Add the following to your project/plugins.sbt file:
addSbtPlugin("com.gilt.sbt" % "sbt-aws-lambda" % "0.6.6")Add the AwsLambdaPlugin auto-plugin to your build.sbt:
enablePlugins(AwsLambdaPlugin)sbt configureLambda Creates a new AWS Lambda if it doesn't exist yet, or updates the Lambda configuration, if it has changed.
sbt deployLambda Packages and deploys the current project to an existing AWS Lambda.
The plugin also has the following deprecated tasks:
sbt createLambda creates a new AWS Lambda function from the current project.
sbt updateLambda updates an existing AWS Lambda function with the current project.
sbt-aws-lambda can be configured using sbt settings, environment variables or by reading user input at deploy time
| sbt setting | Environment variable | Description |
|---|---|---|
| s3Bucket | AWS_LAMBDA_BUCKET_ID | The name of an S3 bucket where the lambda code will be stored |
| s3KeyPrefix | AWS_LAMBDA_S3_KEY_PREFIX | The prefix to the S3 key where the jar will be uploaded |
| lambdaName | AWS_LAMBDA_NAME | The name to use for this AWS Lambda function. Defaults to the project name |
| handlerName | AWS_LAMBDA_HANDLER_NAME | Java class name and method to be executed, e.g. com.gilt.example.Lambda::myMethod |
| roleArn | AWS_LAMBDA_IAM_ROLE_ARN | The ARN of an IAM role to use when creating a new Lambda |
| region | AWS_REGION | The name of the AWS region to connect to. Defaults to us-east-1 |
| awsLambdaTimeout | AWS_LAMBDA_TIMEOUT | The Lambda timeout in seconds (1-900). Defaults to AWS default. |
| awsLambdaMemory | AWS_LAMBDA_MEMORY | The amount of memory in MB for the Lambda function (128-1536, multiple of 64). Defaults to AWS default. |
| lambdaHandlers | Sequence of Lambda names to handler functions (for multiple lambda methods per project). Overrides lambdaName and handlerName if present. |
|
| deployMethod | AWS_LAMBDA_DEPLOY_METHOD | The preferred method for uploading the jar, either S3 for uploading to AWS S3 or DIRECT for direct upload to AWS Lambda |
| deadLetterArn | AWS_LAMBDA_DEAD_LETTER_ARN | The ARN of the Lambda function's dead letter SQS queue or SNS topic, to receive unprocessed messages |
| vpcConfigSubnetIds | AWS_LAMBDA_VPC_CONFIG_SUBNET_IDS | Comma separated list of subnet IDs for the VPC |
| vpcConfigSecurityGroupIds | AWS_LAMBDA_VPC_CONFIG_SECURITY_GROUP_IDS | Comma separated list of security group IDs for the VPC |
| environment | Seq[(String, String)] of environment variables to set in the lambda function |
An example configuration might look like this:
retrieveManaged := true
enablePlugins(AwsLambdaPlugin)
lambdaHandlers := Seq(
"function1" -> "com.gilt.example.Lambda::handleRequest1",
"function2" -> "com.gilt.example.Lambda::handleRequest2",
"function3" -> "com.gilt.example.OtherLambda::handleRequest3"
)
// or, instead of the above, for just one function/handler
//
// lambdaName := Some("function1")
//
// handlerName := Some("com.gilt.example.Lambda::handleRequest1")
s3Bucket := Some("lambda-jars")
awsLambdaMemory := Some(192)
awsLambdaTimeout := Some(30)
roleArn := Some("arn:aws:iam::123456789000:role/lambda_basic_execution")
(note that you will need to use a real ARN for your role rather than copying this one).
By default, sbt-aws-lambda uses sbt-assembly to package a fat jar that includes your project and all its dependencies.
In some cases you may want to override this and provide a different task that sbt-aws-lambda should call to package the jar. To do this, override the packageLambda task key with a different task that produces an package file,
like sbt-native-package or the built-in packageBin task:
// Use sbt-native-packager
packageLambda := (packageBin in Universal).value
// Use project jar only
packageLambda := (packageBin in Compile).value
This plugin uses sbt-sonatype to publish to Gilt's account on maven central
sbt publishSigned sonatypeRelease