AWS Lambda

This section helps you to check the list of actions you must perfom before you start working on your function.

You have the latest version of the AWS Lambda CLI.
You have jq installed to extract some information from the JSON responses when executing commands.

Start by setting all the parameters required for deploying your function using the AWS CLI.

# Name of the binary file you downloaded from Cloudera DataFlow. Change it based on the version you downloaded.
FILEKEY="naaf-aws-lambda-1.0.0.2.3.6.0-35-bin.zip"

# Location where the function will be deployed. Change this value to match your needs/requirements.
LOCATION=eu-west-2

# Name of the bucket where the binary and NAR will be uploaded. Change this value to match your needs/requirements.
BUCKET_NAME=mydffresizeimagebucketpvillard

# Name of the function (only lower case letters). Change this value to match your needs/requirements.
FUNCTIONNAME=resizeimagepvillard

# Flow CRN. Add the Customer Resource Number you obtained from the DataFlow Catalog.
FLOW_CRN=***YOUR VALUE HERE***

# Credentials of the machine user
DF_ACCESS_KEY=***YOUR VALUE HERE***
DF_PRIVATE_KEY=***YOUR VALUE HERE***

Download the function template from here and make it available locally.

Execute the below commands:

Create your S3 bucket.

aws s3api create-bucket \
    --bucket $BUCKET_NAME \
    --region $LOCATION \
    --create-bucket-configuration LocationConstraint=$LOCATION

Copy the Lambda DataFlow Function binariy zip file to your bucket.
```
aws s3 cp $FILEKEY s3://$BUCKET_NAME/$FILEKEY
```

Create the role for the Lambda.

ROLEARN=`aws iam create-role --role-name $FUNCTIONNAME\_lambda\_role --assume-role-policy-document '{"Version": "2012-10-17","Statement": [{ "Effect": "Allow", "Principal": {"Service": "lambda.amazonaws.com"}, "Action": "sts:AssumeRole"}]}' | jq -r ".Role.Arn" `
aws iam attach-role-policy --role-name $FUNCTIONNAME\_lambda\_role --policy-arn arn:aws:iam::aws:policy/service-role/AWSLambdaBasicExecutionRole

sleep 10

Create the function.


cp my-function-definition.json template.json

sed -i '.bak' -e 's|BUCKETNAME|'"$BUCKET_NAME"'|g' \
		-e 's|FUNCTIONNAME|'"$FUNCTIONNAME"'|g' \
		-e 's|FLOWCRN|'"$FLOW_CRN"'|g' \
		-e 's|DFACCESSKEY|'"$DF_ACCESS_KEY"'|g' \
		-e 's|DFPRIVATEKEY|'"$DF_PRIVATE_KEY"'|g' \
		-e 's|LAYERARN|'"$LAYERARNVERSION"'|g' \
		-e 's|FILEKEY|'"$FILEKEY"'|g' \
		-e 's|ROLEARN|'"$ROLEARN"'|g' template.json

aws lambda create-function --cli-input-json file://template.json

echo "please wait a bit more (30 sec)..."
sleep 30

aws lambda publish-version --function-name $FUNCTIONNAME

When the execution is complete (it takes a few minutes), the function will be created and a version will be published but without a trigger.

Open the AWS Console UI and go to your Lambda.
Find to the published version of the function and add the API Gateway trigger:
This gives you the endpoint on which the Lambda is listening.

For example: https://aa2q99qo3i.execute-api.eu-west-2.amazonaws.com/default/resizeimage

You can now use a tool like Postman to send your picture.

You can add two custom headers (resize-width and resize-height) to specify the resizing dimensions. If not specified, the default values of the NiFi flow parameters are used.

note
The first attempt may fail due to a cold start and timeout (30 sec timeout on the API Gateway). A subsequent retry after a minute should work and should return the resized image much faster. For more information, see Cold start.

It is possible to send a request to your function using curl.
For example:
```
curl -X POST https://3bftvfa66m.execute-api.us-east-2.amazonaws.com/default/cpresizeimage \
-H "Content-Type: image/png" \
-H "resize-width: 400" \
-H "resize-height: 200" \
--data-binary "@/Users/pierre/Desktop/test.png" \
--output /tmp/my_resized_image.png
```
note
The function will be open to the internet. Make sure to secure it if you want to keep it running (this is done when creating the API Gateway trigger) or delete everything once you are done with this quickstart.
You can delete the function using the following command:
```
aws lambda delete-function --function-name $FUNCTIONNAME --region $LOCATION

aws iam detach-role-policy --role-name $FUNCTIONNAME\_lambda\_role --policy-arn arn:aws:iam::aws:policy/service-role/AWSLambdaBasicExecutionRole
aws iam delete-role --role-name $FUNCTIONNAME\_lambda\_role

aws s3 rm s3://$BUCKET_NAME --recursive --region $LOCATION
aws s3api delete-bucket --bucket $BUCKET_NAME --region $LOCATION
rm template.json template.json.bak
```
You may also want to manually delete the API Gateway that you created earlier.