isabella232 / serverless-image-handler

A solution to dynamically handle images on the fly, utilizing SharpJS

Jobber Custom AWS Serverless Image based on v4.0.0

This custom build is created to address these 2 problems

• Support regex in environment variable SOURCE_BUCKETS ie. jobber\-development\-.* Instead of a comma separate list that includes all 50+ developer's S3 buckets. The CSV is not scalable and not maintainable.
• Fix SmartCrop since v4.0.0 broke on it and origin repo has no fix as time of writing this.

Building a custom image

docker build -t serverless-image-handler .  # this builds the image
docker run -i -t serverless-image-handler /bin/bash  # this starts a container using above image, and /bin/bash into it.

Once you are in the container, run the build script to generate a .zip package.

bash-4.2# cd deployment/
bash-4.2# ./build-s3-dist.sh $DIST_OUTPUT_BUCKET $TEMPLATE_OUTPUT_BUCKET $VERSION

You will see a new folder created call dist inside the deployment folder above. You can confirm by

bash-4.2# ls -al dist
total 33792
drwxr-xr-x 3 root root     4096 Jul 30 23:03 .
drwxr-xr-x 1 root root     4096 Jul 30 23:02 ..
-rw-r--r-- 1 root root   876367 Jul 30 23:03 custom-resource.zip
drwxr-xr-x 2 root root     4096 Jul 30 23:03 demo-ui
-rw-r--r-- 1 root root       88 Jul 30 23:03 demo-ui-manifest.json
-rw-r--r-- 1 root root 33676835 Jul 30 23:03 image-handler.zip
-rw-r--r-- 1 root root    30340 Jul 30 23:02 serverless-image-handler.template

There! You have now built 3 .zip packages for lambda to use.

Extract and upload the images you built to S3

In order for AWS lambda image resizing server to use your lambda functions, you will have to upload the .zip files you built in the step above to S3. This step assumes you have setup your AWS credentials here: https://jobber.atlassian.net/wiki/spaces/JTW/pages/470777891/Deploying+Lambdas+WIP#DeployingLambdas(WIP)-Setup

Extract

To extract files form a docker container, you can do the following, this copies the whole dist folder to the current directory you are in.

docker cp <docker_container_id>:/tmp/deployment/dist .

You can find container id with this command.

docker ps

Find the container ID of image serverless-image-handler.

ie. for container ID c376c5a86485

docker cp c376c5a86485:/tmp/deployment/dist .

Upload to S3

Uploads to 2 buckets, jobber-development-harriswong-lambda-us-east-1 and jobber-development-harriswong-lambda. jobber-development-harriswong-lambda stores the template jobber-development-harriswong-lambda-us-east-1 stores the .zip which is what the lambda server uses.

aws s3 cp ./dist/ s3://jobber-serverless-image-handler-us-east-1/serverless-image-handler/v4.0.0/ --recursive --exclude "*" --include "*.zip"
aws s3 cp ./dist/serverless-image-handler.template s3://jobber-serverless-image-handler/serverless-image-handler/v4.0.0/

Deploying to dev environment

Deploy ./dist/serverless-image-handler.template to CloudFormation with the following

aws cloudformation deploy --template-file serverless-image-handler.template --stack-name ServerlessImageHandler --no-execute-changeset --region=us-east-1 --parameter-overrides CorsEnabled='Yes' CorsOrig='*' SourceBuckets='jobber\-development\-.*' LambdaLogRetentionPeriod=30 DeployDemoUI='No' --capabilities CAPABILITY_NAMED_IAM