timzhangau/dash-serverless-example

dash example repo with zappa deploying to AWS lambda

dash-serverless-example

dash serverless deployment to AWS lambda

THINGS TO NOTE ON AWS API GATEWAY PATHNAME

AWS lambda has different stages, the pathname will mess up the communication between front end and back end. to avoid that, need to set a pathname prefix equal to the environment name corresponding to AWS API gateway

steps to create and deploy flask app to AWS lambda

• detailed instruction is from: https://medium.com/@Twistacz/flask-serverless-api-in-aws-lambda-the-easy-way-a445a8805028

1. make sure you already have AWS account and have awscli installed on your machine

2. install serverless

   npm install -g serverless

3. create python project and virtual environment. (NOTE: use virtualenv to create virtual environment as the python3 default venv is not working properly with serverless deployment)

   # create new dir and jump in
   $ mkdir quotes
   $ cd quotes/
   # create virtualenv, activate it
   $ virtualenv venv -p python3
   $ . venv/bin/activate.fish
   # install Flask and freeze requirements
   $ (venv) pip install Flask
   $ (venv) pip freeze > requirements.txt

4. write your flask codes

5. create a serverless config yaml file with the following format as example:

   service: quotes
   
   provider:
     name: aws
     runtime: python3.6
     stage: dev
     region: us-east-1
     memorySize: 128

6. install two additional sls plugins.

   $ sls plugin install -n serverless-wsgi
   $ sls plugin install -n serverless-python-requirements

7. test app locally with

   sls wsgi serve

8. if good, deploy with

   sls deploy

Additional documentation on serverless can be found at https://serverless.com/framework/docs/

update existing deployment

• use deploy function or package for existing project for speedier deployment

  sls deploy function --function myFunction

  or

  sls deploy --package path-to-package