User Story: MVP Deploy

[DeeMcCart]

EPIC: #7

As a system owner I can deploy and re-deploy the system with MVP functionality so that the site operates as expected with MVP functionality

Assumptions or Pre-Requisites:

• Basic site setup is in place #1
• Debug mode set to FALSE (note that this is now set using an environment variable so longer-term solution)
• First Deploy has been performed and any lessons learned incorporated
• Possibly some additional configuration settings needed in Heroku for MVP deploy (vs first deploy)

Acceptance Criteria: (Must be completed before task is moved to 'Done')

• [x] Must be able to build app inHeroku
• [x] Must be able to run app once built

Tasks - Build ElephantSQL Database

• [x] Start with ElephantSQL and verify database in place... YIKES not there (first deployment took different approach)
• [x] Install psycopg2, dj_database_url - and freeze requirements
• [x] Export data from existing database: python3 manage.py dumpdata products > products.json
• [x] Settings.py: import dj_database_url. Commment out existing DATABASES = and replace with DATABASES= { default: dj_database_url()}
• [x] python3 manage.py showmigrations - should see a list of migrations to be applied to the database I am now connected to (the elephantsql database)
• [x] python3 manage.py migrate - to apply all the database structures to the new database
• [x] IF USING FIXTURES to load JSON data: python3 manage.py loaddata categories; python3 manage.py loaddata products;
• [x] create database superuser(s)
• [x] revert database settings in settings.py, remove URL to new ElephantSQL database and save/commit

Tasks - load Data into ElephantSQL database

• [x] Check migrations in ElephantSQL - open up the ElephantSQL console
• [x] Browser - table queries - auth user - execute - should see superuser(s) setup in previous task
• [ ] OOPs - just realised - becuase I'm not using JSON I will need to manually key the data into my Production database

Tasks - Prepare to Deploy

• [ ] setup secret_key and retrieve it from the environment (refer to 'Deploying to Heroku')
• [x] Add Stripe fields to config variables in Heroku
• [x] If not already done, add a conditional statement into settings.py to link to appropriate database depending on env settings
• [x] Ensure gunicorn is installed to act as web server & freeze
• [x] set Heroku config DISABLE_COLLECTSTATIC to 1 to prevent Heroku from trying to collect static files when we deploy
• [x] create Procfile in root directory to tell Heroku to treat this as a python app/ create a web dyno: web: gunicorn jeweller.wsgi --log-file -
• [x] settings.py ALLOWED_HOSTS = ... Heroku_deployment_name
• [x] commit changes and attempt to build from Heroku

Tasks - Build & Test (without static)

• [x] BUILD IN HEROKU:
• [x] Usual craic with deploy - build

Tasks - Configure AWS for Static files

• [x] CONFIGURE AWS FOR STATIC FILES
• [x] Create bucket,set to public access
• [x] Properties tab - scroll to bottom - static - enable, set default index.html error.html, Save Changes (This gives a new endpoint accessible from internet)
• [x] Permissions tab - scroll to CORS , set to below, then Save Changes [{"AllowedHeaders": ["Authorization"], "AllowedMethods": ["GET"], "AllowedOrigins": ["*"], "ExposeHeaders": []}]
• [x] Permissions tab - Bucket Policy - Edit - Policy generator: type S3 Bucket policy; Allow all Principals ; Action: GetObject; ARN: (paste from other tab); AddStatement button; SaveChanges button; GeneratePolicy button; (copy to other tab); insert / at end of Resource string; SaveChanges button.
• [x] Permissions tab - ACL - Edit - Grant Public: ListAccess; override message; SaveChanges

  Tasks - SET AWS IAM (Identity and Access Management)

• [x] Create Group
• [x] Create Access Policy, attach policy to Group,
• [x] Create user, attach user to Group
• [x] Generate keys

  Tasks - TRANSFER STATIC FILES FROM DEV TO PROD

• [x] Install boto and django-storages, freeze and add 'storages' to installed apps
• [x] Configure settings.py to use AWS for static and media storage
• [x] Create file custom_storages.py and add some content...
• [x] Commit, build deployment
• [ ] Test app

[DeeMcCart]

DMcC 09/02/24 Got instructions from tutor support how to export product dataset from DEV environment (create a json using dumpdata ) Was able to use this to import data into the PROD database YAAY.

[DeeMcCart]

DMcC 10/02/24 deploy-build done but lots of problems with static elements with leading '/' - about 6 of these uncovered and resolved. Big problems with images not being retrieved.... possibly related to removing whitenoise just before deploy (possibly prevengint it from being searched as an image location?)

[DeeMcCart]

DMcC 12/02/24 DONE but errors with static and media files on deployed image prevent its proper use. Closing out this issue as part of Sprint3 review, and adding a fresh task in Sprint4 to resolve MVP deployment problems. Covered in task #56 Tidy Up Deployment