LEMP Re-Deployment (LEMP, AutoScale)

LEMP Re-Deployment Steps

### GitHub ###
- Create new release for master branch after merging
- run production instance
- ssh to prod server

$ ssh -i "your_pem.pem" ubuntu@<your public dns>.compute.amazonaws.com


### Ubuntu Console ###

$ sudo supervisorctl stop all
$ cd /var/www/<laravel app directory>
$ sudo git fetch
$ sudo git checkout <tag branch/ master>

- optional command depend on needs
    - migrate table
      $ sudo php artisan migrate
    - seed class
      $ php artisan db:seed —class=<seeder file>
    - edit env file
      $ sudo nano .env
    - edit scheduler
      $ sudo crontab -e
    - edit queue worker
      $ sudo nano /etc/supervisor/conf.d/laravel-worker.conf
    - has changes in route (clear cache & re-caching)
      $ php artisan route:cache
    - has package change
      $ composer install –no-dev
    - purge revoked and expired tokens and auth codes...
      $ php artisan passport:purge

$ chmod 775 storage/framework/cache
$ chmod 775 storage/framework/cache/data
$ sudo chgrp www-data storage/framework/cache
$ sudo chgrp www-data storage/framework/cache/data
$ php artisan queue:restart
$ php artisan config:clear
$ php artisan cache:clear

- update BUGSNAG_APP_VERSION inside env to latest release version.
- exit and stop server

### Create Branch AMI ###
  ec2 > <deployment instance> > action > image and templates > create image
- image name : PROJECT_NAME-VERSION
- tag 'project': PROJECT_NAME,
- wait created and then copy created AMI id


### Launch Configuration ###
  ec2 > Auto Scaling > Launch Configuration
- create new launch config
- paste AMI id
- name same as AMI
- use same security group as previous LC
- existing key pair: deployment server key
- instance type: same as previous


### Auto Scaling Group ###
  ec2 > Auto Scaling > Auto Scaling Groups
- edit ASG by choose latest LC created
- replace previous auto scaling instance by:
    - increase temporary the group min to 2, which will created 2 auto scaling instance inside instance list
    - wait status running/ complete
    - after complete, change back to 1, it will terminate previous (older) auto scaling instance


### Clean Up ###

- ssh into auto scaling instance

$ chmod 775 storage/framework/cache
$ chmod 775 storage/framework/cache/data
$ sudo chgrp www-data storage/framework/cache
$ sudo chgrp www-data storage/framework/cache/data
$ php artisan queue:restart
$ php artisan config:clear
$ php artisan cache:clear
$ supervisorctl restart all

- ssh back into deployment instance

$ php artisan queue:restart
$ php artisan config:clear
$ php artisan cache:clear
$ supervisorctl restart all


- STOP deployment instance
- optional
    - remove all snapshot except latest 2 as backup

install supervisor
https://www.digitalocean.com/community/tutorials/how-to-install-and-manage-supervisor-on-ubuntu-and-debian-vps