SapphireHWMigration/sapphire_hardware_migration_checklist.md

## Sapphire Migration Checklist

### Meta

  - Migrate stagingsapphire to new hardware (following the procedure below)
  - Migrate stagingsapphire with dump from sapphire
  - Migrate sapphire to new hardware (following the procedure below)
  - Drop databases on stagingsapphire and initialize with dummy data through rake task.

### Webserver

  - switch to nginx
  - ssl certs (?)
  - new.sapphire.isds.tugraz.at
  - https://www.phusionpassenger.com/docs/advanced_guides/install_and_upgrade/nginx/install/oss/el8.html14:37

### Questions (for Matthias and/or Helmut)

  - CentOS 8 Stream is a rolling release? Are there and implications for sapphire because of some security critical OS updates we will have to do?
    (website reads: "**Continuously delivered** distro ...[https://www.centos.org/centos-stream/])
    - How are package updates going to be handled?

### Prerequisites

  - create production db dump
  - **copy** assets to new hardware

### Install

  - Database: PostgreSQL 8.4.20 -> probs to 13.xx.xx
  - Ruby (install via rbenv -> 2.6.5p114 (2019-10-01 revision 67812)) -> 2.7.xx
    - https://medium.com/@adrienjarthon/ruby-jemalloc-results-for-updown-io-d3fb2d32f67f
  - Redis (Redis server v=3.2.12 sha=00000000:0 malloc=jemalloc-3.6.0 bits=64 build=b2d74fe5fff7657d) -> 6.2.5
  - NodeJS (v0.10.35) -> ??
  - Bundler
  - Cap Deploy (check deploy target)

### Setup

  - setup config files:
    - database.yml
    - mail.yml
    - secrets.yml
    - sidekiq.yml
  - gem install bundler
  - import DB dump (lookup for appropriate GIST/snippet TODO)
  - check if assets are correctly linked (i.e. in the correct folder, is this sufficient?)

### Checks

  - check certificate
  - check auto https
  - check mailing
  - check upload (tests sidekiq)
  - check entering of new ratings (tests redis)