jonico

MySQL on RDS configured for import to PlanetScale example

This folder contains an example Terraform configuration that deploys a MySQL database - using RDS in an Amazon Web Services (AWS) account - that is properly configured for PlanetScale's DB import feature.

It will make sure that the RDS database has the binlog exposed, gtid-mode set to ON and is using ROW-based replication. All these are pre-requisites for imports to work without zero downtime.

If you are going to write a lot of data in your database in a very short amount of time, don't forget the only manual step after the Terraform setup.

jonico

PLANETSCALE_DB=brandnewdb
PLANETSCALE_BRANCH=mybranch
PLANETSCALE_ORG=jonico
PLANETSCALE_SERVICE_TOKEN=pscale_tkn_loCzIH7NktDK-GWJ71eX97Qr5D3a9iEO_pgHCSHUtw  
PLANETSCALE_SERVICE_TOKEN_NAME=69xrlIwgs4ms

jonico

#!/bin/sh

# shell script that waits for a PlanetScale branch to be ready for use and increases retry times exponentially

#  usage: wait-for-ps-branch-readiness.sh <db name> <branch name> <max retries>

function wait_for_branch_readiness {
    local retries=$1
    local db=$2
    local branch=$3

jonico

# copy Azure bloc storage files to S3 buckets
#
# Usage:
#   ./copy-azure-blob-storage-to-s3.sh <blob storage container> <s3 bucket>
#
# Example:
#   ./copy-azure-blob-storage-to-s3.sh my-container s3://my-bucket
#
# Note:
#   This script requires a working Azure CLI.

jonico

#!/usr/bin/bash
cd
mkdir -p /tmp/actions
git clone https://github.com/github/platform-samples.git
sdk install groovy
cd platform-samples/api/groovy/
groovy ListReposInOrg.groovy -t $GH_PUBLIC actions > ~/repo-list

cd
wget https://github.com/actions/actions-sync/releases/download/v202009231612/gh_202009231612_linux_amd64.tar.gz 

jonico

apiVersion: v1
items:
- apiVersion: v1
  kind: Pod
  metadata:
    creationTimestamp: "2021-04-29T15:09:09Z"
    labels:
      pod-template-hash: 65f67dcb66
      runner-deployment-name: actions-runner-deployment-epona
      runner-template-hash: 6788f4694c

jonico

# Replace the string "Hyrule" with your search pattern - only run on backup system - will take a LOOOONG time to run through
ghe-console -y <<'ENDSCRIPT'
  Repository.find_each do |repository|
    if repository.active && !repository.empty? then
      host,path = GitHub::Routing.lookup_route(repository.nwo)
        puts "Processing repo #{repository.nwo} ..."
        refs=`git --git-dir=#{path} rev-list --branches --tags | sort | uniq`
        refs.each_line do |ref|
          if system("git --git-dir=#{path} grep 'Hyrule' #{ref}")
            puts "Found pattern in ref #{ref} in repo #{repository.nwo}: "

jonico

ghe-console -y <<'ENDSCRIPT'
ActiveRecord::Base.connected_to(role: :reading) do
  puts "Version 0.2.0"
  emails = Set.new
  start_time = 90.days.ago.beginning_of_day
  Repository.where(active: true).find_each do |repo|
    Push
      .where(repository: repo)
      .where("created_at >= ?", start_time)
      .find_each do |push|

jonico

name: Build and publish Docker image + Container Scanning
on: [push]
jobs:

  build:
    name: Build
    runs-on: ubuntu-latest
    outputs:
      dockerimage: ${{ steps.dockerimage.outputs.dockerimage }}
      enablecontainerscanning: ${{ steps.enablecontainerscanning.outputs.enablecontainerscanning }}

jonico

name: CI
on:
  push:
    branches: [ master ]
  pull_request:
    branches: [ master ]

jobs:

  build: