Skip to content

Use AWS CDK Create Aurora Global Database.

License

Notifications You must be signed in to change notification settings

neilkuan/cdk-aurora-globaldatabase

Folders and files

NameName
Last commit message
Last commit date

Latest commit

1b377bb Â· Dec 27, 2024
Oct 29, 2024
Nov 10, 2024
Feb 23, 2024
Oct 20, 2020
Sep 21, 2024
Sep 16, 2023
Nov 10, 2024
Aug 23, 2024
Jan 6, 2024
Feb 6, 2024
Oct 18, 2024
Jan 6, 2024
Sep 21, 2024
Aug 27, 2024
Apr 21, 2021
Jun 4, 2021
Oct 22, 2022
Dec 21, 2024
Jan 6, 2024
Sep 18, 2020
Apr 21, 2021
Dec 27, 2024

Repository files navigation

NPM version PyPI version Release

Downloads npm PyPI

cdk-aurora-globaldatabase

cdk-aurora-globaldatabase is an AWS CDK construct library that allows you to create Amazon Aurora Global Databases with AWS CDK in Typescript or Python.

Why

Amazon Aurora Global Databases is designed for multi-regional Amazon Aurora Database clusters that span across different AWS regions. Due to the lack of native cloudformation support, it has been very challenging to build with cloudformation or AWS CDK with the upstream aws-rds construct.

cdk-aurora-globaldatabase aims to offload the heavy-lifting and helps you provision and deploy cross-regional Amazon Aurora Global Databases simply with just a few CDK statements.

Install

Use the npm dist tag to opt in CDKv1 or CDKv2:

// for CDKv2
npm install cdk-aurora-globaldatabase
or
npm install cdk-aurora-globaldatabase@latest

// for CDKv1
npm install cdk-aurora-globaldatabase@cdkv1

⛔️ Please do not use cdk v1, because lot of db engine version already not been update in @aws-cdk/aws-rds upstream. ⛔️

Now Try It !!!

Sample for Mysql

import { GlobalAuroraRDSMaster, InstanceTypeEnum, GlobalAuroraRDSSlaveInfra } from 'cdk-aurora-globaldatabase';
import { App, Stack, CfnOutput } from 'aws-cdk-lib';
import * as ec2 from 'aws-cdk-lib/aws-ec2';
// new app .
const mockApp = new App();

// setting two region env config .
const envSingapro  = { account: process.env.CDK_DEFAULT_ACCOUNT, region: 'ap-southeast-1' };
const envTokyo = { account: process.env.CDK_DEFAULT_ACCOUNT, region: 'ap-northeast-1' };

// create stack main .
const stackM = new Stack(mockApp, 'testing-stackM',{env: envTokyo});
const vpcPublic = new ec2.Vpc(stackM,'defaultVpc',{
  natGateways: 0,
  maxAzs: 3,
  subnetConfiguration: [{
    cidrMask: 26,
    name: 'masterVPC2',
    subnetType: ec2.SubnetType.PUBLIC,
  }],
});
const  globaldbM = new GlobalAuroraRDSMaster(stackM, 'globalAuroraRDSMaster',{
  instanceType: InstanceTypeEnum.R5_LARGE,
  vpc: vpcPublic,
  rdsPassword: '1qaz2wsx',
});
globaldbM.rdsCluster.connections.allowDefaultPortFrom(ec2.Peer.ipv4(`${process.env.MYIP}/32`))

// create stack slave infra or you can give your subnet group.
const stackS = new Stack(mockApp, 'testing-stackS',{env: envSingapro});
const vpcPublic2 = new ec2.Vpc(stackS,'defaultVpc2',{
  natGateways: 0,
  maxAzs: 3,
  subnetConfiguration: [{
    cidrMask: 26,
    name: 'secondVPC2',
    subnetType: ec2.SubnetType.PUBLIC,
  }],
});
const globaldbS = new GlobalAuroraRDSSlaveInfra(stackS, 'slaveregion',{vpc: vpcPublic2,subnetType:ec2.SubnetType.PUBLIC });

// so we need to wait stack slave created first .
stackM.addDependency(stackS)


new CfnOutput(stackM, 'password', { value: globaldbM.rdsPassword });
// add second region cluster
globaldbM.addRegionalCluster(stackM,'addregionalrds',{
  region: 'ap-southeast-1',
  dbSubnetGroupName: globaldbS.dbSubnetGroup.dbSubnetGroupName,
});

like this

Sample for Postgres

import { GlobalAuroraRDSMaster, InstanceTypeEnum, GlobalAuroraRDSSlaveInfra } from 'cdk-aurora-globaldatabase';
import { App, Stack, CfnOutput } from 'aws-cdk-lib';
import * as ec2 from 'aws-cdk-lib/aws-ec2';
import * as _rds from 'aws-cdk-lib/aws-rds';

const mockApp = new App();
const envSingapro  = { account: process.env.CDK_DEFAULT_ACCOUNT, region: 'ap-southeast-1' };
const envTokyo = { account: process.env.CDK_DEFAULT_ACCOUNT, region: 'ap-northeast-1' };

const stackM = new Stack(mockApp, 'testing-stackM',{env: envTokyo});
const vpcPublic = new ec2.Vpc(stackM,'defaultVpc',{
  natGateways: 0,
  maxAzs: 3,
  subnetConfiguration: [{
    cidrMask: 26,
    name: 'masterVPC2',
    subnetType: ec2.SubnetType.PUBLIC,
  }],
});

// Note if you use postgres , need to give the same value in engineVersion and  dbClusterpPG's engine .
const globaldbM = new GlobalAuroraRDSMaster(stackM, 'globalAuroraRDSMaster',{
  instanceType: InstanceTypeEnum.R5_LARGE,
  vpc: vpcPublic,
  rdsPassword: '1qaz2wsx',
  engineVersion: _rds.DatabaseClusterEngine.auroraPostgres({
    version: _rds.AuroraPostgresEngineVersion.VER_12_11}),
  dbClusterpPG: new _rds.ParameterGroup(stackM, 'dbClusterparametergroup', {
    engine: _rds.DatabaseClusterEngine.auroraPostgres({
      version: _rds.AuroraPostgresEngineVersion.VER_12_11,
    }),
    parameters: {
      'rds.force_ssl': '1',
      'rds.log_retention_period': '10080',
      'auto_explain.log_min_duration': '5000',
      'auto_explain.log_verbose': '1',
      'timezone': 'UTC+8',
      'shared_preload_libraries': 'auto_explain,pg_stat_statements,pg_hint_plan,pgaudit',
      'log_connections': '1',
      'log_statement': 'ddl',
      'log_disconnections': '1',
      'log_lock_waits': '1',
      'log_min_duration_statement': '5000',
      'log_rotation_age': '1440',
      'log_rotation_size': '102400',
      'random_page_cost': '1',
      'track_activity_query_size': '16384',
      'idle_in_transaction_session_timeout': '7200000',
    },
  }),
});
globaldbM.rdsCluster.connections.allowDefaultPortFrom(ec2.Peer.ipv4(`${process.env.MYIP}/32`))

const stackS = new Stack(mockApp, 'testing-stackS',{env: envSingapro});
const vpcPublic2 = new ec2.Vpc(stackS,'defaultVpc2',{
  natGateways: 0,
  maxAzs: 3,
  subnetConfiguration: [{
    cidrMask: 26,
    name: 'secondVPC2',
    subnetType: ec2.SubnetType.PUBLIC,
  }],
});
const globaldbS = new GlobalAuroraRDSSlaveInfra(stackS, 'slaveregion',{
  vpc: vpcPublic2,subnetType:ec2.SubnetType.PUBLIC, 
});

stackM.addDependency(stackS)


new CfnOutput(stackM, 'password', { value: globaldbM.rdsPassword });
// add second region cluster
globaldbM.addRegionalCluster(stackM,'addregionalrds',{
  region: 'ap-southeast-1',
  dbSubnetGroupName: globaldbS.dbSubnetGroup.dbSubnetGroupName,
});

To deploy

cdk deploy

To destroy

cdk destroy

đź‘Ź Supporters

Stargazers repo roster for @neilkuan/cdk-aurora-globaldatabase Forkers repo roster for @neilkuan/cdk-aurora-globaldatabase