cdk-rds-sql 9.0.1 → 9.1.0

package/.jsii

@@ -9748,7 +9748,7 @@
   },
   "name": "cdk-rds-sql",
   "readme": {
-    "markdown": "# About\n\nThis CDK construct library makes it possible to create databases,\nschemas, and roles in an Aurora Serverless v2, RDS Database Cluster,\nDatabase Instance, or AWS DSQL clusters. PostgreSQL, MySQL, and DSQL\ndatabases are supported.\n\nThis construct library is intended to be used in enterprise\nenvironments, and works in isolated subnets.\n\n<p align=\"left\">\n  <a href=\"https://github.com/semantic-release/semantic-release\"><img src=\"https://img.shields.io/badge/semantic--release-angular-e10079?logo=semantic-release\" alt=semantic-release: angular\"></a>\n  <a href=\"https://github.com/berenddeboer/cdk-rds-sql/actions/workflows/release.yml\"><img src=\"https://github.com/berenddeboer/cdk-rds-sql/actions/workflows/release.yml/badge.svg\" alt=\"Release badge\"></a>\n</p>\n\n# Requirements\n\n- CDK v2.\n- Node.js 22.14 or newer for consumers, as declared in `package.json`'s `engines`.\n  This is the library's supported minimum, independent of its development tools.\n\n# Installation\n\n     npm i cdk-rds-sql\n\n# Usage\n\n## Provider\n\nFirst setup your VPC and create your cluster:\n\n```ts\nimport { Duration, RemovalPolicy } from \"aws-cdk-lib\"\nimport * as ec2 from \"aws-cdk-lib/aws-ec2\"\nimport * as rds from \"aws-cdk-lib/aws-rds\"\n\nconst vpc = ec2.Vpc.fromLookup(this, \"Vpc\", {\n  vpcId: this.node.tryGetContext(\"vpc-id\"),\n})\n\nconst cluster = new rds.ServerlessCluster(this, \"Cluster\", {\n  vpc: vpc,\n  vpcSubnets: {\n    subnetType: ec2.SubnetType.PRIVATE_ISOLATED,\n  },\n  engine: rds.DatabaseClusterEngine.AURORA_POSTGRESQL,\n  parameterGroup: rds.ParameterGroup.fromParameterGroupName(\n    this,\n    \"ParameterGroup\",\n    \"default.aurora-postgresql11\"\n  ),\n  removalPolicy: RemovalPolicy.DESTROY,\n  scaling: {\n    autoPause: Duration.minutes(60),\n    minCapacity: rds.AuroraCapacityUnit.ACU_2,\n    maxCapacity: rds.AuroraCapacityUnit.ACU_2,\n  },\n})\n```\n\nThen create a provider which will connect to your database. For a cluster:\n\n```ts\nimport { Provider } from \"cdk-rds-sql\"\n\nconst provider = new Provider(this, \"Provider\", {\n  vpc: vpc,\n  cluster: cluster,\n  secret: cluster.secret!,\n})\n```\n\nFor an instance:\n\n```ts\nimport { Provider } from \"cdk-rds-sql\"\n\nconst provider = new Provider(this, \"Provider\", {\n  vpc: vpc,\n  instance: instance,\n  secret: cluster.secret!,\n})\n```\n\nThe provider will setup a lambda, which normally lives in the same VPC\nas the database. You can give a different VPC, as long as that VPC has\naccess to the VPC of the database. Only the provider lambda will talk\nto your database.\n\nThe provider will by default use the private isolated subnet of the\nVPC. Your isolated network must have a VPC endpoint to AWS Secrets Manager\nand possibly KMS as well. If you want to use a subnet with egress\naccess in case you have no such VPC endpoints, specify the subnet as\nfollows:\n\n```ts\nimport { Provider } from \"cdk-rds-sql\"\n\nconst provider = new Provider(this, \"Provider\", {\n  vpc: vpc,\n  vpcSubnets: {\n    subnetType: ec2.SubnetType.PRIVATE_WITH_EGRESS,\n  },\n  cluster: cluster,\n  secret: cluster.secret!,\n})\n```\n\n## Roles\n\nCreate a postgres role (or mysql user) for a cluster as follows:\n\n```ts\nimport { Role } from \"cdk-rds-sql\"\n\nconst role = new Role(this, \"Role\", {\n  provider: provider,\n  roleName: \"myrole\",\n  databaseName: \"mydb\",\n})\n```\n\nThis will automatically create a secret just like `ServerlessCluster`\ndoes, with all the connection info needed for this user. It's secret\nvalue is a JSON like:\n\n```json\n{\n  \"dbClusterIdentifier\": \"teststack-clustereb1186t9-sh4wpqfdyfvu\",\n  \"password\": \"very-long-and-boring\",\n  \"dbname\": \"mydb\",\n  \"engine\": \"postgres\",\n  \"port\": 5432,\n  \"host\": \"teststack-clustereb1186t9-sh4wpqfdyfvu.cluster-cgudolabssna.us-east-1.rds.amazonaws.com\",\n  \"username\": \"myrole\"\n}\n```\n\nYou can access the secret via `role.secret`.\n\nInstead of `databaseName` you can also specify `database` to reference\nan existing database. The role will not be created until the database\nhas been created.\n\nIf you want to make the role the owner of a new database, just specify\nthe `databaseName` here, and create the database later.\n\n### Connection limit (MySQL/MariaDB)\n\nCap how many connections a MySQL or MariaDB user may open:\n\n```ts\nconst role = new Role(this, \"Role\", {\n  provider: provider,\n  roleName: \"myrole\",\n  databaseName: \"mydb\",\n  maxUserConnections: 5,\n})\n```\n\nCreate and update then run:\n\n```sql\nALTER USER 'myrole'@'%' WITH MAX_USER_CONNECTIONS 5\n```\n\nOmit `maxUserConnections` to leave the user unlimited, which is the\nexisting behaviour. Omitting it does not clear a limit that was already\napplied; set `0` to restore the server default (unlimited unless the\nglobal `max_user_connections` is set). PostgreSQL and DSQL throw if\nthis property is set.\n\n### IAM Authentication\n\nInstead of password-based authentication, you can create roles that\nuse AWS IAM database authentication. This eliminates the need to store\ndatabase passwords and provides enhanced security through AWS IAM\npolicies.\n\n```ts\nimport { Role } from \"cdk-rds-sql\"\n\nconst iamRole = new Role(this, \"IamRole\", {\n  provider: provider,\n  roleName: \"myiamrole\",\n  databaseName: \"mydb\",\n  enableIamAuth: true,\n})\n```\n\nWhen `enableIamAuth` is set to `true`:\n\n- **No secret is created** - the `role.secret` property will be `undefined`\n- The database role is created without a password\n- Your application must generate an IAM authentication token at runtime\n\nThis is the same behavior as DSQL clusters, which always use IAM authentication.\n\n**Requirements for IAM Authentication:**\n\n- SSL connections are required (enabled by default in this library)\n- Your application must have IAM permissions to connect to the database\n- The database user/role name must match the IAM identity\n\n**IAM Policy Example:**\n\nYour application will need an IAM policy like this to connect:\n\n```json\n{\n  \"Version\": \"2012-10-17\",\n  \"Statement\": [\n    {\n      \"Effect\": \"Allow\",\n      \"Action\": [\"rds-db:connect\"],\n      \"Resource\": [\n        \"arn:aws:rds-db:region:account-id:dbuser:cluster-resource-id/myiamrole\"\n      ]\n    }\n  ]\n}\n```\n\n**Generating an IAM Authentication Token:**\n\nAt runtime, your application needs to generate an authentication token\nusing the AWS SDK. Here's an example using the AWS SDK for JavaScript:\n\n```ts\nimport { Signer } from \"@aws-sdk/rds-signer\"\nimport { Client } from \"pg\"\n\nconst signer = new Signer({\n  hostname: \"your-cluster.cluster-xxxxx.us-east-1.rds.amazonaws.com\",\n  port: 5432,\n  username: \"myiamrole\",\n  region: \"us-east-1\",\n})\n\nconst token = await signer.getAuthToken()\n\nconst client = new Client({\n  host: \"your-cluster.cluster-xxxxx.us-east-1.rds.amazonaws.com\",\n  port: 5432,\n  user: \"myiamrole\",\n  database: \"mydb\",\n  password: token,\n  ssl: { rejectUnauthorized: true },\n})\n\nawait client.connect()\n```\n\nBoth PostgreSQL and MySQL databases support IAM authentication. For more details, see the [AWS RDS IAM Database Authentication documentation](https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/UsingWithRDS.IAMDBAuth.html).\n\n### MySQL support\n\nIn MySQL users are created with '%' as value for the host. It is hard\nto do this better:\n\n- Determine the CIDR blocks used by a VPC is not trivial.\n- For imported VPCs you cannot specify the IPv6 CIDR.\n- CIDRs might change without the system knowing, meaning applications\n  could lose access to the database at random times, such as when a\n  container is restarted.\n\n## Database\n\nCreate a database as follows:\n\n```ts\nimport { Database } from \"cdk-rds-sql\"\n\nconst database = new Database(this, \"Database\", {\n  provider: provider,\n  databaseName: \"mydb\",\n})\n```\n\nYou can provide an owner, which makes it easy to create database owned\nby a new user:\n\n```ts\nconst database = new Database(this, \"Database\", {\n  provider: provider,\n  databaseName: \"mydb\",\n  owner: role,\n})\n```\n\n## Schema\n\nCreate a schema in the default database as follows:\n\n```ts\nimport { Schema } from \"cdk-rds-sql\"\n\nnew Schema(this, \"Schema\", {\n  provider: provider,\n  schemaName: \"myschema\",\n})\n```\n\nOr in another database:\n\n```ts\nconst database = new Database(this, \"Database\", {\n  provider: provider,\n  databaseName: \"mydb\",\n})\n\nnew Schema(this, \"Schema\", {\n  provider: provider,\n  schemaName: \"myschema\",\n  databaseName: database.databaseName,\n})\n```\n\nOne may need a role permitted for using schema:\n\n```ts\nnew Schema(this, \"Schema\", {\n  provider: provider,\n  schemaName: \"myschema\",\n  databaseName: database.databaseName,\n  role: role,\n})\n```\n\n## Sql\n\nYou can insert arbitrary SQL into your database with the `Sql` construct:\n\n```ts\nimport { Sql } from \"cdk-rds-sql\"\n\nconst sql = new Sql(this, \"Sql\", {\n  provider: provider,\n  database: database,\n  statement: \"create table t (i int)\",\n})\n```\n\nCreate a table if it does not exist, and grant a role privileges:\n\n```ts\nconst sql = new Sql(this, \"Sql\", {\n  provider: provider,\n  database: database,\n  statement: `\ncreate table if not exists t (i int);\ngrant select on t to myrole;\n`,\n})\n```\n\nRollback sql on stack deletion:\n\n```ts\nconst sql = new Sql(this, \"Sql\", {\n  provider: provider,\n  database: database,\n  statement: `\ncreate table if not exists t (i int);\ngrant select on t to myrole;\n`,\n  rollback: `\nDO $$BEGIN\n  IF EXISTS (select from pg_database WHERE datname = 't') THEN\n    IF EXISTS (select from pg_catalog.pg_roles WHERE rolname = 'myrole') THEN\n      revoke select t from myrole;\n    END IF;\n    drop table t;\n  END IF;\nEND$$;\n`,\n})\n```\n\nNote that there is no synchronisation between various `Sql`\nconstructs, in particular the order in your code does not determine\nthe order in which your SQL is executed. This happens in parallel,\nunless you specify an explicit dependency via `sql.node.addDepency()`.\n\nThere are a lot of concerns when using `Sql`:\n\n- When you update your Sql, your previous Sql is not \"rolled back\",\n  the new Sql is simply executed again.\n- When you delete your `Sql` construct the rollback is executed if specified\n- When permission are granted via `Sql` they must be removed via\n  rollback to succesfully remove the role\n- Currently the `Sql` constructs has less than 5 minutes to execute\n  its work.\n- It is unknown how large your SQL can be.\n\n# Parameters\n\nSome lambda constructs, in particular [Bref](https://bref.sh/), do not\nsupport secrets out of the box. Use the `parameterPrefix` property to\ncreate SSM parameters in addition to a secret:\n\n```ts\nconst role = new Role(this, \"Role\", {\n  provider: provider,\n  roleName: \"myrole\",\n  databaseName: \"mydb\",\n  parameterPrefix: \"/my-app/\",\n})\n```\n\nThis will create `/my-app/username`, `/my-app/password` and such.\n\nTo access parameters you will need IAM permissions such as:\n\n```ts\ninitialPolicy: [\n  new iam.PolicyStatement({\n\tactions: [\"ssm:GetParameter\", \"ssm:GetParameters\"],\n\tresources: [\n\t  // Grant access to all parameters under the base path\n\t  `arn:aws:ssm:${this.region}:${this.account}:parameter/my-app/*`,\n\t],\n\teffect: iam.Effect.ALLOW,\n  })\n],\n```\n\nNote that your VPC will need an SSM Parameters interface endpoint to support this.\n\n# AWS DSQL Support\n\nThis construct also supports AWS DSQL. DSQL is a great number of\nlimitations around it's postgresql support, so there is some learing\ncurve. This library supports creating roles, schemas, and executing\nSQL on DSQL clusters. Creating databases is not supported by DSQL.\n\n## DSQL Provider Setup\n\nDSQL clusters don't require a VPC or secrets since they use IAM authentication:\n\n```ts\nimport * as dsql from \"aws-cdk-lib/aws-dsql\"\nimport { Provider } from \"cdk-rds-sql\"\n\nconst dsqlCluster = new dsql.CfnCluster(this, \"DsqlCluster\", {\n  deletionProtectionEnabled: false,\n})\n\nconst provider = new Provider(this, \"Provider\", {\n  cluster: dsqlCluster,\n  // No VPC or secret needed for DSQL\n})\n```\n\nYou can access DSQL from within a VPC, but cdk-rds-sql has not been\nspecifically tested with that scenario.\n\n## DSQL Roles\n\nDSQL always uses IAM authentication, so the `enableIamAuth` property is ignored:\n\n```ts\nimport { Role } from \"cdk-rds-sql\"\n\nconst role = new Role(this, \"Role\", {\n  provider: provider,\n  roleName: \"myrole\",\n})\n```\n\n## Using DSQL\n\nConnecting to DSQL requires a couple of things:\n\n1. Your IAM role needs the `dsql:DbConnect` IAM permission to connect to the cluster.\n2. You probably want a normal postgres role (not \"admin\") to use from your apps.\n3. Your IAM role needs to be connected to your postgres role: use the `IamGrant` construct for that:\n\n   ```ts\n   new IamGrant(this, \"LambdaIamGrant\", {\n     provider: provider,\n     roleName: role.roleName,\n     resourceArn: queryLambda.role!.roleArn,\n   })\n   ```\n\n4. From your code use the Dsql signer to create the password:\n\n   ```ts\n   import { DsqlSigner } from \"@aws-sdk/dsql-signer\"\n   import { Client } from \"pg\"\n\n   const region =\n     process.env[\"AWS_REGION\"] || process.env[\"AWS_DEFAULT_REGION\"] || \"us-west-2\"\n\n   // Generate DSQL auth token using AWS SDK\n   const signer = new DsqlSigner({\n     hostname: host,\n     region,\n   })\n\n   const authToken = await signer.getDbConnectAuthToken()\n\n   const client = new Client({\n     host,\n     port: 5432,\n     user: dbUser,\n     database: dbName,\n     password: authToken,\n     ssl: { rejectUnauthorized: false },\n   })\n\n   await client.connect()\n   ```\n\n# IPv6\n\nIf you use the provider in an IPv6 subnet you probably need these settings:\n\n```ts\nimport { Provider } from \"cdk-rds-sql\"\n\nconst provider = new Provider(this, \"Provider\", {\n  ...\n  functionProps: {\n\tipv6AllowedForDualStack: true,\n\tallowAllIpv6Outbound: true,\n  },\n}\n```\n\n# Disabling SSL\n\nThe default connection to RDS is ssl enabled. You can disable ssl by\nsetting the `ssl` option to `false`:\n\n```ts\nconst provider = new Provider(this, \"Provider\", {\n  vpc: vpc,\n  instance: instance,\n  secret: cluster.secret!,\n  ssl: false, // default is true\n})\n```\n\n# Working on this code\n\nThis code is managed by [projen](https://github.com/projen/projen).\n[mise](https://mise.jdx.dev/) installs Node 24, pnpm 12, and\n[hk](https://hk.jdx.dev/) for Git hooks. Biome handles formatting and linting.\n\nDevelopment and CI use Node 24. Testcontainers is a development-only dependency;\nits dependency tree requires newer APIs than the consumer minimum of Node 22.14.\n\nAfter cloning:\n\n```sh\nmise trust\nmise install\nmise run setup\n```\n\nWhen migrating an existing clone, replace the old pre-commit hooks with\n`mise exec -- hk install --mise --force`.\n\nRun all checks with `mise run check`, or format and lint with\n`pnpm exec projen biome`. Tests require Docker. TypeScript stays on the\n6.0 release line supported by jsii 6.0.\n\nThe pre-commit hook runs the full build, tests, and packaging, so Docker must\nbe running. When `VPC_ID` is set, changes under `src/` also run CDK synthesis\nagainst that VPC; this requires AWS credentials. Unset `VPC_ID` to skip synthesis.\n\nThe development tsconfig uses `node16` module resolution, including package\nexports. Jest tasks enable Node's VM module support for native dynamic imports\nsuch as the MySQL driver import.\n\n## Testing\n\nTest code via projen with:\n\n    pnpm exec projen test\n\nTests do not update snapshots automatically. To intentionally refresh them, run\n`pnpm exec projen test:update` and review the snapshot diff.\n\nYou can run the PostgreSQL sample stack with:\n\n    pnpm exec projen integ:deploy:postgresql:serverless\n\nIf you want to use an existing vpc:\n\n    pnpm exec cdk deploy --context vpc-id=vpc-0123456789 TestRdsSqlServerlessV2Stack\n\nYou can also test DSQL support with:\n\n    pnpm exec projen integ:deploy:dsql\n\n# To do\n\n- Update role: will not revoke connect to previous database if database name has changed.\n"
+    "markdown": "# About\n\nThis CDK construct library makes it possible to create databases,\nschemas, and roles in an Aurora Serverless v2, RDS Database Cluster,\nDatabase Instance, or AWS DSQL clusters. PostgreSQL, MySQL, and DSQL\ndatabases are supported.\n\nThis construct library is intended to be used in enterprise\nenvironments, and works in isolated subnets.\n\n<p align=\"left\">\n  <a href=\"https://github.com/semantic-release/semantic-release\"><img src=\"https://img.shields.io/badge/semantic--release-angular-e10079?logo=semantic-release\" alt=semantic-release: angular\"></a>\n  <a href=\"https://github.com/berenddeboer/cdk-rds-sql/actions/workflows/release.yml\"><img src=\"https://github.com/berenddeboer/cdk-rds-sql/actions/workflows/release.yml/badge.svg\" alt=\"Release badge\"></a>\n</p>\n\n# Requirements\n\n- CDK v2.\n- Node.js 22.14 or newer for consumers, as declared in `package.json`'s `engines`.\n  This is the library's supported minimum, independent of its development tools.\n\n# Installation\n\n     npm i cdk-rds-sql\n\nFor .NET (available from the first NuGet release):\n\n```sh\ndotnet add package BerenddeBoer.CdkRdsSql\n```\n\nThe .NET namespace is `BerenddeBoer.CdkRdsSql`. Node.js is also required\nwhen using the .NET bindings.\n\n# Usage\n\n## Provider\n\nFirst setup your VPC and create your cluster:\n\n```ts\nimport { Duration, RemovalPolicy } from \"aws-cdk-lib\"\nimport * as ec2 from \"aws-cdk-lib/aws-ec2\"\nimport * as rds from \"aws-cdk-lib/aws-rds\"\n\nconst vpc = ec2.Vpc.fromLookup(this, \"Vpc\", {\n  vpcId: this.node.tryGetContext(\"vpc-id\"),\n})\n\nconst cluster = new rds.ServerlessCluster(this, \"Cluster\", {\n  vpc: vpc,\n  vpcSubnets: {\n    subnetType: ec2.SubnetType.PRIVATE_ISOLATED,\n  },\n  engine: rds.DatabaseClusterEngine.AURORA_POSTGRESQL,\n  parameterGroup: rds.ParameterGroup.fromParameterGroupName(\n    this,\n    \"ParameterGroup\",\n    \"default.aurora-postgresql11\"\n  ),\n  removalPolicy: RemovalPolicy.DESTROY,\n  scaling: {\n    autoPause: Duration.minutes(60),\n    minCapacity: rds.AuroraCapacityUnit.ACU_2,\n    maxCapacity: rds.AuroraCapacityUnit.ACU_2,\n  },\n})\n```\n\nThen create a provider which will connect to your database. For a cluster:\n\n```ts\nimport { Provider } from \"cdk-rds-sql\"\n\nconst provider = new Provider(this, \"Provider\", {\n  vpc: vpc,\n  cluster: cluster,\n  secret: cluster.secret!,\n})\n```\n\nFor an instance:\n\n```ts\nimport { Provider } from \"cdk-rds-sql\"\n\nconst provider = new Provider(this, \"Provider\", {\n  vpc: vpc,\n  instance: instance,\n  secret: cluster.secret!,\n})\n```\n\nThe provider will setup a lambda, which normally lives in the same VPC\nas the database. You can give a different VPC, as long as that VPC has\naccess to the VPC of the database. Only the provider lambda will talk\nto your database.\n\nThe provider will by default use the private isolated subnet of the\nVPC. Your isolated network must have a VPC endpoint to AWS Secrets Manager\nand possibly KMS as well. If you want to use a subnet with egress\naccess in case you have no such VPC endpoints, specify the subnet as\nfollows:\n\n```ts\nimport { Provider } from \"cdk-rds-sql\"\n\nconst provider = new Provider(this, \"Provider\", {\n  vpc: vpc,\n  vpcSubnets: {\n    subnetType: ec2.SubnetType.PRIVATE_WITH_EGRESS,\n  },\n  cluster: cluster,\n  secret: cluster.secret!,\n})\n```\n\n## Roles\n\nCreate a postgres role (or mysql user) for a cluster as follows:\n\n```ts\nimport { Role } from \"cdk-rds-sql\"\n\nconst role = new Role(this, \"Role\", {\n  provider: provider,\n  roleName: \"myrole\",\n  databaseName: \"mydb\",\n})\n```\n\nThis will automatically create a secret just like `ServerlessCluster`\ndoes, with all the connection info needed for this user. It's secret\nvalue is a JSON like:\n\n```json\n{\n  \"dbClusterIdentifier\": \"teststack-clustereb1186t9-sh4wpqfdyfvu\",\n  \"password\": \"very-long-and-boring\",\n  \"dbname\": \"mydb\",\n  \"engine\": \"postgres\",\n  \"port\": 5432,\n  \"host\": \"teststack-clustereb1186t9-sh4wpqfdyfvu.cluster-cgudolabssna.us-east-1.rds.amazonaws.com\",\n  \"username\": \"myrole\"\n}\n```\n\nYou can access the secret via `role.secret`.\n\nInstead of `databaseName` you can also specify `database` to reference\nan existing database. The role will not be created until the database\nhas been created.\n\nIf you want to make the role the owner of a new database, just specify\nthe `databaseName` here, and create the database later.\n\n### Connection limit (MySQL/MariaDB)\n\nCap how many connections a MySQL or MariaDB user may open:\n\n```ts\nconst role = new Role(this, \"Role\", {\n  provider: provider,\n  roleName: \"myrole\",\n  databaseName: \"mydb\",\n  maxUserConnections: 5,\n})\n```\n\nCreate and update then run:\n\n```sql\nALTER USER 'myrole'@'%' WITH MAX_USER_CONNECTIONS 5\n```\n\nOmit `maxUserConnections` to leave the user unlimited, which is the\nexisting behaviour. Omitting it does not clear a limit that was already\napplied; set `0` to restore the server default (unlimited unless the\nglobal `max_user_connections` is set). PostgreSQL and DSQL throw if\nthis property is set.\n\n### IAM Authentication\n\nInstead of password-based authentication, you can create roles that\nuse AWS IAM database authentication. This eliminates the need to store\ndatabase passwords and provides enhanced security through AWS IAM\npolicies.\n\n```ts\nimport { Role } from \"cdk-rds-sql\"\n\nconst iamRole = new Role(this, \"IamRole\", {\n  provider: provider,\n  roleName: \"myiamrole\",\n  databaseName: \"mydb\",\n  enableIamAuth: true,\n})\n```\n\nWhen `enableIamAuth` is set to `true`:\n\n- **No secret is created** - the `role.secret` property will be `undefined`\n- The database role is created without a password\n- Your application must generate an IAM authentication token at runtime\n\nThis is the same behavior as DSQL clusters, which always use IAM authentication.\n\n**Requirements for IAM Authentication:**\n\n- SSL connections are required (enabled by default in this library)\n- Your application must have IAM permissions to connect to the database\n- The database user/role name must match the IAM identity\n\n**IAM Policy Example:**\n\nYour application will need an IAM policy like this to connect:\n\n```json\n{\n  \"Version\": \"2012-10-17\",\n  \"Statement\": [\n    {\n      \"Effect\": \"Allow\",\n      \"Action\": [\"rds-db:connect\"],\n      \"Resource\": [\n        \"arn:aws:rds-db:region:account-id:dbuser:cluster-resource-id/myiamrole\"\n      ]\n    }\n  ]\n}\n```\n\n**Generating an IAM Authentication Token:**\n\nAt runtime, your application needs to generate an authentication token\nusing the AWS SDK. Here's an example using the AWS SDK for JavaScript:\n\n```ts\nimport { Signer } from \"@aws-sdk/rds-signer\"\nimport { Client } from \"pg\"\n\nconst signer = new Signer({\n  hostname: \"your-cluster.cluster-xxxxx.us-east-1.rds.amazonaws.com\",\n  port: 5432,\n  username: \"myiamrole\",\n  region: \"us-east-1\",\n})\n\nconst token = await signer.getAuthToken()\n\nconst client = new Client({\n  host: \"your-cluster.cluster-xxxxx.us-east-1.rds.amazonaws.com\",\n  port: 5432,\n  user: \"myiamrole\",\n  database: \"mydb\",\n  password: token,\n  ssl: { rejectUnauthorized: true },\n})\n\nawait client.connect()\n```\n\nBoth PostgreSQL and MySQL databases support IAM authentication. For more details, see the [AWS RDS IAM Database Authentication documentation](https://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/UsingWithRDS.IAMDBAuth.html).\n\n### MySQL support\n\nIn MySQL users are created with '%' as value for the host. It is hard\nto do this better:\n\n- Determine the CIDR blocks used by a VPC is not trivial.\n- For imported VPCs you cannot specify the IPv6 CIDR.\n- CIDRs might change without the system knowing, meaning applications\n  could lose access to the database at random times, such as when a\n  container is restarted.\n\n## Database\n\nCreate a database as follows:\n\n```ts\nimport { Database } from \"cdk-rds-sql\"\n\nconst database = new Database(this, \"Database\", {\n  provider: provider,\n  databaseName: \"mydb\",\n})\n```\n\nYou can provide an owner, which makes it easy to create database owned\nby a new user:\n\n```ts\nconst database = new Database(this, \"Database\", {\n  provider: provider,\n  databaseName: \"mydb\",\n  owner: role,\n})\n```\n\n## Schema\n\nCreate a schema in the default database as follows:\n\n```ts\nimport { Schema } from \"cdk-rds-sql\"\n\nnew Schema(this, \"Schema\", {\n  provider: provider,\n  schemaName: \"myschema\",\n})\n```\n\nOr in another database:\n\n```ts\nconst database = new Database(this, \"Database\", {\n  provider: provider,\n  databaseName: \"mydb\",\n})\n\nnew Schema(this, \"Schema\", {\n  provider: provider,\n  schemaName: \"myschema\",\n  databaseName: database.databaseName,\n})\n```\n\nOne may need a role permitted for using schema:\n\n```ts\nnew Schema(this, \"Schema\", {\n  provider: provider,\n  schemaName: \"myschema\",\n  databaseName: database.databaseName,\n  role: role,\n})\n```\n\n## Sql\n\nYou can insert arbitrary SQL into your database with the `Sql` construct:\n\n```ts\nimport { Sql } from \"cdk-rds-sql\"\n\nconst sql = new Sql(this, \"Sql\", {\n  provider: provider,\n  database: database,\n  statement: \"create table t (i int)\",\n})\n```\n\nCreate a table if it does not exist, and grant a role privileges:\n\n```ts\nconst sql = new Sql(this, \"Sql\", {\n  provider: provider,\n  database: database,\n  statement: `\ncreate table if not exists t (i int);\ngrant select on t to myrole;\n`,\n})\n```\n\nRollback sql on stack deletion:\n\n```ts\nconst sql = new Sql(this, \"Sql\", {\n  provider: provider,\n  database: database,\n  statement: `\ncreate table if not exists t (i int);\ngrant select on t to myrole;\n`,\n  rollback: `\nDO $$BEGIN\n  IF EXISTS (select from pg_database WHERE datname = 't') THEN\n    IF EXISTS (select from pg_catalog.pg_roles WHERE rolname = 'myrole') THEN\n      revoke select t from myrole;\n    END IF;\n    drop table t;\n  END IF;\nEND$$;\n`,\n})\n```\n\nNote that there is no synchronisation between various `Sql`\nconstructs, in particular the order in your code does not determine\nthe order in which your SQL is executed. This happens in parallel,\nunless you specify an explicit dependency via `sql.node.addDepency()`.\n\nThere are a lot of concerns when using `Sql`:\n\n- When you update your Sql, your previous Sql is not \"rolled back\",\n  the new Sql is simply executed again.\n- When you delete your `Sql` construct the rollback is executed if specified\n- When permission are granted via `Sql` they must be removed via\n  rollback to succesfully remove the role\n- Currently the `Sql` constructs has less than 5 minutes to execute\n  its work.\n- It is unknown how large your SQL can be.\n\n# Parameters\n\nSome lambda constructs, in particular [Bref](https://bref.sh/), do not\nsupport secrets out of the box. Use the `parameterPrefix` property to\ncreate SSM parameters in addition to a secret:\n\n```ts\nconst role = new Role(this, \"Role\", {\n  provider: provider,\n  roleName: \"myrole\",\n  databaseName: \"mydb\",\n  parameterPrefix: \"/my-app/\",\n})\n```\n\nThis will create `/my-app/username`, `/my-app/password` and such.\n\nTo access parameters you will need IAM permissions such as:\n\n```ts\ninitialPolicy: [\n  new iam.PolicyStatement({\n\tactions: [\"ssm:GetParameter\", \"ssm:GetParameters\"],\n\tresources: [\n\t  // Grant access to all parameters under the base path\n\t  `arn:aws:ssm:${this.region}:${this.account}:parameter/my-app/*`,\n\t],\n\teffect: iam.Effect.ALLOW,\n  })\n],\n```\n\nNote that your VPC will need an SSM Parameters interface endpoint to support this.\n\n# AWS DSQL Support\n\nThis construct also supports AWS DSQL. DSQL is a great number of\nlimitations around it's postgresql support, so there is some learing\ncurve. This library supports creating roles, schemas, and executing\nSQL on DSQL clusters. Creating databases is not supported by DSQL.\n\n## DSQL Provider Setup\n\nDSQL clusters don't require a VPC or secrets since they use IAM authentication:\n\n```ts\nimport * as dsql from \"aws-cdk-lib/aws-dsql\"\nimport { Provider } from \"cdk-rds-sql\"\n\nconst dsqlCluster = new dsql.CfnCluster(this, \"DsqlCluster\", {\n  deletionProtectionEnabled: false,\n})\n\nconst provider = new Provider(this, \"Provider\", {\n  cluster: dsqlCluster,\n  // No VPC or secret needed for DSQL\n})\n```\n\nYou can access DSQL from within a VPC, but cdk-rds-sql has not been\nspecifically tested with that scenario.\n\n## DSQL Roles\n\nDSQL always uses IAM authentication, so the `enableIamAuth` property is ignored:\n\n```ts\nimport { Role } from \"cdk-rds-sql\"\n\nconst role = new Role(this, \"Role\", {\n  provider: provider,\n  roleName: \"myrole\",\n})\n```\n\n## Using DSQL\n\nConnecting to DSQL requires a couple of things:\n\n1. Your IAM role needs the `dsql:DbConnect` IAM permission to connect to the cluster.\n2. You probably want a normal postgres role (not \"admin\") to use from your apps.\n3. Your IAM role needs to be connected to your postgres role: use the `IamGrant` construct for that:\n\n   ```ts\n   new IamGrant(this, \"LambdaIamGrant\", {\n     provider: provider,\n     roleName: role.roleName,\n     resourceArn: queryLambda.role!.roleArn,\n   })\n   ```\n\n4. From your code use the Dsql signer to create the password:\n\n   ```ts\n   import { DsqlSigner } from \"@aws-sdk/dsql-signer\"\n   import { Client } from \"pg\"\n\n   const region =\n     process.env[\"AWS_REGION\"] || process.env[\"AWS_DEFAULT_REGION\"] || \"us-west-2\"\n\n   // Generate DSQL auth token using AWS SDK\n   const signer = new DsqlSigner({\n     hostname: host,\n     region,\n   })\n\n   const authToken = await signer.getDbConnectAuthToken()\n\n   const client = new Client({\n     host,\n     port: 5432,\n     user: dbUser,\n     database: dbName,\n     password: authToken,\n     ssl: { rejectUnauthorized: false },\n   })\n\n   await client.connect()\n   ```\n\n# IPv6\n\nIf you use the provider in an IPv6 subnet you probably need these settings:\n\n```ts\nimport { Provider } from \"cdk-rds-sql\"\n\nconst provider = new Provider(this, \"Provider\", {\n  ...\n  functionProps: {\n\tipv6AllowedForDualStack: true,\n\tallowAllIpv6Outbound: true,\n  },\n}\n```\n\n# Disabling SSL\n\nThe default connection to RDS is ssl enabled. You can disable ssl by\nsetting the `ssl` option to `false`:\n\n```ts\nconst provider = new Provider(this, \"Provider\", {\n  vpc: vpc,\n  instance: instance,\n  secret: cluster.secret!,\n  ssl: false, // default is true\n})\n```\n\n# Working on this code\n\nThis code is managed by [projen](https://github.com/projen/projen).\n[mise](https://mise.jdx.dev/) installs Node 24, pnpm 12, and\n[hk](https://hk.jdx.dev/) for Git hooks. Biome handles formatting and linting.\n\nDevelopment and CI use Node 24. Testcontainers is a development-only dependency;\nits dependency tree requires newer APIs than the consumer minimum of Node 22.14.\n\nAfter cloning:\n\n```sh\nmise trust\nmise install\nmise run setup\n```\n\nWhen migrating an existing clone, replace the old pre-commit hooks with\n`mise exec -- hk install --mise --force`.\n\nRun all checks with `mise run check`, or format and lint with\n`pnpm exec projen biome`. Tests require Docker. TypeScript stays on the\n6.0 release line supported by jsii 6.0.\n\nThe pre-commit hook runs the full build, tests, and packaging, so Docker must\nbe running. When `VPC_ID` is set, changes under `src/` also run CDK synthesis\nagainst that VPC; this requires AWS credentials. Unset `VPC_ID` to skip synthesis.\n\nThe development tsconfig uses `node16` module resolution, including package\nexports. Jest tasks enable Node's VM module support for native dynamic imports\nsuch as the MySQL driver import.\n\n## Testing\n\nTest code via projen with:\n\n    pnpm exec projen test\n\nTests do not update snapshots automatically. To intentionally refresh them, run\n`pnpm exec projen test:update` and review the snapshot diff.\n\nYou can run the PostgreSQL sample stack with:\n\n    pnpm exec projen integ:deploy:postgresql:serverless\n\nIf you want to use an existing vpc:\n\n    pnpm exec cdk deploy --context vpc-id=vpc-0123456789 TestRdsSqlServerlessV2Stack\n\nYou can also test DSQL support with:\n\n    pnpm exec projen integ:deploy:dsql\n\n# To do\n\n- Update role: will not revoke connect to previous database if database name has changed.\n"
   },
   "repository": {
     "type": "git",
@@ -9756,6 +9756,10 @@
   },
   "schema": "jsii/0.10.0",
   "targets": {
+    "dotnet": {
+      "namespace": "BerenddeBoer.CdkRdsSql",
+      "packageId": "BerenddeBoer.CdkRdsSql"
+    },
     "js": {
       "npm": "cdk-rds-sql"
     },
@@ -11285,6 +11289,6 @@
       "symbolId": "src/sql:SqlProps"
     }
   },
-  "version": "9.0.1",
-  "fingerprint": "o7SlNmzy/IZe/BYSlW+9L1JuGFjTax1lAgBFk+yd8f8="
+  "version": "9.1.0",
+  "fingerprint": "OkYYbqoqEMmAS3UeELvqwNawjTxsCwW40LiuBwYsIRY="
 }

package/README.md

@@ -23,6 +23,15 @@ environments, and works in isolated subnets.
 
      npm i cdk-rds-sql
 
+For .NET (available from the first NuGet release):
+
+```sh
+dotnet add package BerenddeBoer.CdkRdsSql
+```
+
+The .NET namespace is `BerenddeBoer.CdkRdsSql`. Node.js is also required
+when using the .NET bindings.
+
 # Usage
 
 ## Provider

package/lib/database.js

@@ -14,7 +14,7 @@ class ImportedDatabase extends constructs_1.Construct {
     }
 }
 class Database extends aws_cdk_lib_1.CustomResource {
-    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.Database", version: "9.0.1" };
+    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.Database", version: "9.1.0" };
     /**
      * Return a Database based upon name only. Use for importing existing databases.
      */

package/lib/iam-grant.js

@@ -6,7 +6,7 @@ const aws_cdk_lib_1 = require("aws-cdk-lib");
 const enum_1 = require("./enum");
 const provider_1 = require("./provider");
 class IamGrant extends aws_cdk_lib_1.CustomResource {
-    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.IamGrant", version: "9.0.1" };
+    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.IamGrant", version: "9.1.0" };
     constructor(scope, id, props) {
         // IAM grants are only supported on DSQL
         if (props.provider.engine !== provider_1.DatabaseEngine.DSQL) {

package/lib/provider.js

@@ -60,7 +60,7 @@ var DatabaseEngine;
     DatabaseEngine["DSQL"] = "dsql";
 })(DatabaseEngine || (exports.DatabaseEngine = DatabaseEngine = {}));
 class Provider extends constructs_1.Construct {
-    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.Provider", version: "9.0.1" };
+    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.Provider", version: "9.1.0" };
     /**
      * Import an existing provider Lambda function
      */

package/lib/role.js

@@ -83,7 +83,7 @@ class Parameters extends constructs_1.Construct {
     }
 }
 class Role extends constructs_1.Construct {
-    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.Role", version: "9.0.1" };
+    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.Role", version: "9.1.0" };
     /**
      * The role name.
      */

package/lib/schema.js

@@ -5,7 +5,7 @@ const JSII_RTTI_SYMBOL_1 = Symbol.for("jsii.rtti");
 const aws_cdk_lib_1 = require("aws-cdk-lib");
 const enum_1 = require("./enum");
 class Schema extends aws_cdk_lib_1.CustomResource {
-    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.Schema", version: "9.0.1" };
+    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.Schema", version: "9.1.0" };
     schemaName;
     constructor(scope, id, props) {
         super(scope, id, {

package/lib/sql.js

@@ -5,7 +5,7 @@ const JSII_RTTI_SYMBOL_1 = Symbol.for("jsii.rtti");
 const aws_cdk_lib_1 = require("aws-cdk-lib");
 const enum_1 = require("./enum");
 class Sql extends aws_cdk_lib_1.CustomResource {
-    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.Sql", version: "9.0.1" };
+    static [JSII_RTTI_SYMBOL_1] = { fqn: "cdk-rds-sql.Sql", version: "9.1.0" };
     constructor(scope, id, props) {
         super(scope, id, {
             serviceToken: props.provider.serviceToken,

package/package.json

@@ -75,7 +75,7 @@
   "publishConfig": {
     "access": "public"
   },
-  "version": "9.0.1",
+  "version": "9.1.0",
   "jest": {
     "coverageProvider": "v8",
     "testMatch": [
@@ -137,6 +137,10 @@
       "python": {
         "distName": "cdk-rds-sql",
         "module": "cdk_rds_sql"
+      },
+      "dotnet": {
+        "namespace": "BerenddeBoer.CdkRdsSql",
+        "packageId": "BerenddeBoer.CdkRdsSql"
       }
     },
     "tsconfig": "tsconfig.json",
@@ -162,6 +166,7 @@
     "integ:destroy:serverless": "projen integ:destroy:serverless",
     "package": "projen package",
     "package-all": "projen package-all",
+    "package:dotnet": "projen package:dotnet",
     "package:js": "projen package:js",
     "package:python": "projen package:python",
     "post-compile": "projen post-compile",