MSSQL database secrets engine

MSSQL is one of the supported plugins for the database secrets engine. This plugin generates database credentials dynamically based on configured roles for the MSSQL database.

See the database secrets engine docs for more information about setting up the database secrets engine.

The following privileges are needed by the plugin for minimum functionality. Additional privileges may be needed depending on the SQL configured on the database roles.

-- Create LoginCREATE LOGIN vault_login WITH PASSWORD = '<password>'; -- Create UserCREATE user vault_user for login vault_login; -- Grant PermissionsGRANT ALTER ANY LOGIN TO "vault_user";GRANT ALTER ANY USER TO "vault_user";GRANT ALTER ANY CONNECTION TO "vault_login";GRANT CONTROL ON SCHEMA::<schema_name> TO "vault_user";EXEC sp_addrolemember "db_accessadmin", "vault_user";

Capabilities

Setup

1. Enable the database secrets engine if it is not already enabled:

   $ vault secrets enable databaseSuccess! Enabled the database secrets engine at: database/

   By default, the secrets engine will enable at the name of the engine. To enable the secrets engine at a different path, use the -path argument.

2. Configure Vault with the proper plugin and connection information:

   $ vault write database/config/my-mssql-database \    plugin_name=mssql-database-plugin \    connection_url='sqlserver://{{username}}:{{password}}@localhost:1433' \    allowed_roles="my-role" \    username="vaultuser" \    password="yourStrong(!)Password"

   Note: The example above demonstrates a connection with SQL server user named vaultuser, although the user vaultuser might be Windows Authentication user part of Active Directory domain, for example: DOMAIN\vaultuser.

   In this case, we've configured Vault with the user "vaultuser" and password "yourStrong(!)Password", connecting to an instance at "localhost" on port 1433. It is not necessary that Vault has the vaultuser login, but the user must have privileges to create logins and manage processes. The fixed server roles securityadmin and processadmin are examples of built-in roles that grant these permissions. The user also must have privileges to create database users and grant permissions in the databases that Vault manages. The fixed database roles db_accessadmin and db_securityadmin are examples or built-in roles that grant these permissions.

3. Configure a role that maps a name in Vault to an SQL statement to execute to create the database credential:

   $ vault write database/roles/my-role \    db_name=my-mssql-database \    creation_statements="CREATE LOGIN [{{name}}] WITH PASSWORD = '{{password}}';\        CREATE USER [{{name}}] FOR LOGIN [{{name}}];\        GRANT SELECT ON SCHEMA::dbo TO [{{name}}];" \    default_ttl="1h" \    max_ttl="24h"Success! Data written to: database/roles/my-role

Usage

After the secrets engine is configured and a user/machine has a Vault token with the proper permission, it can generate credentials.

1. Generate a new credential by reading from the /creds endpoint with the name of the role:

   $ vault read database/creds/my-roleKey                Value---                -----lease_id           database/creds/my-role/2f6a614c-4aa2-7b19-24b9-ad944a8d4de6lease_duration     1hlease_renewable    truepassword           wJKpk9kg-T1Ma7qQfS8yusername           v-vaultuser-my-role-r7kCtKGGr3eYQP1OGR6G-1602542258

Example for Azure SQL database

Here is a complete example using Azure SQL Database. Note that databases in Azure SQL Database are contained databases and that we do not create a login for the user; instead, we associate the password directly with the user itself. Also note that you will need a separate connection and role for each Azure SQL database for which you want to generate dynamic credentials. You can use a single database backend mount for all these databases or use a separate mount for each of them. In this example, we use a custom path for the database backend.

First, we mount a database backend at the azuresql path with vault secrets enable -path=azuresql database. Then we configure a connection called "testvault" to connect to a database called "test-vault", using "azuresql" at the beginning of our path:

$ vault write azuresql/config/testvault \    plugin_name=mssql-database-plugin \    connection_url='server=hashisqlserver.database.windows.net;port=1433;user id=admin;password=pAssw0rd;database=test-vault;app name=vault;' \    allowed_roles="test"

Now we add a role called "test" for use with the "testvault" connection:

$ vault write azuresql/roles/test \    db_name=testvault \    creation_statements="CREATE USER [{{name}}] WITH PASSWORD = '{{password}}';" \    revocation_statements="DROP USER IF EXISTS [{{name}}]" \    default_ttl="1h" \    max_ttl="24h"

We can now use this role to dynamically generate credentials for the Azure SQL database, test-vault:

$ vault read azuresql/creds/testKey             Value---             -----lease_id        azuresql/creds/test/2e5b1e0b-a081-c7e1-5622-39f58e79a719lease_duration  1h0m0slease_renewable truepassword        cZ-BJy-SqO5tKwazAuUPusername        v-token-test-tr2t4x9pxvq1z8878s9s-1513446795

When we no longer need the backend, we can unmount it with vault unmount azuresql. Now, you can use the MSSQL Database Plugin with your Azure SQL databases.

Amazon RDS

The MSSQL plugin supports databases running on Amazon RDS, but there are differences that need to be accommodated. A key limitation is that Amazon RDS doesn't support the "sysadmin" role, which is used by default during Vault's revocation process for MSSQL. The workaround is to add custom revocation statements to roles, for example:

vault write database/roles/my-role revocation_statements="\   USE my_database \   IF EXISTS \     (SELECT name \      FROM sys.database_principals \      WHERE name = N'{{name}}') \   BEGIN \     DROP USER [{{name}}] \   END \    IF EXISTS \     (SELECT name \      FROM master.sys.server_principals \      WHERE name = N'{{name}}') \   BEGIN \     DROP LOGIN [{{name}}] \   END"

API

The full list of configurable options can be seen in the MSSQL database plugin API page.

For more information on the database secrets engine's HTTP API please see the Database secrets engine API page.