Database Monitoring for SQL Server on Google Cloud SQL

Atatus monitors SQL Server on Google Cloud SQL, collecting key data like query metrics, execution plans, and events. Because Google Cloud SQL does not allow direct host access, the Atatus Infra Agent runs on a separate host (typically a GCE VM in the same VPC) that can connect to your Cloud SQL instance.

Before You Begin

Component	Supported Versions / Requirements
SQL Server	2014, 2016, 2017, 2019, 2022
Atatus Infra Agent	3.4.0 or higher

1. Grant the Agent Access

Create a SQL login for the Atatus Infra Agent and assign the required permissions. On Cloud SQL the default admin user is sqlserver; connect as that role to run the following statements.

copy

USE [master];
CREATE LOGIN atatus WITH PASSWORD = '<YourSecurePassword>';
GO

GRANT VIEW SERVER STATE TO atatus AS CustomerDbRootRole;
GRANT VIEW ANY DEFINITION TO atatus AS CustomerDbRootRole;
ALTER SERVER ROLE CustomerDbRootRole DROP MEMBER atatus;
GO

Note: Google Cloud SQL does not permit granting CONNECT ANY DATABASE. You must create the atatus user explicitly in each target database.

For each application database, run:

copy

USE [<your_database_name>];
CREATE USER atatus FOR LOGIN atatus;
GO

For SQL Server Agent Jobs monitoring (SQL Server 2016+), grant the same access to msdb:

copy

USE [msdb];
CREATE USER atatus FOR LOGIN atatus;

GRANT SELECT ON dbo.log_shipping_monitor_primary TO atatus;
GRANT SELECT ON dbo.log_shipping_monitor_secondary TO atatus;

GRANT SELECT ON dbo.sysjobs TO atatus;
GRANT SELECT ON dbo.sysjobhistory TO atatus;
GRANT SELECT ON dbo.sysjobactivity TO atatus;
GO

2. Configure the Atatus Agent

Step 1: Configure the SQL Server Integration

Edit the conf.d/sqlserver.d/sqlserver.yml configuration file in your Atatus Agent installation directory:

copy

metrics:
  - hosts:
      - "<CLOUD_SQL_INSTANCE_IP>"
    port: 1433
    username: atatus
    password: <YourSecurePassword>
    dbm: true

Replace <CLOUD_SQL_INSTANCE_IP> with your Cloud SQL instance's private/public IP, or 127.0.0.1 if connecting through the Cloud SQL Auth Proxy.

Step 2: Restart the Atatus Infra Agent

copy

sudo service atatus-infra-agent restart

Enable Schema Collection for SQL Server 2017+ (Optional)

Atatus Agent can collect schema information from SQL Server 2017 and above.

Requirements

Agent v3.4.0 or higher
collect_database_info: true
collect_settings: true
Use auto_discovery: true to auto-detect all databases

Configuration Example

copy

metrics:
  - hosts:
      - "<CLOUD_SQL_INSTANCE_IP>"
    port: 1433
    username: <DB_USERNAME>
    password: <DB_PASSWORD>
    dbm: true
    dbm_mssqlserver_options:
      collect_settings:
        enabled: true
      collect_database_info:
        enabled: true
        auto_discovery: true
        # include:
        #   - db_name1
        #   - db_name2
        # exclude:
        #   - db_name1
        #   - db_name2

Collect Metrics Using Custom Queries (Optional)

You can collect custom metrics from SQL Server using the custom_queries option in your configuration.

Configuration Example

copy

metrics:
  - hosts:
      - "<CLOUD_SQL_INSTANCE_IP>"
    port: 1433
    username: <DB_USERNAME>
    password: <DB_PASSWORD>
    dbm: true
    dbm_mssqlserver_options:
      additional_metrics_options:
        custom_queries:
          - query: SELECT age, salary, hours_worked, name FROM hr.employees;
            columns:
              - name: custom.employee_age
                type: gauge
              - name: custom.employee_salary
                type: gauge
              - name: custom.employee_hours
                type: count
              - name: name
                type: tag
            tags:
              - 'table:employees'
        max_custom_queries: 20