Connecting to SQL Server ​

Connect Kyomi to Microsoft SQL Server or Azure SQL Database for AI-powered analytics.

Connection Details ​

FieldDescriptionDefault
HostServer hostname or IPRequired
PortSQL Server port1433
EncryptUse TLS encryptiontrue
Trust Server CertificateTrust self-signed certsfalse
DatabaseDatabase to connect toRequired
Default SchemaDefault schema for queriesdbo

Setup Steps ​

Step 1: Configure Connection ​

  1. In the datasource modal, select SQL Server as the datasource type
  2. Enter your server Host (e.g., sqlserver.example.com)
  3. Enter the Port (default: 1433)
  4. Configure encryption options:
    • Encrypt: Enable TLS encryption (recommended)
    • Trust Server Certificate: Enable if using self-signed certificates
  5. Click Connect to test the connection

Step 2: Select Database and Schema ​

  1. Choose your Database from the dropdown
  2. Select a Default Schema (usually dbo)

Step 3: Configure Credentials ​

Enter your SQL Server Username and Password.

Authentication Types

Kyomi uses SQL Server Authentication. Windows Authentication is not currently supported.

Shared vs Personal Credentials

Shared Credentials: Admin configures once, all workspace users share the same database access.

Personal Credentials: Each user provides their own username/password for individual access control.

Step 4: Configure Catalog ​

Select which schemas Kyomi should index:

  • Tables and columns from these schemas will appear in the catalog
  • The AI will use this information to help write queries
  • Leave empty to index all accessible schemas

SSH Tunnel (Optional) ​

For SQL Server behind a firewall, Kyomi supports SSH tunnel connections.

Configuration ​

  1. Enable Use SSH Tunnel in the connection settings
  2. Enter SSH connection details:
    • SSH Host: Bastion server hostname
    • SSH Port: SSH port (default: 22)
    • SSH Username: Your SSH username
  3. The database connection will be tunneled through SSH

Required Permissions ​

Create a read-only user for Kyomi:

sql
-- Create login at server level
CREATE LOGIN kyomi_user WITH PASSWORD = 'SecurePassword123!';

-- Create user in target database
USE your_database;
CREATE USER kyomi_user FOR LOGIN kyomi_user;

-- Grant read access to schema
GRANT SELECT ON SCHEMA::dbo TO kyomi_user;
GRANT SELECT ON SCHEMA::analytics TO kyomi_user;

-- For catalog indexing
GRANT VIEW DEFINITION ON SCHEMA::dbo TO kyomi_user;

For broader read access:

sql
-- Add to db_datareader role (all tables)
ALTER ROLE db_datareader ADD MEMBER kyomi_user;

Troubleshooting ​

"Cannot connect to server" error ​

  • Verify the host and port are correct
  • Check that SQL Server is accepting remote connections
  • Ensure TCP/IP protocol is enabled in SQL Server Configuration Manager
  • Verify firewall allows connections on port 1433

"Login failed for user" error ​

  • Verify username and password are correct
  • Check SQL Server Authentication is enabled (mixed mode)
  • Verify the login exists at server level

"Certificate chain was issued by an authority that is not trusted" ​

  • Enable Trust Server Certificate for self-signed certificates
  • Or install the server's certificate in your trust store

"Cannot open database" error ​

  • Verify the database exists
  • Check the user has access to the database

Azure SQL Database ​

For Azure SQL Database:

  1. Get the server name from Azure portal (e.g., myserver.database.windows.net)
  2. Use port 1433
  3. Enable Encrypt (required by Azure)
  4. May need to Trust Server Certificate depending on your setup
  5. Configure Azure firewall to allow Kyomi's IP addresses

Connection String Format ​

Azure SQL uses this format:

Server=myserver.database.windows.net,1433
Database=mydb
User Id=myuser
Password=mypassword
Encrypt=True

Named Instances ​

For SQL Server named instances:

  • Default instance: Use just the hostname (e.g., sqlserver.example.com)
  • Named instance: Use hostname\instancename format
  • Port may differ from 1433 - check SQL Server Configuration Manager

SQL Server Versions ​

Kyomi supports:

  • SQL Server 2012 and later (for OFFSET-FETCH pagination)
  • Azure SQL Database
  • Azure SQL Managed Instance

Additional Resources ​

← Back to Datasources | Back to Docs