ChatDBA

ChatDBA is NineData's intelligent assistant for database Q&A, knowledge retrieval, SQL troubleshooting, and everyday AI-assisted inquiries.

What ChatDBA Can Do

• Knowledge Q&A: Ask questions in natural language and quickly get answers about databases, platform features, and troubleshooting topics.
• Knowledge-base-enhanced answers: ChatDBA can combine enterprise knowledge bases with general AI reasoning. For more information, see Knowledge Base Management.
• Data source access: Select a data source that has been added to NineData so ChatDBA can answer questions or help locate issues with data source context.
• SQL execution Skill: In a data source context where the current account has permission, ChatDBA can help initiate SQL execution and work with SQL execution OpenAPI for automation scenarios. For more information, see Skill.
• Text2SQL in conversations: Describe query requirements in natural language during a ChatDBA conversation, and ChatDBA can generate SQL based on the selected data source context. For more information, see Text2SQL / Chat2SQL.
• SQL troubleshooting: When AI diagnosis is triggered from the SQL Console, ChatDBA analyzes the error and suggests follow-up actions.
• Performance diagnosis: Provides a performance diagnosis Agent for MySQL-compatible, PostgreSQL-compatible, Oracle, and SQL Server databases, including database health checks, real-time session diagnosis, slow SQL governance, intelligent SQL optimization, lock diagnosis, and long-transaction diagnosis.
• Efficient retrieval: Retrieves relevant information from knowledge bases, historical conversations, and the current question to provide answers with fuller context.
• Cross-session retrieval memory: Retrieves relevant information from previous conversations to reduce repeated context input.
• Multi-turn conversation: A single chat keeps context, making it suitable for follow-up questions and step-by-step troubleshooting.

Supported Data Sources for Performance Diagnosis

The performance diagnosis capability currently supports MySQL, RDS MySQL, PolarDB MySQL, TDSQL-C MySQL, TaurusDB MySQL, Aurora MySQL, PostgreSQL-compatible databases, Oracle, and SQL Server.

Performance Diagnosis Scenarios

Scenario	Description
Instance inspection	Analyzes database instance status and identifies risks related to connections, workload, waits, and resource usage.
Session diagnosis	Analyzes real-time sessions to locate abnormal sessions, blocking chains, or sessions with high resource usage.
Slow query governance	Analyzes slow SQL or high-consumption SQL and provides risk assessment and governance suggestions.
AI SQL diagnosis and optimization	Diagnoses SQL statements and generates optimization suggestions. For more information, see AI SQL Diagnosis and Optimization.
Lock diagnosis	Analyzes lock waits, blocking relationships, and potential deadlock risks, and provides handling suggestions.
Long-transaction diagnosis	Analyzes the impact of long-running, large, or uncommitted transactions on database status.

Usage Limits

• Unregistered or logged-out users have a limit of 10 Q&A sessions per day.
• Unregistered or logged-out users cannot start new sessions.

Prerequisites

• You are logged in to the NineData console.
• If you want ChatDBA to answer with enterprise knowledge, configure and enable a knowledge base first.

Getting Started

1. Log in to the NineData Console.

2. Click the icon at the top of the page to open ChatDBA. The page opens in full-screen mode by default, and you can switch between full-screen and window mode as needed.

3. To let ChatDBA answer with the context of a specific data source, click Select Datasource and select the target data source.

4. Enter your question in the input box at the bottom, then click to send it. If the question requires SQL generation or execution, make sure the selected data source and account permissions are correct.

5. To start a new conversation, click New Chat on the left. Previous chats stay in the history list for later review.

   tip

   Context is preserved within the same session. For more coherent answers, keep related follow-up questions in a single conversation.

6. To delete a chat, hover over the target conversation, click , and then confirm with Delete Chat.

   caution

   Deleted conversations cannot be restored.

View Knowledge Sources

If an answer is based on knowledge base content, ChatDBA displays the View knowledge sources button below the answer. Click it to see which sources were referenced.