This document was written for FastGPT v4.17.0.
What Data Looks Like for This Product Category
Data for this product category is generated for individual transformers and high-voltage equipment. Data sources include design BOMs, silicon steel sheet and insulating material procurement data, vacuum and drying equipment operation logs, oil test reports, partial discharge and withstand voltage test reports from test stations, and product nameplate parameter tables.
Data is updated on a per-product basis, with a complete record set for each product, and batch data generated synchronously for each raw material batch. Single-product data includes fields such as material batch numbers, assembly process parameters, test detection values, and nameplate identifiers. Some detection parameters use dedicated units.
Constraints Imposed on Database and Operations Workflows
Full end-to-end traceability requirements for individual products require the database to use unique product identifiers for indexing, to avoid lag during cross-record queries.
Multi-source associated data storage requires the database to support multi-table join queries, to accommodate structured and semi-structured data in different formats.
High-frequency single-product or batch data writes require reasonable configuration of database connection limits, to prevent concurrent write conflicts.
Storage of high-precision test parameters requires matching dedicated numeric field types, to ensure data precision is not lost.
Complex data relationships and severe potential consequences of errors require regular full and incremental backups, to ensure data recoverability.
How to Set Configurations
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
MONGODB_URI | mongodb://[username]:[password]@[database host]:27017/transformer_trace?authSource=admin | Adapts to semi-structured test report and assembly record storage, uses a dedicated database name per business scenario for data isolation |
PG_URL | postgresql://[username]:[password]@[database host]:5432/material_bom?sslmode=require | Stores structured BOM, material batch, and nameplate data; SSL mode ensures secure transmission of sensitive parameters |
DB_MAX_LINK | 20–30 | Meets concurrent demands of high-frequency single-product writes and multi-table join queries, prevents connection exhaustion |
REDIS_URL | redis://[password]@[cache host]:6379/0 | Caches frequently queried basic product information and test data, reduces database access load |
| Database connection whitelist | Configured using actual deployed device IP ranges | Restricts access to the database from unauthorized devices, protects core business data security |
Three Common Configuration Errors
- Symptom: Database connection failure, with errors indicating unresolvable hostnames or connection timeouts. Cause: The IP address of the Docker container host running the database or the local loopback address was not added to the connection configuration, or Docker Compose port mapping was not configured correctly.
- Symptom: Syntax errors returned when executing multi-line query statements for MySQL databases, with prompts for unclosed statements or insufficient permissions. Cause: The database configuration allowing batch statement execution was not enabled, or batch query support was not enabled in connection parameters.
- Symptom: After restoring a FastGPT project backup file, the backend knowledge base and agents do not display existing data. Cause: Only the project configuration file was restored, associated full database backups were not synchronized, or the correct database recovery node was not selected during the restore process.
How to Verify Proper Configuration
- Run a database connection test, confirm that the configured
MONGODB_URIandPG_URLcan establish normal connections with no authentication or network errors. - Write a simulated single-product assembly record and test data, then verify that the data can be properly associated and retrieved, with field precision and format meeting requirements.
- Trigger a database backup and restore operation, confirm that backup files can be generated normally, and data remains complete and intact after restoration.
- Check the database connection whitelist configuration, confirm that only authorized devices can access the database port, with no abnormally open ports.
The values provided here are common starting points, and should be evaluated against the reader's own samples.