What data for this category looks like
Daily documents for this category include assembly process cards, BOMs, torque records, clearance adjustment records, commissioning reports, factory inspection certificates, and packing lists. Data sources include PLM/ERP, part inspection data, assembly tool collection, test benches, MES, and customer technical agreements. Data is updated once per device or per order. Fields include material codes, torque values (unit: newton meters), clearance values (unit: millimeters), coaxiality parameters, and more. Some documents have multi-level nested structures, requiring data to be associated according to assembly hierarchy.
What constraints these characteristics impose on deployment and upgrades
This category uses multi-variety order-based assembly. Data is updated per device or order and comes from scattered sources. Deployment must support multi-data source connection and order-level data isolation. The structure of multi-level BOMs and multiple associated documents requires the knowledge base to support retrieval and recall of nested documents. Professional fields must retain units and precision, so corresponding data parsing rules must be configured during deployment. Data update frequency is high, so data synchronization must not be interrupted during upgrades. For intranet deployment scenarios, HTTP proxy configuration with username and password must be supported to connect to external data sources or update services.
How to configure settings
These configurations are written for FastGPT v4.17.0.
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
MONGODB_URI | mongodb://{username}:{password}@{host}:27017/fastgpt?authSource=admin | Stores historical assembly documents, BOMs, and inspection records, supports order-level data isolation |
PG_URL | postgresql://{username}:{password}@{host}:5432/fastgpt | Stores system configurations, user sessions, and workflow metadata, ensures stable service operation |
DB_MAX_LINK | 20–30 | Adapts to concurrent access requirements for multiple associated documents per device, prevents database connection overflow |
REDIS_URL | redis://:{password}@{host}:6379/0 | Caches professional term thesaurus, search results, and workflow node data, improves response speed |
FILE_TOKEN_KEY | Randomly generated 32-character string with no special characters | Protects download permissions for sensitive documents such as assembly process cards and test reports, prevents unauthorized access |
STORAGE_S3_ENDPOINT | Configured according to the actual address of the intranet object storage | Stores attachment data such as torque records and clearance adjustment records, adapts to intranet deployment scenarios |
Three common configuration errors
- An error indicating missing configuration fields occurs when creating a new knowledge base. The cause is that authentication parameters for
MONGODB_URIorPG_URLare not filled correctly, leading to failed database connection verification. - A
Cannot read properties of undefinederror appears when opening a workflow. The cause is incorrectREDIS_URLconfiguration, where cached workflow node configuration fails to load normally, leading to abnormal page rendering. - Professional terms are not correctly identified, and search result matching rate is low. The cause is that the category-specific professional term thesaurus has not been imported, and search matching rules adapted to the assembly scenario have not been configured.
How to verify successful configuration
- Run the built-in database connectivity test script to verify the connection status of
MONGODB_URIandPG_URL, confirm there are no address, port, or authentication errors. - Access the domain configured in
FE_DOMAINafter starting the service, check whether the knowledge base management and workflow editing modules load normally, with no loading failure prompts. - Upload an assembly process card document, verify that the file upload process works normally, and that file content can be accessed normally through the generated download link.
- Trigger a manual data synchronization task, confirm that BOM data from PLM/ERP can be synchronized to the knowledge base according to order dimensions, with no synchronization failure logs.
The configuration values provided are common starting points and should be evaluated against individual deployment samples.