What Data for This Category Looks Like
Daily documents for this category include bill of materials (BOM), assembly work instructions, refrigerant charging records, electrical safety test reports, performance sampling inspection reports, and repair records. Data sources include ERP/MES systems, component batch ledgers, tightening and charging equipment, safety compliance testers, performance laboratory data, and after-sales fault records. Data updates follow individual product units, shift production batches, or sampling inspection schedules. Each data entry includes fields such as model identifier, process parameters, test results, and batch number. Some parameters have defined units of measurement.
Constraints Imposed by These Characteristics During Deployment and Upgrade
Data sources for this category are scattered and updates occur frequently. During deployment, multi-format document import and incremental synchronization configurations must be supported to prevent system lag from bulk data imports. Multi-model mixed-line assembly business scenarios require the knowledge base to split indexes by product line. During deployment, independent database connection pool parameters must be configured to prevent confusion of process data across different models. Storage of sensitive data such as safety compliance and refrigerant charging records must enable encryption mechanisms. During deployment, correct S3 storage endpoint and access key configurations must be applied. Frequently accessed documents such as assembly work instructions must use a caching service to speed up queries. During deployment, reasonable cache connection parameters must be set. Shift-updated production data must have scheduled synchronization tasks configured to ensure consistency between knowledge base data and on-site production.
How to Configure Settings
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
MONGODB_URI | mongodb://fastgpt:yourpassword@mongodb-host:27017/fastgpt?authSource=admin | Stores structured and unstructured documents such as process specifications and sampling inspection reports. Specifies the authentication source and database name to ensure data security |
STORAGE_S3_ENDPOINT | http://your-s3-host:9000 | Stores files such as refrigerant charging records and raw safety compliance test data. Matches the standard port configuration for S3 services |
REDIS_URL | redis://:redispassword@redis-host:6379/0 | Caches frequently accessed assembly work instructions and BOM data to reduce query load on the primary database |
DB_MAX_LINK | 20–30 | Handles process data queries for multiple home appliance models simultaneously, preventing service interruptions from exhausted connection pools |
FILE_TOKEN_KEY | Random 32-character string | Encrypts file access tokens to prevent unauthorized access to stored process documents |
Three Common Configuration Mistakes
- Issue: After deploying FastGPT v4.17.0, the error
Failed to parse URL from /model/getProvidersappears, and the frontend cannot load the model channel selection interface. Cause: TheFE_DOMAINparameter is not configured correctly, preventing the frontend from assembling a complete model interface access address. - Issue: Stored process documents in the S3 service cannot be retrieved. The interface prompts that the file does not exist or the connection timed out. Cause: Only port 9000 is opened on the firewall, but the correct internal network address for
STORAGE_S3_ENDPOINTis not configured, or permissions for the FastGPT container to access the S3 service are not assigned. - Issue: After connecting to a locally deployed MinerU PDF parsing service, uploaded PDF documents cannot generate knowledge base indexes. Cause: The local API address of MinerU is not specified in the FastGPT configuration, or the port of the MinerU service is not opened for network interoperability with the FastGPT container.
How to Verify Successful Configuration
- Run the curl command
curl http://your-fastgpt-domain.com/model/getProviders, check that the returned result includes a normal list of model channels, with no parsing error prompts. - Upload a PDF file of refrigerant charging records to the knowledge base, verify that the file can be stored normally and returns a correct content link during retrieval.
- View the database connection pool monitoring panel, confirm that the current active connection count does not exceed the value set for
DB_MAX_LINK. - Submit a query about the electrical safety test standard for a specified home appliance model, verify that the returned result includes document links that can be accessed normally, and the content matches the configured process documents.
The values provided are common starting points and should be evaluated against the reader's own samples.