What this category of data looks like
Data for this category is generated and maintained by data governance teams, security consultants, and legal compliance teams. Daily documents include data catalogs, classification and grading rules, important data identification tables, and review minutes. Data sources include data asset inventory results, database metadata, business system description documents, and general compliance regulatory documents. The update frequency is quarterly or triggered each time a business system is changed. Data is stored in separate databases by document type, with fields including data asset identifiers, sensitivity levels, compliance tags, and others. There is no unified fixed unit, and values adjust based on the document type.
What constraints do these characteristics bring to the database and operations link
The data sources for this category are scattered. They cover multiple file types including database metadata and business system descriptions. The database and operations link must support multi-source data access. Update frequencies vary. Both incremental synchronization and full refresh modes must be supported to avoid repeated processing of old data. Document structures change with business needs. Database table structures must be scalable. Sensitive field identification and rule review generate a large number of concurrent requests. Set proper connection limits to avoid resource exhaustion. Data compliance attributes require the operations link to have audit and backup capabilities. This prevents important data from being unmanaged or failing compliance rectification. Data for this category covers assets other than personal information. Store this data separately from personal information data to avoid cross-contamination.
How to set configurations
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
MONGODB_URI | mongodb://fastgpt:securepass@127.0.0.1:27017/fastgpt?authSource=admin | Matches the standard connection format for the official MongoDB deployment of FastGPT v4.17.0 |
PG_URL | postgresql://fastgpt:securepass@127.0.0.1:5432/fastgpt | Complies with standard PostgreSQL database connection specifications, supporting structured storage of classification and grading rules |
DB_MAX_LINK | 10–15 | Adapts to concurrent requirements for batch sensitive field identification, avoiding database connection exhaustion on single instances |
REDIS_URL | redis://:securepass@127.0.0.1:6379/0 | Used for caching classification and grading rules and session data, conforming to FastGPT cache configuration standards |
| Data synchronization trigger rules | Per system change or every 7 days | Matches the quarterly or on-demand update frequency requirements of business data, reducing invalid synchronization overhead |
Three common configuration mistakes
- Phenomenon: After upgrading from 4.9.13 to 4.10.1, classification and grading rules in the knowledge base do not display synchronously. Database logs show permission verification failures. Cause: Connection parameters for
PG_URLandMONGODB_URIwere not updated correctly. The database permission configuration from the old version cannot meet the deployment requirements of the new version. - Phenomenon: External service requests fail when running locally, and logs show Redis connection refused. Cause:
REDIS_URLwas not configured to point to an external cache service. Local Redis was not started despite relying on it, causing the classification and grading rule cache to fail to load normally. - Phenomenon: Batch sensitive field identification tasks time out, returning 504 status code. Cause: The value set for
DB_MAX_LINKis too low, which cannot support concurrent database requests, leading to connection pool exhaustion and task failure.
How to verify configurations are correct
- Run the built-in database connectivity detection script of FastGPT to verify whether the connection status of
MONGODB_URIandPG_URLis normal. - Upload a test classification and grading rule document to confirm that data can be written to the database normally and the corresponding content can be queried through the retrieval interface.
- Initiate 10 concurrent sensitive field identification tasks, and observe that the database connection count does not exceed the set threshold of
DB_MAX_LINK. - Check whether cache entries for classification and grading rules exist in the Redis cache to verify that the
REDIS_URLconfiguration is effective.
The values provided are common starting points and should be measured against the reader's own samples.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-10-19.