What Data for This Category Looks Like
Data sources for internal office and employee services in coal mining production include internal OA systems, equipment operation and maintenance management platforms, human resource management systems, safety supervision department submission reports, and underground operation logs. Update frequencies vary: safety supervision daily reports and attendance records are updated daily. Equipment operation and maintenance ledgers are updated quarterly. Employee shift schedules are updated weekly. Safety regulation documents are updated consistently over long periods. Document structures cover long text types (safety operation procedures, training courseware), structured table types (operation and maintenance ledgers, attendance reports), and single record types (operation logs, inspection records). Fields include equipment ID, inspection time, gas concentration, employee ID, job title, shift information, and more. Some data carries sensitive identifiers.
Constraints During Deployment and Upgrade
The coexistence of multi-source heterogeneous data requires supporting connection configurations for both structured databases and unstructured storage during deployment. Plan parameter configurations for MONGODB_URI and STORAGE_S3_ENDPOINT in advance. Large differences in data update frequencies mean incremental synchronization tasks must adapt to different update cycles. During upgrades, adjust the DB_MAX_LINK parameter to accommodate high-concurrency requests and avoid timeouts caused by connection pool exhaustion. Some data has specific unit and field identifiers. Retain metadata when embedding into the knowledge base. Configure FILE_TOKEN_KEY to encrypt and store sensitive documents. Internal coal mine networks may have access restrictions. Ensure service ports can connect to internal systems during deployment. During upgrades, verify network permissions for plugin pulling.
How to Set Configurations
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
DB_MAX_LINK | 20–30 connections | Adapt to concurrent retrieval requests for multi-source coal mine data, avoid timeouts caused by connection pool exhaustion |
MONGODB_URI | mongodb://[username]:[password]@[host IP]:27017/fastgpt?authSource=admin | Store structured internal coal mine data such as operation and maintenance ledgers and attendance reports |
STORAGE_S3_ENDPOINT | http://192.168.100.200:9000 | Connect to internal object storage services, store unstructured data such as underground monitoring videos and safety training courseware |
REDIS_URL | redis://:[custom password]@[host IP]:6379/0 | Cache frequently accessed employee shift schedules and safety reminder data to reduce database load |
FILE_TOKEN_KEY | Randomly generated 32-bit string | Encrypt and store employee personal documents and confidential inspection records to ensure data security |
Three Common Mistakes
- Phenomenon: After upgrading to FastGPT v4.17.0, knowledge base retrieval time increases significantly with the same knowledge base and embedding model. Cause: The
DB_MAX_LINKparameter is not adjusted. Insufficient database connection pool size causes retrieval requests to queue, extending retrieval latency. - Phenomenon: Plugin pulling fails during Docker deployment, and the service cannot start normally. Cause:
STORAGE_S3_ENDPOINTis not configured or the parameter is filled incorrectly. Resource packages required by storage plugins cannot be pulled, causing service initialization failure. - Phenomenon: When deploying locally, the database connection node port cannot be entered. Cause: Connection parameters for
PG_URLorMONGODB_URIare not formatted correctly. System parameter validation blocks invalid port input formats, preventing the input box from activating.
How to Confirm Proper Configuration
- Run the built-in database connection verification tool to verify connectivity for
MONGODB_URIorPG_URL. Confirm no connection timeout or permission error messages appear. - Upload a single copy of coal mine safety regulation documents. Verify that the
FILE_TOKEN_KEYencryption upload and metadata extraction functions work normally, with no upload failure prompts. - Check service runtime logs. Confirm that the
REDIS_URLcache connection is normal, with no connection disconnection or authentication failure error messages. - Initiate a retrieval request that includes equipment ID and inspection time. Confirm that returned results contain field identifiers unique to coal mine data, with no missing fields.
The values given 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 2026-09-28.