What this category of data looks like
Ledger data for mine mining and beneficiation originates from mining scheduling systems, mineral processing DCS control systems, weighbridge weighing terminals, and on-site safety inspection equipment. The system updates data per production shift, and completes batch synchronization after each shift concludes. The system pushes key production node data in real time. Single ledger documents include fields such as work area number, extraction volume, equipment runtime, material conveyance volume, and inspection record number. Corresponding units are tons, hours, cubic meters, and times. Documents primarily use structured forms, with stable per-entry data volume. Single-batch data volume during batch summary fluctuates with production scale. Reference only; verify against the target FastGPT release and deployment documentation.
Constraints during deployment and upgrade
The structured ledger data and mixed update rhythm of mine mining and beneficiation create multiple deployment constraints. Large batch sync data requires sufficient database connection pools to avoid write blocking. The mixed scenario of real-time node data and post-shift batch updates requires properly configured cache expiration and database timeout parameters. Most mines deploy their systems in intranet environments, so teams must use intranet-specific storage and database connection configurations. Ledger fields may adjust alongside production processes, so upgrade processes must retain old field compatibility logic to prevent business disruption from forced changes.
How to set configurations
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
DB_MAX_LINK | 20–30 | Matches the large connection demand of mine mining and beneficiation batch ledger synchronization, prevents connection pool exhaustion |
MONGODB_URI | mongodb://intranet MongoDB address:port/mine ledger database | Stores unstructured inspection records and historical ledger backups, adapts to intranet deployment scenarios |
REDIS_URL | redis://intranet Redis address:port/0 | Caches real-time production node data, reduces database query pressure |
STORAGE_S3_ENDPOINT | intranet MinIO address:port | Stores PDF and Excel report files exported from ledgers, adapts to intranet environments without public network access |
ROOT_KEY | Custom key according to enterprise intranet security specifications | Ensures identity verification security for internal system interface calls, complies with mine intranet security requirements |
Three common mistakes
- A database connection error occurs after deployment, with the prompt
Access denied for user 'xxx'@'xxx'. The cause is that the FastGPT container network is not configured to communicate with the database network, or the database has not opened the corresponding intranet port. - The ledger generation interface responds slowly, while direct curl calls to the model interface work normally. The cause is that
REDIS_URLis not configured to cache real-time data, leading to repeated full ledger database queries for each request, orDB_MAX_LINKis set too low, causing connection wait timeouts. - Ledger reports cannot be generated after upgrade, with exported fields empty. The cause is that compatibility mapping for old ledger fields was not retained during upgrade, and the new configuration overwrote original field definitions, preventing the system from reading existing ledger data.
How to confirm configurations are correct
- Run
docker-compose logsto view database and FastGPT container logs, confirm there are no database connection timeout or rejection errors. - Submit a single ledger summary request, check if the generated report file contains field data for the corresponding work area after receiving a response.
- Use curl to call FastGPT's internal interface in the intranet environment, confirm that the returned ledger data fields match the source system.
- Trigger a batch ledger upgrade sync task, check if the storage service generates backup files for the corresponding batch.
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 2026-09-28.