Reference only; verify against the target FastGPT release and deployment documentation.
What the Data for This Category Looks Like
Coal chemical ledger data comes from production distributed control systems (DCS), laboratory physical and chemical analysis reports, equipment inspection records, and team shift change logs. Update frequencies include real-time (parameters such as process pressure, temperature), daily (shift production ledgers), and weekly (monthly summary drafts). Each record includes fields such as device number, sampling time, parameter name, measured value, legal unit, exception flag, and operator number. Some ledgers include PDF format laboratory report attachments.
Constraints Imposed by These Characteristics on Deployment and Upgrade
Data sources for coal chemical ledgers are scattered and have diverse formats. Multi-data source connection and adaptation must be configured during deployment. High-frequency updates of real-time parameters require adjustments to database connection pool parameters to avoid connection overflow. Large-capacity unstructured report attachments included in ledgers require adaptation to large-capacity object storage. Old storage paths must be compatible during upgrades. Special parameter units such as MPa, ℃, and m³/h require configuration of field mapping rules during knowledge base import to avoid unit confusion. Historical ledger data migration must retain the original field structure. Data integrity must be verified after upgrade.
Configuration Settings
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
PG_URL | postgresql://{account}:{password}@{host}:5432/coal_chem_ledger_db | Structured data for coal chemical ledgers is mostly stored in PostgreSQL. Specify the connection string for the corresponding business database |
MONGODB_URI | mongodb://{account}:{password}@{host}:27017/coal_chem_attachment_library | Stores unstructured data such as laboratory reports and inspection photos attached to ledgers, adapting to MongoDB's document storage characteristics |
STORAGE_S3_ENDPOINT | http://{internal object storage address}:9000 | Attachment files for coal chemical ledgers have large capacity. Connect to internal object storage to implement persistent storage |
DB_MAX_LINK | 20–30 connections | Concurrent access volume for coal chemical ledger databases is moderate. Set a reasonable connection count to avoid excessive system resource occupation |
REDIS_URL | redis://:{password}@{host}:6379/0 | Caches query results for real-time process parameters to reduce access pressure on structured databases |
FE_DOMAIN | http://{deployment node domain}:8080 | Unified front-end access entry to ensure normal loading of localized interface configurations |
Three Common Misconfigurations
- Phenomenon: Login interface and knowledge base interface display in English. Cause: Localized configuration was not retained during image migration, or
FE_DOMAINconfiguration error causes front-end language pack loading failure. - Phenomenon: Locally deployed large models cannot be called. Cause: The interface address of the local large model was not correctly specified in system configuration, or network access permissions for the corresponding port were not opened.
- Phenomenon: Ledger attachments referenced by the knowledge base cannot be downloaded. Cause:
STORAGE_S3_ENDPOINTconfiguration is incorrect, or object storage access keys were not correctly bound, resulting in failure to read attachment files.
How to Verify Successful Configuration
- Access the address specified by
FE_DOMAINto confirm the interface language is Chinese. - Run database connection tests to verify connectivity for
PG_URL,MONGODB_URI, andREDIS_URL. - Upload a test attachment for a coal chemical ledger to confirm normal download and full content viewing.
- After adjusting the
DB_MAX_LINKparameter, check the system monitoring panel to confirm database connection counts do not exceed the set limit.
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.