What data this category includes
Data for this category centers on the full approval process of engineering construction projects. Daily documents include construction permit applications, construction drawing review qualification certificates, quality and safety supervision procedures, and completion acceptance record forms. Data sources include engineering construction project approval management systems, construction drawing review systems, enterprise qualification databases, and credit platforms. Data is updated daily. Data fields include project unique identifiers, participating unit qualification information, status of each approval link node, application material lists, and similar items. Most formats are text and number types, with no dedicated unit fields.
Constraints for deployment and upgrade
Data sources for this category are scattered and updated daily, and daily document formats vary. This creates the following constraints for deployment and upgrade:
- Configure a stable database connection to support persistent storage of multi-source approval data, maintain stable daily synchronization links, and avoid data synchronization interruptions.
- Configure file storage and access permissions to adapt to storage and retrieval of various application materials, given the diverse formats of daily documents.
- Ensure compatibility of external system data pulling during upgrades to prevent approval node data synchronization interruptions caused by version updates.
- Configure sufficient connection pool resources to support operation of batch parallel approval process guidance tasks.
How to set configurations
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
PG_URL | postgresql://admin:securepass@db-host:5432/fastgpt?schema=public | Adapts to persistent storage requirements for multi-source approval data of this category, supports compatibility with Docker-deployed database containers |
MONGODB_URI | mongodb://admin:securepass@mongo-host:27017/fastgpt | Stores dynamic status of approval nodes and parallel running task data |
DB_MAX_LINK | 8–12 | Supports connection pool management for large volumes of daily updated approval data, prevents connection exhaustion |
REDIS_URL | redis://:securepass@redis-host:6379/0?timeout=600 seconds | Caches frequently accessed approval node data to reduce database query pressure |
FILE_TOKEN_KEY | 32-bit randomly generated secure string | Secures access permissions for documents such as construction permits and completion acceptance records |
STORAGE_S3_ENDPOINT | http://s3.your-domain.com | Stores various approval documents and application materials |
Three common misconfigurations
- Phenomenon: After local Docker deployment of FastGPT v4.17.0, database connection errors occur after several hours. Recovery requires deleting the PG container and redeploying. Cause: No connection pool timeout setting configured for
PG_URL, or connections are not automatically released, leading to database connection exhaustion.
Reference only; verify against the target FastGPT release and deployment documentation.
- Phenomenon: Calls to overseas large models to generate approval process guidance content fail. Cause:
ROOT_KEYor the access permission for the corresponding model is not configured, and docking configuration for overseas large models is not adapted.
How to verify successful configuration
- Run a database connection test script to verify that the connection status of
PG_URLandMONGODB_URIis normal, and the connection pool connection count matches theDB_MAX_LINKconfiguration value. - Upload a construction permit application document, generate an access link via
FILE_TOKEN_KEYto confirm normal download, and confirm that documents stored viaSTORAGE_S3_ENDPOINTcan be read normally. - Start 10 parallel approval process guidance test tasks to verify that tasks can execute normally in parallel, with no connection exhaustion or timeout errors.
- Access the business guidance page configured at
https://fastgpt.your-domain.comto confirm that the page loads normally and approval node information is displayed correctly.
The values provided are common starting points and should be validated against actual deployment 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.