What data for this category consists of
Data for this category comes from resin batches and material property sheets, injection molding machine parameters, mold temperature controllers, MES systems, and dimensional and appearance inspection equipment. Daily documents include injection molding process cards, raw material drying records, mold histories, first-piece inspection forms, patrol inspection forms, and material change and machine cleaning records. Data fields include packing pressure, holding time, mold temperature, raw material moisture content, and additional parameters. Units include seconds, degrees Celsius, percentages, and other standard industrial units. Data updates occur per production batch, each mold or material change, and each shift patrol checkpoint. Each production batch includes multi-dimensional process parameters and inspection results.
What constraints these characteristics impose on deployment and upgrade
FastGPT v4.17.0 deployment must strictly follow official configuration specifications. Data sources cover multiple industrial devices and business systems, so deployment must support multi-protocol data source connection configuration. High-frequency data update requirements demand that database and cache connection pools maintain sufficient capacity to avoid connection exhaustion and data loss. Support for multiple document types and differentiated field requirements necessitates flexible metadata adaptation rule configuration. During the upgrade process, original data source configurations and administrator permissions must not be altered. The upgrade must also be compatible with new inspection data parsing logic to avoid interrupting real-time audit processes in production links.
How to set configurations
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
MONGODB_URI | mongodb://admin:password@host1:port,host2:port/database?authSource=admin | Persistently stores structured data such as multi-batch injection molding process cards and patrol records. High-availability cluster configuration ensures data stability |
REDIS_URL | redis://:cachepass@host:port/0 | Caches real-time injection molding machine parameters and mold temperature data to reduce database query pressure, and adapts to high-frequency update business scenarios |
DB_MAX_LINK | 200–300 | Adapts to the large volume of inspection and process data generated per shift, maintains sufficient database connection pool capacity, and calibrates connection timeout based on actual measurements |
STORAGE_S3_ENDPOINT | Configured according to the enterprise's internal object storage address | Stores large files such as mold histories and raw material property sheets, and adapts to multi-terminal file access requirements |
FE_DOMAIN | https://fastgpt.yourdomain.com | Configures the access domain name for the login-free window, adapts to cross-device usage scenarios |
ROOT_KEY | Custom complex random string | Configures system administrator permissions, prevents unauthorized access, and ensures system configuration security |
Three common configuration mistakes
- Phenomenon: After version upgrade, the configured specified large model switches to another model when called after saving, and cannot be used. Cause: The upgrade process does not retain the original application's model configuration, or the
FE_DOMAINconfiguration does not match the original deployment domain name, causing the front-end loaded configuration to be reset. - Phenomenon: After Docker deployment, an error is prompted when entering the workflow, and the system cannot be used normally. Cause:
MONGODB_URIorREDIS_URLis not configured correctly, causing database or cache connection failure, and the necessary configuration directory is not mounted when the container starts. - Phenomenon: After local deployment, another computer cannot use the service via the login-free link. Cause: The
FE_DOMAINconfiguration does not include a valid domain name for cross-device access, or the port for local deployment is not open to the public, causing external devices to fail to resolve the access address normally.
How to verify successful configuration
- Run a database connection test to confirm that the
MONGODB_URIandREDIS_URLconfigurations can establish normal connections and read stored injection molding process and inspection data. - Create a simple application, configure the specified large model, save the configuration, and initiate a test conversation to confirm that the called model matches the configured one.
- Generate a login-free access link, use a separate device to access the link, and confirm that the application interface loads normally and interactions can be completed.
- Trigger the system upgrade process to confirm that original data source configurations and administrator permissions remain unaltered after the upgrade.
The values provided are common starting points and should be measured against the reader's own samples.