What the Data for This Category Looks Like
Data for this category comes from raw material batch ledgers, production MES systems, slurry and coating testing equipment, workshop dew point monitoring devices, formation and grading cabinets, and laboratory testing systems. Daily documents include batching records, coating and rolling parameter ledgers, winding or laminating process records, liquid injection operation logs, formation and grading data reports, and OCV and safety test reports. Data updates occur in real time per production batch and per individual cell. Fields include process numbers, process parameter values, environmental monitoring values, test results, and batch association identifiers. No additional statistical percentage data is included.
Constraints Imposed by These Characteristics on Deployment and Upgrade
This category has high data update frequency, and single-batch data volume grows with the number of produced cells. Adapt storage configurations for high-frequency writes during deployment to avoid connection blocking. Multi-source heterogeneous production data requires unified access formatting. Ensure upgrade steps are compatible with existing equipment protocol interfaces to prevent interruptions to real-time data collection links. Electrochemical production has strict requirements for data accuracy. Configure data validation rules during deployment to ensure deviations between detected values and process standards are identified promptly. Single records have many fields and close associations. Reserve sufficient context storage resources to prevent data truncation or loss of associated information during queries.
Configuration Settings
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
DB_MAX_LINK | 100–150 connections | Adapts to high-frequency real-time write production data, avoids request failures caused by exhausted connection pools |
MONGODB_URI | Points to a dedicated database instance storing process records and test reports | Data for this category must be archived by batch and cell dimensions. A dedicated instance can isolate business data from other business databases |
REDIS_URL | Configured as a local cache node, with cache duration set to 300 seconds | Used to temporarily store high-frequency read production data such as real-time dew point and rolling parameters, reducing database query pressure |
STORAGE_S3_ENDPOINT | Bound to the access address of the enterprise’s internal object storage service, with a single file maximum limit set to 1000 MB | Used to store large-capacity formation and grading data and test report files, meeting storage requirements for large-volume data |
FILE_TOKEN_KEY | Generate a 32-bit random string in accordance with enterprise internal key management specifications | Used to verify the legitimacy of uploaded process record files, preventing unauthorized files from accessing the system |
PG_URL | Points to a relational database storing process parameter baseline values | Used to compare deviations between real-time detected data and process standards, enabling rapid compliance verification |
Three Common Configuration Errors
- Phenomenon: Incomplete results are returned when querying inspection records after deployment, or the inference output length is insufficient to cover complete inspection report content. Cause: The
DB_MAX_LINKparameter is not configured correctly, with too low a connection pool upper limit causing some data requests to be rejected, or the context storage configuration is not adjusted after upgrade to adapt to large-capacity inspection data. - Phenomenon: A request failure prompt appears when uploading formation and grading data or test reports, and the issue cannot be resolved after restarting the server. Cause: The
STORAGE_S3_ENDPOINTconfiguration is incorrect, or the cache node corresponding toREDIS_URLmalfunctions, causing the object storage service to fail to write large-capacity test files normally. - Phenomenon: When configuring the MCP service, the parameter type is fixed as a string and cannot be modified. Cause: The custom parameter type switch is not enabled in accordance with the FastGPT v4.17.0 official deployment documentation, or the deployed version does not support this function.
How to Confirm Proper Configuration
- Log in to the database connection management page in the FastGPT backend, verify that the instances configured with
MONGODB_URIandPG_URLcan establish normal connections. - Upload a simulated liquid injection record file, check whether the file can be stored normally and displayed correctly in the document list.
- Initiate a batch inspection record query request, confirm that the number of returned results does not exceed the connection pool upper limit configured by
DB_MAX_LINK. - View the system real-time logs, confirm that there are no connection timeout or read/write failure error reports for the cache node corresponding to
REDIS_URL.
The values provided are common starting points and should be measured against the reader's own samples.