What this category of data looks like
Data for this category originates from hot metal and scrap steel data, L2 process control systems, laboratories, MES, energy medium systems, and equipment status. Daily documents include heat plans, batching lists, converter smelting records, composition reports, continuous casting process records, and slab judgment sheets. Data update frequencies are per heat, real-time, and per shift. Document fields cover raw material ratios, process temperatures, composition test values, equipment operating parameters, and other metrics. Units include tons, degrees Celsius, meters per minute, and others. Each document has a large number of field dimensions that are tightly bound to the production process.
What constraints do these characteristics impose during deployment and upgrade?
Data for this category comes from multiple dispersed sources with varied update rhythms. During deployment, adapt access logic for multi-source heterogeneous data. This prevents synchronization delays that disrupt production scheduling. Document structures are fixed but include many field dimensions. Configure parsing rules that match existing process document formats in advance. This ensures accurate extraction of core fields. Production processes run continuously, and heats are tightly coupled. Use a rolling update strategy during upgrades. This avoids full downtime that causes continuous casting interruptions or heat scrap. High-frequency real-time data processing requires reserving sufficient resource bandwidth during deployment. This supports real-time data flow and rapid response.
How to configure the settings
This configuration applies to FastGPT v4.17.0.
| Configuration Item | Recommended Setting | Rationale for This Setting |
|---|---|---|
MONGODB_URI | mongodb://username:password@address:port/fastgpt?authSource=admin | Process records and exception report data for this category has a large volume. Configure a dedicated MongoDB instance to ensure data storage and retrieval performance |
REDIS_URL | redis://:password@address:port/0, with memory configuration no less than 16 GB | Cache real-time heat data and scheduling commands. A high memory configuration supports fast read/write of high-frequency real-time data |
DB_MAX_LINK | 100–150 connections | Multi-source data synchronization and document parsing generate many concurrent requests. Adjust the maximum database connection count to match concurrent load |
STORAGE_S3_ENDPOINT | Fill in according to the enterprise's internal object storage address | Process documents and test reports for this category require unified storage. Adapt to the enterprise's existing object storage service |
FE_DOMAIN | Enterprise internal access domain name | Configure the internal access address to ensure personnel from production departments can normally access the deployed service |
ROOT_KEY | Custom complex key string of 32 or more characters | Set a high-strength key to secure access to internal service interfaces |
Three common misconfigurations
- Symptom: After deploying with docker-compose, the
aiproxycontainer reports an error shortly after startup. Logs show curl request execution failed. Cause: TheREDIS_URLorMONGODB_URIparameters are not configured correctly. This prevents the service from connecting to dependent components. - Symptom: Calling a custom environment variable in a workflow returns an empty value. Cause: The environment variable pass-through switch is not enabled in the deployment configuration, or the variable name does not match the calling format used in the workflow.
- Symptom: After executing an index increment by 1 operation in a loop, the output result is null. Cause: The variable initialization logic is not configured correctly in the workflow node. This causes the index variable to not update as expected.
How to confirm the configuration is complete
- Run the docker-compose ps command. Confirm all service containers are in a running state, with no abnormal exit records.
- Upload a converter smelting record document. Verify that the system correctly extracts core fields from the batching list and composition report.
- Add an environment variable call node to a workflow. Enter a preset test variable. Confirm that the configured value is returned normally.
- Initiate a single heat data synchronization test. Verify that process records complete synchronization within the preset time and are stored in the specified database.
The values provided here are common starting points. They should be evaluated against the reader's own samples.