What data for this category includes
This category of data comes from design documents, standard instrument and environmental monitoring data, calibration software outputs, laboratory information systems, and customer range requirements. Each calibration updates data for the associated device. Included documents are assembly and commissioning sheets, calibration procedures, original calibration records, error curves, metrology certificates, and software version records. Core fields include device serial number, standard instrument number, calibration date, measurement error value (unit: Pa, ℃, MPa, and similar units), range parameters, and audit traceability identifiers.
Constraints on Database and Operations Maintenance
Each calibration updates data for the associated device. This data includes both structured records and unstructured curves and certificates. These factors create two key constraints: high data write frequency, and large variation in single-record data volume. Index core fields along the traceability chain to support rapid traceability. Professional data units and calibration identifiers require strict field validation rules to prevent invalid data from being stored. Batch calibration scenarios generate concurrent write requests. Adapt deployments to database connection count limits while ensuring atomicity of data writes, to prevent traceability chain breaks.
Configuration Settings
Use the following configurations, which align with the database operations and maintenance logic of FastGPT v4.17.0.
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
MONGODB_URI | Point to a replica set address, enable read-write separation | Store unstructured error curves and metrology certificates. Replica sets ensure high availability, and read-write separation distributes read pressure |
PG_URL | Configure a partition table-enabled instance, enable SSL encryption | Store structured calibration records and traceability chain data. Partition tables adapt to high-frequency writes sorted by calibration date, and SSL ensures secure data transmission |
DB_MAX_LINK | 100–150 connections | Adapt to concurrent write requests from batch calibration scenarios, avoid connection pool exhaustion |
REDIS_URL | Point to an independent cache instance, set key expiration time to 3600–7200 seconds | Cache frequently accessed calibration procedures and standard instrument parameters, reduce repeated database queries |
Three Common Configuration Errors
- Symptom: Database connection error, returns
Failed to connect to 192.168.xx.xx.:1433. Cause: Deployments omit configuringDB_MAX_LINKto adapt to batch calibration concurrent requests, leading to connection pool exhaustion. - Symptom: Single node becomes unresponsive or crashes at 50 concurrent requests. Cause: Deployments do not enable database read-write separation, leading unstructured data reads to occupy too many primary node resources.
- Symptom: Unit mismatch appears in the error value field of calibration records. Cause: Deployments omit configuring field validation rules at the database layer, leading to invalid data being stored.
How to Verify Successful Configuration
- Execute batch simulated calibration requests, monitor database connection pool usage, confirm connection count does not reach the configured limit.
- Import a calibration certificate and error curve data, verify successful data storage and passage of field validation.
- Access cached calibration procedure documents, confirm that subsequent requests retrieve data directly from the cache after the first access.
- Review database logs, confirm that read-write separation configuration is active, and read requests for unstructured data are assigned to secondary nodes.
The values provided are common starting points and should be measured against the reader’s own samples.