What data looks like for this category
This category’s data comes from design documents, standard instrument and environmental monitoring data, calibration software, laboratory information systems, and customer range requirements. Maintain daily documents including assembly and commissioning sheets, calibration procedures, original calibration records, error curves, metrology certificates, and software version records. Update data after each calibration for each device. Fields include tested device model, calibration parameters, error values, traceability numbers, environmental conditions, and more. Units are mostly flow units, pressure units, temperature units, and metrology-related standard units.
Constraints Imposed by These Characteristics on Deployment and Upgrade
This category’s data sources are scattered and updated frequently. During deployment, support multi-source data docking, including calibration software, laboratory information systems, and range requirement documents submitted by customers. Store unstructured daily documents such as error curves and metrology certificates, so configure storage capabilities that support multi-format files. Treat calibration records as core deliverables, so ensure data storage traceability and permission isolation during deployment. When each calibration generates new data for a device, divide knowledge base entries by device to avoid mixing data from different batches. During upgrades, maintain compatibility with the format of original calibration data to prevent historical documents from being unsearchable.
All configurations below are adapted for FastGPT v4.17.0.
How to Set Configurations
| Configuration Item | Recommended Value | Rationale for This Value |
|---|---|---|
DB_MAX_LINK | 10–15 connections | This category generates multiple records per calibration, so database connection count must be limited to avoid overload under high-frequency access |
STORAGE_S3_ENDPOINT | Fill in according to the local object storage address | This category needs to store unstructured files such as error curves and metrology certificates, so a compatible storage endpoint must be connected |
REDIS_URL | Local Redis service address, default port 6379 | Caching of high-frequency calibration data requires low-latency memory storage to ensure retrieval speed |
FILE_TOKEN_KEY | Randomly generated 32-bit string | Calibration certificates for this category are sensitive deliverables, so file access permissions must be verified via tokens |
MONGODB_URI | Fill in according to the local MongoDB cluster address | The data volume of this category grows with calibration frequency, and MongoDB is suitable for storage and retrieval of unstructured documents |
ROOT_KEY | Only used during initial deployment, disabled after deployment | The knowledge base for this category requires strict permission management, so the root key is only used for initial configuration to avoid permission leaks |
Three Common Mistakes
- A WARNING: current commit information was not captured by the error occurs when performing a local image build. The cause is that the Git repository was not initialized in the project root directory, causing the build script to fail to obtain version information.
- After local deployment, knowledge base retrieval results remain fixed with no changes. The cause is that
REDIS_URLwas not configured correctly, causing the cache to not expire, so each call returns the initial retrieval result. - Original records exported from calibration software cannot be retrieved by the knowledge base. The cause is that
STORAGE_S3_ENDPOINTwas not configured correctly, resulting in incorrect file storage paths, making calibration certificates and error curves unreadable.
How to Confirm Configuration is Complete
- Perform a database connection test to verify that the connection count limit set by the
DB_MAX_LINKconfiguration is effective, with no frequent connection timeout errors. - Upload a calibration certificate and error curve file to verify that the file can be read normally via the storage address configured by
STORAGE_S3_ENDPOINT. - Initiate a calibration record retrieval request to verify that the returned results include correct fields such as tested device model and error values, with no missing fields.
- Generate a temporary file access token to verify that only requests carrying the correct
FILE_TOKEN_KEYcan access sensitive calibration documents.
The values provided are common starting points and should be measured against the reader's own samples.