What data looks like for this category
Source data includes Gerber files, drill files, and BOMs provided by clients, plus factory in-house CAM output files, equipment operating parameters, bath test records, AOI inspection reports, electrical test data, and laboratory slice reports. Data updates occur per production batch and shift frequency; some equipment parameters are collected continuously. Daily documents cover engineering production instructions, Gerber parsing records, drilling and lamination parameters, chemical analysis sheets, AOI reports, electrical test reports, and slice reports. These documents include dedicated fields and corresponding units such as drilling aperture, lamination temperature, and copper thickness.
Constraints during deployment and upgrade
Multi-source heterogeneous business data formats require pre-processing flows that support multiple document types during deployment. This ensures unified handling of files in different formats such as Gerber and inspection reports. Frequently collected continuous equipment parameters need stable caching and storage capabilities. This places high availability requirements on REDIS_URL and MONGODB_URI. Bid document preparation requires fast matching of historical production data. Database connection latency and stability directly affect overall efficiency. Minor differences exist in document structures across different production batches. Upgrades must support importing old-format data and field mapping, to avoid breaking existing bid template configuration logic.
How to set configurations
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
REDIS_URL | redis://:your_secure_pass@192.168.1.100:6379/0 | Caches frequently called production parameters and bid templates, reduces database access pressure, and adapts to high-frequency data retrieval requirements |
MONGODB_URI | mongodb://admin:your_secure_pass@10.0.0.5:27017/fastgpt?authSource=admin | Stores unstructured business data such as Gerber parsing records and inspection reports, adapts to multi-format document storage requirements |
FILE_TOKEN_KEY | 32-bit randomly generated hexadecimal string | Encrypts bid-related uploaded files such as Gerber and BOM, prevents unauthorized access, and ensures business data security |
STORAGE_S3_ENDPOINT | https://s3-cn-north-1.example.com | Stores batch-uploaded historical production documents and bid document backups, ensures data traceability |
FE_DOMAIN | https://fastgpt.pcb-manufacturing.example.com | Configures the frontend access domain, adapts to access requirements of the factory’s internal office network |
ROOT_KEY | Custom strong password string | Used to initialize the administrator account, ensures permission security after deployment |
Three common configuration errors
- Symptom: After deploying with Docker Compose, the frontend loads normally, but OneApi interface calls return invalid URL errors. Cause: Incorrect configuration of the access address for
REDIS_URLorMONGODB_URIprevents internal services from connecting to external dependent components. - Symptom: Uploading bid-related files such as Gerber and BOM fails, and an error occurs after creating a simple application and enabling the file upload function. Cause: Incorrect encryption key setup for
FILE_TOKEN_KEY, or insufficient permissions forMONGODB_URIto write uploaded file metadata. - Symptom: Knowledge base retrieval reordering times out, causing bid document preparation to lag. Cause: Improper
REDIS_URLcaching policy configuration means frequently retrieved production data is not cached. Each request directly queries the database, increasing response latency.
How to confirm correct configuration
- Log in to the FastGPT v4.17.0 backend, navigate to the storage configuration page, and confirm
REDIS_URLandMONGODB_URIconnection addresses match locally deployed service addresses. - Upload a test Gerber file, verify successful upload and parsing with no error prompts.
- Call the OneApi interface to test production parameter retrieval, confirm no timeout errors in responses.
- Check system logs for no abnormal errors such as database connection failures or file encryption errors.
The values provided here are common starting points and should be evaluated against local samples.