Data for This Category
Core data for sheet metal cutting and bending scenarios comes from CAD/CAM software, order BOM systems, sheet material inventory ledgers, laser cutters, bending machines, and workshop MES systems. Daily produced documents include part unfolding drawings, CNC cutting programs, bending process cards, nesting diagrams, first-piece inspection records, and process flow cards. Data fields include bending compensation values, hole position coordinates, sheet material thickness, leftover material dimensions, and more. Units are mostly millimeters and square meters. Data updates synchronize with order generation, daily inventory checks, and each process changeover.
Constraints During Deployment and Upgrade
Data sources for this category are scattered and have varied formats. Integration with non-standard documents such as industrial CAD files and CNC programs is required. During deployment, storage and database connections that support parsing multiple file types must be configured. The high-frequency update rhythm requires sufficient real-time synchronization bandwidth and cache resources to be reserved during deployment, to avoid data delays. The multi-variety, small-batch business model leads to large fluctuations in single-batch data volume. During upgrade, database and storage resources must be able to expand on demand. Complete historical process data backups must also be retained, to prevent new data after process changeover from overwriting existing records.
Configuration Settings
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
MONGODB_URI | mongodb://admin:securepass@db-host:27017/fastgpt?authSource=admin | Process flow cards, first-piece inspection records and other data for this category have large volume. A permission-verified database connection must be configured to ensure data security |
REDIS_URL | redis://:redispass@cache-host:6379/0 | Frequently read bending process cards and nesting diagram metadata must be cached, to reduce repeated query load on the database |
STORAGE_S3_ENDPOINT | http://s3-cn-north-1.example.com | Large industrial files such as unfolding drawings and CNC programs must be stored. An S3-compatible object storage endpoint must be configured to support batch upload requirements |
FE_DOMAIN | https://fastgpt-manufacture.example.com | A public network access domain name must be uniformly configured for workshop terminals, to resolve cross-domain and permission restriction issues with local IP access |
FILE_TOKEN_KEY | 32-character random encrypted string | Secure access tokens must be generated for uploaded sheet metal drawings, to prevent unauthorized personnel from accessing production drawing data |
ROOT_KEY | Exclusive key generated by administrators | Only core operation and maintenance personnel must be allowed to modify core configurations for industrial data synchronization, to prevent misoperations from affecting production data flow |
Three Common Mistakes
- The symptom is that accessing the public IP plus port of the deployment node fails to load the page, while access via local 127.0.0.1 works normally. The cause is that
FE_DOMAINis not configured as a publicly accessible domain name, or the server firewall has not opened inbound rules for the corresponding port. - The symptom is that the knowledge base list is empty after upgrade, or the version number displayed on the page does not match the actual upgraded version. The cause is that the original database data pointed to by
MONGODB_URIwas not properly backed up and restored, or the upgrade script did not overwrite the original configuration file, resulting in lost parameters. - The symptom is an empty field error when clicking the knowledge base module. The cause is that
FILE_TOKEN_KEYis not configured, so valid access links cannot be generated for uploaded sheet metal drawings, or theREDIS_URLconnection failed, resulting in metadata cache failure.
How to Verify Successful Configuration
- Run the local health check command
curl http://localhost:3000/api/health, confirm that the connection status of the database, cache and storage services are all normal. - Upload a sheet metal unfolding drawing to the knowledge base, verify that the file can be previewed normally without permission-related errors.
- Log in with the permission account corresponding to
ROOT_KEY, confirm that core configuration items for data synchronization can be modified, and regular accounts cannot access the corresponding settings page. - After restarting the FastGPT service, access the front-end page via public IP plus port or the configured
FE_DOMAIN, confirm that the page loads normally and the version number displays FastGPT v4.17.0.
The values provided are common starting points and should be measured against the reader's own samples.