What this category of data looks like
Usually, data for this category comes from ERP, WMS systems, barcode or RFID identification devices, purchase work orders, storage location management data, AGV transportation systems and weighing equipment. Documents such as receiving slips, putaway slips, material picking and returning slips, warehouse transfer slips, inventory counts, lot ledgers, first-in, first-out (FIFO) and expiration date records are generated daily. Data update frequencies include real-time synchronization of material circulation information, daily inventory data summary, and full monthly inventory counts. Core fields include material codes, lot numbers, storage location information, expiration dates, inspection statuses and complete circulation chain records.
What constraints do these characteristics impose during deployment and upgrade
The multi-source, heterogeneous data sources of this category require configuring adaptation rules for multi-system data docking during the deployment phase to avoid field mapping errors. Real-time updated material circulation data requires adjusting database connection parameters to ensure synchronization stability. The structured field requirements for lots and expiration dates require clarifying the mapping logic of associated fields during deployment. Processing monthly bulk inventory count data requires adapting to storage capacity and bulk write configuration constraints. During the upgrade process, the associated chain of lot circulation must be retained to prevent interruption of the traceability chain, which would affect subsequent quality problem traceability.
How to set configurations
The following configuration items apply to FastGPT v4.17.0.
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
DB_MAX_LINK | 20–50 concurrent connections | Meets the concurrent connection requirements for real-time synchronization of multi-source data, avoiding synchronization interruptions caused by exhausted connections |
MONGODB_URI | Point to a dedicated MongoDB instance that stores lot ledgers and circulation records | This category of data contains a large amount of time-series circulation information associated with lots, and a dedicated instance ensures performance for complex queries |
PG_URL | Point to a PostgreSQL instance that stores structured documents such as receiving slips and inventory counts | Structured document data is suitable for relational database storage, facilitating multi-condition queries by material code and lot number |
REDIS_URL | Configure to the locally deployed Redis service address | Used to cache frequently queried material lot and storage location information, reducing query pressure on backend databases |
FE_DOMAIN | Fill in the public or internal network access domain name of the deployment node | Ensures normal loading of the front-end page, adapting to internal or external enterprise access scenarios |
STORAGE_S3_ENDPOINT | Configure to the endpoint address of the enterprise's self-built object storage | Used to store attachments such as barcode images and weighing data, adapting to the enterprise's existing storage architecture |
Three common mistakes
- The symptom is that the front-end page cannot be accessed after the container starts, returning a connection timeout or 403 status code. The cause is that the address configured in
FE_DOMAINdoes not match the actual deployed public or internal network IP, or the network access rules for the corresponding port are not opened. - The symptom is that a third-party logistics system cannot call FastGPT's query interface, prompting a connection refused error. The cause is that the API listening address was not configured correctly, and it was only bound to the local loopback address, preventing external systems from accessing it.
- The symptom is that after upgrading the version, circulation records for historical lots cannot be queried normally. The cause is that the index rebuilding script for the associated database was not executed during the upgrade, or the associated constraints for lot fields in the original database were not retained.
How to confirm the configuration is correct
- Perform a front-end page access test, confirm that the system login interface can be loaded normally, and verify that the address configured in
FE_DOMAINmatches the actual access address. - Perform a multi-source data docking test, import a batch of receiving slip data from ERP and WMS systems, confirm that field mapping is correct, and core fields such as lot numbers and storage location information are not missing.
- Perform a bulk data import test, upload a monthly inventory count file, confirm that the file can be parsed normally and stored in the corresponding database instance.
- Perform an API interface call test, use a third-party system to call the material lot query interface, confirm that correct circulation record information is returned.
The values given are common starting points and should be measured against the reader's own samples.