What this category of data looks like
Data for this category covers the full production workflow across manufacturing enterprise departments including PMC, planning, and shop scheduling. Sources include ERP orders and inventory, BOM lists, process routes, equipment capacity, staff shifts, work-in-progress inventory, and supplier delivery information. Daily generated documents include sales orders, master production schedules, material requirement plans, weekly and daily scheduling tables, work orders, material shortage lists, and delivery change notices. Data fields include order numbers, process codes, equipment IDs, shift time periods, material codes, work-in-progress quantities, and additional relevant fields. Units include pieces, hours, days, and other standard production units. Data update frequency is not fixed, triggered by daily scheduling, weekly plan adjustments, order placement, or exceptional events.
What constraints these characteristics impose on deployment and upgrades
This category has complex data structures and dispersed data sources, covering production information across multiple departments and stages. Deployment must accommodate concurrent requirements for multi-source data synchronization. Structured ledger data such as work orders and scheduling tables must be compatible with ERP system database formats, requiring dedicated relational storage. Frequently accessed data including equipment capacity and shift rules must use a caching service to reduce core database load. Data updates have no fixed cycle, requiring flexible trigger configurations to adapt to temporary synchronization scenarios such as order placement or exceptional events. Large-volume process documents and change notice attachments require independent storage services to avoid overloading the core database. During upgrades, existing field mapping configurations must be preserved to prevent breaks in ledger data association relationships.
How to set the configurations
All following configuration items are compatible with FastGPT v4.17.0.
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
PG_URL | postgresql://admin:securepass@192.168.1.100:5432/fastgpt_prod | Stores structured ledger data such as work orders and scheduling tables, compatible with ERP system database formats, and supports multi-field associated production data |
REDIS_URL | redis://:redispass@192.168.1.101:6379/1 | Caches frequently accessed data including equipment capacity and shift rules, reducing repeated query pressure on the core database |
DB_MAX_LINK | 20–30 | Meets concurrent connection requirements during multi-source data synchronization, preventing synchronization interruptions caused by exhausted connections |
STORAGE_S3_ENDPOINT | http://192.168.1.102:9000 | Stores large-volume process route documents and exceptional change notice attachments, compatible with internal enterprise object storage architectures |
ROOT_KEY | 3a7f2d9c1e4b6a8d0f5e7c9b2a4d6f8e | Configures system administrator permissions, uses a randomly generated 32-bit string to avoid default key exposure risks |
FE_DOMAIN | https://prod-plan.yourcompany.com | Binds the front-end access address, ensuring shop scheduling personnel can normally access the ledger generation interface |
Three common mistakes
- Symptom: Unable to synchronize ERP work order data during initial deployment, with logs indicating incompatible data formats. Cause: Failed to select
PG_URLas the primary storage based on the structured nature of production ledgers, using a non-relational database instead which caused field mapping failures. - Symptom: Ledger reports generated after upgrades cannot associate operator information, with the userId field empty. Cause: Failed to update custom user identifier mapping configurations after upgrading, continuing to use the legacy
tmbIdparameter without adaptation. - Symptom: Storage volume mounting errors occur after Sealos deployment, making it impossible to save ledger data. Cause: Failed to clarify storage permission configurations for Sealos deployments, confusing local volume mounting logic with Docker deployments, and failing to correctly bind the data directory.
How to confirm configurations are correct
- Run a database connectivity test script compatible with FastGPT v4.17.0, verify connectivity for
PG_URLandREDIS_URL, with no timeout or authentication failure prompts. - Import a test work order data set, confirm that ledger fields are correctly mapped and an initial scheduling table is generated, with no missing fields or format error prompts.
- Access the address configured in
FE_DOMAIN, confirm that the front-end page loads normally and can submit test scheduling generation tasks. - Upload a test process route document, verify that the storage service configured in
STORAGE_S3_ENDPOINTcan save and read files normally.
The values provided are common starting points and should be measured against the reader's own samples.