This document applies to FastGPT v4.17.0.
What data for this category looks like
Data for this category is generated around bidding requirements and enterprise qualifications. Daily documents include bid documents, qualification certificates, business deviation forms, and technical deviation forms. Data is sourced from bidding announcements, bidding documents, enterprise qualification databases, and supplier quotations. It is updated independently per order. A single bid document contains structured content across multiple modules. Fields include bidding clause response items, qualification document numbers, quotation details, and seal status identifiers. No unified fixed units apply. Values adjust dynamically based on bidding requirements.
Constraints on database and operations workflows
Data for this category is updated independently per order, and its structure adjusts dynamically based on bidding requirements. This requires the database to support flexible non-fixed schema storage, to adapt to multi-module, multi-field document structures. A single data entry links to multiple source information. This requires consistency across cross-source data links. Because errors directly impact bidding outcomes, the database must guarantee stable read/write operations and timely data backups. The per-order update rhythm also requires operations workflows to support on-demand resource configuration. This avoids resource waste or performance bottlenecks, and ensures business continuity during parallel processing of multiple orders.
How to set configurations
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
MONGODB_URI | mongodb://fastgpt:bidpass@db-host:27017/fastgpt_bid?authSource=admin | Bid documents and qualification certificates for this category are mostly unstructured. MongoDB supports flexible schema storage. This configuration specifies a dedicated database instance to isolate business data for this category |
PG_URL | postgresql://fastgpt:bidmeta@pg-host:5432/fastgpt_bid_meta | Stores structured metadata such as bidding clause response items and qualification numbers. PostgreSQL supports complex associated queries and transactions, ensuring consistency checks for cross-source data |
DB_MAX_LINK | 15–25 | This category processes business per order. Concurrent connection counts for a single instance do not need to be high. This value range meets connection requirements for parallel processing of multiple orders, and prevents database connection exhaustion |
REDIS_URL | redis://:redispass@redis-host:6379/1 | Caches bidding announcement pull results and qualification file verification status, reducing repeated database queries and improving data response speed during bid document preparation |
| Database automatic backup trigger timing | Triggered after a single order is completed | Errors in data for this category can lead to bid invalidation. Backing up per order enables rapid recovery of single business data entries, avoiding impact on other orders |
Three common error cases
- Symptom: Error
connection refusedappears on startup. Logs show database connection failure. Cause: The address, account, or password forMONGODB_URIorPG_URLis not configured correctly, or the database instance is not running, preventing connection establishment. - Symptom: After uploading a test bid document, some fields of qualification materials are empty. Cause:
DB_MAX_LINKis set too low. During parallel processing of multiple orders, database connections are exhausted, preventing complete reading and storage of file metadata. - Symptom: External tools cannot directly manage vector database data. Cause: External access permissions for the vector database are not enabled, or corresponding access keys are not configured, preventing third-party tools from connecting to the database instance.
How to verify correct configuration
- Run a database connection test script to verify connectivity for
MONGODB_URI,PG_URL, andREDIS_URL, and confirm no connection errors occur. - Upload a test bid document that meets bidding requirements, check that metadata is fully written to the database, and confirm field storage matches business needs.
- Simulate a parallel processing scenario for multiple orders, observe whether database connection counts stay within the preset range, and confirm no connection exhaustion occurs.
- Trigger a database backup operation, confirm that backup files are generated normally, and that data can be restored successfully.
The values provided here are common starting points. They should be evaluated against the reader's own samples.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-10-19.