What the data for this category looks like
Survey and design internal office data mainly comes from project survey reports, structural design drawings, BIM model files, project ledgers, review feedback comments, and similar sources. Data update rhythm adjusts with project nodes, and is synced immediately when drawings are revised, survey points are added, or review comments are updated. Document structure includes two categories: structured and unstructured. Structured data includes fields such as project number, survey coordinates, load parameters, design floor height, and involved personnel, with units including meters, megapascals, man-days, and similar units. Unstructured data mostly consists of large files such as PDF documents and CAD drawings.
What Constraints Do These Characteristics Impose on Deployment and Upgrade
Reference only; verify against the target FastGPT release and deployment documentation.
How to Set the Configuration
| Config Item | Recommended Value | Basis for This Value |
|---|---|---|
PG_URL | postgresql://admin:securepass@db-host:5432/survey_and_design_database?schema=public | Stores structured data such as project ledgers and review feedback, supports multi-field associated queries for relational data |
REDIS_URL | redis://default:securepass@redis-host:6379/0?timeout=30 | Caches session logs and temporary embedding vector data, supports high-concurrency internal office session access, timeout set to 30 seconds |
STORAGE_S3_ENDPOINT | https://s3-cn-south-1.mycloud.com | Stores large files such as CAD drawings and survey reports, configures chunked upload threshold at 100 MB |
DB_MAX_LINK | 200–300 connections | Adapts to database connection needs for parallel multiple projects, set connection pool upper limit based on actual measurements |
FE_DOMAIN | https://fastgpt.yourcompany.com | Configures frontend access domain name, used to generate valid callback addresses for file access tokens |
FILE_TOKEN_KEY | 32-character random secure string | Encrypts file access tokens, prevents unauthorized access to survey and design exclusive documents |
Three Common Misconfigurations
- The session log detail page fails to load, displaying "cache field is empty". The cause is incorrect configuration of
REDIS_URL, which prevents temporary session data from being cached properly. - Fails to add project data indexes, with empty retrieval results. The cause is failure to configure database indexes for special fields such as survey coordinates and project number, so retrieval cannot match target data.
- The frontend page has no response when initiating office process requests, and container resource monitoring shows the connection pool is exhausted. The cause is
DB_MAX_LINKbeing set too low, which cannot support database access needs for parallel multiple projects.
How to Confirm the Configuration Is Successful
- Run the docker-compose ps command, confirm all deployment containers have a running status.
- Log in to the system management backend, attempt to create a database index for survey project ledgers, confirm no errors occur and the index is created successfully.
- Upload a standard CAD drawing file, confirm the file upload succeeds and can be previewed normally via the generated access link.
- Initiate a test session, enter the session log detail page, confirm the log content is complete and can be viewed normally.
The values provided here are common starting points, and should be evaluated against local deployment samples.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-09-28.