What the data for this category looks like
This category of data primarily comes from tracking systems, data warehouses, CRM systems, and ticket systems. The update rhythm is daily or weekly. Regularly produced documents include usage analysis reports, tracking plans, indicator specification tables, and dashboard requirement sheets. Common data fields include customer unique identifiers, function module names, operation durations, abnormal behavior tags, renewal-related nodes, and more. Units are mostly times, minutes, percentages, etc. Statistics must strictly follow preset indicator specifications.
What constraints do these characteristics impose on the deployment and upgrade link
This document adapts to the FastGPT v4.17.0 deployment process. Data sources include multiple heterogeneous systems. Deployment must adapt to connection protocols and permission configurations for different data sources. It ensures connectivity of each link. The data update rhythm is daily or weekly. After deployment, configure reasonable synchronization cycle parameters. Avoid data lag caused by overly long synchronization intervals. Avoid excessive system resource occupation caused by overly short intervals. The requirement for unified indicator specifications requires loading preset specification configuration files during deployment. Ensure that the statistical logic is consistent with business documents. Avoid data statistical errors caused by configuration deviations.
How to Set Configurations
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
MONGODB_URI | mongodb://admin:securepass@db-host:27017/fastgpt_analytics?authSource=admin | Stores raw tracking data for product usage behavior, must match the authentication information and target database name of the data source |
PG_URL | postgresql://analytics_user:dbpass@pg-host:5432/analytics_db | Stores structured business documents such as indicator specification tables and dashboard requirement sheets, adapts to relational database connection rules |
REDIS_URL | redis://:redispass@redis-host:6379/1 | Caches intermediate results of real-time statistics, set an independent database number to avoid conflicts with other business caches |
DB_MAX_LINK | 80–150 | Limits the maximum concurrent connection count of the database connection pool, avoids exhausting server connection resources during high-concurrency synchronization |
FE_DOMAIN | https://analytics.your-domain.com | Configures the access domain name of the frontend page, used to generate valid jump links for reports and perform permission verification |
STORAGE_S3_ENDPOINT | Set based on actual testing | If using object storage to store analysis report attachments, must match the endpoint address of the storage service provider |
Three Common Mistakes
- Accessing the deployed frontend page shows a spinning blank screen with no clear error logs. The cause is incorrect configuration of the
FE_DOMAINparameter. This causes cross-origin verification failure for frontend resources. It prevents normal loading of page resources. - The model test step returns empty results with no clear error information. The cause is that the local port configuration of the ollama model does not match the FastGPT proxy parameters. Cross-origin access permission for ollama is not enabled.
- Dashboard data for product usage analysis is missing after server restart. The cause is that the persistence parameter for
REDIS_URLis not configured. Cached temporary statistical data is lost after restart. Historical data must be resynchronized.
How to Confirm Configurations Are Correct
- Run the database connection test command. Verify that configurations for
MONGODB_URIandPG_URLcan establish normal connections. No authentication or address errors are present. - Access the frontend address configured in
FE_DOMAIN. Confirm that the page loads normally and displays the basic dashboard framework. No cross-origin related console errors are present. - Manually trigger a data synchronization task. Check that the cache for
REDIS_URLgenerates temporary statistical data. The storage path matches the configured requirements. - View system logs. Confirm that the
DB_MAX_LINKconnection count limit does not trigger resource exhaustion alerts. The database connection pool operates normally.
The values provided are common starting points and should be measured 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.