What data this category includes
Data for this category originates primarily from computer room power and environment monitoring systems, access control systems, UPS monitoring equipment, and video inspection tools. Daily generated documents include inspection records, temperature and humidity logs, alarm event records, and duty shift handover logs. Data fields include temperature and humidity (units: degrees Celsius and relative humidity percentage), input and output voltage, device operating status, alarm level, and trigger time. Data update frequencies fall into two categories: real-time (when an alarm triggers) and daily (when inspection records are aggregated).
Constraints on deployment and upgrade
Real-time alarm data for this category generates high-frequency write requests. Inspection records and log data require stable long-term storage. For offline deployment scenarios, dependent service images must be copied in full. Missing required environment variables cause service startup verification to fail. During version upgrades, cross-version database schema compatibility must be maintained to avoid exceptions when reading inspection records due to data structure changes. Real-time alarm processing logic must remain uninterrupted during upgrades to prevent critical alarm omissions.
How to configure settings
| Configuration Item | Recommended Value | Rationale |
|---|---|---|
DB_MAX_LINK | 20–30 connections | Handles high-frequency real-time alarm writes and inspection record queries, prevents database connection exhaustion |
MONGODB_URI | Points to a standalone MongoDB service address | Stores alarm records and duty logs, adapts to unstructured data storage requirements |
PG_URL | Points to a performance-matched PostgreSQL service address | Stores structured data such as inspection tables and temperature and humidity logs, supports multi-condition combined queries |
REDIS_URL | redis://:custom password@service address:6379/0 | Caches real-time alarm temporary status and session information, reduces database read/write pressure |
FILE_TOKEN_KEY | 16–32 byte random string | Used to generate access tokens for inspection files, ensures access security for internal documents |
FE_DOMAIN | Intranet or public network deployment domain | Used to generate access links for internal tickets, ensures normal access for operation and maintenance personnel |
Three Common Configuration Mistakes
- Symptom: After deploying using an offline copied image, the AI Proxy service fails to start. Logs indicate that required environment variables are missing. Troubleshooting: Verify correct configuration of
AIPROXY_API_ENDPOINT(root address of the AI Proxy service) andAIPROXY_API_TOKEN(AI Proxy administrator token, must match the AI Proxy server configuration and cannot be empty). This configuration is a mandatory dependency starting from FastGPT v4.17.0. - Symptom: After cross-version upgrade, inspection records cannot load normally, and the interface displays missing fields. Cause: Schema migration was not performed per official documentation, and the image was overwritten directly, resulting in incompatible data structures.
- Symptom: After configuring
PG_URL, database connection times out, and inspection queries return no results. Cause: Database firewall rules were not opened for the FastGPT deployment node, causing connection requests to be blocked.
How to Verify Configuration Completion
- Run the startup verification script for the FastGPT service, check that all required environment variables are fully configured and have correct formatting.
- Send a simulated alarm data write request to verify database connections and data storage function normally.
- Access the generated internal ticket link, confirm normal navigation to the corresponding page without permission errors.
- View the AI Proxy service runtime logs, confirm no startup failure or connection exception error messages.
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.