What the data for this category looks like
Core data for this category originates from incident timelines, operation logs, monitoring metrics, and code commit records. Update triggers include major incident launches and fixed monthly retrospective updates. Daily deliverables include RCA reports, incident tickets, retrospective meeting minutes, and rectification tracking sheets. Document structures include root cause analysis modules, associated log fields, rectification tracking items, and more. Fields cover incident occurrence timestamp, root cause type, rectification measures, and verification results. Units are mostly milliseconds, count, and items. Data association relationships are complex, requiring cross-document and cross-data source associated queries to support root cause analysis and rectification tracking.
What constraints these characteristics impose on the database and operations link
Data sources for this category are scattered and have high real-time requirements. This demands the database and operations link to maintain stable long connections and low-latency retrieval capabilities. Data updates are mostly triggered by sudden incidents, supplemented by monthly retrospectives. This requires support for batch writing and scheduled archiving mechanisms to adapt to storage needs across different update rhythms. Multiple field association relationships exist between documents. This requires ensuring the execution efficiency of multi-table associated queries to avoid excessive time spent on root cause analysis. The evidence collection link relies on rapid retrieval of historical logs and monitoring data. A reasonable indexing strategy must be configured to shorten retrieval time. The scope of impact from incident consequences is broad. Database high availability and regular backup processes must be configured to prevent data loss from compromising the completeness of root cause analysis.
How to configure settings
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
MONGODB_URI | mongodb://sre-admin:secure-pass@db-ops:27017/fastgpt-rca?authSource=admin | Standard connection format adapted for storing unstructured data such as RCA reports and incident tickets, meets requirements for multi-document associated queries |
PG_URL | postgresql://sre-user:monitor-pass@db-ops:5432/fastgpt-ops?sslmode=require | Supports structured storage of incident timelines, monitoring metrics, and code commit records, ensures efficiency of multi-table associated queries |
DB_MAX_LINK | 8–12 | Balances concurrent query capability and database resource usage, adapts to dual access peaks for sudden incident troubleshooting and monthly retrospectives |
REDIS_URL | redis://:cache-pass@redis-ops:6379/1 | Caches frequently retrieved incident log fragments to shorten response time during the evidence collection link |
| Database Archival Cycle | 1st day of every month | Adapts to the monthly retrospective update rhythm, cleans up expired non-core data, ensures database storage space utilization |
Three common misconfigurations
- Symptom: Unable to enter a port number when configuring a database connection node. Cause: The advanced configuration switch for database connection nodes in FastGPT v4.17.0 was not enabled, causing the port number input field to be hidden.
- Symptom: No mongo/data folder exists in the current working directory after startup. Cause: The data storage path was not specified in the startup command, the default path was not automatically created, manual creation or adjustment of startup parameters is required.
- Symptom: Database connection timeout error occurs during workflow invocation, status code 503 is returned. Cause: The
DB_MAX_LINKconfiguration value is too low, causing the connection pool to be exhausted and unable to respond to new query requests.
How to confirm successful configuration
- Run the built-in database connectivity test tool included with FastGPT v4.17.0 to verify the connection status of
MONGODB_URI,PG_URL, andREDIS_URL, with no authentication failure or address error prompts. - View database connection pool monitoring metrics to confirm that the current number of connections is lower than the
DB_MAX_LINKconfiguration value, with no connection overflow occurrences. - Manually trigger a batch write operation for incident logs to verify that data can be properly stored to the corresponding database instance.
- Check scheduled task logs to confirm that the database archival cycle task executes normally at the preset time.
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.