Log Management (Beta)

HertzBeat's log management feature provides comprehensive log data management capabilities, including log querying, statistical analysis, and batch deletion operations. Users can precisely search logs through various filter conditions, view detailed statistical charts, and clean up unnecessary log data.

:::warning The log management feature is currently in Beta (experimental) stage. There may be potential defects and limitations. The feature is under active development and iteration. :::

Prerequisites

Before using the log management feature, you must first configure a database that supports log storage.

Feature Overview

In the HertzBeat interface, navigate to "Logs" -> "Log Management" page, where you'll see a comprehensive log management interface.

Log Query Features

Filter Conditions

Log management supports the following filter conditions for precise querying:

Filter Condition	Field Name	Data Type	Purpose	Example Value	Use Case
Time Range	timestamp	DateTime	Specify query time window	2024-01-01 00:00:00 to 2024-01-02 00:00:00	Issue time period location, performance analysis
Trace ID	traceId	String	Request tracing in distributed systems	1234567890abcdef	Link tracing, request flow analysis
Span ID	spanId	String	Identifier for specific operations in traces	abcdef1234567890	Microservice call analysis
Severity Number	severityNumber	Number	OpenTelemetry standard level	1-24	Filter by numeric level
Severity Text	severityText	String	Human-readable log level	ERROR, WARN, INFO	Quick filtering by level

Visualization Analysis Charts

Visualization analysis charts include basic statistical indicator charts and aggregated statistical charts. Click the "Show Statistics" button to expand and view them.

Basic Statistical Indicators

Statistical Indicator	Description	Application Value	Anomaly Threshold Reference
Total Logs	Total number of logs matching query criteria	Evaluate system activity and data scale	Sudden increase/decrease over 50% needs attention
FATAL Logs	Fatal error level log statistics	Identify system critical failures	Any FATAL logs need immediate handling
ERROR Logs	Error level log statistics	Monitor system anomalies	Over 5% of total logs needs attention
WARN Logs	Warning level log statistics	Discover potential issues and performance bottlenecks	Continuous growth trend needs analysis
INFO Logs	Information level log statistics	Understand normal system operation status	Should be the main body of logs
DEBUG Logs	Debug level log statistics	Development debugging and issue troubleshooting	Should control quantity in production environment

Aggregated Statistical Charts

Chart Name	Chart Type	Display Content	Main Purpose	Included Metrics/Levels
Severity Distribution Chart	Pie chart	Quantity distribution of different log levels	Quickly understand log severity distribution	FATAL, ERROR, WARN, INFO, DEBUG
Trace Coverage Chart	Pie chart	Proportion of logs with trace information	Analyze system trace coverage	Number of logs with Trace ID
Number of logs with Span ID
Number of logs with both Trace and Span
Number of logs without trace information
Log Trend Chart	Timeline chart	Hourly statistics of log quantity trends	Analyze log generation time patterns and anomaly peaks	Time granularity: Hourly statistics
Analysis dimension: Time pattern recognition, anomaly peak detection

Column Display Configuration

Click the "Column Settings" button to customize table column display:

• Time: Log generation timestamp
• Observed Time: Time when log was observed
• Severity: Log level label
• Log Content: Main message content
• Attributes: Log additional attribute information
• Resource: Resource-related information
• Trace ID: Distributed tracing identifier
• Span ID: Operation span identifier
• Trace Flags: Trace flag information
• Instrumentation Scope Info: Instrumentation scope information
• Dropped Count: Number of dropped attributes

Log Details View

Click any log entry in the table to open the details modal:

Basic Information Section

• Severity: Display level label and color identifier
• Timestamp: Formatted detailed time display
• Trace ID: Complete distributed tracing identifier
• Span ID: Complete operation span identifier

Complete JSON Data

• Raw Data: Display complete JSON format of log entry
• Copy Function: Support one-click copy to clipboard
• Formatted Display: JSON data is formatted for easy reading

Batch Delete Feature

1. Select Logs: Use checkboxes on the left side of the table to select logs for deletion
2. Select All: Use the checkbox in the table header to select all logs on the current page
3. Batch Delete: Click the "Batch Delete" button to delete selected logs
4. Delete Count: Button shows the number of currently selected logs

Troubleshooting

Unable to View Logs

Symptoms: Page shows "No Data" or loading failure Solutions:

1. Check if time-series database is properly configured
2. Confirm database service is running normally
3. Verify database connection configuration is correct
4. Check if log data is being written to the database

For more information about log management features or technical issues, feel free to engage with the community through GitHub Issues.