This section describes how to query Log4j logs in Cloud Log Service.

This section takes the most common Java project as an example to describe the log analysis operation process for Log4j.

Background

Log4j is an open-source project of Apache. Log4j allows you to specify the level, output destination, and output format of logs. The log levels in descending order are ERROR, WARN, INFO, and DEBUG. The output destination determines whether logs are displayed in consoles or files. The output format determines the format of logs that are displayed. Similar components also include logback, etc.

For example, in a Java project, the exception information for a certain period needs to be queried to see if there are any system errors so that the fault can be quickly located. For this scenario, Cloud Log Service provides a quick query function. For example, the information of a Java log is as follows:

level:ERROR location:cn.ctyun.log4jtest.getUserInfo message:getconnectionfrompooltimeout,poolisbusy,rejecttasktime:2023-07-2010:20:30.437

Before You Begin

Log4j logs have been collected.

Collection rule configuration has been defined.

Procedure

Log in to the Cloud Log Service console.
In Log Unit and Project, select the project where the logs are stored.
In the project, select or filter out the corresponding log units.
Select ERROR through Quick Mode and set the time range for query analysis to count errors in the selected time period.

Or enter other specific business-related keywords in the retrieval box for quick retrieval.

Click the Context Search icon to quickly locate the context of the log line. This facilitates troubleshooting log information for you to obtain the system running status.

SQL query and analysis can be performed, such as statistics on the distribution of error types in the last 7 days. The SQL query analysis statement is as follows:

SELECTLevel,count(*)asNumbergroupbyLevel

Cloud Log Service

Querying Log4j logs