Cloud Log Service provides a log analysis function that supports constructing analysis scenarios through the interaction mode, or you can directly write SQL statements for query and analysis. This section describes the basic operations of the analysis function.

Analyzing with SQL Statements

Structured Query Language (SQL) is a programming language used to control database access and manage data in databases. Cloud Log Service provides a set of SQL statements for analyzing structured data in log units, which can be used for statistical analysis on logs that meet your query conditions and return statistical analysis results.

For example, the query conditions can be set as:

status=404

And set the SQL analysis statement as:

selectcount(*)aslogCounts

It represents the statistics of the number of logs with response status code 404.

For more about the statements and syntax, see SQL Syntax.

Procedure

Go to the Cloud Log Service console, select the target log project and log unit, and click Retrieve Log to enter the retrieval page. Switch to the statement mode or interaction mode to set log query conditions based on your needs. For specific operations, see Log Query. If you do not set the query conditions, all log data of the current log unit will be analyzed by default.
In Configure Statistical Analysis, enter an SQL statement. For SQL statements and syntax, see SQL Syntax.
Click the Statistics/Analysis button to perform statistical analysis based on the currently set query conditions and SQL statements.

Constructing an Analysis Scenario Using Interaction Mode

Cloud Log Service supports the rapid construction of SQL statements using the interaction mode. Click on the scene name to view specific scenario descriptions and operations.

Scenario	Description
Field Screening & Filtering	Set the filter fields, filter conditions, sort conditions and number of results.
Metric Statistics	Multiple statistical metrics preset, including the number of logs, log fields, mathematical computing, mathematical statistics, and estimation functions.
Group Statistics	Supports grouping by fields and statistics on multiple metrics.
Log Proportion	Statistics on the number and proportion of logs that meet the query conditions.
TopN	Statistics on the Top N value distribution of a specified field.
Time Trend	Supports statistics on time trends of specified metrics based on a specified time granularity.

As an example, the following describes how to query and analyze Nginx raw logs in LTS.

Procedure

Under the target log unit, click Retrieve Log.
In the input box for configuration statistical analysis, enter an SQL statement for statistical analysis. The following are two specific statistical analysis cases.

Top 3 Access Sources

The SQL analysis statement entered is as follows:

selectcount(1)aspv,http_referergroupbyhttp_refererorderbypvdesclimit3

Using a Pie Chart for Statistics on the Proportion of the Status of Each Request

The SQL analysis statement entered is as follows:

select"status",count(*)as"num"fromloggroupby"status"orderby"num"

Cloud Log Service

Log Analysis