Axiom Docs

Log4j is a Java logging framework developed by the Apache Software Foundation and widely used in the Java community. This page covers how to get started with Log4j, configure it to forward log messages to Fluentd, and send logs to Axiom.

Prerequisites

Create an Axiom account.
Create a dataset in Axiom where you send your data.
Create an API token in Axiom with permissions to update the dataset you have created.

Configure Log4j

Log4j is a flexible and powerful logging framework for Java applications. To use Log4j in your project, add the necessary dependencies to your pom.xml file. The dependencies required for Log4j include log4j-core, log4j-api, and log4j-slf4j2-impl for logging capability, and jackson-databind for JSON support.

Create a new Maven project:

mvn archetype:generate -DgroupId=com.example -DartifactId=log4j-axiom-test -DarchetypeArtifactId=maven-archetype-quickstart -DinteractiveMode=false

cd log4j-axiom-test

Open the pom.xml file and replace its contents with the following:

<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
  xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/maven-v4_0_0.xsd">
  <modelVersion>4.0.0</modelVersion>
  <groupId>com.example</groupId>
  <artifactId>log4j-axiom-test</artifactId>
  <packaging>jar</packaging>
  <version>1.0-SNAPSHOT</version>
  <name>log4j-axiom-test</name>
  <url>http://maven.apache.org</url>

  <properties>
    <maven.compiler.source>11</maven.compiler.source>
    <maven.compiler.target>11</maven.compiler.target>
    <log4j.version>2.19.0</log4j.version>
  </properties>

  <dependencies>
    <dependency>
      <groupId>junit</groupId>
      <artifactId>junit</artifactId>
      <version>4.12</version>
      <scope>test</scope>
    </dependency>
    <dependency>
      <groupId>org.apache.logging.log4j</groupId>
      <artifactId>log4j-core</artifactId>
      <version>${log4j.version}</version>
    </dependency>
    <dependency>
      <groupId>org.apache.logging.log4j</groupId>
      <artifactId>log4j-api</artifactId>
      <version>${log4j.version}</version>
    </dependency>
    <dependency>
      <groupId>org.apache.logging.log4j</groupId>
      <artifactId>log4j-slf4j2-impl</artifactId>
      <version>${log4j.version}</version>
    </dependency>
    <dependency>
      <groupId>com.fasterxml.jackson.core</groupId>
      <artifactId>jackson-databind</artifactId>
      <version>2.13.0</version>
    </dependency>
  </dependencies>

  <build>
    <plugins>
      <plugin>
        <groupId>org.apache.maven.plugins</groupId>
        <artifactId>maven-shade-plugin</artifactId>
        <version>3.2.4</version>
        <executions>
          <execution>
            <phase>package</phase>
            <goals>
              <goal>shade</goal>
            </goals>
            <configuration>
              <transformers>
                <transformer implementation="org.apache.maven.plugins.shade.resource.ManifestResourceTransformer">
                  <mainClass>com.example.App</mainClass>
                </transformer>
              </transformers>
              <createDependencyReducedPom>false</createDependencyReducedPom>
            </configuration>
          </execution>
        </executions>
      </plugin>
    </plugins>
  </build>
</project>

This pom.xml file includes the necessary Log4j dependencies and configures the Maven Shade plugin to create an executable JAR file.

Create a new file named log4j2.xml in your root directory and add the following content:

<?xml version="1.0" encoding="UTF-8"?>
<Configuration status="WARN">
  <Appenders>
    <Socket name="Socket" host="127.0.0.1" port="24224" protocol="TCP">
      <JsonLayout complete="false" compact="true" eventEol="true" properties="true" includeTimeMillis="true"/>
    </Socket>
    <Console name="Console" target="SYSTEM_OUT">
      <PatternLayout pattern="%d{HH:mm:ss.SSS} [%t] %-5level %logger{36} - %msg%n"/>
    </Console>
  </Appenders>
  <Loggers>
    <Root level="info">
      <AppenderRef ref="Socket"/>
      <AppenderRef ref="Console"/>
    </Root>
  </Loggers>
</Configuration>

This configuration sets up two appenders:

A Socket appender that sends logs to Fluentd, running on localhost:24224. Is uses JSON format for the log messages, which makes it easier to parse and analyze the logs later in Axiom.
A Console appender that prints logs to the standard output,

Set log level

Log4j supports various log levels, allowing you to control the verbosity of your logs. The main log levels, in order of increasing severity, are the following:

TRACE: Fine-grained information for debugging.
DEBUG: General debugging information.
INFO: Informational messages.
WARN: Indications of potential problems.
ERROR: Error events that might still allow the app to continue running.
FATAL: Severe error events that might lead the app to cancel.

In the configuration above, the root logger level is set to INFO which means it logs messages at INFO level and above (WARN, ERROR, and FATAL).

To set the log level, create a simple Java class to demonstrate these log levels. Create a new file named App.java in the src/main/java/com/example directory with the following content:

package com.example;

import org.apache.logging.log4j.LogManager;
import org.apache.logging.log4j.Logger;
import org.apache.logging.log4j.ThreadContext;
import org.apache.logging.log4j.core.config.Configurator;
import org.apache.logging.log4j.Level;

import java.util.Random;

public class App {
    // Define loggers for different purposes
    private static final Logger logger = LogManager.getLogger(App.class);
    private static final Logger securityLogger = LogManager.getLogger("SecurityLogger");
    private static final Logger performanceLogger = LogManager.getLogger("PerformanceLogger");

    public static void main(String[] args) {
        // Configure logging levels programmatically
        configureLogging();

        Random random = new Random();

        // Infinite loop to continuously generate log events
        while (true) {
            try {
                // Simulate various logging scenarios
                simulateUserActivity(random);
                simulateDatabaseOperations(random);
                simulateSecurityEvents(random);
                simulatePerformanceMetrics(random);

                // Simulate a critical error with 10% probability
                if (random.nextInt(10) == 0) {
                    throw new RuntimeException("Simulated critical error");
                }

                Thread.sleep(1000);  // Sleep for 1 second
            } catch (InterruptedException e) {
                logger.warn("Sleep interrupted", e);
            } catch (Exception e) {
                logger.error("Critical error occurred", e);
            } finally {
                // Clear thread context after each iteration
                ThreadContext.clearAll();
            }
        }
    }

    private static void configureLogging() {
        // Set root logger level to DEBUG
        Configurator.setRootLevel(Level.DEBUG);

        // Set custom logger levels
        Configurator.setLevel("SecurityLogger", Level.INFO);
        Configurator.setLevel("PerformanceLogger", Level.TRACE);
    }

    // Simulate user activities and log them
    private static void simulateUserActivity(Random random) {
        String[] users = {"Alice", "Bob", "Charlie", "David"};
        String[] actions = {"login", "logout", "view_profile", "update_settings"};

        String user = users[random.nextInt(users.length)];
        String action = actions[random.nextInt(actions.length)];

        // Add user and action to thread context
        ThreadContext.put("user", user);
        ThreadContext.put("action", action);

        // Log different user actions with appropriate levels
        switch (action) {
            case "login":
                logger.info("User logged in successfully");
                break;
            case "logout":
                logger.info("User logged out");
                break;
            case "view_profile":
                logger.debug("User viewed their profile");
                break;
            case "update_settings":
                logger.info("User updated their settings");
                break;
        }
    }

    // Simulate database operations and log them
    private static void simulateDatabaseOperations(Random random) {
        String[] operations = {"select", "insert", "update", "delete"};
        String operation = operations[random.nextInt(operations.length)];
        long duration = random.nextInt(1000);

        // Add operation and duration to thread context
        ThreadContext.put("operation", operation);
        ThreadContext.put("duration", String.valueOf(duration));

        // Log slow database operations as warnings
        if (duration > 500) {
            logger.warn("Slow database operation detected");
        } else {
            logger.debug("Database operation completed");
        }

        // Simulate database connection loss with 5% probability
        if (random.nextInt(20) == 0) {
            logger.error("Database connection lost", new SQLException("Connection timed out"));
        }
    }

    // Simulate security events and log them
    private static void simulateSecurityEvents(Random random) {
        String[] events = {"failed_login", "password_change", "role_change", "suspicious_activity"};
        String event = events[random.nextInt(events.length)];

        ThreadContext.put("security_event", event);

        // Log different security events with appropriate levels
        switch (event) {
            case "failed_login":
                securityLogger.warn("Failed login attempt");
                break;
            case "password_change":
                securityLogger.info("User changed their password");
                break;
            case "role_change":
                securityLogger.info("User role was modified");
                break;
            case "suspicious_activity":
                securityLogger.error("Suspicious activity detected", new SecurityException("Potential breach attempt"));
                break;
        }
    }

    // Simulate performance metrics and log them
    private static void simulatePerformanceMetrics(Random random) {
        String[] metrics = {"cpu_usage", "memory_usage", "disk_io", "network_latency"};
        String metric = metrics[random.nextInt(metrics.length)];
        double value = random.nextDouble() * 100;

        // Add metric and value to thread context
        ThreadContext.put("metric", metric);
        ThreadContext.put("value", String.format("%.2f", value));

        // Log high resource usage as warnings
        if (value > 80) {
            performanceLogger.warn("High resource usage detected");
        } else {
            performanceLogger.trace("Performance metric recorded");
        }
    }

    // Custom exception classes for simulating errors
    private static class SQLException extends Exception {
        public SQLException(String message) {
            super(message);
        }
    }

    private static class SecurityException extends Exception {
        public SecurityException(String message) {
            super(message);
        }
    }
}

This class demonstrates the use of different log levels and also shows how to add context to your logs using ThreadContext.

Forward log messages to Fluentd

Fluentd is a popular open-source data collector used to forward logs from Log4j to Axiom. The Log4j configuration is already set up to send logs to Fluentd using the Socket appender. Fluentd acts as a unified logging layer, allowing you to collect, process, and forward logs from various sources to different destinations.

Configure the Fluentd.conf file

To configure Fluentd, create a configuration file. Create a new file named fluentd.conf in your project root directory with the following content:

<source>
  @type forward
  bind 0.0.0.0
  port 24224
  <parse>
    @type multi_format
    <pattern>
      format json
      time_key timeMillis
      time_type string
      time_format %Q
    </pattern>
  </parse>
</source>

<filter **>
  @type record_transformer
  <record>
    tag java.log4j
  </record>
</filter>

<match **>
  @type http
  endpoint https://AXIOM_DOMAIN/v1/datasets/DATASET_NAME/ingest
  headers {"Authorization":"Bearer API_TOKEN"}
  json_array true
  <buffer>
    @type memory
    flush_interval 5s
    chunk_limit_size 5m
    total_limit_size 10m
  </buffer>
  <format>
    @type json
  </format>
</match>

Replace API_TOKEN with the Axiom API token you have generated. For added security, store the API token in an environment variable.

Replace DATASET_NAME with the name of the Axiom dataset where you want to send data.

Replace AXIOM_DOMAIN with api.axiom.co if your organization uses the US region, and with api.eu.axiom.co if your organization uses the EU region. For more information, see Regions.

This configuration does the following:

Set up a forward input plugin to receive logs from Log4j.
Add a java.log4j tag to all logs.
Forward the logs to Axiom using the HTTP output plugin.

Create the Dockerfile

To simplify the deployment of the Java app and Fluentd, use Docker. Create a new file named Dockerfile in your project root directory with the following content:

# Build stage
FROM maven:3.8.1-openjdk-11-slim AS build

WORKDIR /usr/src/app
COPY pom.xml .
COPY src ./src
COPY log4j2.xml .
RUN mvn clean package

# Runtime stage
FROM openjdk:11-jre-slim

WORKDIR /usr/src/app

RUN apt-get update && \
    apt-get install -y --no-install-recommends \
    ruby \
    ruby-dev \
    build-essential && \
    gem install fluentd --no-document && \
    fluent-gem install fluent-plugin-multi-format-parser && \
    apt-get clean && \
    rm -rf /var/lib/apt/lists/*

COPY --from=build /usr/src/app/target/log4j-axiom-test-1.0-SNAPSHOT.jar .
COPY fluentd.conf /etc/fluent/fluent.conf
COPY log4j2.xml .

# Create startup script
RUN echo '#!/bin/sh\n\
fluentd -c /etc/fluent/fluent.conf &\n\
sleep 5\n\
java -Dlog4j.configurationFile=log4j2.xml -jar log4j-axiom-test-1.0-SNAPSHOT.jar\n'\
> /usr/src/app/start.sh && chmod +x /usr/src/app/start.sh

EXPOSE 24224

CMD ["/usr/src/app/start.sh"]

This Dockerfile does the following:

Build the Java app.
Set up a runtime environment with Java and Fluentd.
Copy the necessary files and configurations.
Create a startup script to run both Fluentd and the Java app.

Build and run the Dockerfile

To build the Docker image, run the following command in your project root directory:
```
docker build -t log4j-axiom-test .
```

Run the container with the following:

docker run -p 24224:24224 log4j-axiom-test

This command starts the container, running both Fluentd and your Java app.

View logs in Axiom

Now that your app is running and sending logs to Axiom, you can view them in the Axiom dashboard. Log in to your Axiom account and go to the dataset you specified in the Fluentd configuration.

Logs appear in real-time, with various log levels and context information added.

Logging in Log4j best practices

Use appropriate log levels: Reserve ERROR and FATAL for serious issues, use WARN for potential problems, and INFO for general app flow.
Include context: Add relevant information to your logs using ThreadContext or by including important variables in your log messages.
Use structured logging: Log in JSON format to make it easier to parse, and later, analyze the logs using APL.
Log actionable information: Include enough detail in your logs to understand and potentially reproduce issues.
Use parameterized logging: Instead of string concatenation, use Log4j’s support for parameterized messages to improve performance.
Configure appenders appropriately: Use asynchronous appenders for better performance in high-throughput scenarios.
Regularly review and maintain your logs: Periodically check your logging configuration and the logs themselves to ensure they’re providing value.

Axiom 101

Send data to Axiom

Understand data

Process data

Management and billing

LLMs

Legal

Send logs from Apache Log4j to Axiom

Prerequisites

Configure Log4j

Set log level

Forward log messages to Fluentd

Configure the Fluentd.conf file

Create the Dockerfile

Build and run the Dockerfile

View logs in Axiom

Logging in Log4j best practices

Axiom 101

Send data to Axiom

Understand data

Process data

Management and billing

LLMs

Legal

​Prerequisites

​Configure Log4j

​Set log level

​Forward log messages to Fluentd

​Configure the Fluentd.conf file

​Create the Dockerfile

​Build and run the Dockerfile

​View logs in Axiom

​Logging in Log4j best practices

Prerequisites

Configure Log4j

Set log level

Forward log messages to Fluentd

Configure the Fluentd.conf file

Create the Dockerfile

Build and run the Dockerfile

View logs in Axiom

Logging in Log4j best practices