Java Runtime Reference¶

This guide provides technical details on the Java runtime environment and optimization strategies for running Spring Boot applications on Azure Container Apps.

Runtime Environment¶

Azure Container Apps uses the Eclipse Temurin distribution for the OpenJDK runtime. The reference application uses Java 21 (LTS).

graph TD
    JVM[JVM Runtime] --> GC[Garbage Collector]
    JVM --> MEM[Memory Management]
    JVM --> LOGS[Structured Logging]
    GC --> G1GC[G1 Garbage Collector]
    MEM --> CGROUPS[Cgroups v2]
    LOGS --> LOGBACK[Logback JSON]

JVM Memory Management¶

Java applications on Azure Container Apps are aware of the container's memory limits (Cgroups v2). The JVM automatically adjusts its heap size based on the available container memory.

1. Default Memory Settings¶

By default, the JVM sets the maximum heap size to approximately 25% of the container's memory limit.

Container Memory	Default Max Heap (approx.)
512 MiB	128 MiB
1 GiB	256 MiB
2 GiB	512 MiB

2. Customizing Memory Limits¶

You can override the default JVM behavior using the JAVA_TOOL_OPTIONS environment variable.

az containerapp update \
  --resource-group $RG \
  --name $APP_NAME \
  --set-env-vars "JAVA_TOOL_OPTIONS=-Xms512m -Xmx1024m"

Multi-stage Docker Build Optimization¶

To minimize the container image size and improve deployment speed, use a multi-stage Docker build.

1. Build Stage¶

Uses a full JDK and Maven to compile the application.

FROM docker.io/library/maven:3.9-eclipse-temurin-21 AS build
WORKDIR /app
COPY pom.xml .
RUN mvn dependency:go-offline -B
COPY src ./src
RUN mvn package -DskipTests -B

2. Runtime Stage¶

Uses a lightweight JRE (Java Runtime Environment) for the final image.

FROM docker.io/library/eclipse-temurin:21-jre-alpine
WORKDIR /app
COPY --from=build /app/target/*.jar app.jar
EXPOSE 8000
ENTRYPOINT ["java", "-jar", "app.jar"]

Spring Boot Best Practices¶

Spring Profiles: Use SPRING_PROFILES_ACTIVE=prod for production workloads.
Actuator Endpoints: Enable health and info endpoints for platform integration.
Graceful Shutdown: Enable server.shutdown=graceful in application.yml to handle SIGTERM signals.
Logback JSON: Use the LogstashLogbackEncoder for structured, parseable logs in Log Analytics.

Runtime Verification¶

You can verify the runtime details by calling the /info endpoint of the reference application.

curl https://$FQDN/info

Expected output

{
  "runtime": {
    "vendor": "Eclipse Adoptium",
    "java": "21.0.10"
  },
  "app": "azure-container-apps-java-guide",
  "version": "1.0.0"
}

Runtime Checklist¶

[x] Application uses a supported LTS version of Java (17 or 21)
[x] Dockerfile uses multi-stage builds to minimize image size
[x] JVM is aware of container memory limits (Cgroups v2)
[x] Graceful shutdown is enabled for zero-downtime releases
[x] Structured logging is enabled for better observability

Choosing the right memory limit

Spring Boot applications typically require at least 512 MiB of container memory to run efficiently. For production workloads, 1 GiB or more is recommended to avoid OutOfMemory (OOM) errors during startup or under load.