Exploring the Command 'jstack' (with examples)

The jstack command is a powerful debugging tool used to print Java stack traces, allowing developers and system administrators to diagnose issues related to thread activity in Java applications. This command is essential when analyzing performance bottlenecks or unexpected behavior in Java processes by providing insights into the execution threads. Utilizing jstack, you can effortlessly obtain a snapshot of what every thread in a Java process is doing, which aids in understanding the overall state of the application at a given point in time.

Print Java stack traces for all threads in a Java process

Code:

jstack java_pid

Motivation:

When running a Java application, you may encounter performance issues or unexpected behavior. At such times, it’s crucial to understand what individual threads in your Java process are doing. Using jstack in this context allows you to capture and examine the stack trace of all threads, helping you to identify threads that are consuming CPU time, waiting on locks, or blocked.

Explanation:

• jstack: The command itself, used to obtain stack traces.
• java_pid: The process identifier (PID) of the Java process whose stack traces you want to examine. It is a unique number assigned to each running process in the system. To find the PID, you can use commands like ps or jps.

Example output:

Thread 1:
   at java.lang.Object.wait(Native Method)
   at java.util.TimerThread.mainLoop(Timer.java:552)
   at java.util.TimerThread.run(Timer.java:505)

Thread 2:
   at java.net.PlainSocketImpl.socketAccept(Native Method)
   at java.net.AbstractPlainSocketImpl.accept(AbstractPlainSocketImpl.java:409)
   ...

In this output, you can see the call stack for each thread, including the method being executed and line numbers, which can help diagnose performance or deadlocking issues.

Print mixed mode (Java/C++) stack traces for all threads in a Java process

Code:

jstack -m java_pid

Motivation:

Sometimes, the performance bottlenecks in a Java application aren’t within the Java code itself but occur in the native C++ libraries that Java code calls. In such cases, understanding both Java and native stack traces is essential. The mixed mode stack traces include native C++ frames in addition to the Java frames, giving you a complete picture of what each thread is doing, which is crucial for diagnosing JNI-related issues or performance bottlenecks involving native libraries.

Explanation:

• jstack: The core command used to print stack traces.
• -m: The flag that tells jstack to include mixed mode (Java and C++) frames. It provides insights into both the Java layer and the native system calls or library functions being executed.
• java_pid: The Java process identifier you’re diagnosing.

Example output:

Thread 1:
   at java.lang.Object.wait(Native Method)
   at java.util.TimerThread.mainLoop(Timer.java:558)
   at java.util.TimerThread.run(Timer.java:505)
   ...
   Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
   J  com/example/MyClass.myMethod()V
   C  [libnative.so+0x4e10]  nativeMethod

Thread 2:
   at java.net.PlainSocketImpl.socketAccept(Native Method)
   at java.net.AbstractPlainSocketImpl.accept(AbstractPlainSocketImpl.java:409)
   ...

Here, we see both Java and native stack traces, aiding in a more comprehensive understanding of thread execution paths.

Print stack traces from Java core dump

Code:

jstack /usr/bin/java file.core

Motivation:

A core dump is created when a Java application crashes, providing a snapshot of the process’s memory at the time of the crash. Analyzing this core dump is critical when diagnosing the root cause of crashes or unexpected termination. The jstack command allows you to extract stack traces from a core dump file, which can be especially useful when dealing with production environments where live debugging is not an option.

Explanation:

• jstack: The command for generating stack traces.
• /usr/bin/java: The path to the Java executable that created the core dump. This is necessary to interpret the core dump file correctly.
• file.core: The core dump file containing the memory snapshot of the crashed Java process. This file helps in recreating the state of the process just before it crashed.

Example output:

2023-08-01 12:30:45
Full thread dump Java HotSpot(TM) 64-Bit Server VM (build 25.301-b09, mixed mode):

...

"main" #1 prio=5 os_prio=0 tid=0x00007fdcc000b000 nid=0xabc runnable [0x00007fdcc1459000]
   java.lang.Thread.State: RUNNABLE
	at com.example.Main.main(Main.java:23)
	- locked <0x00000000c182ed50> (a java.lang.Object)
	at java.lang.Thread.run(Thread.java:748)

...

This output enables you to investigate the state of each thread at the moment the core dump was taken, aiding in pinpointing the cause of application failure.

Conclusion:

The jstack command is an invaluable tool for diagnosing and debugging Java applications by providing insightful traces of thread activity. Whether it’s printing Java stack traces, exploring mixed mode traces that involve native code, or analyzing core dumps after a crash, jstack equips developers with the necessary data to refine performance, troubleshoot deadlocks, and maintain the smooth operation of their Java programs.