Java Exeutor framework's exception handling - java

I wrote an android app for killing background running processes which is performed in a background thread. And I use below class to create my own simple thread framework.
public final class ThreadPool {
private static final ExecutorService sES = Executors.newCachedThreadPool();
public static Future<?> runNow(Runnable task) {
return sES.submit(task);
}
}
However, a serious problem occured. That is the exception(unchecked exception) would be consumed by Executor framework quitely. So I don't know why the ActivityManager.killBackgroundProcesses() method does not work. After spending 2 or 3 hours and I wrote some log at almost every method invoke point, I found this method requires android.permission.KILL_BACKGROUND_PROCESSES permission, otherwise, it would throw a SecurityException that is an unchecked exception. The key is this exception is consumed by Excecutor framework, so I cannot see any exception information at logcat and the app does not crash at all and runs weird.
Of course, I don't know that at first, so I spent a lot of time to find out that reason, mainly depending on two posts:
Handling exceptions from Java ExecutorService tasks and
Catching thread exceptions from Java ExecutorService
So I changed the my ThreadPool class as:
public final class ThreadPool {
private static final ExecutorService sES = Executors.newCachedThreadPool();
/*
* submit(Runnable) and execute(Runnable) method has
* big difference. Especially in Exception handling!!!
* You have to pay attention.
*/
public static Future<?> submitNow(Runnable task) {
return sES.submit(task);
}
public static void executeNow(Runnable task) {
sES.execute(task);
}
}
But I still have below question:
Why Sun/Oracle decide to consume the exception instead of transfer to users to handle if submit(Runnable command) method is used?
How can I change this behavior to handle unchecked exception according to my own need if I insist using submit() method?
And my doubts are:
If submit(Runnable command) method is used, I know the exception result can be get by Future.get() method. But, if we use Future.get() method to judge if an exception occured, the thread Future object located in would be block. That is not what we expect in most case, I suppose.
I also learned Executor.execute() method handles exception like common Thread.start(). But there is no return value. So the task cannot be shut down at any time. Users have no ability to shut down any running thread by Future.cancel() method when leaving activity.

If you call get() on the future, you will get an ExecutionException if the underlying operation (callable) threw an exception. See the docs.
You can't change this behavior (from point 1. )
The reason why this is implemented this way is the following: submit is a non blocking call. The job gets posted in the executor and executed at a later time.
Only when the job is executed do you know if it crashed or not, so only when you try to access the result of the job do you get the exception.

Finally, I find a good solution.
We can extend Thread and invoke setUncaughtHandler() in the constructor like below.
public class MyThread1 extends Thread {
public MyThread1(Runnable task) {
super(task);
setUncaughtExceptionHandler(new UncaughtExceptionHandler() {
#Override
public void uncaughtException(Thread t, Throwable e) {
System.out.println("thread throws an uncaught exception at thread id: " + t.getId());
}
});
}
}
And then customize a ThreadFactory like following.
public class MyThreadFactory1 implements ThreadFactory {
#Override
public Thread newThread(Runnable r) {
return new MyThread1(r, "Peace");
}
}
So we can call the factory method in Executors like following.
ExecutorService es = Executors.newSingleThreadExecutor(new MyThreadFactory1());
So we can detect the uncaught exception happened in thread.

In a threaded environment, unchecked exceptions are known to be notorious and weird behaviors could occur like threads dying, no exception log etc.
One good way is to wrap the runnable object in a thread. Create a thread group and add the thread to the thread group.
final ThreadGroup group = new ThreadGroup("<a name for the thread group>");
public static Future<?> submitNow(Runnable task) {
//Create a thread wrapping the runnable task and specify the thread group
Thread t = new Thread(group,task);
return sES.submit(task);
}
The ThreadGroup class has its uncaughtException(Thread, Throwable) method which is automatically called by the JVM if a thread encountered an exception and is uncaught in your code. See http://developer.android.com/reference/java/lang/ThreadGroup.html
You can also change the behavior by creating your own ThreadGroup object and overriding the uncaughtException method:
public class MyThreadGroup extends ThreadGroup {
#Override
public void uncaughtException(Thread t, Throwable e) {
//do what you need to do to handle the exception
}
}
Or you can assign an UncaughtExceptionHandler to the current Thread.
public class MyUncaughtExceptionHandler implements UncaughtExceptionHandler {
#Override
public void uncaughtException(Thread t, Throwable t2) {
//Implement
}
}
Thread.currentThread().setUncaughtExceptionHandler(new MyUncaughtExceptionHandler());
Or set the default exception handler:
Thread.setDefaultUncaughtExceptionHandler(new MyUncaughtExceptionHandler());

Related

Block Java application from exiting until ThreadPool is empty

I've got an ExecutorService sitting inside a singleton class which receives tasks from many different classes. On application shutdown, I need to wait for the pool to be empty before I allow the application to exit.
private static NotificationService instance = null;
private ExecutorService executorService = Executors.newFixedThreadPool(25);
public static synchronized NotificationService getInstance() {
if (instance == null) {
instance = new NotificationService(true);
}
return instance;
}
While using this NotificationService, it frequently happens that I restart the application and the executorService hasn't finished processing all the notifications.
For Testing, I can manually shutdown the executorService and wait until all tasks are completed.
public static boolean canExit() throws InterruptedException {
NotificationService service = getInstance();
service.executorService.shutdown();
service.executorService.awaitTermination(30, TimeUnit.SECONDS);
return service.executorService.isTerminated();
}
Is it reliable and safe to override the finalize method and wait there until the pool is empty? From what I've read, finalize is not always called, especially not when using a singleton class.
#Override
protected void finalize() throws Throwable {
while (!canExit()){
Thread.sleep(100);
}
super.finalize();
}
This code is included in a library that will be included in another application, so there's no main method where I can wait until the pool is empty, unless I force the person using it to do so which is not great.
What is the correct way to stall the application (for a reasonable amount of time) from terminating until the pool is empty?
You can use addShutdownHook to catch the process termination event and wait for the pool there.
example:
Runtime.getRuntime().addShutdownHook(new Thread() {
public void run() {
NotificationService service = getInstance();
service.executorService.shutdown();
service.executorService.awaitTermination(30, TimeUnit.SECONDS);
}
});
Answered here: Java Finalize method call when close the application
Finalizers do not run on exit by default and the functionality to do this is deprecated.
One common advice is to use the Runtime.addShutdownHook but be aware of the following line of documentation:
Shutdown hooks should also finish their work quickly. When a program invokes exit the expectation is that the virtual machine will promptly shut down and exit. When the virtual machine is terminated due to user logoff or system shutdown the underlying operating system may only allow a fixed amount of time in which to shut down and exit. It is therefore inadvisable to attempt any user interaction or to perform a long-running computation in a shutdown hook.
In all honesty the best way to ensure everything gets properly cleaned up is to have your own application lifecycle which you can end before you even ask the VM to exit.
Don't use blocking shutdown hooks or anything similar in a library. You never know how the library is meant to be used. So it should always be up to the code that is using your library to take sensible actions on shut down.
Of course, you have to provide the necessary API for that, e.g. by adding lifecycle-methods to your class:
public class NotificationService {
...
public void start() {
...
}
/**
* Stops this notification service and waits until
* all notifications have been processed, or a timeout occurs.
* #return the list of unprocessed notification (in case of a timeout),
or an empty list.
*/
public List<Notification> stop(long timeout, TimeUnit unit) {
service.shutdown();
if (!service.awaitTermination(timeout, unit)) {
List<Runnable> tasks = service.shutdownNow();
return extractNotification(tasks);
}
return Collections.emptyList();
}
private List<Notification> extractNotification(List<Runnable> tasks) {
...
}
}
Then, the application code can take the required actions to handle your service, e.g.:
public static void main(String[] args) {
NotificationService service = new NotificationService(...);
service.start();
try {
// use service here
} finally {
List<Notification> pending = service.stop(30, TimeUnit.SECONDS);
if (!pending.isEmpty()) {
// timeout occured => handle pending notifications
}
}
}
Btw.: Avoid using singletons, if feasible.

How to be notified when thread goes down?

In java, I have ExecutorService that runs with while true, and Throwable catch clouse. I find out that from time to time the thread goes down. That means the system stop function.
So my question is, first of all, how can I catch the "thread killed" event (in order to send me email on such case)?
Also, how can this thread goes down?
the code is:
ExecutorService changesTrackerThread = Executors.newSingleThreadExecutor();
changesTrackerThread.submit(queueUpdater());
private Runnable queueUpdater() {
return new Runnable() {
#Override
public void run() {
while (true)
{
try
{
// do some code, then sleep
Thread.sleep(2000L);
} catch (Throwable t)
{
_log.error("something bad happened, but the loop should keep running", t);
}
}
}
};
Well first of all, why are you using a while loop here!?
You should use a scheduled executor:
ExecutorService changesTrackerThread = Executors.newSingleThreadScheduledExecutor()();
changesTrackerThread.scheduleAtFixedRate(new queueUpdater(), 0, 2, TimeUnit.SECONDS);
private Runnable queueUpdater() {
return new Runnable() {
#Override
public void run() {
try
{
// do some code
} catch (Throwable t)
{
_log.error("something bad happened", t);
}
}
};
I do not know why your thread dies, show us the full code.
But this way even if the thread dies the Excecutor will rerun it after the given period(2 seconds in this example.
As others have noted, you could replace your while (true) and sleep() loop with a ScheduledExecutorService. Scheduling a repeating task on such a service will return a ScheduledFuture which you can use to check the status of this task or to cancel it if you have a need for that. This will enable you to remove the try/catch block from the code.
Start the service like this:
ScheduledExecutorService svc = Executors.newScheduledThreadPool(1);
I would use newScheduledThreadPool() instead of newSingleThreadScheduledExecutor() since the former will restart threads if necessary.
Then, schedule the work like this:
void doSomeCode() {
// do some code
}
ScheduledFuture<?> sf = svc.scheduleAtFixedRate(this::doSomeCode, 0L, 2L, TimeUnit.SECONDS);
(Or if you wish you can inline doSomeCode() as a lambda or an anonymous inner class.)
Now what happens if the task fails with an exception? The ScheduledFuture object returned allows you to check status in a variety of ways. If you have a thread that you can dedicate to waiting for failures, you can have it call sf.get() which will throw an ExecutionException that wraps the exception that caused the task to fail. Otherwise, it blocks indefinitely. ScheduledFuture.get() is a bit weird in that unlike an ordinary Future.get() call, it never returns a value; it always throws an exception.
When/if the task fails, the caller of sf.get() can log the exception and resubmit the task, or whatever. If you don't want to block a thread indefinitely, you can poll for failure using sf.isDone() or sf.get(0L, TimeUnit.SECONDS). Note that both overloads of sf.get() communicate all of their return information via the type of a thrown exception, which may make them somewhat inconvenient to use.
You could put exception handling within the task itself, catching Throwable and continuing no matter what, and this will probably work. It does bake the logging/restart/resubmission policy into the task itself, which may be unpleasant. Using ScheduledFuture lets you separate these policies from the actual work performed by the task.

Exception handling in thread

We are working on a project to handle virtual machine managers from a Java application.
At the moment, we faced a problem where some instructions must be considered as a task because they require more time. Theses taks can also end with errors or be a success, and we can't know the result from the application unless we request the state of the task to the VM hypervisor.
To have a smooth application we wanted to have a CommandManager to handle the differents request to theses hypervisors in separated threads. The problem is that these Command might return specific errors such as that we used to catch in the view to display revelant information to the user but when we implement Runnableon our Commands interface, we can't throw any exception out of that thread back to reach the view.
What could I do to keep notifying the user of the errors that happened and their nature?
Sorry for my english!
Lets see the code below for a brief example :
First the Command
class ChangeMemorySize extends Command {
private String name;
private int memorySizeMb;
public ChangeMemorySize(Hypervisor hypervisor, String server,
String name, int memory) {
super(hypervisor, server);
this.name = name;
this.memorySizeMb = memory;
}
protected void execute() throws ConnectionException,OperationException{
//Some code here
}
public void run() //CANT THROW ANYTHING HERE :throws VmConfigFault,
try{
execute();
}catch{Exception e){
// I have to catch it all here!
}
This isn't real code this is just an example. This command would then be passed to a manager who would run it in another thread. But I loose all the specifics exception that I used to notify the user of what really happened in there!
Use an ExecutorService and do one of two things.
First you can store all results in a Future and when you want to know if an Exception occurs just invoke Future.get() and handle any exception.
Second you can create the ExecutorService with a ThreadFactory in which you set it a knew UncaughtExceptionHandler like
final UncaughtExceptionHandler handler = new UncaughtExceptionHandler() {
public void uncaughtException(Thread t, Throwable e) {
// notify error
}
});
Executor executor = Executors.newFixedThreadPool(1, new ThreadFactory() {
public Thread newThread(Runnable r) {
Thread thread = new Thread(r);
thread.setUncaughtExceptionHandler(handler);
return thread;
}
});
While I agree with #John Vint that ExecutorService with an UncaughtExceptionHandler is a good approach, if for some reason you really want to stick with Runnables, I'd suggest that your base class, Command, have some fields/getters for status and Exception. e.g.
Enum Status {FAILED, CANCELLED, SUCCESS};
// protected so subclasses can set them
protected Status status;
protected Exception exception; // null means none
public Status getStatus();
public Exception getException();
You might even want to add a result field to really mimic a Callable.
For simplicity, then add a method that subclasses can call in the catch clause
protected void s***Happenned(Exception e) {
this.exception = e;
status = FAILED;
// any standard cleanup can go here too...
}

Catching exceptions from another threads?

I am writing an app that will make use of multiple threads. There is main thread that is launching another threads. What i want to acomplish is when one of the launched threads throws an exception, the main thread should stop launching threads. It looks more or less like this:
class SomeClass {
boolean launchNewThread = true;
public static void main() {
while (launchNewThread) {
try {
AnotherClass.run();
} catch (CrossThreadException e) {
launchNewThread = false;
}
}
}
}
class AnotherClass implements Runnable {
public void run() {
if (a=0) throw new CrossThreadException();
}
}
You should do it yourself - catch the exception and pass it somehow into the launching thread.
Also, there is Future concept, which does it already. You should launch your threads as futures and check isDone(), and catch ExecutionException from get(), this exception will be thrown if a future's task thrown an exception.
You can also use a listener as described in How to throw a checked exception from a java thread?
When an exception is thrown inside one of the child threads, you could call a method like listener.setLaunchNewThread(false) from the child thread which will change the value of your boolean flag in the parent thread.
On a side note, calling AnotherClass.run() does not start a new thread but only call the run method from AnotherClass within the same thread. Use new Thread(new AnotherClass()).start() instead.

How do I test this method's expected behavior in Java: it spawn a thread, and throws an exception under certain conditions

Suppose that I have a method which spawns a new thread and do some work. Under certain conditions, the newly spawn thread would throw a certain type of exception, which terminates the entire process. I would like to write JUnit tests to verify this behavior. Is there a way to do it?
The method is:
private void foo() {
new Thread() {
#Override void run() {
throw new CertainException("exception messages");
}
}.start();
}
In test (conceptually):
public testExceptionThrownFromNewThread() throws Exception {
try {
foo();
Thread.sleep(5000); // wait for the exception to be thrown
fail();
} catch (CertainException e) {
assertEquals(e.message, "exception messages");
}
}
This test doesn't work because the exception spawn from the other thread cannot be caught.
If you want to test just the code inside of the run() method, refactor it ouf of the foo() method (probably into a Runnable) and test it separately without running it from a thread.
private void foo() {
new Thread(new MyRunnable()).start();
}
public class MyRunnable implements Runnable {
public void run() {
....
}
}
Now you can instantiate a MyRunnable object and call the run() method from your test without needing to start a thread.
EDIT
Testing of the thread creation could be done by using a ThreadFactory Mock. (as Jon Skeet pointed out).
You could overwrite the default UncaughtExceptionHandler for Threads. It gets called whenever a Thread throws an exception. In this handler, you can check whether the expected exception is equal to the thrown exception and e.g. test for messages or count the occurences of the exception. By using a CountDownLatch, you can also check whether the exceptions are thrown in time and how many of them you expect.
This works even if you do not have access to the Thread created by the class under test. If you have access to it though, there is certainly an easier approach, e.g. refactoring the class under test and introduce an Exception Listener or alike. Make the class under test better testable also improves the design, e.g. by removing the dependency on Threads and directly test the body of the run() method which you could externalize.
public class ThreadExceptionTest {
private void foo() {
new Thread(new Runnable() {
#Override
public void run() {
throw new RuntimeException("exception messages");
}
}).start();
}
#Test
public void testFoo() throws Exception {
final CountDownLatch latch = new CountDownLatch(1);
final RuntimeException expectedException = new RuntimeException("exception messages");
UncaughtExceptionHandler eh = new UncaughtExceptionHandler() {
#Override
public void uncaughtException(Thread t, Throwable e) {
if (e.getMessage().equals(expectedException.getMessage()))
latch.countDown();
}
};
Thread.setDefaultUncaughtExceptionHandler(eh);
foo();
assertTrue(latch.await(100,TimeUnit.MILLISECONDS));
}
}
Well, unit tests are supposed to verify results of method calls, not implementation details.
In your library, if thread terminates, how does it affect library user? Maybe computations won't be finished and end results won't be recored in database? Then check database. Maybe thread will stop doing some periodic tasks (like cleanup)? Then check whether cleanup is still being done.
And if exception thrown won't affect user in any way, then there's nothing to check. Because whether exception is thrown or not is just an implementation details (user will never see it).
One option is to make the capability to start a thread a dependency - which you can specify using the existing ThreadFactory interface. Then in your unit test you can provide a specialist ThreadFactory which wraps the given Runnable in order to record exceptions etc.
You'll be able to test that:
The ThreadFactory was used
The thread was started
The operation threw an exception

Categories