Develop Reference

Ensure that java thread is really suspended

I have following class:
public class PawnThread implements Runnable {
public void start() {
thread.start();
}
#Override
public void run() {
try {
while (... some finish condition ...) {
move();
synchronized (this) {
while (suspendFlag) {
wait();
}
}
}
} catch (InterruptedException e) {
System.err.println(pawn.toString() + ": thread interrupted :(");
}
}
void move() {
... some blocking actions
}
synchronized void suspend() {
suspendFlag = true;
}
synchronized void resume() {
suspendFlag = false;
notify();
}
}
Now I have a list of its objects:
private final List<PawnThread> pawnThreadList;
I defined some helper method to suspend all of them:
public void suspendAll() {
pawnThreadList.forEach(PawnThread::suspend);
}
Now suspend() method is only about changing flag. The requirement is, that when I leave suspendAll() method, all threads should be actually paused (they cannot be in RUNNABLE state) - for now it is not a case, beacause for some of them, it may take some time to actually finish their job before pause.
I would be grateful for advice what is correct design for this soulution.
Regards

Make PawnThread#suspend() wait for suspension to be completed:
public class PawnThread implements Runnable {
private final Waiter suspender = new Waiter();
private final Waiter suspending = new Waiter();
#Override
public void run() {
try {
while (...) {
suspending.suspend();
move();
suspending.resume();
suspender.await();
}
} catch (InterruptedException e) {
...
}
}
void suspend() throws InterruptedException {
suspender.suspend();
suspending.await();
}
void resume() {
suspender.resume();
}
}
public class Waiter {
private boolean waiting;
public synchronized void await() throws InterruptedException {
while (waiting) {
wait();
}
}
public synchronized void suspend() {
waiting = true;
}
public synchronized void resume() {
waiting = false;
notify();
}
}

The requirement is impossible to satisfy, but also makes no sense. In order for the thread to communicate the fact that it has suspended, the thread must be running. There is no way to ensure the thread has completed the suspension process.
But this is also not a sensible requirement. How can it possibly matter whether the thread has suspended itself or is about to suspend itself, so long as it has nothing left to do but suspend itself?
A sensible requirement should be satisfied by having each thread set some indication somewhere that it has received the suspend request and is about to stop executing. Then the calling thread can wait for all threads to have provided that indication.

Universal correct design for any parallel solution is to define streams of tokens and firing rule (see Petry Net tedminology). Most simple and useful firing rule is to start an action when all input tokens are ready. I your case, input tokens are hidden in whle condition and in suspend condition. Your mistake is you defined suspend condition as negative, while all tokens must be defined as positive. That is, a thread works where there are enough tokens, and stops when they are exhausted, and then thread waits while the number of tokens is increased by external threads.
Tokens may be of 2 kinds - black (pure permissions), passed by Semaphores, and color (messages), passed by BlockingQueues. These 2 communicator classes cover most of use cases. In some complex cases, user can create custom communicators using synchronized/wait/notify.
So canonical way to design any parallel program is as follows:
design Petry Net, with places for tokens (communicators), and transitions (actions).
map places to Semaphores/BlockingQueues/CustomCommunicators, and transition to threads (or Actors).

Interrupting unknown thread

Consider the following (simplified) class, designed to allow my entire component to enter some interim state before completely stopping. (The purpose of the interim state is to allow the component to complete its existing tasks, but reject any new ones).
The component might be started and stopped multiple times from any number of threads.
class StopHandler {
boolean isStarted = false;
synchronized void start() {isStarted = true;}
//synchronized as I do want the client code to block until the component is stopped.
//I might add some async method as well, but let's concentrate on the sync version only.
synchronized void stop(boolean isUrgent) {
if (isStarted) {
if (!isUrgent) {
setGlobalState(PREPARING_TO_STOP); //assume it is implemented
try {Thread.sleep(10_000L);} catch (InterruptedException ignored) {}
}
isStarted = false;
}
}
The problem with the current implementation is that if some client code needs to urgently stop the component while it is in the interim state, it will still have to wait.
For example:
//one thread
stopHandler.stop(false); //not urgent => it is sleeping
//another thread, after 1 millisecond:
stopHandler.stop(true); //it's urgent, "please stop now", but it will wait for 10 seconds
How would you implement it?
I might need to interrupt the sleeping thread, but I don't have the sleeping thread object on which to call 'interrupt()'.

How about storing a reference to current Thread (returned by Thread.currentThread()) in a field of StopHandler directly before you call sleep? That would allow you you to interrupt it in the subsequent urgent call in case the Thread is still alive.

Couldn't find a better solution than the one suggested by Lars.
Just need to encapsulate the sleep management for completeness.
class SleepHandler {
private final ReentrantLock sleepingThreadLock;
private volatile Thread sleepingThread;
SleepHandler() {
sleepingThreadLock = new ReentrantLock();
}
void sleep(long millis) throws InterruptedException {
setSleepingThread(Thread.currentThread());
Thread.sleep(millis);
setSleepingThread(null);
}
void interruptIfSleeping() {
doWithinSleepingThreadLock(() -> {
if (sleepingThread != null) {
sleepingThread.interrupt();
}
});
}
private void setSleepingThread(#Nullable Thread sleepingThread) {
doWithinSleepingThreadLock(() -> this.sleepingThread = sleepingThread);
}
private void doWithinSleepingThreadLock(Runnable runnable) {
sleepingThreadLock.lock();
try {
runnable.run();
} finally {
sleepingThreadLock.unlock();
}
}
}
With this helper class, handling of the original problem is trivial:
void stop(boolean isUrgent) throws InterruptedException {
if (isUrgent) {sleepHandler.interruptIfSleeping();} //harmless if not sleeping
try {
doStop(isUrgent); //all the stuff in the original 'stop(...)' method
} catch (InteruptedException ignored) {
} finally {
Thread.interrupted(); //just in case, clearing the 'interrupt' flag as no need to propagate it futher
}

Alternative method to kill thread

I have been looking for ways to kill a thread and it appears this is the most popular approach
public class UsingFlagToShutdownThread extends Thread {
private boolean running = true;
public void run() {
while (running) {
System.out.print(".");
System.out.flush();
try {
Thread.sleep(1000);
} catch (InterruptedException ex) {}
}
System.out.println("Shutting down thread");
}
public void shutdown() {
running = false;
}
public static void main(String[] args)
throws InterruptedException {
UsingFlagToShutdownThread t = new UsingFlagToShutdownThread();
t.start();
Thread.sleep(5000);
t.shutdown();
}
}
However, if in the while loop we spawn another another object which gets populated with data (say a gui that is running and updating) then how do we call back - especially considering this method might have been called several times so we have many threads with while (running) then changing the flag for one would change it for everyone?
thanks

One approach with these problems is to have a Monitor class which handles all the threads. It can start all necessary threads (possibly at different times/when necessary) and once you want to shutdown you can call a shutdown method there which interrupt all (or some) of the threads.
Also, actually calling a Threads interrupt() method is generally a nicer approach as then it will get out of blocking actions that throw InterruptedException (wait/sleep for example). Then it will set a flag that is already there in Threads (which can be checked with isInterrupted() or checked and cleared with interrupted(). For example the following code can replace your current code:
public class UsingFlagToShutdownThread extends Thread {
public void run() {
while (!isInterrupted()) {
System.out.print(".");
System.out.flush();
try {
Thread.sleep(1000);
} catch (InterruptedException ex) { interrupt(); }
}
System.out.println("Shutting down thread");
}
public static void main(String[] args)
throws InterruptedException {
UsingFlagToShutdownThread t = new UsingFlagToShutdownThread();
t.start();
Thread.sleep(5000);
t.interrupt();
}
}

i added a utlility class which essentially had a static map and methods.
the map was of type Long id, Thread thread. I added two methods one to add to the map and one to stop the thread via the use of interrupt. This method took the id as a parameter.
I also changed my loop logic from while true, too while ! isInterrupted. Is this approach ok or is this bad programming style/convention
thanks

Using boolean var for stopping threads

I have a Java book I'm learning from and in one of the examples, I saw something suspicious.
public class ThreadExample extends MIDlet {
boolean threadsRunning = true; // Flag stopping the threads
ThreadTest thr1;
ThreadTest thr2;
private class ThreadTest extends Thread {
int loops;
public ThreadTest(int waitingTime) {
loops = waitTime;
}
public void run() {
for (int i = 1; i <= loops; i++) {
if (threadsRunning != true) { // here threadsRunning is tested
return;
}
try {
Thread.sleep(1000);
} catch(InterruptedException e) {
System.out.println(e);
}
}
}
}
public ThreadExample() {
thr1 = new ThreadTest(2);
thr2 = new ThreadTest(6);
}
public void startApp() throws MIDletStateChangeException {
thr1.start();
thr2.start();
try {
Thread.sleep(4000); // we wait 4 secs before stopping the threads -
// this way one of the threads is supposed to finish by itself
} catch(InterruptedException e) {
System.out.println(e);
}
destroyApp();
}
public void destroyApp() {
threadsRunning = false;
try {
thr1.join();
thr2.join();
} catch(InterruptedException e) {
System.out.println(e);
}
notifyDestroyed();
}
}
As it is a MIDlet app, when it's started, the startApp method is executed. To keep it simple, the startApp method itself calls destroyApp and so the program destroys, stopping the threads and notifying the destruction.
The question is, is it safe to use this 'threadsRunning' variable and would its use inside both threads and in the destroyApp method cause any trouble at some point? Would 'volatile' keyword put in front of the declaration help to synchronize it?

Setting a boolean value is atomic, and there is no "read then modify" logic in this example, so access to the variable doesn't need to be synchronised in this particular case.
However, the variable should at least be marked volatile.
Marking the variable volatile does not synchronise the threads' access to it; it makes sure that a thread doesn't miss another thread's update to the variable due to code optimisation or value caching. For example, without volatile, the code inside run() may read the threadsRunning value just once at the beginning, cache the value, and then use this cached value in the if statement every time, rather than reading the variable again from main memory. If the threadsRunning value gets changed by another thread, it might not get picked up.
In general, if you use a variable from multiple threads, and its access is not synchronised, you should mark it volatile.

Wait until any of Future<T> is done

I have few asynchronous tasks running and I need to wait until at least one of them is finished (in the future probably I'll need to wait util M out of N tasks are finished).
Currently they are presented as Future, so I need something like
/**
* Blocks current thread until one of specified futures is done and returns it.
*/
public static <T> Future<T> waitForAny(Collection<Future<T>> futures)
throws AllFuturesFailedException
Is there anything like this? Or anything similar, not necessary for Future. Currently I loop through collection of futures, check if one is finished, then sleep for some time and check again. This looks like not the best solution, because if I sleep for long period then unwanted delay is added, if I sleep for short period then it can affect performance.
I could try using
new CountDownLatch(1)
and decrease countdown when task is complete and do
countdown.await()
, but I found it possible only if I control Future creation. It is possible, but requires system redesign, because currently logic of tasks creation (sending Callable to ExecutorService) is separated from decision to wait for which Future. I could also override
<T> RunnableFuture<T> AbstractExecutorService.newTaskFor(Callable<T> callable)
and create custom implementation of RunnableFuture with ability to attach listener to be notified when task is finished, then attach such listener to needed tasks and use CountDownLatch, but that means I have to override newTaskFor for every ExecutorService I use - and potentially there will be implementation which do not extend AbstractExecutorService. I could also try wrapping given ExecutorService for same purpose, but then I have to decorate all methods producing Futures.
All these solutions may work but seem very unnatural. It looks like I'm missing something simple, like
WaitHandle.WaitAny(WaitHandle[] waitHandles)
in c#. Are there any well known solutions for such kind of problem?
UPDATE:
Originally I did not have access to Future creation at all, so there were no elegant solution. After redesigning system I got access to Future creation and was able to add countDownLatch.countdown() to execution process, then I can countDownLatch.await() and everything works fine.
Thanks for other answers, I did not know about ExecutorCompletionService and it indeed can be helpful in similar tasks, but in this particular case it could not be used because some Futures are created without any executor - actual task is sent to another server via network, completes remotely and completion notification is received.

simple, check out ExecutorCompletionService.

ExecutorService.invokeAny

Why not just create a results queue and wait on the queue? Or more simply, use a CompletionService since that's what it is: an ExecutorService + result queue.

This is actually pretty easy with wait() and notifyAll().
First, define a lock object. (You can use any class for this, but I like to be explicit):
package com.javadude.sample;
public class Lock {}
Next, define your worker thread. He must notify that lock object when he's finished with his processing. Note that the notify must be in a synchronized block locking on the lock object.
package com.javadude.sample;
public class Worker extends Thread {
private Lock lock_;
private long timeToSleep_;
private String name_;
public Worker(Lock lock, String name, long timeToSleep) {
lock_ = lock;
timeToSleep_ = timeToSleep;
name_ = name;
}
#Override
public void run() {
// do real work -- using a sleep here to simulate work
try {
sleep(timeToSleep_);
} catch (InterruptedException e) {
interrupt();
}
System.out.println(name_ + " is done... notifying");
// notify whoever is waiting, in this case, the client
synchronized (lock_) {
lock_.notify();
}
}
}
Finally, you can write your client:
package com.javadude.sample;
public class Client {
public static void main(String[] args) {
Lock lock = new Lock();
Worker worker1 = new Worker(lock, "worker1", 15000);
Worker worker2 = new Worker(lock, "worker2", 10000);
Worker worker3 = new Worker(lock, "worker3", 5000);
Worker worker4 = new Worker(lock, "worker4", 20000);
boolean started = false;
int numNotifies = 0;
while (true) {
synchronized (lock) {
try {
if (!started) {
// need to do the start here so we grab the lock, just
// in case one of the threads is fast -- if we had done the
// starts outside the synchronized block, a fast thread could
// get to its notification *before* the client is waiting for it
worker1.start();
worker2.start();
worker3.start();
worker4.start();
started = true;
}
lock.wait();
} catch (InterruptedException e) {
break;
}
numNotifies++;
if (numNotifies == 4) {
break;
}
System.out.println("Notified!");
}
}
System.out.println("Everyone has notified me... I'm done");
}
}

As far as I know, Java has no analogous structure to the WaitHandle.WaitAny method.
It seems to me that this could be achieved through a "WaitableFuture" decorator:
public WaitableFuture<T>
extends Future<T>
{
private CountDownLatch countDownLatch;
WaitableFuture(CountDownLatch countDownLatch)
{
super();
this.countDownLatch = countDownLatch;
}
void doTask()
{
super.doTask();
this.countDownLatch.countDown();
}
}
Though this would only work if it can be inserted before the execution code, since otherwise the execution code would not have the new doTask() method. But I really see no way of doing this without polling if you cannot somehow gain control of the Future object before execution.
Or if the future always runs in its own thread, and you can somehow get that thread. Then you could spawn a new thread to join each other thread, then handle the waiting mechanism after the join returns... This would be really ugly and would induce a lot of overhead though. And if some Future objects don't finish, you could have a lot of blocked threads depending on dead threads. If you're not careful, this could leak memory and system resources.
/**
* Extremely ugly way of implementing WaitHandle.WaitAny for Thread.Join().
*/
public static joinAny(Collection<Thread> threads, int numberToWaitFor)
{
CountDownLatch countDownLatch = new CountDownLatch(numberToWaitFor);
foreach(Thread thread in threads)
{
(new Thread(new JoinThreadHelper(thread, countDownLatch))).start();
}
countDownLatch.await();
}
class JoinThreadHelper
implements Runnable
{
Thread thread;
CountDownLatch countDownLatch;
JoinThreadHelper(Thread thread, CountDownLatch countDownLatch)
{
this.thread = thread;
this.countDownLatch = countDownLatch;
}
void run()
{
this.thread.join();
this.countDownLatch.countDown();
}
}

If you can use CompletableFutures instead then there is CompletableFuture.anyOf that does what you want, just call join on the result:
CompletableFuture.anyOf(futures).join()
You can use CompletableFutures with executors by calling the CompletableFuture.supplyAsync or runAsync methods.

Since you don't care which one finishes, why not just have a single WaitHandle for all threads and wait on that? Whichever one finishes first can set the handle.

See this option:
public class WaitForAnyRedux {
private static final int POOL_SIZE = 10;
public static <T> T waitForAny(Collection<T> collection) throws InterruptedException, ExecutionException {
List<Callable<T>> callables = new ArrayList<Callable<T>>();
for (final T t : collection) {
Callable<T> callable = Executors.callable(new Thread() {
#Override
public void run() {
synchronized (t) {
try {
t.wait();
} catch (InterruptedException e) {
}
}
}
}, t);
callables.add(callable);
}
BlockingQueue<Runnable> queue = new ArrayBlockingQueue<Runnable>(POOL_SIZE);
ExecutorService executorService = new ThreadPoolExecutor(POOL_SIZE, POOL_SIZE, 0, TimeUnit.SECONDS, queue);
return executorService.invokeAny(callables);
}
static public void main(String[] args) throws InterruptedException, ExecutionException {
final List<Integer> integers = new ArrayList<Integer>();
for (int i = 0; i < POOL_SIZE; i++) {
integers.add(i);
}
(new Thread() {
public void run() {
Integer notified = null;
try {
notified = waitForAny(integers);
} catch (InterruptedException e) {
e.printStackTrace();
} catch (ExecutionException e) {
e.printStackTrace();
}
System.out.println("notified=" + notified);
}
}).start();
synchronized (integers) {
integers.wait(3000);
}
Integer randomInt = integers.get((new Random()).nextInt(POOL_SIZE));
System.out.println("Waking up " + randomInt);
synchronized (randomInt) {
randomInt.notify();
}
}
}