Closed GoogleCodeExporter closed 9 years ago
I am sorry,what's your meaning?
Original comment by killme2...@gmail.com
on 26 Aug 2012 at 7:40
for example the following:
public static void main(String ...strings) throws TimeoutException, InterruptedException, MemcachedException, IOException {
MemcachedClientBuilder builder = new XMemcachedClientBuilder(
AddrUtil.getAddresses("localhost:11211"));
builder.setCommandFactory(new BinaryCommandFactory());//use binary protocol
builder.setConnectionPoolSize(5);
//set the hashing alg.
builder.setSessionLocator(new KetamaMemcachedSessionLocator());
MemcachedClient client = builder.build();
client.get("example");
System.out.println("DONE, process should exit now, but doesn't");
}
This program will never exit due to the xmemcached threads not being daemonized.
Original comment by dust...@gmail.com
on 27 Aug 2012 at 8:25
@dustinn
You must call shutdown method to stop the client.
Original comment by killme2...@gmail.com
on 28 Aug 2012 at 5:33
Yes, I know. Problem is I don't always have control of where to shutdown the
xmemcached client. This really needs a daemon option for the threadpool.
found this in WorkerThreadFactory
if (t.isDaemon()) {
t.setDaemon(false);
}
Any way you can make this a config option?
Original comment by dust...@gmail.com
on 28 Aug 2012 at 5:33
Experiencing the same issue where our jobs don't exit after completion with the
use of xmemcached client.
We don't have access to the client either as it's created by another
intermediary library.
Setting threads as a daemon won't hurt your functionality. It will just let the
JVM know that those threads are supporting threads and not main functionality,
so when the main application threads are done, the JVM can ignore them and
safely exit.
Many other supporting threads such as "MySQL Statement Cancellation Timer" and
"Finalizer" are daemons.
Xmemcached-Reactors and xmemcached-read-thread should be as well.
You can register Runtime.addShutdownHook if you'd like to do any custom cleanup
on JVM exit.
It should be a safe change that we'd welcome!
Thank you for your consideration.
Original comment by cyril.bo...@koinzmedia.com
on 2 May 2013 at 5:10
1.4.2 released.It was fixed.
Original comment by killme2...@gmail.com
on 19 Jul 2013 at 6:49
Original issue reported on code.google.com by
dust...@gmail.com
on 22 Aug 2012 at 5:42