[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #25 from Mark Thomas  ---
This has been fixed in 9.0.x although the connector refactoring means that a
direct back-port may not be possible.

Note that two tests were removed because they were invalid as per section 2.1.5
of the WebSocket 1.1 specification.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

Mark Thomas  changed:

   What|Removed |Added

 Resolution|--- |FIXED
 Status|NEW |RESOLVED

--- Comment #26 from Mark Thomas  ---
This issue identified in comment #22 did apply to 8.0.x and 7.0.x along with
the issue identified in 9.0.x.

The back-ports have been completed and this fix will be included in 9.0.0.M2,
8.0.31 and 7.0.68 onwards.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #22 from Barry Coughlan  ---
You've probably got this far already, but it seems that the cause for all of
the test failures is this block in AbstractServletInputStream.java which
swallows the exception:

try {
ready = Boolean.valueOf(doIsReady());
} catch (IOException e) {
onError(e);
}

Throwing the exception in the catch block causes the tests to pass (although
some awaitOnError() statements need to be modified).

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #23 from Remy Maucherat  ---
This looks like what it should do however: call onError. Rethrowing in that
method is not allowed.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #24 from Barry Coughlan  ---
(In reply to Remy Maucherat from comment #23)
> This looks like what it should do however: call onError. Rethrowing in that
> method is not allowed.

Sorry for not being clear - I wasn't suggesting that throwing the exception was
a good fix (I'm not very familiar with the code), I was just using it to
demonstrate that the issue lies there. Other read errors that occur in
WsFrameServer.onDataAvailable() throw up the IOException so that
onError/onClose is called, but swallowing the IOException in isReady() means
that the onClose is never called.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

Mark Thomas  changed:

   What|Removed |Added

 Status|NEEDINFO|NEW

--- Comment #21 from Mark Thomas  ---
Thanks for the test cases. I'm taking a look.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #19 from Barry Coughlan  ---
Created attachment 33322
  --> https://bz.apache.org/bugzilla/attachment.cgi?id=33322=edit
Websocket client for testing TCP behaviour

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #18 from Barry Coughlan  ---
Created attachment 33321
  --> https://bz.apache.org/bugzilla/attachment.cgi?id=33321=edit
Test cases for closing websockets

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #17 from Barry Coughlan  ---
I tried creating a test for this today. It turns out that this problem only
happens when a websocket is not closed with a proper closing handshake.

Most Websocket clients are quite polite when closing websockets, they will send
a close frame and then wait for the handshake before closing the TCP
connection. This means that the problem can't be replicated by doing
session.close() in a Tomcat websocket client.

The reason that I can replicate this with Tyrus is that Tyrus sends a websocket
close frame but then closes the TCP connection immediately without completing
the handshake.

The steps with the example in my previous post and Tyrus CLI are:
1. Open connection
2. Send message, breaking in the @OnMessage
3. Close connection in Tyrus (or kill -9 the process to send just a TCP RST,
the result is the same)
4. Resume @OnMessage (which sends a message)

Any suggestions on how to automate the test? Can I force a Tomcat client's
socket to close without a handshake?

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #16 from Barry Coughlan  ---
I just built the trunk and confirmed the same behaviour. Note that this only
works with the NIO connector. I'll get back to you with a test case.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #15 from Mark Thomas  ---
I can't reproduce this using the current 8.0.x code. There is a fix that looks
related: bug 58624

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

Mark Thomas  changed:

   What|Removed |Added

 Status|REOPENED|NEEDINFO

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

Barry Coughlan  changed:

   What|Removed |Added

 Resolution|FIXED   |---
 Status|RESOLVED|REOPENED

--- Comment #14 from Barry Coughlan  ---
I can reproduce this on 8.0.29. I will create a proper test case that doesn't
need breakpoints during the week.

Reproduced with the following:

@OnMessage
public void onMessage(Session session, ModelObject msg) {
try {
session.getBasicRemote().sendText("Test"); // <-- Breakpoint here
} catch (IOException e) {
e.printStackTrace();
}
}
1. Open websocket connection to the server
2. Send message from client to server, hits the breakpoint
3. Close the connection
4. Continue from the @OnMessage breakpoint

The following happens:

- Step 4 throws an exception as expected (java.io.IOException: An established
connection was aborted by the software in your host machine) (we catch and log
this exception)


- @OnError is called with a java.nio.channels.ClosedChannelException:
java.nio.channels.ClosedChannelException
at sun.nio.ch.SocketChannelImpl.ensureReadOpen(SocketChannelImpl.java:256)
at sun.nio.ch.SocketChannelImpl.read(SocketChannelImpl.java:299)
at org.apache.tomcat.util.net.NioChannel.read(NioChannel.java:137)
at
org.apache.coyote.http11.upgrade.NioServletInputStream.fillReadBuffer(NioServletInputStream.java:136)
at
org.apache.coyote.http11.upgrade.NioServletInputStream.doIsReady(NioServletInputStream.java:49)
at
org.apache.coyote.http11.upgrade.AbstractServletInputStream.isReady(AbstractServletInputStream.java:66)
at
org.apache.tomcat.websocket.server.WsFrameServer.onDataAvailable(WsFrameServer.java:49)
at
org.apache.tomcat.websocket.server.WsHttpUpgradeHandler$WsReadListener.onDataAvailable(WsHttpUpgradeHandler.java:183)
at
org.apache.coyote.http11.upgrade.AbstractServletInputStream.onDataAvailable(AbstractServletInputStream.java:198)
at
org.apache.coyote.http11.upgrade.AbstractProcessor.upgradeDispatch(AbstractProcessor.java:96)
at
org.apache.coyote.AbstractProtocol$AbstractConnectionHandler.process(AbstractProtocol.java:669)
at
org.apache.tomcat.util.net.NioEndpoint$SocketProcessor.doRun(NioEndpoint.java:1500)
at
org.apache.tomcat.util.net.NioEndpoint$SocketProcessor.run(NioEndpoint.java:1456)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at
org.apache.tomcat.util.threads.TaskThread$WrappingRunnable.run(TaskThread.java:61)
at java.lang.Thread.run(Thread.java:745)


- Further attempts to send data on this session results in the "Key must be
cancelled" exception:
java.io.IOException: java.util.concurrent.ExecutionException:
java.io.IOException: Key must be cancelled
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.startMessageBlock(WsRemoteEndpointImplBase.java:282)
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.sendPing(WsRemoteEndpointImplBase.java:174)
at
org.apache.tomcat.websocket.WsRemoteEndpointBase.sendPing(WsRemoteEndpointBase.java:55)
at com.myapp.WsOutbound.sendPing(WsOutbound.java:295)
.

The @OnClose method is never called.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

Mark Thomas ma...@apache.org changed:

   What|Removed |Added

 Resolution|--- |FIXED
 Status|NEEDINFO|RESOLVED

--- Comment #13 from Mark Thomas ma...@apache.org ---
No response for over two months. I'm going to assume that this is fixed. Please
feel free to re-open if that is not the case. If the bug is still present, a
test case to reproduce the issue will make it a lot easier to fix.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://bz.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #12 from Mark Thomas ma...@apache.org ---
A handful of additional fixes that might be related to this have gone into the
8.0.20 release candidate. An update on whether or not the issue is still
present would be appreciated.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #11 from Mark Thomas ma...@apache.org ---
I've fixed another issue. I don't think it will make any different but if you
could do a quick test it would be appreciated.

I've just noticed that the stack trace passes through the onOpen method of the
WebSocket Endpoint. That looks odd. I'm wondering if Atmosphere is doing
something we didn't anticipate with the API. I'll look further but I don't
expect to get this fixed for the next 8.0.x release. What I really need is a
simple test case to reproduce this issue.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #10 from wodenc...@gmail.com ---
Hi Mark,

I built Tomcat from Trunk and tested again.

Unfortunately the issue happened again, here is the Stack Trace:

2015-02-10 16:01:13,056 CST ERROR [null] - java.io.IOException: 
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.startMessageBlock(WsRemoteEndpointImplBase.java:282)
at
org.apache.tomcat.websocket.WsSession.sendCloseMessage(WsSession.java:570)
at org.apache.tomcat.websocket.WsSession.doClose(WsSession.java:478)
at org.apache.tomcat.websocket.WsSession.close(WsSession.java:445)
at org.apache.tomcat.websocket.WsSession.close(WsSession.java:439)
at
org.atmosphere.container.version.JSR356WebSocket.close(JSR356WebSocket.java:104)
at org.atmosphere.websocket.WebSocket.close(WebSocket.java:264)
at org.atmosphere.cpr.AtmosphereResponse.close(AtmosphereResponse.java:836)
at
org.atmosphere.cpr.AtmosphereResourceImpl.cancel(AtmosphereResourceImpl.java:720)
at
org.atmosphere.cpr.AtmosphereResourceImpl.close(AtmosphereResourceImpl.java:785)
at
com.vaadin.server.communication.AtmospherePushConnection.disconnect(AtmospherePushConnection.java:294)
at
com.vaadin.server.communication.AtmospherePushConnection.connect(AtmospherePushConnection.java:232)
at com.vaadin.server.communication.PushHandler$2.run(PushHandler.java:135)
at
com.vaadin.server.communication.PushHandler.callWithUi(PushHandler.java:253)
at
com.vaadin.server.communication.PushHandler.access$200(PushHandler.java:56)
at
com.vaadin.server.communication.PushHandler$1.onRequest(PushHandler.java:74)
at
org.atmosphere.cpr.AsynchronousProcessor.action(AsynchronousProcessor.java:174)
at
org.atmosphere.cpr.AsynchronousProcessor.suspended(AsynchronousProcessor.java:95)
at
org.atmosphere.container.Servlet30CometSupport.service(Servlet30CometSupport.java:66)
at
org.atmosphere.cpr.AtmosphereFramework.doCometSupport(AtmosphereFramework.java:1802)
at
org.atmosphere.websocket.DefaultWebSocketProcessor.dispatch(DefaultWebSocketProcessor.java:432)
at
org.atmosphere.websocket.DefaultWebSocketProcessor.open(DefaultWebSocketProcessor.java:186)
at org.atmosphere.container.JSR356Endpoint.onOpen(JSR356Endpoint.java:160)
at
org.apache.tomcat.websocket.server.WsHttpUpgradeHandler.init(WsHttpUpgradeHandler.java:138)
at
org.apache.coyote.AbstractProtocol$AbstractConnectionHandler.process(AbstractProtocol.java:687)
at
org.apache.coyote.http11.Http11NioProtocol$Http11ConnectionHandler.process(Http11NioProtocol.java:223)
at
org.apache.tomcat.util.net.NioEndpoint$SocketProcessor.doRun(NioEndpoint.java:1558)
at
org.apache.tomcat.util.net.NioEndpoint$SocketProcessor.run(NioEndpoint.java:1515)
at java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)
at
org.apache.tomcat.util.threads.TaskThread$WrappingRunnable.run(TaskThread.java:61)
at java.lang.Thread.run(Unknown Source)
Caused by: java.util.concurrent.ExecutionException: java.io.IOException: Key
must be cancelled
at
org.apache.tomcat.websocket.FutureToSendHandler.get(FutureToSendHandler.java:102)
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.startMessageBlock(WsRemoteEndpointImplBase.java:275)
... 31 more
Caused by: java.io.IOException: Key must be cancelled
at
org.apache.coyote.http11.upgrade.NioServletOutputStream.doWriteInternal(NioServletOutputStream.java:84)
at
org.apache.coyote.http11.upgrade.NioServletOutputStream.doWrite(NioServletOutputStream.java:61)
at
org.apache.coyote.http11.upgrade.AbstractServletOutputStream.writeInternal(AbstractServletOutputStream.java:165)
at
org.apache.coyote.http11.upgrade.AbstractServletOutputStream.write(AbstractServletOutputStream.java:132)
at
org.apache.tomcat.websocket.server.WsRemoteEndpointImplServer.onWritePossible(WsRemoteEndpointImplServer.java:99)
at
org.apache.tomcat.websocket.server.WsRemoteEndpointImplServer.doWrite(WsRemoteEndpointImplServer.java:81)
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.writeMessagePart(WsRemoteEndpointImplBase.java:450)
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.startMessage(WsRemoteEndpointImplBase.java:338)
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.startMessageBlock(WsRemoteEndpointImplBase.java:270)
... 31 more
Caused by: java.util.concurrent.ExecutionException:
at
org.apache.tomcat.websocket.FutureToSendHandler.get(FutureToSendHandler.java:102)
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.startMessageBlock(WsRemoteEndpointImplBase.java:275)
at
org.apache.tomcat.websocket.WsSession.sendCloseMessage(WsSession.java:570)
at org.apache.tomcat.websocket.WsSession.doClose(WsSession.java:478)
at org.apache.tomcat.websocket.WsSession.close(WsSession.java:445)
at 

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #9 from Mark Thomas ma...@apache.org ---
OK. I've made various changes to 8.0.x that should at least make this less
likely and hopefully prevent. If you could test with the current 8.0.x and let
us know the results that would be great.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #8 from Christopher Schultz ch...@christopherschultz.net ---
(In reply to wodencafe from comment #7)
 I ran svn update, then ant, and finally copied the built
 libs into my Tomcat 8 server for testing.

That's the correct procedure. If you are using split CATALINA_BASE and
CATALINA_HOME (a great thing to do IMO), you can just point your CATALINA_HOME
to the directory you have your svn working copy in, under output/build,
something like this:

$ export CATALINA_BASE=[...]
$ export CATALINA_HOME=/path/to/tomcat-svn/output/build
$ $CATALINA_HOME/bin/catalina.sh start

You can always go back to your previous Tomcat (8.0.18, etc.) by switching
CATALINA_HOME back to its previous value.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #5 from wodenc...@gmail.com ---
Hi Mark,

I have been testing with the latest Trunk of 8.0.19, and verified that is the
case with this line in the logs:

2015-02-04 15:06:02,796 CST INFO  [null] - Atmosphere is using async support:
org.atmosphere.container.JSR356AsyncSupport running under container: Apache
Tomcat/8.0.19-dev using javax.servlet/3.0 using javax.servlet/3.0 and
jsr356/WebSocket API

Unfortunately, after interacting with the application for a brief amount of
time, it hangs and we see the same exception:

2015-02-04 15:07:01,691 CST ERROR [null] - java.io.IOException: 
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.startMessageBlock(WsRemoteEndpointImplBase.java:282)
at
org.apache.tomcat.websocket.WsSession.sendCloseMessage(WsSession.java:570)
at org.apache.tomcat.websocket.WsSession.doClose(WsSession.java:478)
at org.apache.tomcat.websocket.WsSession.close(WsSession.java:445)
at org.apache.tomcat.websocket.WsSession.close(WsSession.java:439)
at
org.atmosphere.container.version.JSR356WebSocket.close(JSR356WebSocket.java:104)
at org.atmosphere.websocket.WebSocket.close(WebSocket.java:264)
at org.atmosphere.cpr.AtmosphereResponse.close(AtmosphereResponse.java:836)
at
org.atmosphere.cpr.AtmosphereResourceImpl.cancel(AtmosphereResourceImpl.java:720)
at
org.atmosphere.cpr.AtmosphereResourceImpl.close(AtmosphereResourceImpl.java:785)
at
com.vaadin.server.communication.AtmospherePushConnection.disconnect(AtmospherePushConnection.java:294)
at
com.vaadin.server.communication.AtmospherePushConnection.connect(AtmospherePushConnection.java:232)
at com.vaadin.server.communication.PushHandler$2.run(PushHandler.java:135)
at
com.vaadin.server.communication.PushHandler.callWithUi(PushHandler.java:253)
at
com.vaadin.server.communication.PushHandler.access$200(PushHandler.java:56)
at
com.vaadin.server.communication.PushHandler$1.onRequest(PushHandler.java:74)
at
org.atmosphere.cpr.AsynchronousProcessor.action(AsynchronousProcessor.java:174)
at
org.atmosphere.cpr.AsynchronousProcessor.suspended(AsynchronousProcessor.java:95)
at
org.atmosphere.container.Servlet30CometSupport.service(Servlet30CometSupport.java:66)
at
org.atmosphere.cpr.AtmosphereFramework.doCometSupport(AtmosphereFramework.java:1802)
at
org.atmosphere.websocket.DefaultWebSocketProcessor.dispatch(DefaultWebSocketProcessor.java:432)
at
org.atmosphere.websocket.DefaultWebSocketProcessor.open(DefaultWebSocketProcessor.java:186)
at org.atmosphere.container.JSR356Endpoint.onOpen(JSR356Endpoint.java:160)
at
org.apache.tomcat.websocket.server.WsHttpUpgradeHandler.init(WsHttpUpgradeHandler.java:138)
at
org.apache.coyote.AbstractProtocol$AbstractConnectionHandler.process(AbstractProtocol.java:687)
at
org.apache.coyote.http11.Http11NioProtocol$Http11ConnectionHandler.process(Http11NioProtocol.java:223)
at
org.apache.tomcat.util.net.NioEndpoint$SocketProcessor.doRun(NioEndpoint.java:1558)
at
org.apache.tomcat.util.net.NioEndpoint$SocketProcessor.run(NioEndpoint.java:1515)
at java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)
at
org.apache.tomcat.util.threads.TaskThread$WrappingRunnable.run(TaskThread.java:61)
at java.lang.Thread.run(Unknown Source)
Caused by: java.util.concurrent.ExecutionException: java.io.IOException: Key
must be cancelled
at
org.apache.tomcat.websocket.FutureToSendHandler.get(FutureToSendHandler.java:102)
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.startMessageBlock(WsRemoteEndpointImplBase.java:275)
... 31 more
Caused by: java.io.IOException: Key must be cancelled
at
org.apache.coyote.http11.upgrade.NioServletOutputStream.doWriteInternal(NioServletOutputStream.java:84)
at
org.apache.coyote.http11.upgrade.NioServletOutputStream.doWrite(NioServletOutputStream.java:61)
at
org.apache.coyote.http11.upgrade.AbstractServletOutputStream.writeInternal(AbstractServletOutputStream.java:162)
at
org.apache.coyote.http11.upgrade.AbstractServletOutputStream.write(AbstractServletOutputStream.java:129)
at
org.apache.tomcat.websocket.server.WsRemoteEndpointImplServer.onWritePossible(WsRemoteEndpointImplServer.java:99)
at
org.apache.tomcat.websocket.server.WsRemoteEndpointImplServer.doWrite(WsRemoteEndpointImplServer.java:81)
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.writeMessagePart(WsRemoteEndpointImplBase.java:450)
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.startMessage(WsRemoteEndpointImplBase.java:338)
at
org.apache.tomcat.websocket.WsRemoteEndpointImplBase.startMessageBlock(WsRemoteEndpointImplBase.java:270)
... 31 more
Caused by: java.util.concurrent.ExecutionException:
at
org.apache.tomcat.websocket.FutureToSendHandler.get(FutureToSendHandler.java:102)
at

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #7 from wodenc...@gmail.com ---
Hi Mark,

I am fine with building SVN, as long as I am doing it correctly :)

What I did today was I went into the directory where I had Tomcat trunk checked
out, I ran svn update, then ant, and finally copied the built libs into my
Tomcat 8 server for testing.

This is the correct process, right?

Let me know when to test again.

Thank you,

Chris Boyd

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #6 from Mark Thomas ma...@apache.org ---
Thanks for the feedback.

The CI system has highlighted a couple of issues with these fixes in place.
I've just found what I think is another instance of multiple write
registrations in trunk.

My plan is to get the CI system building trunk cleanly, port the fixes back to
8.0.x and then get the CI system for 8.0.x running cleanly. I don't think there
is much benefit in you testing any more until the CI system says the build is
clean. I'll ping this ticket when that is done.

Do you need me to build 8.0.x for you to test or are you happy building from
svn?

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

Mark Thomas ma...@apache.org changed:

   What|Removed |Added

 Status|NEW |NEEDINFO

--- Comment #4 from Mark Thomas ma...@apache.org ---
I've just fixed some issues that might have triggered this sort of issue in
trunk and 8.0.x. There is a development build of 8.0.19-dev here:
http://people.apache.org/~markt/dev/v8.0.19-dev/

Any feedback you can provide on testing with this or 8.0.x trunk would be
great.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #3 from Jean-Francois Arcand jfarc...@apache.org ---
@Mark,

I did hack a little to prevent that error (which I may not fixed completely)

https://github.com/Atmosphere/atmosphere/issues/1646

The stack trave will probably help you.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

--- Comment #2 from Mark Thomas ma...@apache.org ---
I've added some debig logging to trunk, 8.0.x (for 8.0.19 onwards) and 7.0.x
(for 7.0.58 onwards). Any further information you can provide as to what is
going on would be very helpful.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

Mark Thomas ma...@apache.org changed:

   What|Removed |Added

 OS||All

--- Comment #1 from Mark Thomas ma...@apache.org ---
Hmm. Whatever is going wrong, is going wrong earlier than this stack trace.

Something causes the WebSocket connection to close. It appears that Atmosphere
isn't notified about this / doesn't notice it.

Something happens later that causes Atmosphere to realise there is a dead
WebSocket connection. Atmosphere triggers a 'dirty' close.

This 'dirty' close fails because the WebSocket connection has already been
closed. That triggers an error because, as far as I can tell, a normal
WebSocket close reason (1000) is used. If abnormal closure was used (1006),
Tomcat would swallow any error.

What we really need to figure out is why is this connection closed in the first
place. There isn't much in the way of debug logging for WebSocket session
closure in Tomcat - I'll fix that for the next release - but it is worth seeing
is Atmosphere's debug logging tells you anything useful.

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org

[Bug 57489] java.util.concurrent.ExecutionException: java.io.IOException: Key must be cancelled

[bugzilla]

https://issues.apache.org/bugzilla/show_bug.cgi?id=57489

wodenc...@gmail.com changed:

   What|Removed |Added

 OS|Linux   |Windows 8

-- 
You are receiving this mail because:
You are the assignee for the bug.

-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org