Mailing-List: contact httpclient-users-help@hc.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "HttpClient User Discussion" <httpclient-users@hc.apache.org>
Received-SPF: neutral (athena.apache.org: local policy)
Message-ID: <1344372755.2021.6.camel@ubuntu>
Subject: Re: Strange ClassCastException
From: Oleg Kalnichevski <olegk@apache.org>
To: HttpClient User Discussion <httpclient-users@hc.apache.org>
Cc: Harold Rosenberg <hrosenbe@vmware.com>
Date: Tue, 07 Aug 2012 22:52:35 +0200
In-Reply-To: <CC46CE28.11F05%hrosenbe@vmware.com>
References: <CC46CE28.11F05%hrosenbe@vmware.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Mime-Version: 1.0

On Tue, 2012-08-07 at 10:52 -0700, Harold Rosenberg wrote:
> I am using the AsyncClient (4.0-beta1) as part of an HTTP workload driver=
, along with HttpCore 4.2.1 and HttpClient 4.2.1.  My primary entry-point i=
s through calls to execute(HttpUriRequest<eclipse-javadoc:%E2%98%82=3Drain-=
workload-toolkit/%5C/Users%5C/Hal%5C/.m2%5C/repository%5C/org%5C/apache%5C/=
httpcomponents%5C/httpasyncclient%5C/4.0-beta1%5C/httpasyncclient-4.0-beta1=
.jar%3Corg.apache.http.nio.client(HttpAsyncClient.class%E2%98%83HttpAsyncCl=
ient~execute~Lorg.apache.http.client.methods.HttpUriRequest;~Lorg.apache.ht=
tp.protocol.HttpContext;~Lorg.apache.http.concurrent.FutureCallback%5C%3CLo=
rg.apache.http.HttpResponse;%3E;%E2%98%82HttpUriRequest> request,HttpContex=
t<eclipse-javadoc:%E2%98%82=3Drain-workload-toolkit/%5C/Users%5C/Hal%5C/.m2=
%5C/repository%5C/org%5C/apache%5C/httpcomponents%5C/httpasyncclient%5C/4.0=
-beta1%5C/httpasyncclient-4.0-beta1.jar%3Corg.apache.http.nio.client(HttpAs=
yncClient.class%E2%98%83HttpAsyncClient~execute~Lorg.apache.http.client.met=
hods.HttpUriRequest;~Lorg.apache.http.protocol.HttpContext;~Lorg.apache.htt=
p.concurrent.FutureCallback%5C%3CLorg.apache.http.HttpResponse;%3E;%E2%98%8=
2HttpContext> context, FutureCallback<eclipse-javadoc:%E2%98%82=3Drain-work=
load-toolkit/%5C/Users%5C/Hal%5C/.m2%5C/repository%5C/org%5C/apache%5C/http=
components%5C/httpasyncclient%5C/4.0-beta1%5C/httpasyncclient-4.0-beta1.jar=
%3Corg.apache.http.nio.client(HttpAsyncClient.class%E2%98%83HttpAsyncClient=
~execute~Lorg.apache.http.client.methods.HttpUriRequest;~Lorg.apache.http.p=
rotocol.HttpContext;~Lorg.apache.http.concurrent.FutureCallback%5C%3CLorg.a=
pache.http.HttpResponse;%3E;%E2%98%82FutureCallback><HttpResponse<eclipse-j=
avadoc:%E2%98%82=3Drain-workload-toolkit/%5C/Users%5C/Hal%5C/.m2%5C/reposit=
ory%5C/org%5C/apache%5C/httpcomponents%5C/httpasyncclient%5C/4.0-beta1%5C/h=
ttpasyncclient-4.0-beta1.jar%3Corg.apache.http.nio.client(HttpAsyncClient.c=
lass%E2%98%83HttpAsyncClient~execute~Lorg.apache.http.client.methods.HttpUr=
iRequest;~Lorg.apache.http.protocol.HttpContext;~Lorg.apache.http.concurren=
t.FutureCallback%5C%3CLorg.apache.http.HttpResponse;%3E;%E2%98%82HttpRespon=
se>> callback) on a DefaultHttpAsyncClient.  I am sporadically getting the =
following exception:
>=20
> 12/08/06 08:31:21 ERROR protocol.HttpAsyncRequestExecutor: http-outgoing-=
143466 [ACTIVE] HTTP protocol exception: org.apache.http.impl.nio.conn.Defa=
ultClientAsyncConnection cannot be cast to org.apache.http.conn.HttpRoutedC=
onnection
> java.lang.ClassCastException: org.apache.http.impl.nio.conn.DefaultClient=
AsyncConnection cannot be cast to org.apache.http.conn.HttpRoutedConnection
> at org.apache.http.client.protocol.RequestAddCookies.process(RequestAddCo=
okies.java:123)
> at org.apache.http.protocol.ImmutableHttpProcessor.process(ImmutableHttpP=
rocessor.java:109)
> at org.apache.http.nio.protocol.HttpAsyncRequestExecutor.requestReady(Htt=
pAsyncRequestExecutor.java:157)
> at org.apache.http.impl.nio.client.LoggingAsyncRequestExecutor.requestRea=
dy(LoggingAsyncRequestExecutor.java:89)
> at org.apache.http.impl.nio.DefaultNHttpClientConnection.produceOutput(De=
faultNHttpClientConnection.java:225)
> at org.apache.http.impl.nio.DefaultHttpClientIODispatch.onOutputReady(Def=
aultHttpClientIODispatch.java:130)
> at org.apache.http.impl.nio.DefaultHttpClientIODispatch.onOutputReady(Def=
aultHttpClientIODispatch.java:50)
> at org.apache.http.impl.nio.reactor.AbstractIODispatch.outputReady(Abstra=
ctIODispatch.java:141)
> at org.apache.http.impl.nio.reactor.BaseIOReactor.writable(BaseIOReactor.=
java:181)
> at org.apache.http.impl.nio.reactor.AbstractIOReactor.processEvent(Abstra=
ctIOReactor.java:346)
> at org.apache.http.impl.nio.reactor.AbstractIOReactor.processEvents(Abstr=
actIOReactor.java:320)
> at org.apache.http.impl.nio.reactor.AbstractIOReactor.execute(AbstractIOR=
eactor.java:280)
> at org.apache.http.impl.nio.reactor.BaseIOReactor.execute(BaseIOReactor.j=
ava:106)
> at org.apache.http.impl.nio.reactor.AbstractMultiworkerIOReactor$Worker.r=
un(AbstractMultiworkerIOReactor.java:601)
> at java.lang.Thread.run(Thread.java:662)
>=20
> This shows up in my code as a call to failed(Exception) in my FutureCallb=
ack.
>=20
> This only occurs very rarely (e.g. 7 such exceptions during a run with 1.=
4 million calls to execute), and I can handle the exception and retry the o=
peration, but I find it very puzzling.  Using a debugger, I have looked at =
the state of the application on both successful and unsuccessful calls thro=
ugh this code path, and I can't see any difference.
>=20
> Does anyone have any ideas about what might be causing this?  I have trie=
d with both JDK 1.6.0_33 and 1.7.0_5 and get the same exception on both.
>=20
> Thanks,
>=20
> Hal
>=20

Hi Hal

This looks like a bug in HttpAsyncClient. Please raise a JIRA for this
defect. It is a shame this report did not come a little earlier. We just
released 4.0-beta2.

Oleg=20


---------------------------------------------------------------------
To unsubscribe, e-mail: httpclient-users-unsubscribe@hc.apache.org
For additional commands, e-mail: httpclient-users-help@hc.apache.org