hc-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Oleg Kalnichevski (JIRA)" <j...@apache.org>
Subject [jira] [Resolved] (HTTPASYNC-36) SocketTimeoutException can result in failed() being called multiple times
Date Thu, 25 Apr 2013 11:14:16 GMT

     [ https://issues.apache.org/jira/browse/HTTPASYNC-36?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Oleg Kalnichevski resolved HTTPASYNC-36.
----------------------------------------

    Resolution: Cannot Reproduce

Clinton
I do see how this issue could be a result of a race condition given that timeout event can
be triggered by an I/O dispatch thread only. I may well be missing something but without a
reproducer there is no way for me to find that out.

On the other hand there might be a simpler explanation for multiple log entries. This is how
timeout handling code looks like:
---
    public synchronized void failed(final Exception ex) {
        try {
            if (!this.requestSent) {
                this.requestProducer.failed(ex);
            }
            this.responseConsumer.failed(ex);
        } finally {
            try {
                this.resultCallback.failed(ex, this);
            } finally {
                close();
            }
        }
    }
--- 
As you can see the same exception instance can be potentially passed to three different callbacks
(request producer / response consumer / future callback). If all of them log that exception
there can be three entires for the same exception. Could that be the reason?

Oleg
                
> SocketTimeoutException can result in failed() being called multiple times
> -------------------------------------------------------------------------
>
>                 Key: HTTPASYNC-36
>                 URL: https://issues.apache.org/jira/browse/HTTPASYNC-36
>             Project: HttpComponents HttpAsyncClient
>          Issue Type: Bug
>    Affects Versions: 4.0-beta3
>            Reporter: Clinton Nielsen
>
> Appears to be a race condition, so I am having trouble duplicating, but we have observed
the failed callback method being called twice for a single request that timed out.
> Following stack trace was logged twice for a single request:
> HttpAsyncRequestExecutor.java:279:org.apache.http.nio.protocol.HttpAsyncRequestExecutor:timeout
> LoggingAsyncRequestExecutor.java:128:org.apache.http.impl.nio.client.LoggingAsyncRequestExecutor:timeout
> DefaultHttpClientIODispatch.java:136:org.apache.http.impl.nio.DefaultHttpClientIODispatch:onTimeout
> DefaultHttpClientIODispatch.java:50:org.apache.http.impl.nio.DefaultHttpClientIODispatch:onTimeout
> AbstractIODispatch.java:169:org.apache.http.impl.nio.reactor.AbstractIODispatch:timeout
> BaseIOReactor.java:257:org.apache.http.impl.nio.reactor.BaseIOReactor:sessionTimedOut
> AbstractIOReactor.java:494:org.apache.http.impl.nio.reactor.AbstractIOReactor:timeoutCheck
> BaseIOReactor.java:207:org.apache.http.impl.nio.reactor.BaseIOReactor:validate
> AbstractIOReactor.java:284:org.apache.http.impl.nio.reactor.AbstractIOReactor:execute
> BaseIOReactor.java:106:org.apache.http.impl.nio.reactor.BaseIOReactor:execute
> AbstractMultiworkerIOReactor.java:604:org.apache.http.impl.nio.reactor.AbstractMultiworkerIOReactor$Worker:run
> Thread.java:636:java.lang.Thread:run | class java.net.SocketTimeoutException

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

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


Mime
View raw message