[
https://issues.apache.org/jira/browse/HTTPASYNC-39?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13638138#comment-13638138
]
Clinton Nielsen commented on HTTPASYNC-39:
------------------------------------------
I'm trying to reproduce this locally, and am also having a hard time.
For reference, the code I'm using looks something like this (cleaned for
brevity):
final long start = System.currentTimeMillis();
cachingHttpAsyncClient.execute(appInfo.httpMethod, new BasicHttpContext(), new
FutureCallback<HttpResponse>() {
public void failed(final Exception e) {
logger.error(e + ", Time (ms): " + (System.currentTimeMillis() - start));
...
}
...
});
When trying locally, with SocketTimeout set to 30,000 ms, when the failed
method is called, the output in the logs consistently shows time taken to be
between 30,000 and 31,000 milliseconds.
On production, however, using the same code, the logs look something like this.
Notice that the time taken is never anywhere near the 30,000ms mark. It's
always much shorter, or much longer:
java.net.SocketTimeoutException, Time(ms): 0
java.net.SocketTimeoutException, Time(ms): 0
java.net.SocketTimeoutException, Time(ms): 1
java.net.SocketTimeoutException, Time(ms): 0
java.net.SocketTimeoutException, Time(ms): 46194
java.net.SocketTimeoutException, Time(ms): 50692
java.net.SocketTimeoutException, Time(ms): 49626
java.net.SocketTimeoutException, Time(ms): 45256
java.net.SocketTimeoutException, Time(ms): 0
java.net.SocketTimeoutException, Time(ms): 1
java.net.SocketTimeoutException, Time(ms): 10
java.net.SocketTimeoutException, Time(ms): 1
java.net.SocketTimeoutException, Time(ms): 1
java.net.SocketTimeoutException, Time(ms): 1
java.net.SocketTimeoutException, Time(ms): 1
java.net.SocketTimeoutException, Time(ms): 1
java.net.SocketTimeoutException, Time(ms): 1
java.net.SocketTimeoutException, Time(ms): 1
java.net.SocketTimeoutException, Time(ms): 1
java.net.SocketTimeoutException, Time(ms): 50656
java.net.SocketTimeoutException, Time(ms): 46603
java.net.SocketTimeoutException, Time(ms): 88
On a maybe related not, when load goes up we're also seeing a lot of the
following, none of which were present when using the non-async client:
java.net.ConnectException: Connection timed out (usually around the 21,000 ms
mark)
java.io.IOException: Connection timed out (this occurs at the ~15 minute mark!)
Either way, I'll keep playing with it and maybe get more information and
hopefully reproduce in controlled conditions.
Thanks.
> SocketTimeoutException occurs too soon
> --------------------------------------
>
> Key: HTTPASYNC-39
> URL: https://issues.apache.org/jira/browse/HTTPASYNC-39
> Project: HttpComponents HttpAsyncClient
> Issue Type: Bug
> Affects Versions: 4.0-beta3
> Reporter: Clinton Nielsen
>
> In a highly concurrent environment, where SocketTimeout is set to 30,000
> milliseconds, we are often seeing a timeoutexception after only 0 or 1 or 2
> milliseconds has actually passed.
> I get the impression that in the AbstractIOReactor timeoutCheck function,
> session.getLastAccessTime is being called on the session before the
> lastAccessTime is set on the session for the current session (ie. the
> lastAccessTime being retrieved is the time that was set on the session object
> for the previous http session)
--
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: [email protected]
For additional commands, e-mail: [email protected]
