New HttpClient PUT request fails oddly - is it size is it number of invocations!

Chris Hegarty chris.hegarty at oracle.com
Tue Jun 26 14:04:06 UTC 2018


Hi Jaikiran,

I filed the following JIRA issue to track this:
  https://bugs.openjdk.java.net/browse/JDK-8205684 <https://bugs.openjdk.java.net/browse/JDK-8205684>

Thanks for the test and initial investigation. We’ll look into it.

-Chris.

> On 26 Jun 2018, at 14:52, Jaikiran Pai <jai.forums2013 at gmail.com> wrote:
> 
> So it looks like there some race condition somewhere in the HttpClient implementation. I just addeda smalldelay between the PUT requests, in the test that I sent in the patch and it's now started passing:
> 
> diff -r 955a66f0f04a test/jdk/java/net/httpclient/PUTRequestSizeTest.java
> --- a/test/jdk/java/net/httpclient/PUTRequestSizeTest.java    Tue Jun 26 18:33:18 2018 +0530
> +++ b/test/jdk/java/net/httpclient/PUTRequestSizeTest.java    Tue Jun 26 19:19:54 2018 +0530
> @@ -59,6 +59,9 @@
>              for (int i = 0; i < 10; i++) {
>                  System.out.println("Sending PUT request " + (i +1));
>                  issuePUT(httpClient, requestURL);
> +                // TODO: This shouldn't be needed but is here to demonstrate a race
> +                // condition somewhere in the HttpClient implementation
> +                Thread.sleep(500);
>              }
>          } finally {
>              server.stop(0);
> 
> 
> 
> -Jaikiran
> On 26/06/18 7:12 PM, Jaikiran Pai wrote:
>> In my random experimentation with the new HttpClient API usage, I have ended up running into an odd and hard to decipher exception when dealing with PUT requests. I am noticing that if I issue multiple PUT requests using the same HttpClient instance, the first 2 invocations succeed while the 3rd one fails with an exception[1]. Initially when I ran into this, I saw this happening depending on the size of the data being uploaded via PUT request. As soon as it hit 16385 bytes (16 KB + 1), it would end up throwing the odd exception. However, when I decided to narrow it down to a testcase, I was able to reproduce this without the size of the data playing a role. And weirdly, it now keeps failing for the third invocation.
>> 
>> I have now isolated this into a jtreg testcase and created a patch against the latest upstream jdk to reproduce this issue. I've attached the patch in this mail. In the test, I create a local server and keep sending PUT requests to the server using the same instance of HttpClient. As soon as it hits the 3rd invocation, I end up seeing the exception[1].
>> 
>> It's really odd, since, initially I thought it could be a size based issue, which would have been more understandable. However with the way it's failing now in this jtreg test, I am starting to wonder if I have got some basics wrong.
>> 
>> 
>> [1]
>> 
>> java.io.IOException: HTTP/1.1 header parser received no bytes
>>     at java.net.http/jdk.internal.net.http.HttpClientImpl.send(HttpClientImpl.java:546)
>>     at java.net.http/jdk.internal.net.http.HttpClientFacade.send(HttpClientFacade.java:113)
>>     at PUTRequestSizeTest.issuePUT(PUTRequestSizeTest.java:76)
>>     at PUTRequestSizeTest.main(PUTRequestSizeTest.java:61)
>>     at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
>>     at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
>>     at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
>>     at java.base/java.lang.reflect.Method.invoke(Method.java:566)
>>     at com.sun.javatest.regtest.agent.MainWrapper$MainThread.run(MainWrapper.java:115)
>>     at java.base/java.lang.Thread.run(Thread.java:834)
>> Caused by: java.io.IOException: HTTP/1.1 header parser received no bytes
>>     at java.net.http/jdk.internal.net.http.common.Utils.wrapWithExtraDetail(Utils.java:293)
>>     at java.net.http/jdk.internal.net.http.Http1Response$HeadersReader.onReadError(Http1Response.java:646)
>>     at java.net.http/jdk.internal.net.http.Http1AsyncReceiver.checkForErrors(Http1AsyncReceiver.java:297)
>>     at java.net.http/jdk.internal.net.http.Http1AsyncReceiver.flush(Http1AsyncReceiver.java:263)
>>     at java.net.http/jdk.internal.net.http.common.SequentialScheduler$SynchronizedRestartableTask.run(SequentialScheduler.java:175)
>>     at java.net.http/jdk.internal.net.http.common.SequentialScheduler$CompleteRestartableTask.run(SequentialScheduler.java:147)
>>     at java.net.http/jdk.internal.net.http.common.SequentialScheduler$SchedulableTask.run(SequentialScheduler.java:198)
>>     at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128)
>>     at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628)
>>     ... 1 more
>> Caused by: java.io.EOFException: EOF reached while reading
>>     at java.net.http/jdk.internal.net.http.Http1AsyncReceiver$Http1TubeSubscriber.onComplete(Http1AsyncReceiver.java:587)
>>     at java.net.http/jdk.internal.net.http.SocketTube$InternalReadPublisher$ReadSubscription.signalCompletion(SocketTube.java:629)
>>     at java.net.http/jdk.internal.net.http.SocketTube$InternalReadPublisher$InternalReadSubscription.read(SocketTube.java:830)
>>     at java.net.http/jdk.internal.net.http.SocketTube$SocketFlowTask.run(SocketTube.java:175)
>>     at java.net.http/jdk.internal.net.http.common.SequentialScheduler$SchedulableTask.run(SequentialScheduler.java:198)
>>     at java.net.http/jdk.internal.net.http.common.SequentialScheduler.runOrSchedule(SequentialScheduler.java:271)
>>     at java.net.http/jdk.internal.net.http.common.SequentialScheduler.runOrSchedule(SequentialScheduler.java:224)
>>     at java.net.http/jdk.internal.net.http.SocketTube$InternalReadPublisher$InternalReadSubscription.handleSubscribeEvent(SocketTube.java:687)
>>     at java.net.http/jdk.internal.net.http.AsyncTriggerEvent.handle(AsyncTriggerEvent.java:54)
>>     at java.net.http/jdk.internal.net.http.HttpClientImpl$SelectorManager.run(HttpClientImpl.java:796)
>> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/net-dev/attachments/20180626/ad4f2994/attachment-0001.html>


More information about the net-dev mailing list