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

[Jaikiran Pai]

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)
>