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

Hi Daniel, On 26/06/18 8:26 PM, Daniel Fuchs wrote: Hi Jaikiran, On 26/06/2018 15:41, Jaikiran Pai wrote: So is this a case of the server implementation misbehaving? I haven't checked what the expectations are from a spec point of view for handling such requests on the server. Or is it a comb

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

Hi Jaikiran, On 26/06/2018 15:41, Jaikiran Pai wrote: So is this a case of the server implementation misbehaving? I haven't checked what the expectations are from a spec point of view for handling such requests on the server. Or is it a combination of a misbehaving server plus the HttpClient i

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

Hi Daniel, You are right- I just added: 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 20:05:20 2

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

Hi Jaikiran, I believe you need to drain the request body in the server before sending the response. If you pass -ea -esa options to jtreg, you should see the following assertion printed by the server: java.lang.AssertionError: State is not RESPONSE (REQUEST) at jdk.httpserver/sun.net.http

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

Thanks Chris! -Jaikiran On 26/06/18 7:34 PM, Chris Hegarty wrote: Hi Jaikiran, I filed the following JIRA issue to track this: 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

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

Hi Jaikiran, I filed the following JIRA issue to track this: 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 wrote: > > So

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

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/j

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

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 3r