[
https://issues.apache.org/jira/browse/HTTPCORE-635?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Pedro Lamarão updated HTTPCORE-635:
-----------------------------------
Attachment: integrationTest.log
> AbstractBinAsyncEntityProducer.produceData called again after
> StreamChannel.endStream
> -------------------------------------------------------------------------------------
>
> Key: HTTPCORE-635
> URL: https://issues.apache.org/jira/browse/HTTPCORE-635
> Project: HttpComponents HttpCore
> Issue Type: Bug
> Components: HttpCore
> Affects Versions: 5.0.1
> Reporter: Pedro Lamarão
> Priority: Major
> Attachments: InputStreamServerHandler.java, integrationTest.log,
> stack-01.txt, stack-02.txt, stack-03.txt
>
>
> We extend AbstractBinAsyncEntityProducer to stream from an InputStream. While
> investigating a conceptually higher-level problem, we have hit the following
> situation. In the attached code file, our program raises the assertion at
> line 66.
> The documentation for produceData asserts "the ability of the underlying byte
> channel to accept more data" and the documentation for endStream asserts
> "terminates the underlying data stream". Though not explicitly stated, it
> seems obvious that, after endStream, the underlying stream cannot accept more
> data.
> Our original problem is a test case where the server program apparently does
> not actually send the response entity to the client program. Client logs
> indicate there is no response body, and debugging revealed that produce
> eventually calls endStream twice in a row.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
