https://issues.apache.org/bugzilla/show_bug.cgi?id=54998

--- Comment #3 from Tim Kosse <[email protected]> ---
Created attachment 31593
  --> https://issues.apache.org/bugzilla/attachment.cgi?id=31593&action=edit
Proposed patch that fixes this bug.

I figured out what seems to be the problem: The closure alert is being
generated, but is stuck in some buffers without ever making it to the wire, let
alone the peer. Flushing the output BIO seems to solve the problem.

The attached patch fixes the bug for me. The patch has been created against the
2.4.x branch.

Current trunk (rev 1592635) is also affected by this bug. I've tested that this
patch also applies to trunk and fixes the problem there.



As testcase I used the following command:
(echo -e "GET / HTTP/1.0\n\n"; sleep 1) | gnutls-cli --insecure -p 443
127.0.0.1 --crlf

If the bug occurs, gnutls-cli will at the end display a fatal error:
*** Fatal error: The TLS connection was non-properly terminated.
*** Server has terminated the connection abnormally.

-- 
You are receiving this mail because:
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to