https://issues.apache.org/bugzilla/show_bug.cgi?id=51881
Mark Thomas <ma...@apache.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|REOPENED |RESOLVED
Resolution| |FIXED
--- Comment #11 from Mark Thomas <ma...@apache.org> 2011-10-03 17:03:17 UTC ---
This took a while to track down but was eventually relatively simple.
When using NIO and a comet request is closed from a different thread then the
poller is used to trigger the processing of the close event. The problem was
that that request was marked as non-comet before this poller task was triggered
meaning that when it happened the request was treated as a non-Comet request
rather than a comet request. This meant that the necessary clean-up was not
make and then next request received on the connection (i.e. from the same
client) was corrupted.
This has been fixed - along with some additional clean-up - in trunk and 7.0.x
and will be included in 7.0.23 onwards.
Thanks again for the test case. Bugs that involve multiple threads are always
tricky to track down but the test case made it a lot easier.
--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
