https://issues.apache.org/bugzilla/show_bug.cgi?id=52121
--- Comment #2 from Mark Thomas <ma...@apache.org> 2011-11-03 23:14:00 UTC --- Many thanks for the excellent test case. It looks like it was very difficult to track down. We really do appreciate your efforts on this. I have converted the stand-alone reproduction JAR to a Tomcat test case. It isn't yet in the standard test suite as it currently always fails. I noticed that the following is sufficient to reproduce the issue: - send parts 19 to 28, flush, 29, flush, close I have therefore merged parts 19 to 28 for the test case. The failure is a CRC error. At this point this looks like a JVM bug. We don't normally address JVM bugs in the Tomcat codebase but if we can figure out exactly what causes this bug (it seems to be related to a very precise state) then we might be able to detect the state and prevent the flush that triggers the bug. Failing that, it should be possible to make use of the flushable GZIP filter configurable. As a workaround, you may be able to add an explicit flush() to the page(s) where this occurs. -- 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
