mina-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Rupert Smith" <rupertlssm...@googlemail.com>
Subject Bug in IdentityHashMap
Date Tue, 02 Oct 2007 15:27:06 GMT

Hi, I am contributing to the Qpid incubator project, and have discovered a
bug in IdentityHashMap, that Mina is exposed to. I notice that
IdentityHashMap is used a fair amount within the Mina code base, so thought
I would take the additional step of mailing your dev list.

Occasionally, certain sequences of put and iterator.remove() can fool the
map into allowing put() to be called with a full table. It then loops
forever looking for a free slot in the table. See the JIRA for sample code
to expose the bug.

This is a problem for Qpid as thgouh the VmPipeIdleStatusChcker class, but
there may be other usages of IdentityHashMap.iterator.remove() in the Mina
code base.

Is there any possibility of creating a new 1.0.x release with a fix for this



  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message