Mailing-List: contact etch-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: etch-dev@incubator.apache.org
Received-SPF: pass (nike.apache.org: domain of jliau@cisco.com designates
 171.71.176.71 as permitted sender)
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: [vote] include the fix for etch-25 in release 1.0.2
Date: Wed, 14 Jan 2009 07:54:39 -0800
Message-ID: 
 <4C0FAAC489C8B74F96BEAD85EAEB262507D3CC58@xmb-sjc-215.amer.cisco.com>
In-Reply-To: <496E058A.8070004@cisco.com>
Thread-Topic: [vote] include the fix for etch-25 in release 1.0.2
Thread-Index: Acl2XXBZS0uRhcycT3i0bYIEkvo6OgAAukhw
References: <496E058A.8070004@cisco.com>
From: "J.D. Liau (jliau)" <jliau@cisco.com>
To: <etch-dev@incubator.apache.org>

Yes, I think this should be included in 1.0.2.=20

-----Original Message-----
From: Scott Comer (sccomer)=20
Sent: Wednesday, January 14, 2009 9:32 AM
To: etch-dev@incubator.apache.org
Subject: [vote] include the fix for etch-25 in release 1.0.2

DefaultDeliveryService ends call with mb.CloseDelivery() and it should
end it with mb.CloseRead()
------------------------------------------------------------------------
-------------------------

                 Key: ETCH-25
                 URL: https://issues.apache.org/jira/browse/ETCH-25
             Project: Etch
          Issue Type: Bug
          Components: csharp-binding, java-binding
    Affects Versions: 1.0.1, 1.0.0
            Reporter: scott comer


when done reading the mailbox, the right method to call is CloseRead().
somehow the code actually calls CloseDelivery(), a method used to
shutdown the writing side of the mailbox. CloseRead() calls
CloseDelivery() first off, but then takes the additional steps of
redelivering any messages left in the mailbox.

The distinction is rather fine, and this bug was found by reading the
source code. There are certain ways you could exercise this bug, but it
would be very rare and not have any real impact. So, it isn't an
emergency to fix it, but it is also a relatively benign fix. I've
already made the change in my tree and tested it. The patch is attached.

------------------

there is little risk. all unit tests have passed.

scott out