Hi Justin,

Thanks for approving QPID-4164 (now merged). Could you take a look at
QPID-4170 as well?

(I realise you might have spotted it was still RtR when I sent the email,
sorry; I forgot to ask Keith to press the button, he had actually reviewed
it before I committed it)

Thanks,
Robbie

On 30 July 2012 15:20, Robbie Gemmell <[email protected]> wrote:

> Hi Justin,
>
> I would like to request that the following commits be considered for
> inclusion in 0.18:
>
> https://issues.apache.org/jira/browse/QPID-4164
> http://svn.apache.org/viewvc?view=revision&revision=1366339
> http://svn.apache.org/viewvc?view=revision&revision=1365832
> http://svn.apache.org/viewvc?view=revision&revision=1367095
>
> The above commits fix / add tests for an issue whereby moving/copying
> persistent messages between queues (either manually via the management
> interface, or automatically using a feature such as maximum-redelivery on
> queues configured with with DLQs) can result in loss of the payload if the
> message was originally recovered from the message store. The change is very
> isolated (to the message stores) and trivial as-is but would look even
> smaller if we hadn't inverted the meaning of a boolean.
>
> https://issues.apache.org/jira/browse/QPID-4170
> http://svn.apache.org/viewvc?view=revision&revision=1367084
>
> The above commit fixes an issue where JMX management threads can enter a
> infite loop when viewing the content of messages affected by the issue
> above, consuming CPU resources and creating additional lock contention in
> the message store. The change is again very isolated (to the Queue MBean)
> and trivial.
>
> The changes were made either by Keith, Phil, or myself, and have then been
> reviewed by at least one but usually both of the others.
>
> Robbie
>

Reply via email to