[
https://issues.apache.org/jira/browse/AMQCPP-598?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15308576#comment-15308576
]
ASF subversion and git services commented on AMQCPP-598:
--------------------------------------------------------
Commit e846cab46e864b08e4a810fd082c03673829ce0d in activemq-cpp's branch
refs/heads/3.9.x from [~tabish121]
[ https://git-wip-us.apache.org/repos/asf?p=activemq-cpp.git;h=e846cab ]
https://issues.apache.org/jira/browse/AMQCPP-598
Add null check
(cherry picked from commit 4754bb810f171f2a180b96c2a87b19afd1458016)
> CmsTemplate::createConsumer possible dereference of null pointer
> ----------------------------------------------------------------
>
> Key: AMQCPP-598
> URL: https://issues.apache.org/jira/browse/AMQCPP-598
> Project: ActiveMQ C++ Client
> Issue Type: Bug
> Components: CMS Impl
> Reporter: Daniel L. Robertson
> Assignee: Timothy Bish
> Priority: Minor
> Attachments: CmsTemplate.cpp.patch
>
>
> If `CMSTemplate::takeSession()` returns NULL in `CMSTemplate::execute`, it is
> dereferenced in `CmsTemplate::createConsumer`. As far as I could tell, this
> is a plausible scenario, but I could be wrong. It looks like
> `ProducerExecutor::doInCms` checks for a NULL session, but
> `ReceiveExecutor::doInCms` does not. If this is because it is known that the
> session cannot be NULL, please close this issue. It just looked a bit off to
> me.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
