[
https://issues.apache.org/jira/browse/ARTEMIS-2082?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16611207#comment-16611207
]
ASF subversion and git services commented on ARTEMIS-2082:
----------------------------------------------------------
Commit e065e3e960ac8cc6505a71ec4f67b58b4ffb5990 in activemq-artemis's branch
refs/heads/master from [~tabish121]
[ https://git-wip-us.apache.org/repos/asf?p=activemq-artemis.git;h=e065e3e ]
ARTEMIS-2082 Reset buffer valid flag after re-encoding the message
Once a re-encode of the message is done the buffer is not being marked
as valid and so subsequent checks on the buffer are all assuming the
message data is not valid and re-encoding over and over. This can lead
to poor performance in some cases and corrupted data in others.
> AMQP: Message corruption possible on redelivered message that have moved
> ------------------------------------------------------------------------
>
> Key: ARTEMIS-2082
> URL: https://issues.apache.org/jira/browse/ARTEMIS-2082
> Project: ActiveMQ Artemis
> Issue Type: Improvement
> Components: AMQP
> Affects Versions: 2.6.3
> Reporter: Timothy Bish
> Assignee: Timothy Bish
> Priority: Major
> Fix For: 2.7.0
>
>
> In some cases when a message moves from one Queue to another either by
> expiration or other policy defined reason the messages that are sent to
> clients on the new location of the message can become corrupted if the the
> client consumes but does not settle the message. This can occur for instance
> with a transacted consumer that continually rolls back the message and then
> consumes it again. The message eventually becomes corrupt as the broker's
> internal AMQPMessage object loses track of the location of various portions
> of the message data and re-encodes the header to add a delivery count etc.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
