JMS Transformer performance improvements and bug fixes
Trim unnecessary code and improve overall performance of the JMS
Transformer codecs. Remove legacy Qpid JMS client related code from the
transformer as these are no longer supported. Fix outgoing message that
do not match the structure of the incoming message that created them such
as message with had only a body being sent out with Headers and message
Properties.
The temporary store will now delete the old temp directory on start up
if lazyInit is true instead of waiting for the store to initialize to
clear up space. This prevents space on the disk from being wasted with
old data if the temp store isn't initialized
Modifying patch so that only stores that persist the noLocal flag will
check if this flag has changed to prevent a subscription from being
deleted by mistake
The test I'm adding was back ported from Artemis.
It will validate if the ACKs are nacked in case of a connection.close();
To avoid a situation where the TX would sit on a Transaction Resource Manager somewhere like an XID.
When a nolocal durable consumer reconnects the new connectionId is properly captured for
the NoLocal expression so that nolocal works on reconnect. Also fixed
the detection of the nolocal value changing on consumer connect.
Adds ability to not settle accepted messages on the client to enable
creation of tests that are equivalent to the AmqpNetLite client's
transaction tests which hold settlement and expect the resource to
handle it on successful discharge.
Adds some transaction tests ported from AMQP .NET client with some
variances based on the way the test client works and limitations in the
brokers handling of Transacted sends.
Adds support for doing sends and receives that are enrolled in a
transaction created in a session other than the session that created the
sender or receiver. Adds some tests that show this in action.
The test client can allow for quick tracing of the frame traffic via a
call to setTraceFrames on the client or connection instance before
connection to the remote. This allows for tests to easily switch on /
off tracing. The log4j.properties is also updated to output frame
tracing with the URI option is put on the AMQP transport or the client
value is enabled.
I've made a few minor test changes and added a couple more cases. Under
heavy CPU load I'm able to get test,
testReceiveMessageAndRefillCreditBeforeAcceptOnQueue to fail on the
second receive call where it should get the second message since it
granted credit.
Properly copying wire format options map before applying to the
transport when using auto transports so that the options do not get
cleared and will be used for all future connections
Changing the nio+ssl transports to trigger a serviceRead after start up
to prevent blocking. The prevents the channels from not reading in
certain cases, most notably with the auto+nio+ssl transport when used
for a network bridge. Also added a couple tests and changed a network
bridge test to test out auto+nio+ssl.
Christopher L. Shannon (cshannon)
Krzysztof Sobkowiak
Dejan Bosanac
Timothy Bish
Clebert Suconic
gtully