Calling beforeMarshall on messages when they async stored before the
store task is run and before consumer dispatch to prevent two threads
from trying to mutate the message state at the same time.
(cherry picked from commit b9f9f03829)
Update for some added thread safety. Adds method healthStatus that will
regenrate the status from the healthList data which is more intuitive
than the getCurrentStatus which doesn't update state and requires
periodic calls to healthList to capture current metrics.
(cherry picked from commit 19fd084a83)
Don't log the pull timed out messages that might still be in the
unconsumed list during rollback of unconsumed messages
(cherry picked from commit aaecdff8b4)
Expose configuration on the persistence adapter level.
Double the test timeout and increase the compaction frequency to account
for very slow CI boxes.
(cherry picked from commit a9521dcebf)
Moving the bytes copy to the parent Message class to solve this this
issue for all message types as that is the root cause
(cherry picked from commit e69c2cbad6)
Disabling async queue message store when persistJMSRedelivered is turned
on for a destination. That flag will cause a sync update later on
dispatch which can cause a race condition if the original message add is
processed after the update. This can cause a duplicate message to be
stored.
(cherry picked from commit b2327db3b7)
Moving clearedMarshalledState execution to the async listener on an
async add to the message store. This is necessary to make sure this
logic doens't execute until after the message is marshalled for the
store.
(cherry picked from commit 75990ef14a)
Synchronizing ActiveMQText message on state changes for the content and
text fields so that they are always changed together. This will prevent
race conditions where data can be lost when using concurrent store and
dispatch.
(cherry picked from commit e0c5499964)
Fixing copy method of ActiveMQTextMessage to prevent a race condition
when concurrent store and dispatch is used with Queues
We may need explicity synchronization between the text and content fields
in the future if other issues pop up
(cherry picked from commit ea09159a40)
Fixing ReloadableProperties so that the groups file will be properly
reloaded on file modification. Added a test to verify.
Thanks to Nanchang Yang for providing the fix.
(cherry picked from commit 2788bd5584)
Update the threads created by the SelectorManager to be daemon threads.
(cherry picked from commit e2b4ca2c59)
Conflicts:
activemq-client/src/main/java/org/apache/activemq/transport/nio/SelectorManager.java
Fixing the removal logic on virtual destination remove inside of
Advisory Broker to clean up virtual destination maps properly. Added a
test to verify. Also added new debug logging to help track down any
future issues.
(cherry picked from commit a2781e3966)
Reworking patch so that the http trace method is also turned off by
default for the HttpTransport, besides just for the Websocket transport.
Also added SSL tests for both transports.
(cherry picked from commit 473b3284d4)
Disabling the HTTP trace method on the websocket port by default. It
can be enabled by setting the parameter "http.enableTrace=true" on the
connection uri if it is needed for debugging.
(cherry picked from commit 4fb8083977)
Fixing CronParserTest so that it checks 24 hour format properly when
using a timezone shifted by 30 minutes
Thanks to Jamie Goodyear for the patch
(cherry picked from commit 0f445a535f)
Fixing X-Frame-Options header so that is applied for all content served
by Jetty. The previous patch wasn't correct because it only applied it
to Servlets and JSPs and not static content. This also reverts AMQ-6113
(cherry picked from commit 24ad367785)
Computing messageSize for a durable subscription in KahaDB now runs much
faster (n vs n^2) which is noticable when there are a large number of
pending messages for a durable subscription.
(cherry picked from commit 25ff5699f1)
Christopher L. Shannon (cshannon)
Timothy Bish
Brian D. Johnson
gtully
Robert Gemmell
Jeff Genender
Claus Ibsen
Hiram Chirino