For Queues, that's right - all messages are stored until consumed (once and
only once consumption).
For Topics, it's a little more complicated. It's easiest to think of it
this way: topics never store messages themselves; topic subscriptions may
store messages. Durable subscriptions do store mes
There could be another issue, but as long as the GC issue persists, it will
make tracking down any other problem very challenging. Attacking that first
seems like the best approach.
--
View this message in context:
http://activemq.2283324.n4.nabble.com/high-GC-activity-yields-a-high-percentage
Any update on this issue?
https://issues.apache.org/jira/browse/AMQ-5459
Regards,
Pankaj.
On Mon, Dec 1, 2014 at 6:16 PM, pankajtakawale
wrote:
> Hi,
>
> I faced following exception while testing ActiveMQ leveldb cluster.
> On this exception, master demotes itself to slave. New master faces sam
Selectors will make dispatching slightly slow*er*, but will it make it
*slow*? That depends on your setup, and no one but you will be able to
answer performance questions about your usecase, your configuration, and
your hardware. Try it and see; that's the only way you're going to know
whether yo
For queues, all messages will be stored, irrespective of whether there is
currently a consumer for them. That's how queues work.
For topics, I expect that the message would be discarded (topic messages
are delivered to all applicable subscribers, of which there are zero in
this case), but I don't
James,
To the point you made about whether LevelDB is production-ready, I think
the consensus from recent threads on this mailing list is that LevelDB in
5.10 (and also 5.11 snapshots, I think) still has more problems than most
of the people who've used it are comfortable with. Those bugs will ge
it is redundant when a datasource is specified. If there is no
datasource, it will use an embedded derby instance in the data
directory.
On 6 January 2015 at 09:54, James Green wrote:
> Is
> http://activemq.apache.org/maven/apidocs/org/apache/activemq/store/jdbc/JDBCPersistenceAdapter.html#setDir
JDBC Isn’t replicated, it picks the first master as the one that locks the DB
, but offers you another way of scaling (albeit slower) the security
of message persistence to say a commercial RDBMS or open source cluster.
It is also allows you to do master slave without a “san” - something many
Is
http://activemq.apache.org/maven/apidocs/org/apache/activemq/store/jdbc/JDBCPersistenceAdapter.html#setDirectory(java.io.File)
not redundant?
Looking at http://activemq.apache.org/jdbc-master-slave.html the example
configures a JDBC persistence adapter with both a data directory and a data
sour
Looking at http://activemq.apache.org/persistence.html and wondering about
building a cluster of brokers for high availability.
Seems JDBC was implemented quite some time ago and has been eclipsed by
KahaDB and now LevelDB but these are local only. There's something very
shiny and new about LevelD
Lets say I have topic: VirtualTopic.A and a Queue Consumer.B.VirtualTopic.A,
There are two consumer one subscribing to topic VirtualTopic.A and other on
queue Consumer.B.VirtualTopic.A both are using selector?
Will there be any problem because of using selectors?
Will selector make dispatching fro
11 matches
Mail list logo