Deadlock:
BrokerService[test] Task-1 is waiting to lock java.lang.Object@95cc0b which
is held by NettyWorker-thread-1
NettyWorker-thread-1 is waiting to lock java.lang.Object@126c1c8 which is
held by BrokerService[test] Task-1
Thread stacks
BrokerService[test] Task-1 [BLOCKED; waiting to lo
I finally moved away from the idea to use the session threads. Now my
listener schedules the handling of each message into a thread pool (see
SimpleMessageListenerContainer.setThreadExecutor)
In addition i use a different connection uri with new parameters:
alwaysSessionAsync=false makes sure tha
currently i'm using a CachedConnectionFactory from spring to limit the number
of open sessions.
together with the connection uri parameters this means that act
please note that if you are using activemq 5.5 there is a bug in the failover
mechanism that can keep your network bridges from beeing created.
see: https://issues.apache.org/jira/browse/AMQ-3575
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Failover-with-networkConnecto
so the behavior i am experiencing is (or for trunk was) a bug?
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Network-of-Brokers-dropps-messages-if-a-hop-fails-tp4403987p4406890.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
as far as i know there are no fixes. there only was a licence problem between
activemq and a dependency that was resolved.
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Issues-resolved-in-ActiveMQ-5-5-1-tp4397181p4404286.html
Sent from the ActiveMQ - User mailing list arc
the later. temp space is full.
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Network-of-Brokers-dropps-messages-if-a-hop-fails-tp4403987p4404040.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
I have a network of 3 brokers running but sometimes the brokers fail in a
rather unique and annoying way.
They still accept connections but otherwise stop to communicate with the
client (even those still connected).
For new connections created in java this means:
con = factory.getConnection(
I finally figured out what went wrong. because I'm using replication in my
cluster nodes the nodes use the failover transport for the network
connectors (see my config). Unfortunately this is buggy in this scenario
(see activemq jira https://issues.apache.org/jira/browse/AMQ-3575)
To fix this i wi
i'm trying to create a hypercube (6 servers in 3 nodes (master+slave)).
3 has connections to 1,2
2 has connections to 1,3
1 has connections to 2,3
but the activemq doesn't create bridges to all servers so it doesn't get
advisories from them. no advisories means no awarenes of consumers on these
s
My Config for Cluster-Node 03 (Replication-Master)
http://pastebin.com/9iztG67D
TTL is default (documentation says thats 1), no duplex
every cluster node uses pure master slave.
--
View this message in context:
http://activemq.2283324.n4.nabble.com/No-network-bridges-started-in-Network-of-broker
Hi,
i have a Network of 6 nodes in 3 datacenters, each one connected to every
other server.
upon startup some (currently 2) servers don't start network bridges for some
other servers. As a consequence those servers are unaware of consumers on
other nodes and don't distribute the messages they get
Hi,
does activemq currently support AMQP?
Last i head the guys at fusesource were working on it.
Greetings,
Alexander
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Support-of-AMQP-by-ActiveMq-tp4306276p4306276.html
Sent from the ActiveMQ - User mailing list archive at N
Hi,
today one of our activemq nodes required more than double the memory than we
configured for its jvm.
While this behaviour might not strictly be activmq related, maybe someone
can explain to me how a java application can "break out" of its configured
java heap.
Greetings,
Alexander
--
View th
I'm using a mdb as a consumer inside a jboss 6 with the activemq resource
adapter.
currently i experience a lack of consumer performance and wanted to solve
this by using more mdb instances (the server itself still has resources to
spare, so this should work).
In a blog post i found a note saying
When i use a wildcart when requesting the destination statistics from the
plugin i get one message for every destination that matches my wildcart.
thats great but how can i identify which destination the message is for?
so far i can only say that its a destination that matches.
Greetings,
Alexand
Hi,
currently i have a destination with several thousand messages pending.
When i try to connect a consumer to this destination the consumer doesn't
get any messages. instead i find a warning in the log:
2012-01-10 04:26:53,633 | WARN | Failed to register MBean:
org.apache.activemq:BrokerName=br
after reading through the documentation i found:
storeUsageHighWaterMark (a destination policy)
my idea is to configure a watermark of 95% for ">" and reconfigure it to
100% for "ActiveMQ.Advisory.>".
This way there should always be space for consumer and connection advisories
even when no normal
(my broker is running with deactivated producer flow control at the moment)
when a activemq broker gets flooded with messages or the consumer fails it
will stop accepting messages once certain (configurable) limits are reached.
In Broker Networks this effect can take down the whole cluster.
I'm c
i'm using pure master slave and the easiest way to be notified that the
master has failed is to listen to the slave for the inevitable: i am master
now advisory.
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Creating-a-Connection-to-an-activeMQ-in-Slave-Mode-tp4095183p418
short question because i just thought of it:
if a slave doesn't accept any connections while in slave mode:
how does the web console work for slaves?
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Creating-a-Connection-to-an-activeMQ-in-Slave-Mode-tp4095183p4185911.html
Hi,
i have two clients. one calls createTemporaryQueue and sends the name of the
queue to the second client. now they can communicate using this destination.
When client two (the one that used the name) calls connection.close() he
gets:
javax.jms.JMSException: A consumer is consuming from the te
Hi,
i have embedded an additional component into activemq by adding a bean
configuration to the activemq.xml config.
now when activemq is started a bean for my component is created and
initialized with the configured init-method. BUT when activemq is stopped
the destroy-method is not executed.
m
ueues?
is there a simple reason why the advisories are not consumed?
Thanks in advance,
Laures
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Broker-Flooded-by-unconsumed-Advisory-TempQueue-messages-tp3911512p3911512.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
i'm guessing at this point, but could it be that your temp storage/memory
storage for non persistent messages is exhausted? the default config of
activemq in this case is to block the message add until the message can be
added to the destination.
--
View this message in context:
http://activemq.2
so i have to manually set the memory and store limits by hand to something
more sensible and the problem will vanish?
currently i have a broker networks with 6 servers that are not talking to
the server with the consumer and the only new logmessage i can find is this.
--
View this message in cont
here it is:
http://activemq.2283324.n4.nabble.com/file/n3897243/amqconfig.xml
amqconfig.xml
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Don-t-understand-this-memory-error-tp3897082p3897243.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Hi,
2011-10-12 09:29:35,502 | INFO | TopicSubscription:
consumer=ID:myconsumer-54613-1318337363878-0:9:-1:2, destinations=2,
dispatched=100, delivered=795093, matched=2, discarded=0: Pending message
cursor
[org.apache.activemq.broker.region.cursors.VMPendingMessageCursor@52d5d9c5]
is full, temp u
could
at least try to do something (like adding consumers at unexpected peaks and
such).
Greetings,
Laures
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Advisory-when-nearly-full-tp3824589p3824589.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Hi,
i'm using a broker network of 6 servers. All servers have a hardcoded
connection to every other server (hyptercube) the connections are not
duplex. Advisory topics are on and the servers authenticate using a super
user.
My consumer is connected to server A
Currently messages that are send to
Hi,
i'm currently setting up an ActiveMQ Broker Network and have some Problems
with User Authentication.
My User (TestClient) has full read write and admin rights on the queue
tracking.test.message.
The Queue is not created at startup time, thats done on the fly. Since the
User has admin rights,
Default configuration with STOMP until:
In the end i want to inject my own authorization map into the default
authorizationPlugin.
PS: if i deactivate the validation i get the exception i described here:
http://activemq.2283324.n4.nabble.com/Configuration-of-Custom-Plugins-wi
The Documentation says its possible to configure your own plugins like that:
I tried this with on ActiveMQ 5.5 the following way:
I am aware that I'm not configuring an authentication map, this was only a
test. In the end i want to configure another bean tha
Hi,
i just tried to install the authenticationPlugin with my map by defining the
plugin with a bean-tag inside the plugins tag.
The result (with activemq 5.5) is
org.xml.sax.SAXParseException: cvc-complex-type.2.4.a: Invalid content was
found starting with element 'bean'. One of
'{"http://activ
Hi,
i implemented my own Authentication Map class and want to configure ActiveMQ
so it uses it. Can anybody explain how i can do that?
Greetings,
Alexander
--
View this message in context:
http://activemq.2283324.n4.nabble.com/Integrate-own-Authentication-map-tp3579459p3579459.html
Sent from th
Hi,
I'm randomly getting the problem that my ActiveMQ doesn't accept connections
on the configured stomp+nio port. There is no log or anything. The admin
panel still works i can still send messages with it, openwire still works...
The problem is resolved after a restart but on shutdown i get:
Ca
Hi,
for my MessageDrivenBean inside of a JBoss6 I'm using the activemq resource
adapter. At runtime i need to change the adapters ra.xml. JBoss is smart
enough to redeploy the adapter when this happens, but the mdb looses the
connection to the broker and doesn't get a new one.
Instead my log is
Hi,
I'm currently load testing the native REST Api of activemq on our hardware.
I'm using pylot to generate several users.
Some time into the test i start to get:
WARN | Error for /rest/message/FOO/BAR
java.lang.OutOfMemoryError: unable to create new native thread
at java.lang.Thread.st
38 matches
Mail list logo
