Use CACHE_CONSUMER in Spring Message listener container, this will not create
session and consumer everytime its polls.
Adam Lewandowski wrote:
>
> At what level does the redelivery policy get applied? I've got a
> transactional consumer (Spring MDP) that is performing a rollback and I
> a
I see u are using 4.1.1 and we are using DMX storage(SAN) to persist data and
its very fast.
i see that u mentioned that performance degrades overtime( is it mins or
hours or days?). What is ur throughput rate and how much memory did u
configure for MemoryUsageManager?
naga007 wrote
which version of activemq are u using and Can u post ur activemq.xml file?.
Anthrope wrote:
>
> Hello,
> Our application is essentially a pipeline of processing engines that
> communicate via JMS queues, and so far, ActiveMQ has been the JMS
> implementation of our choice. In our applic
I would like to know if there is any side affects if the checkpoint intreval
is changed from 5 mins to some thing less so that full journal recovery
incase of abnormal failover is speeded up?.
--
View this message in context:
http://www.nabble.com/checkpoint-interval-tf3729094s2354.html#a1043774
When running under master slave mode in 4.1.1, the slave broker has a thread
that periodically wakes up to aquire a lock. It looks like when it it tries
to aquire a lock it doesn't close the file handles properly and eventually
runs out file handles.To reproduce this problem in Unix, you can run "
Currently iam running activemq version 4.1.1 in Master Slave mode. When i
failover from master to slave it seems like some of the messages are being
redelivered. Is this a known issue or any work arounds possible.
Note: My consumers and producers use AUTO_ACKNOWELDE MODE and No
Transactions.
-
just backported the code to the 4.1 branch so it should appear in
> tomorrows nightly build. (Or if you're really keen you could try build
> the 4.1 branch...)
>
> http://activemq.apache.org/source.html
> https://svn.apache.org/repos/asf/activemq/branches/activemq-4.1/
&g
.
James.Strachan wrote:
>
> On 3/14/07, naga007 <[EMAIL PROTECTED]> wrote:
>>
>> Thare are no errors, but both become the master.
>
> Could you post the log please? I wonder which jdbc driver is being
> used (we try to auto-detect the driver and use the correct J
Thare are no errors, but both become the master.
James.Strachan wrote:
>
> Is there any kind of error in the log; or do both brokers just become
> the master?
>
> On 3/14/07, naga007 <[EMAIL PROTECTED]> wrote:
>>
>>
>> March 8th and used the fo
March 8th and used the following url
http://people.apache.org/~chirino/activemq-4.1.1-RC1/maven2/org/apache/activemq/apache-activemq/4.1.1/
James.Strachan wrote:
>
> On 3/14/07, naga007 <[EMAIL PROTECTED]> wrote:
>>
>> Iam using mysql 5.1 and it seems like master s
Iam using mysql 5.1 and it seems like master slave lock issue is still not
fixed on 4.1.1 latest snapshot. would appreciate any feedback from activemq
guys.
--
View this message in context:
http://www.nabble.com/Master-Slave-locking-with-MYSQL-is-not-working-with-4.1.1-tf3399897s2354.html#a94675
In my application i was using Master-Slave configuration with JounaledJDBC
persistence with MySQL as secondary storage. Under high load when I shutdown
MasterBroker, the Slave Broker takes a Long time to Startup because of time
spent in recovering messages from journal files. I would like to know
nable async sending...
>
> http://activemq.apache.org/async-sends.html
>
>
> On 3/5/07, naga007 <[EMAIL PROTECTED]> wrote:
>>
>> Let me explain in more detail.
>>
>> As i said, this application is a Framework for SOA based
>> Applications(K
across
mutliple destinations, there seems to be some concurrency issues also i
didnot see much CPU spikes, but i noticed after few requests go through, i
see pauses and thread dumps show that threads are struck on
oneWayRequestor.send() method.
James.Strachan wrote:
>
> On 3/5/07, n
We are noticing significant performance degradation with persistence
turned on, when the no of concurrent users is more than 50. Our
application(which is like is SOA Framework) is using active 4.1 running on
Linux Redhat with 2 Cores, 8GB Ram and using SAN for journal files storage.
Our curre
turn on jmx-connector and use jconsole to watch the queue statistics.
Gaurav Hariani wrote:
>
> Is there a way for a client to determine which broker it is connected to
> (when using a failover URI)?
>
>
>
>
--
View this message in context:
http://www.nabble.com/Failover-URI%3A-Connec
which version of activemq are using, iam using 4.1 and i could failover also
send me ur config.
naga007 wrote:
>
>
> use failover:(tcp://host1:61616,tcp://host2:61616) uri format for the
> client.s
>
>
> Gaurav Hariani wrote:
>>
>> I'm trying to
use failover:(tcp://host1:61616,tcp://host2:61616) uri format for the
client.s
Gaurav Hariani wrote:
>
> I'm trying to setup a cluster of Brokers (standalone) which clients can
> connect to (ideally I would like them to be discovered)
> listening to a TOPIC. In the case of a broker failing, th
change this in includedestinations and try to see if t works.
David Borja-2 wrote:
>
> Hi ...
>
> This is my scenario :
>
> I hava a broker called "dllo-aborja" and other one called "localhost2".
>
> I need that all messages that i put on some queue in dllo-aborja broker,
> forward to l
false ?
>
> On 17 Feb 2007, at 01:48, naga007 wrote:
>
>>
>> I have complex application which has a producer and multiple consumers
>> consuming messages from a Queue. The application runs on two different
>> servers connects to a activeMQ broker which has dema
I have complex application which has a producer and multiple consumers
consuming messages from a Queue. The application runs on two different
servers connects to a activeMQ broker which has demand forwarding bridge to
another activeMQ broker running on a different server. When running a simple
loa
21 matches
Mail list logo
