What I forgot to ask is: Any ideas?
Thanks in advance!
Uli
--
View this message in context:
http://activemq.2283324.n4.nabble.com/discovery-multicast-multihomed-tp4669362p4669405.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.
Which is the best protocol for activemq networks of brokers.tcp/ip, failover,
Discovery ..
It is important for me.Cuz For ex. I have 20 brokers.10 of them on cluster1
and 10 of them on cluster2.
If I put a firewall between these two cluster does it work on performance
--
View this message in c
I also thought so, but creating the ConnectionFactory repeatedly has caused
the issues for me.
It is very easy to reproduce. Just put the creation of ConnectionFactory in
a loop and send
messages from the connections obtained. You will be able to reproduce if my
observation is correct with ActiveM
Just tried test case on 5.7.0, works fine.
Maybe you can check your JMX console and see whether the broker is
dispatching, eg, can you see the "inflight" count go up correctly?
You could also try setting prefetch to 0 on your consumer to force the
consumer to pull a message. That's probably not th
core jar manifest says 5.7.0
[]
Leo
On Tue, Jul 16, 2013 at 5:06 PM, Christian Posta
wrote:
> Trunk.
>
> What version is in tomEE 1.5.2?
>
>
> On Tue, Jul 16, 2013 at 3:56 PM, Leonardo K. Shikida wrote:
>
>> I am using activemq from tomEE 1.5.2
>>
>> probably not the latest
>>
>> what version
I am using activemq from tomEE 1.5.2
probably not the latest
what version are you using?
[]
Leo
On Tue, Jul 16, 2013 at 4:51 PM, Christian Posta
wrote:
> This unit test works:
>
>
> @Test
> public void remove() throws JMSException, InterruptedException {
>
> boolean messageCo
Trunk.
What version is in tomEE 1.5.2?
On Tue, Jul 16, 2013 at 3:56 PM, Leonardo K. Shikida wrote:
> I am using activemq from tomEE 1.5.2
>
> probably not the latest
>
> what version are you using?
>
> []
>
> Leo
>
>
> On Tue, Jul 16, 2013 at 4:51 PM, Christian Posta
> wrote:
> > This unit tes
This unit test works:
@Test
public void remove() throws JMSException, InterruptedException {
boolean messageConsumed = false;
Connection connection = null;
Session session = null;
try {
ConnectionFactory connectionFactory = new
ActiveMQConnect
Hi
I've tried
consumer.receive(1000)
without success
[]
Leo
On Tue, Jul 16, 2013 at 4:25 PM, Christian Posta
wrote:
> Try giving a brief moment between creating your consumers and the call to
> receiveNoWait():
>
> MessageConsumer consumer =
> session.createC
Take a look at the documentation:
http://activemq.apache.org/networks-of-brokers.html
If you're concerned about firewalls, etc, try out the http transport and
assign it to a well-known non-firewalled port.
On Mon, Jul 15, 2013 at 4:53 PM, turkuaz07 wrote:
> I want to learn how to communicate
Try giving a brief moment between creating your consumers and the call to
receiveNoWait():
MessageConsumer consumer =
session.createConsumer(queue, "JMSMessageID='" +
jmsMessageID + "'");
Message message = null;
Thread.sleep(500);
You should write a test that shows this is the case.
Creating the connection factory in a script like you describe shouldn't
have anything to do with messages becoming queued up in a destination.
On Tue, Jul 16, 2013 at 2:34 PM, pkorwar wrote:
> Yes, after 2 days of testing, the messages are no
Yes, after 2 days of testing, the messages are not getting stuck in a pending
state.
So, this means that, creating and destroying the ActiveMQConnectionFactory
repeatedly is
creating some problems. So, this should be either documented or fixed so
that
users won't get caught in a situation where it
Hi
I am trying to remove (consume) a message from the queue, without success.
Here's my code and the output
public boolean remove(String jmsMessageID) throws JMSException{
boolean messageConsumed = false;
Connection connection = null;
Session session = null;
try {
ConnectionFactory connectionF
I want to learn how to communicate two brokers and if I put one of the broker
behind the firewall, Does it work on performance
--
View this message in context:
http://activemq.2283324.n4.nabble.com/communication-of-two-brokers-tp4669326.html
Sent from the ActiveMQ - User mailing list archive at
Hi Tim,
Thanks for taking the time to respond. I think you meant to write that the
http: and https: protocols are NOT currently supported in NMS, correct?
Although I'm not familiar with the NMS internals, I'm happy to have a look
at what it would take to implement them. Can you point me in the
On 07/16/2013 10:15 AM, prgtrdr wrote:
Hi Tim,
Thanks for taking the time to respond. I think you meant to write that the
http: and https: protocols are NOT currently supported in NMS, correct?
Although I'm not familiar with the NMS internals, I'm happy to have a look
at what it would take to
>I don't think there is such algorithm. Its best to run load test with your
>estimated settings and fine tune them if necessary.
I agree. However, what I am looking for specifically is guidance in
the form of what diagnostics to use (other than raw throughput). I
cannot 'fine tune' something unle
I've set up this well known configuration on three Windows 7 machines:
step 1 machines running
Machine 1 listening VLAN57
Machine 3 listening VLAN62
Machine 2 listening on both VLANs
step 2 start all brokers
2 & 3 connect
1 & 2 don't connect
step 3 disconnect 2 from VLAN62
2 & 3 disconnect
1 &
On 07/16/2013 09:24 AM, prgtrdr wrote:
I've since learned that the sandbox restricts the use of all communication
protocols except http: and https:. Is there a way to get NMS to communicate
with ActiveMQ via those protocols?
--
View this message in context:
http://activemq.2283324.n4.nabble.
I've since learned that the sandbox restricts the use of all communication
protocols except http: and https:. Is there a way to get NMS to communicate
with ActiveMQ via those protocols?
--
View this message in context:
http://activemq.2283324.n4.nabble.com/NMS-NET-4-0-Sandbox-Security-Question
ActiveMQSession creates a new local transaction because there isn't an XA trx
in the trasactioncontext.
Logs:
/09:22:32,326 DEBUG [org.apache.activemq.TransactionContext] Start: <
formatId=131076, gtrid_length=29, bqual_length=28,
tx_uid=0:c0a80ad8:126a:51e5393b:43, node_name=1,
branch_uid=0:f
I found the issue just this time. It worked fine for about two month with the
ActiveMQ M/S cluster.
I've thought of the network glitch; What I feel strange most is there is no any
error/exception on both the ActiveMQ client side and server side.
I went through the related codes and found there is
Is this something that happens all the time? Maybe your network had a
glitch and didn't deliver the response back to the client.
On Tue, Jul 16, 2013 at 8:18 AM, lzr wrote:
> I just check the configuration file, actually I've disabled the flow
> control:
>
>
>
>
>
I just check the configuration file, actually I've disabled the flow control:
Yes, on the first time PFC kicked in you would see the log messages. Not on
subsequent times.
On Tue, Jul 16, 2013 at 8:01 AM, lzr wrote:
> Based on my experiences if the issue caused by the producer flow control
> configuration, the logs will have some warning messages, right? In fact I
> didn
I've tried to reproduce this without calling the stateless ejb but it works
well.
The ActiveMQSession class should retain the original transactionContext, as
you mentioned.
I'll create an issue.
--
View this message in context:
http://activemq.2283324.n4.nabble.com/ActiveMQ-XA-JBoss-6-1-problem
Based on my experiences if the issue caused by the producer flow control
configuration, the logs will have some warning messages, right? In fact I
didn't found any warning or errors in the activemq.log.
Anyway, I'll check the configuration files right now.
thanks lot,
Zhuran.
At 2013-07-16 19
btw: in the jboss logs or on the console, do you see:
WARN [org.apache.activemq.ra.ServerSessionImpl:0]
(default-threads - 2) Local transaction had not been commited.
Commiting now.
On 15 July 2013 21:10, gastonscapusio wrote:
> Hi,
> We have some XA problems using ActiveMQ ra within Jbos
This means the client didn't' receive a response from the broker.
Did the broker producer flow control this client? ie, was the destination
to which the client sending full?
On Tue, Jul 16, 2013 at 7:23 AM, lzr wrote:
> Dear all,
>
>
> I found a one of the ActiveMQ thread got stuck unexpectedly
Dear all,
I found a one of the ActiveMQ thread got stuck unexpectedly for several days.
Customer was very angry to the ActiveMQ for the bad behavior. Did anybody
experienced this issue? It would be much appreciated if any clue!!!
My application is normal Java JMS application like the following:
this is a bug for sure. I came across the same thing yesterday and am
investigating. I think the problem can also happen independent of the
call to a stateless ejb. It is an issue with the session auto ack
level and the absence of a transaction context after an abort due to a
timeout.
It looks like
Yep good catch!
On Tuesday, July 16, 2013, Marc Ende wrote:
> Aaaarg... Okay I've found the point myself:
>
> Within the docs: *"All the broker nodes that are part of the same
> replication set
> should have matching brokerName XML attributes."*
>
> Now I've got it...
>
> Thanks for your help, Ch
On 07/15/2013 08:55 PM, James Carr wrote:
I would raise an issue on the original github project so it can get added.
That is certainly a good idea...
node-amqp should be a one stop place for working with amqp rather than
having to make a bunch of modules for each concrete implementation.
..
On 07/15/2013 06:46 PM, razeone wrote:
Thanks a lot Gordon, your answer gave me a very good alternative to STOMP and
it looks like works fine. Now I have been connected, the question I have now
is:
How do I establish content based routing? I mean, pass selector headers, it
is possible with AMQP?
I don't think there is such algorithm. Its best to run load test with your
estimated settings and fine tune them if necessary.
memoryUsage should be below the JVM heap size to leave enough heap for other
broker tasks that storing messages. How much there is needed for other tasks
depends on the
We have certainly seen other clients having 2000+ clients into one broker. That
is possible.
You need to tune your broker accordingly (JVM heap, threading, broker
systemUsage) and I recommend using the nio transport connector in your broker
configuration.
There is an ActiveMQ broker tuning gu
37 matches
Mail list logo
