I saw this bug right before my week vacations (along with a performance
improvement I made)
I raised a Jira and justin Bertram made a fix. But there is something that
still needs a fix apparently.
I saw this using the chat example under examples/openwire. (I do hope it's the
same issue you ar
On 08/21/2015 04:24 PM, mfan wrote:
> I am new to the ActiveMQ-CPP, here below was the problem I used the example
> from here - http://activemq.apache.org/cms/example.html.
>
> I tested it using a simple code in below for producer. About right after the
> producer finished reading, the customer cod
I am new to the ActiveMQ-CPP, here below was the problem I used the example
from here - http://activemq.apache.org/cms/example.html.
I tested it using a simple code in below for producer. About right after the
producer finished reading, the customer code stops without finish, do
anybody know what
If you could, that would be absolutely fantastic!!
Many thanks,
Steve.
Clebert Suconic wrote
> I will fix this on Monday. Ok?
>
> -- Clebert Suconic typing on the iPhone.
>
>> On Aug 21, 2015, at 12:45, slew77 <
> stephen.lewis77@.co
> > wrote:
>>
>> Hi Clebert,
>>
>> Unfortunately I get
I would modify my last statement to add that there is one consumer that still
processes data while everything else is paused. What is unique about that is
that all messages published to that queue are delayed (scheduled) messages
and I am guessing that they are handled differently internally in AMQ
I will fix this on Monday. Ok?
-- Clebert Suconic typing on the iPhone.
> On Aug 21, 2015, at 12:45, slew77 wrote:
>
> Hi Clebert,
>
> Unfortunately I get similar results with the master branch:
>
> 17:20:02,800 WARN [org.apache.activemq.artemis.core.server] AMQ222067:
> Connection failure
Hi Clebert,
Unfortunately I get similar results with the master branch:
17:20:02,800 WARN [org.apache.activemq.artemis.core.server] AMQ222067:
Connection failure has been detected: AMQ119014: Did not receive data from
/127.0.0.1:55098. It is likely the client has exited or crashed without
closin
Just copied the broker configuration file.
It's just a classic configuration and the only change we made is
1. The broker has persistent enabled because we have a few durable messages.
2. But for most of the messages, we don't want them go to the Temp Store, so
we used the memoryCursor.
3. To pr
We haven't got a yokuit setup yet. Only using the JMC and MAT to found this.
We'll try if possible.
Just from the 5.11.1 source code, I see the callbacks list is being added
from this method:
Usage.notifyCallbackWhenNotFull()
Is there a possible guess how it can be called 200 million times with
Thank you Tim ! this clarification helps me a lot in my current development
effort.
By the way, your book "Apache ActiveMQ Messaging Application Development
How-to" is the perfect
complement for my other ActiveMQ books. great job !
--
View this message in context:
http://activemq.2283324.n4.n
We have fixed this on master.
1.0.0 was the first release won openwire. We are doing a lot of work on open
wire for next ones.
Perhaps you could build master and try it out before we release ?
-- Clebert Suconic typing on the iPhone.
> On Aug 21, 2015, at 06:37, slew77 wrote:
>
> Hi,
>
Hi,
I have setup a client (Jmeter) using an Openwire connection to an Artemis
broker.
With one thread continually publishing and consuming messages, everything is
fine for 60 seconds, then the broker issues the messages:
11:31:42,665 WARN [org.apache.activemq.artemis.core.server] AMQ222061:
Cli
On 08/21/2015 10:23 AM, yang.yang.zz wrote:
> Hi Tim,
>
> We are using non-durable messaging. Do you think the fix can cover this
> case?
>
>
>
>
> --
> View this message in context:
> http://activemq.2283324.n4.nabble.com/org-apache-activemq-usage-MemoryUsage-consumes-95-of-the-memory-tp4701250p4
Hi Tim,
We are using non-durable messaging. Do you think the fix can cover this
case?
--
View this message in context:
http://activemq.2283324.n4.nabble.com/org-apache-activemq-usage-MemoryUsage-consumes-95-of-the-memory-tp4701250p4701282.html
Sent from the ActiveMQ - User mailing list archiv
Pretty much everything grinds to a halt... all the consumers stop processing.
Each 'application' has multiple consumer threads pulling from multiple
queues, they all pause. and if there are 2 application instances they
both pause. It's as if there is a single thread feeding all consumers and i
Sounds a lot like this one I fixed for 5.12.0:
https://issues.apache.org/jira/browse/AMQ-4814
On 08/21/2015 09:45 AM, Christopher Shannon wrote:
> The callbacks are in the parent class, Usage. They are used to signal when
> the Usage is no longer full.
>
> On Fri, Aug 21, 2015 at 1:02 AM, Tim Bai
The callbacks are in the parent class, Usage. They are used to signal when
the Usage is no longer full.
On Fri, Aug 21, 2015 at 1:02 AM, Tim Bain wrote:
> I see no references to a Map or to a field called callbacks in
>
> http://www.grepcode.com/file/repo1.maven.org/maven2/org.apache.activemq/a
I don't see an obvious causal link between those two, though I haven't
looked at that code (my usual trick of going to GrepCode didn't work; maybe
GC doesn't index Scala code?) so I can't say for sure. Unless you see a
causal link I'm missing and want to explain it?
Hopefully someone who knows th
James,
Thanks for the additional information about the test you ran. Would you
mind re-running it with a storeUsage limit of 2 MB? I'd like to test
whether PFC isn't kicking in at all, or just isn't respecting the
per-destination limit you set via the policy. I know you're interpreting
the wiki
When this occurs, are some consumers "paused" while others merrily churn
through their backlog, or are all consumers "paused"?
Is the rate of publishing messages to each message group relatively
uniform, both across time and from group to group?
JConsole u s an easy way to get a rough idea of whe
Correct: putting messages in a group ensures that they will be consumed in
order by the same single consumer, which is not what you want.
On Aug 21, 2015 3:34 AM, "Straal" wrote:
> I adapted the code from the examples on the Apache website, but this is for
> single consumer.
>
> Removing the foll
You can tell if a consumer is slower than the rest via JMX. Look at each
subscription for the topic and see if any of them has lots of pending
messages while the rest do not. If you discover that to be the case,
configure a slow consumer abortion strategy to kick them off and throw away
their pen
We have a finite set of message groups - 20 (0-19)
We are monitoring the consumers via JMX and at one time were having issues
with memory leaks and GC operation that have since been solved. Now
everything looks good on the consumer side.
There are variations in how long messages take to process -
I have not been seeing anything in the broker logs indicating that PFC is
occurring and I have also tried adding to disable it entirely without effect.
I'm not currently seeing any detail in the GC operation, only that there is
no obvious sign of memory issues when this happens. I'll see if we c
I adapted the code from the examples on the Apache website, but this is for
single consumer.
Removing the following line solved my problem:
This apparently makes the messages belong to the same group and hence they
stay together.
Thanks for listening.
--
View this message in context:
htt
I want to create a system where ActiveMQ is used with 1 producer and multiple
concurrent consumers.
The intention is to have parallel processing by the consumers.
Requirement is that the Stomp library for .NET (Apache.NMS.Stomp) is used.
I thought using multiple sessions, setting AutoAcknowledge
I think I meant to ask if this can be mapped to a known EIP. Do we need to
identify the timeouts ourselves ? That wouldn't be easy without a timer. Isn't
it ?
Thanks,
Mohan
From: Radhakrishnan, Mohan (Cognizant)
Sent: Wednesday, August 19, 2015 8:01 PM
To: users@activemq.apache.org
Subject: Rec
I noticed you use message groups. What does your consumer logic do? Is it
possible that some messages take longer to process?
Do you have a finite set of message groups? How do you start your
consumers? Do you start them when the queue already contains messages?
Are consumers fixed? Or do they co
28 matches
Mail list logo
