Here comes more performance test information:
Without transaction, I can perform the flow about 500 times per second.
With transaction, it's only 20 times per second.
At 2011-11-29 09:35:30,lzr wrote:
Thanks a lot for your timely response!
I'll try it again following up your advice.
I make fur
I make 3 threads in my program and each maintain a Connection.
There are four queues in my flow, two for request and two for response.
At 2011-11-28 19:18:27,"Gary Tully" wrote:
>are you already caching connections? That will help a bit.
>
>What version are you on? there was some contention ov
How accurate is this document wrt ActiveMQ 5.5.1? Is there a better place to
find relevant details about thread usage/meaning/configuration?
http://fusesource.com/wiki/display/ProdInfo/Understanding+the+Threads+Allocated+in+ActiveMQ
I see threads with names like:
ActiveMQ Transport tcp:...
Act
JIRA should have all the detail:
https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12311210&version=12315621
On 28 November 2011 18:49, joe7pak wrote:
> Hi -
>
> I'm unable to find the release notes for ActiveMA 5.3.1. Does anyone have
> a link? If I do it through the ActiveMQ w
Turned out to be a permission problem in my package, the top level
directory had permission to only allow root to access it so once I stated
drilling down I was able to pin point it.
-Chris
On Mon, Nov 28, 2011 at 10:14 AM, Christopher Johnston
wrote:
> I actually got this working on Friday by c
Hi -
I'm unable to find the release notes for ActiveMA 5.3.1. Does anyone have
a link? If I do it through the ActiveMQ website, I just get a "The selected
version does not exist" message. I've googled, I've downloaded the
distribution ( no love ), I've tried non-ActiveMQ links. I can't find the
On Nov 28, 2011, at 3:34 PM, Ishitori wrote:
> I've encountered a problem of hanging up of the producers and I found that it
> is related to "Producer Flow Control". I made a quick fix via disabling
> producerFlowControl at all. But now I wonder what should I do, because I
> don't really want pro
I actually got this working on Friday by changing to use root, so I am
thinking there is a permission problem somewhere. I need to do a bit more
testing today to see where or what file it can not access.
Strangely I had no problems getting this going on Fedora 14, but on Fedora
10 I had issues.
Is your wrapper.conf setup correctly?
Does the account that ActiveMQ is being started under have read access to the
ActiveMQ installation?
What OS are you running on?
Torsten Mielke
tors...@fusesource.com
tmie...@blogspot.com
On Nov 25, 2011, at 5:43 PM, Christopher Johnston wrote:
> The cl
Hello,
Have seen these as well in the past.
This should all come from your OS, not from ActiveMQ. Its tcp level
retransmission of tcp packets. Higher network layers should not even see these.
Typically this happens on dodgy network connection, but AFAIK can also happen
if your connection crosse
I've encountered a problem of hanging up of the producers and I found that it
is related to "Producer Flow Control". I made a quick fix via disabling
producerFlowControl at all. But now I wonder what should I do, because I
don't really want producers to use all available disk space.
The questions
are you already caching connections? That will help a bit.
What version are you on? there was some contention over an internal
destination map that could be causing some trouble, a few periodic
thread dumps of the broker would help identify if that is the problem.
see: https://issues.apache.org/j
thanks
--Kaustubh
On Mon, Nov 28, 2011 at 12:45 PM, Yuvaraj Vanarase <
yuvaraj.vanar...@synechron.com> wrote:
> Sender does have timeout. "ActiveMQQueueSender"
>
> Regards,
> Yuvaraj
>
> Yuvaraj Vanarase,
> Lead Technology - Software
> Phone: +91.20.40262000 Ext 2305|Mobile: +91.9850818870 |
> h
13 matches
Mail list logo
