Through some trials found that its not enough to increase the Executors thread 
count(maxThreads for Executor element) but also 
need to increase the Connectors thread count(maxThreads for Connector 
element). This behavior is not actually clearly captured in tomcat 
documentation says 

"A reference to the name in an Executor element. If this 
attribute is set, and the named executor exists, the connector will use 
the executor, and all the other thread attributes will be ignored. Note 
that if a shared executor is not specified for a connector then the 
connector will use a private, internal executor to provide the thread 
pool."

"all the other thread attributes will be ignored" makes impression that its 
enough to increase the Executor thread count. As executor is a shared thread 
pool and connector is using that but what I seen is 


Referencing the Executor in Connector element and setting maxThreads for 
Executor is not enough. If we do that tomcat creates no. of threads as per 
maxThread configuration of Executor but not all threads are actually used for 
request processing. 

Even if we put more load the activeCount will never go above 200 that is 
default maxThreads value of Connector. 

To get threads more than 200 we have to additionally set the maxThread 
attribute for Connectors element as well. (irrespective of Executor reference)


When I explicitly set maxThreads for Connectors activeCount actually went UP to 
the maxThreads limit.

Now I'm able to reach activeCounts beyond 250 see attachment.




On Thursday, January 23, 2014 8:45 PM, Christopher Schultz 
<ch...@christopherschultz.net> wrote:
 
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

akshay,

On 1/23/14, 9:07 AM, akshay hiremath wrote:
> We have this HP load runner running a load of requests on this
> system. We don't see request rejected by Tomcat but if I monitor
> the activeCount attribute of Mbean
> "Catalina:type=Executor,name=tomcatThreadPool" over the period of
> test why the activeCount is not going above 200. if I continuously
> track activeCount and currentThreadsBusy of MBean 
> "Catalina:type=ThreadPool,name="http-bio-8080"" I see the graph
> reahes 200 and flat lines there. Please find monitored graph
> attached during one of the tests. See time frm start of graph to
> the 13:30 when test ended. ignore the in between part and rest
> graph that is of another ongoing test.

Attachments are stripped from the list. Find another way to express
your thoughts.

Is it possible that you are simply not generating enough load? If
Tomcat is responding to all of your requests (e.g. none of them are
failing to connect, failing to respond, or responding very slowly --
as if queued) than it sounds like your Tomcat instance is handling all
the load you are throwing at it.

- -chris

> On Thursday, January 23, 2014 7:06 PM, Mark Thomas
> <ma...@apache.org> wrote: On 23/01/2014 13:30, akshay hiremath
> wrote:
> 
>> Why tomcat is not able to have more than 200 active Threads
>> (parallel threads) processng my requests?
> 
> 
> It can. The issue is that the combination of the requests you are
> making (which you fail to describe), your load testing framework
> (which you fail to describe) and the scheduling in the CPUs of your
> hardware (which you also fail to describe) mean that the chances of
> there actually being 300 concurrent requests for Tomcat to process
> is pretty much zero.
> 
> Mark
> 
> 
> ---------------------------------------------------------------------
>
> 
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> <mailto:users-unsubscr...@tomcat.apache.org> For additional
> commands, e-mail: users-h...@tomcat.apache.org 
> <mailto:users-h...@tomcat.apache.org>
> 
> 
> 
> 
> 
> 
> 
> ---------------------------------------------------------------------
>
> 
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
> For additional commands, e-mail: users-h...@tomcat.apache.org
> 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
Comment: GPGTools - http://gpgtools.org
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
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=TIpT
-----END PGP SIGNATURE-----


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Reply via email to