How to make a photo smaller?
ConnectionWaitTimeoutException
Hello Guest
  
  • Login
• Register…
• Start blog
  • Who, Where, When
• What can I do?
• What to Read?
  • Polls
• Avatars
• Interests
  • Cities and Countries
• Random blog
• Users search
  • Search
• Games
• Tests
• QAIX
  • Сообщества
• Talxy Chat
• Horoscope
• Online
 
Зарегистрируйся!

QAIX > Software, Computer Help > ConnectionWaitTimeou­tException 21 April 2008 20:44:12

  Recent blog posts: 
  They have birthday today: 
  Forums:   
  Discuss: 
  Recent forum topics: 
  Recent forum comments:
  Moderators:

ConnectionWaitTimeou­tException

Guest 21 April 2008 20:44:12
 Hi
I have deployed the application on Websphere Application Server 6.0 ND (clustered Environment).

Now the Testing Team has started Load Testing (500 concurrent users) on Application server using Load Runner Test Software.

During run the Load Runnner it is observed that maximum 90 users can able to accesses the application and after that application is throwing below ConnectionWaitTimeo­utException:

com.ibm.websphere.c­e.cm.ConnectionWaitT­imeoutException: Connection not available, Timed out waiting for 191613
at com.ibm.ws.util.Thr­eadPool$Worker.run(T­hreadPool.java(Compi­led Code))
Caused by: com.ibm.websphere.c­e.j2c.ConnectionWait­TimeoutException: Connection not available, Timed out waiting for 191613
at com.ibm.ejs.j2c.poo­lmanager.FreePool.cr­eateOrWaitForConnect­ion(FreePool.java:12­97)
at com.ibm.ejs.j2c.poo­lmanager.PoolManager­.reserve(PoolManager­.java:2094)
at com.ibm.ejs.j2c.Con­nectionManager.alloc­ateMCWrapper(Connect­ionManager.java:807)­
at com.ibm.ejs.j2c.Con­nectionManager.alloc­ateConnection(Connec­tionManager.java:583­)
at com.ibm.ws.rsadapte­r.jdbc.WSJdbcDataSou­rce.getConnection(WS­JdbcDataSource.java:­333)

CPU Utilization is also reached upto 70% percent.

After went through performance guides I have made the below changes in the JDBC Connection pool and Connection Timeout settings etc.:-
JDBC Connection Pool (min 20 and max 50)
Web Container Thread Pool (min 20 / max 50)
ORB Thread Pool ? (min 10 / max 50)
JVM Heap Size ? (Initial ? 1500 and Maximum ? 1500)

But after making the above changes it seems that application server (cluster) is taking more time to Stop and Start the Application Server (App. 20 min. to Stop and 20 min. Start).

Before made the above changes clustered and server was started within 10 min and it seems that now it will be taking 40 min to stop and start the servers and also CPU utilization is consistent 35 to 40% on the both servers (only time at start and stop).

I want to improve the application performance atleast 500 to 700 users concurrently access hit the application and also want to improve the start/stop timestamp as per below the sizing of server hardware.

PowerPC_POWER5 - Number of Processors: 2
Processor Clock Speed: 1654 MHz
CPU Type: 64-bit
Physical Memory ? 3.5 GB

It would be very great if you put your experts comments on this performance issues. Also give me idea about Web Container Thread Pool and ORB Thread sizing.

vz.
Add comment
Paul Ilechko 3 August 2006 19:29:50 permanent link ]
 vazeamit@hotmail.com­ wrote:

It would be very great if you put your experts comments on this> performance issues. Also give me idea about Web Container Thread Pool> and ORB Thread sizing.

The only way to know the appropriate size of these pools is test in your
environment. You want to have enough threads available to drive CPU to
near maximum, and no more than that. Of course, you also need to make
sure that the number of threads you have are not exhausting memory, as
the system will die a painful death at that point.

Once you understand how many web container threads you can drive, you
need to watch carefully how many DB connections and Orb threads are
actually being used. I would start with low min and high max values, and
tune the max down once you see what you actually need.

Also you need to understand where your application is bottlenecking.
What shared resources are the issue, if any? This is not necessarily an
external resource such as a database connection. If your code has any
synchronization you may be blocking threads. Make sure you release
resources as soon as you are done with them. Don't put too much stuff in
the HTTP session. If you are doing any DRS replication, make sure there
are sufficient threads in the default thread pool. Analyze your SQL to
make sure it isn't doing dumb things like table scans. Make sure you
test in an environment that is exactly like production - you CAN NOT
extrapolate performance from a smaller to a larger environment.

A tool like ITCAM that provides lock analysis is useful for
understanding and resolving contention issues.
Add comment
 

Add new comment

As:
Login:  Password:  
 
 
  
 
Пожалуйста, относитесь к собеседникам уважительно, не используйте нецензурные слова, не злоупотребляйте заглавными буквами, не публикуйте рекламу и объявления о купле/продаже, а также материалы нарушающие сетевой этикет или УК РФ.


QAIX > Software, Computer Help > ConnectionWaitTimeou­tException 21 April 2008 20:44:12

see also:
Some virtual hosts serve as expected…
some files not showing in directory…
Reverse proxing through apache where…
пройди тесты:
What is Your Temperament?
see also:
Hi! ^^

  Copyright © 2001—2008 QAIX
Idea: Miсhael Monashev
Помощь и задать вопросы можно в сообществе support.qaix.com.
Сообщения об ошибках оставляем в сообществе bugs.qaix.com.
Предложения и комментарии пишем в сообществе suggest.qaix.com.
Информация для родителей.
Write us at:
If you would like to report an abuse of our service, such as a spam message, please .