Hi, we are observing high CPU utilization rates on JBoss 4.0.1sp1 even while 
the server is IDLE and doing nothing. The rates we see are at %40. 

We do have couple background processes, but they wake up every 30 seconds, and 
finish very quickly. An output from hprof display the performance killer is 
PlainSocketImpl.socketAccept call. Now, we use OSCache but have distributed 
caching disabled. The output is below. 

CPU SAMPLES BEGIN (total = 211354) Fri Jan 21 12:56:38 2005
rank   self  accum   count trace method
   1 74.04% 74.04%  156483   479 java.net.PlainSocketImpl.socketAccept
   2 16.46% 90.50%   34792  1363 java.net.SocketInputStream.socketRead0
   3  7.10% 97.60%   14997  1182 java.lang.System.currentTimeMillis

My question is, if you see high CPU utilization on listener type threads with 
low priority, is it possible that threads with higher priroty can still get 
their CPU share and the overall system can still perform acceptably? Would 
socketAccept() be one of those methods/classes? If not, which JBoss subsystem 
could be using this call?

We use JMS/queuesMDBs, have Hibernate 3.0 with OSCache.

Regards, 



View the original post : 
http://www.jboss.org/index.html?module=bb&op=viewtopic&p=3871254#3871254

Reply to the post : 
http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=3871254


-------------------------------------------------------
This SF.net email is sponsored by: 2005 Windows Mobile Application Contest
Submit applications for Windows Mobile(tm)-based Pocket PCs or Smartphones
for the chance to win $25,000 and application distribution. Enter today at
http://ads.osdn.com/?ad_id=6882&alloc_id=15148&op=click
_______________________________________________
JBoss-user mailing list
JBoss-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jboss-user

Reply via email to