It has nothing to do with the scheduler service.  You should post
ProxyService config instead.  What is the metrics format like? 

Also you should reduce the minThreadHandle to 1 since it doesn't do any
good for you. 

-----Original Message-----
From: Kari M. Scott [mailto:[EMAIL PROTECTED] 
Sent: Wednesday, June 15, 2005 4:34 PM
To: JRun-Talk
Subject: many jrunx.scheduler.JSemaphore threads in wait state

Hello, 

We currently have six Solaris boxes running JRun 4 Build 92909 with
Apache2. Everything runs smoothly for the most part, but we've started
to experience an occasional logjam of threads which coincide with a JRun
performance hit. Most of the time JRun recovers, but occasionally we
need to bounce it to clear things up.

Here are the details. Via a script, we monitor the number of established
JRun connections on port 9000 every minute. This number coincides with
the number of Idle threads reported in the metrics log and generally
hovers between 13 and 40 during an average traffic day. 

Here's a snapshot of the metrics log during the last occurrence. Notice
the jump in idle threads from 32 to 140:

06/15 14:00:25 metrics Thrds: L=25 I=32 Q=0 R=3 TOT=60; Rq: Q=0 X=0
S=314; Shd: L=23 Q=0 R=1 X=0 TOT=24; (Free/Tot)=579626/1572672; Sess:
(InMem/Tot)=868/868
06/15 14:01:25 metrics Thrds: L=41 I=140 Q=0 R=0 TOT=181; Rq: Q=0 X=0
S=278; Shd: L=23 Q=0 R=1 X=0 TOT=24; (Free/Tot)=565261/1572672; Sess:
(InMem/Tot)=867/867
06/15 14:02:25 metrics Thrds: L=55 I=126 Q=0 R=0 TOT=181; Rq: Q=0 X=0
S=334; Shd: L=23 Q=0 R=1 X=0 TOT=24; (Free/Tot)=741200/1572672; Sess:
(InMem/Tot)=876/876
06/15 14:03:25 metrics Thrds: L=74 I=106 Q=0 R=1 TOT=181; Rq: Q=0 X=0
S=326; Shd: L=23 Q=0 R=1 X=0 TOT=24; (Free/Tot)=709294/1572672; Sess:
(InMem/Tot)=883/883


I took a series of thread dumps during this occurrence and found that
the majority of these threads are these JSemaphore things in a wait
state:

"jrpp-4220" prio=5 tid=0x009cd008 nid=0xabf5 in Object.wait()
[7c67f000..7c67fc28]
        at java.lang.Object.wait(Native Method)
        - waiting on <0xa22547d0> (a jrunx.scheduler.JSemaphore)
        at jrunx.scheduler.JSemaphore.acquire(JSemaphore.java:74)
        - locked <0xa22547d0> (a jrunx.scheduler.JSemaphore)
        at
jrun.servlet.network.NetworkService.accept(NetworkService.java:348)


Does anyone know what would cause a backup of these threads? 

Here's what we have set in our jrun.xml file for the scheduler:

<service class="jrunx.scheduler.SchedulerService"
name="SchedulerService">     
  <attribute name="bindToJNDI">true</attribute>     
  <attribute name="activeHandlerThreads">25</attribute>    
  <attribute name="maxHandlerThreads">1000</attribute>
  <attribute name="minHandlerThreads">20</attribute>     
  <attribute name="threadWaitTimeout">180</attribute>  
  <attribute name="timeout">600</attribute>   
</service>


TIA, 
Kari



-- 
--------------------------------------------------
Kari M. Scott                   [EMAIL PROTECTED]
Software Engineer               608 298-1223
Berbee
5520 Research Park Drive
Madison, WI 53711





~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~|
Find out how CFTicket can increase your company's customer support 
efficiency by 100%
http://www.houseoffusion.com/banners/view.cfm?bannerid=49

Message: http://www.houseoffusion.com/lists.cfm/link=i:8:5697
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/8
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:8
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.8
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Reply via email to