Running CF9 on Solaris 10 with the oracle webserver7

Our Jrun process is hanging several times a day with the unresponsive server message. I've got monitoring and tracing turned on and also getting stack dumps. 

Here's a section of the log:

07/08 00:05:42 Information [scheduler-2] - Alert: Unresponsive server state dete

cted. 10 or more threads have been busy for over 60000 milliseconds

07/08 00:05:45 metrics Web threads (busy/total/delay/idle/listen): 19/20/0/0/1 R

equests(delay/drop/handle): 0/0/0 Sessions: 0 Total Memory=438016 Free=138098

07/08 00:05:45 Information [scheduler-2] - Mail: 'ColdFusion: Unresponsive Serve

r Alert active at Sun, 8 Jul 2012 00:05:42' From:'cf9@web2...' To:'...' was successfully sent using smtp...

07/08 00:05:45 Information [scheduler-2] - Alert: unresponsiveserveralert: Email

notification sent.

07/08 00:05:45 Information [scheduler-2] - Aborting request null handled by thre

ad jrpp-7 CFML stack:   Java stack:   java.lang.Object.wait(Native Method)     

at java.lang.Object.wait(Object.java:485)      at coldfusion.util.AbstractCache.

fetch(AbstractCache.java:46)      at coldfusion.util.SoftCache.get_statsOff(Soft

For some reason we now have busy threads. But not all the time. It's intermittent. All Data Sources appear to be responding as they should and I see no hardware issues(disks going bad or interfaces going flaky).

Here's what my threads look like when the server goes unresponsive:

07/10 14:28:26(busy/total/delay/idle/listen):0/3/0/2/1

07/10 14:28:46(busy/total/delay/idle/listen):0/3/0/2/1

07/10 14:29:06(busy/total/delay/idle/listen):0/3/0/2/1

07/10 14:29:26(busy/total/delay/idle/listen):0/3/0/2/1

07/10 14:29:46(busy/total/delay/idle/listen):0/3/0/2/1

07/10 14:30:06(busy/total/delay/idle/listen):0/3/0/2/1

07/10 14:30:26(busy/total/delay/idle/listen):0/3/0/2/1

07/10 14:30:46(busy/total/delay/idle/listen):0/3/0/2/1

07/10 14:31:06(busy/total/delay/idle/listen):0/3/0/2/1

07/10 14:31:26(busy/total/delay/idle/listen):4/5/0/0/1

07/10 14:31:46(busy/total/delay/idle/listen):11/12/0/0/1

07/10 14:32:06(busy/total/delay/idle/listen):13/14/0/0/1

07/10 14:32:26(busy/total/delay/idle/listen):19/20/0/0/1

07/10 14:32:46(busy/total/delay/idle/listen):25/26/0/0/1

07/10 14:33:06(busy/total/delay/idle/listen):6/33/0/15/12

07/10 14:33:26(busy/total/delay/idle/listen):1/33/0/1/31

07/10 14:33:46(busy/total/delay/idle/listen):0/33/0/2/31

07/10 14:34:06(busy/total/delay/idle/listen):0/33/0/2/31

07/10 14:34:26(busy/total/delay/idle/listen):0/33/0/2/31

07/10 14:34:46(busy/total/delay/idle/listen):0/33/0/2/31

07/10 14:35:06(busy/total/delay/idle/listen):0/33/0/2/31

07/10 14:35:26(busy/total/delay/idle/listen):0/33/0/2/31

Any ideas on how to resolve this issue would be greatly appreciated.