Can you paste your cxf-se configuration. Maybe we will be able to spot something. Also did you make any changes to your /conf/activemq.xml?
Below our changes in activemq.xml:
Line 50 changed to
<amq:transportConnector uri="tcp://localhost:61616" />
Line 54 commented auto discover the other brokers:
Below our CXF-SE xbean.xml
<cxfse:proxy id="greeterProxy" service="grt:GreeterService"
context="#context" type="ws.greeter.GreeterPortType" />
<bean class="helloconsumer.HelloConsumerPortTypeImpl" autowire="false">
<property name="greeter" ref="greeterProxy" />
Thank you very much.
It looks ok. If possible can you capture a thread dump of when the esb is spinning @100% and attach it... Maybe also try and remove the #context and see if that helps.
Thank you very much Dave,
Yes we followed the doc at ESB
Attached is a little Maven project containing one CXF SE SU and one SA, when I deploy this SA the CPU spins to 100%.
Please test the attached SU / SA. To deploy it:
cpu-overload.zip 12.0 K
Did you test the attached code?
Let me know if you need something else or if there was any error in attached code.
Hi all !
I am a DiSiD company worker like eruiz user.
We can not find the problem to the 100% CPU spin.
Did you test the attached code ?
We are interested in to buy the HQ tool, but first need to resolve this problem.
Hi all !
We have found a solution to this error, but we cannot fix the source problem.
The error is placed on the xbean.xml file of components that use the cxfse:proxy tag. In this tag, the attribute context="#context" causes an overload and finally the system crash.
Another related problem can be found when the JAVA_OPTS environment variable is setted on servicemix startup because some servicemix default memory configurations are not established.