Hi.
I have an application using Camel that seems to over time builds up heap
space until it starts using 100%CPU time on a windows server.
The problem occures when there is a connection exception in an outbound
endpoint. I have a retry each second, and when this continues for several
days, the heap is building up.
Doing a heap dump shows me that there is a lot of StackTraceElements. This
is, I guess, from the stack trace from every time the connection failed, but
it seems like something is holding the reference to the stacktrace elements
after the exception has been handled.
Could there be something with the deafult error hanlder?
I am using logback and the application is running in Tomcat.
Here is the route configuration:
from("jms:CRM.MO?acknowledgementModeName=CLIENT_ACKNOWLEDGE")
.id("CRM-MO-Receive ( < CRM.MO )")
.errorHandler(defaultErrorHandler().maximumRedeliveries(-1).redeliveryDelay(1000))
.handleFault().transacted("jmsTransactionPolicy")
.setProperty("OriginalInBody", body())
.process(aMessageTranslator)
.process(aCorrelationEnricher)
.removeHeaders("JMS*")
.throttle(configuration.getThrottle().getMessagesPerPeriod()).timePeriodMillis(configuration.getThrottle().getTimePeriodMillis())
.inOut("cxf:bean:receiveEndpoint?defaultOperationName=ReceiveMessage")
.process(receiveResponseTranslator)
.end();
I'm not sure if the problem really is Camel but I have never experienced
something like it before.
Regards
Trond
--
View this message in context:
http://camel.465427.n5.nabble.com/StackTraceElement-using-heap-tp5745623.html
Sent from the Camel - Users mailing list archive at Nabble.com.
