Are you sending 1000 requests per file? Maybe you can consider to disable the error handler processing if there are too much errors are thrown.
Camel cannot get the stack size, but it should be able to get the inflight exchange size from the management consoler, but I don’t think it helps you to resolve the issue. -- Willem Jiang Red Hat, Inc. Web: http://www.redhat.com Blog: http://willemjiang.blogspot.com (English) http://jnn.iteye.com (Chinese) Twitter: willemjiang Weibo: 姜宁willem On March 23, 2015 at 5:44:33 PM, dermoritz (tantea...@hotmail.com) wrote: > Thanks for the hints, > > you are right here are the cxf part/ to-part of route: > > rtdmProcess.to(rtdmEndpoint).to(tpLoggerToRtdm).process(handleRtdmAnswer).to(tpLoggerFromRtdm) > > > rtdmEndpoint is cxf > > it is created this way: > > CxfEndpoint endpoint = new CxfEndpoint(rtdmUrl, new > CxfComponent(context)); > endpoint.setServiceClass(EventPortType.class); > > My guess is that the cause of stackoverflow could be, that on the one hand > we take zipped/jsoned file with 1000 object in each and send this to soap, > and on the other hand for some reason (problem on soap server side) we get > an Fault for every message. This combination - 1000 in a row and Fault > ("handled" by custom error handler) fills up the stack. > > -Xss256 could be a solution. But i want to be sure or at least need some > evidence that stack size correlates to our file size/ messages per file and > the faults (we never saw this problem even with much bigger files) > > Are there any hints you can give on how to monitor (probably with some jmx > metrics camel/cxf/errorhandlers offers) stack size or work load? > > > > > > > > -- > View this message in context: > http://camel.465427.n5.nabble.com/StackOverflowException-for-route-using-cxf-component-sending-to-soap-endpoint-tp5764460p5764630.html > > Sent from the Camel - Users mailing list archive at Nabble.com. >
