java.lang.IllegalStateException: EJBCLIENT000032: Cannot retry a request which hasn't previously been completed

[nlsurfman]

I am getting this stacktrace (below) on a remote client from a Wildfly 8.0.0.Alpha4 instance. While debugging I can see the proxy being loaded on the client, but the method call to createTaskList() does never arrive at the server. (The breakpoint in that method never gets hit). Pasting this stacktrace (title) into Google yields no results (other than Grepcode pages). Does anybody know where this message comes from and what I can do about it?

java.lang.IllegalStateException: EJBCLIENT000032: Cannot retry a request which hasn't previously been completed

    at org.jboss.ejb.client.EJBClientInvocationContext.retryRequest(EJBClientInvocationContext.java:200)

    at org.jboss.ejb.client.EJBInvocationHandler.sendRequestWithPossibleRetries(EJBInvocationHandler.java:256)

    at org.jboss.ejb.client.EJBInvocationHandler.sendRequestWithPossibleRetries(EJBInvocationHandler.java:265)

    at org.jboss.ejb.client.EJBInvocationHandler.doInvoke(EJBInvocationHandler.java:198)

    at org.jboss.ejb.client.EJBInvocationHandler.doInvoke(EJBInvocationHandler.java:181)

    at org.jboss.ejb.client.EJBInvocationHandler.invoke(EJBInvocationHandler.java:144)

    at com.sun.proxy.$Proxy14.createTaskList(Unknown Source)

[jaikiran]

For some reason that request is being retried even before the previous/original request has been considered complete. Can you attach an application which reproduces this?

[nlsurfman]

jaikiran pai wrote:

For some reason that request is being retried even before the previous/original request has been considered complete. Can you attach an application which reproduces this?

I don't think I can, because it's only one EJB that does this, and all the others are working fine. The strange thing however is that all EJB's are designed the same. All @Stateless, nothing special, no obscure frameworks, no security, pretty straightforward. Can you describe situations where this retry is being done?

Thanks for thinking along!

[jaikiran]

The retry is done when the client side (EJB client API) thinks that a particular server can handle the EJB invocation (i.e. app-name, module-name, distinct-name combination is applicable for the target server) and sends across the request to the server, but the server then responds back that there is no such EJB (appname, modulename, distinct name, bean name combination) on that server. So the EJB client API then attempts retries on other possible servers which advertise that they can handle this request.

The strange part is, that the retry seems to have been tried even before the server has responded back saying that it cannot handle the original request. Or there is some bug in the EJB client API.

[nlsurfman]

Just so you know, this error has vanished since I upgraded to a snapshot of WildFly-8.0.0.Beta1. Still weird, but that fixed it for me!