We got an Apache load balancer which forward requests to both WebLogic and JBoss clusters. Both clusters work fine until the load balancer gets flooded and there are waiting requests. When this happen, Weblogics aren't affected but JBoss start to fail reporting this error (it has an Spanish sentence):

09:29:24,144 ERROR [org.apache.catalina.connector.CoyoteAdapter] Ha tenido lugar una excepción o error en el contenedor
durante el procesamiento del requerimiento: java.lang.RuntimeException: Caught TimeoutException acquiring ownership of XZPnYgCZCUvTa9XXwO3YYw__

I have read this discussion in which they're talking about this problem, but it seems the solution of enabling sticky sessions is not for me as we have the sticky sessions already enabled.

Any ideas? As I said, it only fails when the load balancer reaches the maximum of requests it can handle.

  • If this happens very often, have you thought about tuning http parameters to allow more connections? – Sree Oct 13 '14 at 13:22
  • @Sree it doesn't happen very often, few times a week. I already suggested the same but management said that the current value (256) is more than enough, otherwise it would affect the performance of the server. – yzT Oct 13 '14 at 13:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.