OSB 10.3 not Timing Out if Service Using HTTP Transport is not Available (Doc ID 859261.1)

Symptoms

You have a OSB business service with HTTP transport. The external service to which the business service connects is completely down and does not respond to a TCP connect.

1. Timeout on the business set to 10 seconds.
2. OSB tries to call the external service but it is not even able to connect.
3. OSB seems to wait for a long time and does not timeout after the defined 10 seconds. 

This scenario does not seem to work as expected. Instead what happens is, we do not get a timeout, the request only times out after the socket times out after about 3 minutes.

The log files are showing the following messages:

<Error> <OSB Kernel> <BEA-398204> < [OSB Tracing] Internal Error while tracing message null>


The SOAP response is:

<soapenv:Fault>
  <faultcode>soapenv:Server</faultcode>
  <faultstring>
       BEA-380002: Tried all: '1' addresses, but could not connect over HTTP to server:
       'xxx.xxx.xxx.xxx', port: 'xxx'
  </faultstring>
  <detail>
   <con:fault xmlns:con="http://www.bea.com/wli/sb/context">
     <con:errorCode>BEA-380002</con:errorCode>
     <con:reason>
        Tried all: '1' addresses, but could not connect over HTTP to server: 'xxx.xxx.xxx.xxx', port: 'xxx'
     </con:reason>
     <con:location>
       <con:node>RouteTo_Infranet Service</con:node>
       <con:path>request-pipeline</con:path>
     </con:location>
   </con:fault>
  </detail>
</soapenv:Fault>

Cause

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.