Hi Sathwik, I've been using Riftsaw on JBoss EAP 6 - Riftsaw is ODE 1.3.5. I've debugged through, and I don't see the alarm ever try to cancel the invoke. I've set up ODE 1.3.5 under Tomcat as well to try to verify that this exists there as well, but I'm still working on transforming one of the examples into this scenario.
Is this a known issue? Does this happen under 1.3.6? --Tom On Thu, Apr 21, 2016 at 10:47 PM, Sathwik B P <[email protected]> wrote: > Hi Thomas, > > Can you share your process for testing. > > regards, > sathwik > > On Wed, Apr 20, 2016 at 10:55 PM, Thomas Cunningham <[email protected]> > wrote: > > > Hi, > > > > I'm seeing an issue where a long running invoke (the service behind it > does > > a Thread.sleep() for 30 seconds but does not hit a 25 second alarm. > > Issue is decribed here https://issues.jboss.org/browse/ENTESB-5167 > > > > The invoke here : > > > > > > > https://github.com/apache/ode/blob/APACHE_ODE_1.3.5/bpel-runtime/src/main/java/org/apache/ode/bpel/runtime/INVOKE.java#L100 > > > > waits for the full 30 seconds, and returns a successful response, rather > > than timing out and hitting the alarm. > > > > What I'm seeing looks a lot like > > https://issues.apache.org/jira/browse/ODE-263 and this : > > > > > https://mail-archives.apache.org/mod_mbox/ode-user/201105.mbox/%3Cvmime.4dd2574c.4405.2e079b14f5563e0c@noriswww%3E > > > > Wondering if there's workaround here or if this is a known issue. > Should > > this be redesigned as a more asynchronous <invoke/><receive/>? > > >
