Did you wait for 120 seconds after recovery? This is the default timeout for message exchange recovery.
alex On Tue, Dec 2, 2008 at 10:19 AM, Rosalin Pattnaik < [EMAIL PROTECTED]> wrote: > The invoke has > > <ext:failureHandling xmlns:ext="http://ode.apache.org/activityRecovery";> > > <ext:faultOnFailure>true</ext:faultOnFailure> > </ext:failureHandling> > > with a fault handler that sends a mail notification and retries. > > In case it goes down when the invoke activity has started but no exception > is yet thrown, shd it not timeout when the server starts > and then go to the fault handler and the rest of the process resumed. > > > > > On Tue, Dec 2, 2008 at 11:34 PM, Alex Boisvert <[EMAIL PROTECTED]> > wrote: > > > Most likely the invoke activity has gone into activity > > recovery<http://ode.apache.org/activity-failure-and-recovery.html> > > . > > > > alex > > > > > > On Tue, Dec 2, 2008 at 10:02 AM, Rosalin Pattnaik < > > [EMAIL PROTECTED]> wrote: > > > > > Hi, > > > > > > I tried with the new ode.Now in case the server is shutdown once the > > > exception is thrown or it has gone to the fault handler then the flow > is > > > resumed but in case i kill the server when something has been invoked > and > > > it > > > hasnt yet timeout waiting for a response/ hasnt gone to the fault > handler > > > and then restart the server again , the process is not resumed. > > > > > > Regards, > > > Rosalin > > > > > > On Mon, Dec 1, 2008 at 9:30 PM, Matthieu Riou <[EMAIL PROTECTED] > > > >wrote: > > > > > > > On Mon, Dec 1, 2008 at 5:08 AM, Rosalin Pattnaik < > > > > [EMAIL PROTECTED] > > > > > wrote: > > > > > > > > > Hi, > > > > > > > > > > Does Ode persist all its state and can resume from where it had > left > > at > > > > the > > > > > time of shutdown? > > > > > > > > > > > > It does as much as possible. In most cases only running and > uncommitted > > > > transactions will be lost and those will be rollbacked to their > > previous > > > > "stable" state. > > > > > > > > > > > > > > > > > > I tried with a simple Bpel flow. In my case it worked if i killed > the > > > > > process at a Wait stage and when i restarted the server , the > process > > > > > instance was completed succesfully. > > > > > > > > > > In the same process i also have an invoke(which throws an > > > > activityFailure) > > > > > and is handled by a fault handler.If i kill the process when its in > > the > > > > > invoke state and then restart the server , the process is not > resumed > > > and > > > > > completed. > > > > > > > > > > > > It's a bug that I've fixed about a month ago I believe. If you can > use > > > the > > > > latest version of the 1.X branch, it should work. > > > > > > > > Cheers, > > > > Matthieu > > > > > > > > > > > > > > > > > > > > > > > Regards, > > > > > Rosalin > > > > > > > > > > > > > > >
