What is the recommended approach to give a process schedule resilience whilst also remaining as efficient as possible. What I have currently is; initial task Check Logged In, on completion start process, on exception, run Check Logged In and Log In, on completion delay 600 seconds and start process, on exception Abort.
That logic works most the time, however I have noticed that occasionally when running the schedule on multiple resources if the Check Logged In session terminates on one machine, any other running sessions appear as Stopped and do not continue to the next task. I understand this is because I have Fail Fast set as True on the Check Logged In task, however this is necessary, as if any of the machines are logged out I would want the schedule to take the exception path.
I've avoided just having a Check Logged In and Login task as this would mean we'd hit a 10 minute post completion delay even if the resource is logged in and therefore ready to run the process, which cant be afforded.
Any help/advice would be appreciated, thanks.
