Can someone explain how the persistence works for Async tasks/jobs running through JobExecutor?

[rjintur]

What happens when an Async job is executing through JobExecutor and Engine is stopped/crashed for some reason? Does the Workflow start from beginning or from Async node?

[salaboy21]

Hi R SJ,

It will start from the last async node that was executing at the moment of the crash/shutdown.

[rjintur]

And this will be true for all the instances that were started by JobExecutor when my Asyc node was part of a muti-instance subroutine. is it? And what would happen if say out of 10 in multi-instance, 8 were started and picked up and before starting remaining 2 instances, crash happened?

[salaboy21]

If you have a Multi Instance Node that iterates 10 times on top of an async task it will work as follow:

1) As soon as it go into the Multi Instance node it will create the 10 async jobs (it will create all of them or non of them)

2) the executor will start picking from these 10 jobs

3) the server crash:

     a) all the completed jobs before the server crash will not be executed again

     b) all the pending jobs will be started again