Restartability within the StartLoop and Endloop stages

[abc123]

How does restartability happen within the StartLoop and Endloop stages of a a sequencer? Let's say we have 5 stages within the StartLoop and EndLoop stages. If the job fails on the second stage, would during restart,
the loop counter be preserve? Has anybody done something like this?

[narasimha]

There is no inbuilt mechanism or checkpoint to the level of Loops in the Job Sequence.
It would be interesting to know if anybody has implemented any way to handle this.

[chulett]

They most certainly are restartable with checkpointing enabled, we're doing that now. I can vouch for the fact that it will pick up from the failed job in the loop, just like it would for any other sequence of objects in a Sequence job.

I cannot, however, confirm if the loop counter would be smart enough to pick up where it left off. In my case, the loop counter I'm using doesn't do anything except force it to loop until a routine in the loop decides it is time to exit the loop. I've never tested the counter value during failures as I haven't needed to yet - in my case it didn't matter.

That being said, I'd suggest you build yourself a small test 'harness', a little loop that dies after X loops and then restart it and see what happens. Best way to learn is to do, as they say.

[ray.wurlod]

I think you'd need to parameterize the StartLoop properties.