I think it would be nice to have a slightly different node selection by introducing the notion of "was last run successful ?" and discard a node if it failed to successfully run the previous execution of this step.
This should lower the chance to stick with a faulty node.
What do you think ?
Description
Hi,
I think it would be nice to have a slightly different node selection by introducing the notion of "was last run successful ?" and discard a node if it failed to successfully run the previous execution of this step.
This should lower the chance to stick with a faulty node.
What do you think ?