[Feature request] Auto-restart failed job

[mario45211]

Hello Jet Team! :)

I'm developing a stateless stream pipeline under Jet engine. One of my requirements is its continuous working, regardless of errors from any operation within it.

From the Jet's documentation I read: One option for dealing with failure in immutable-state jobs is simply restarting them, but have you ever consider adding auto-restarting option ? Once my job fails I must restart it manually, which is highly undesirable.

Of course, there is a workaround to implement external watcher to monitor and do restart job if failed one was detected, but mantaining and deploying additional watcher is cumbersome, especially when job is deployed to Jet's cluster.

Please, leave below some comment what you think about such feature.

Best regards