- Support Checkpoints
I need to ensure that my job can be restarted after an interruption (e.g. stopped or by error) at a specified checkpoint with the same variables and arguments. Checkpoints ensure that the status is serialized (variables, arguments, files and rows from the result set) and the job can be continued after a safe state that is defined at design time. This avoids unwanted runs of previous tasks that succeeded successful.
- I can define checkpoints within my job
- After a job failed, I can resume a job after the following job entry of the last successful checkpoint
- Add a clear warning box to the user (similar to the warning box for running job entries in parallel).
- checkpoints need an expiration date/time or combine with dependencies (PDI-7910)
- Have an attempt_id or similar
- Provide a Attempts-Log table
- Save the state in the file system (e.g. attempt_state folder)