Explanation
The answer is,
Autoloader automatically re-processes data that was not loaded using the checkpoint.

Explanation
The answer is,
Under jobs UI select the job you are interested, under runs we can see current active runs and last 60 days historical run

Explanation
This is the best configuration for scheduling Structured Streaming jobs for production, as it automatically recovers from query failures and keeps costs low. A new job cluster is created for each run of the job and terminated when the job completes, which saves costs and avoids resource contention. Retries are not needed for Structured Streaming jobs, as they can automatically recover from failures using checkpointing and write-ahead logs. Maximum concurrent runs should be set to 1 to avoid duplicate output or data loss. Verified References: Databricks Certified Data Engineer Professional, under "Monitoring & Logging" section; Databricks Documentation, under "Schedule streaming jobs" section.