Exercise 1: Why "Job Succeeded" and "Data Is Restorable" Are Different Claims — Possible Solution ==================================================================== WHAT THIS CHAPTER SAYS ------------------------------ Per this chapter, "a backup job reporting 'success' confirms exactly one thing: data was copied somewhere. It confirms nothing about whether that copy is complete, uncorrupted, or genuinely restorable into a working state." WHAT A "SUCCESS" STATUS ACTUALLY MEASURES ------------------------------ A backup job's own success status is generated by the backup software itself, based on whether the copy process ran to completion without throwing an error - it's a statement about the process, not about the result. It answers "did the job finish running," not "did it produce something usable." WHY COMPLETION DOESN'T GUARANTEE USABILITY ------------------------------ A copy operation can finish without error while still producing data that's corrupted, incomplete, or missing something the application actually needs to function once restored (a database backed up mid-write, a file locked and silently skipped, a dependency that wasn't included). None of these failure modes necessarily throw an error the backup software would report - they're failures of the *result*, not the *process*, and a status check only ever verifies the process. WHY THIS REQUIRES A DIFFERENT VERIFICATION METHOD ------------------------------ Since a "success" status can't distinguish a genuinely restorable backup from a completed-but-broken one, confirming which one you actually have requires a fundamentally different check: not asking the backup software "did you finish," but actually attempting to restore the data and confirming it works. This is exactly why the chapter's tip-box points ahead to Chapter 5's test-restore material as "the only thing that actually" proves usability. WHY THIS WORKS AS AN ANSWER ------------------------------ It explains what a success status actually measures (process completion), explains why that doesn't guarantee the result is usable, and explains why proving usability requires an entirely different check - a real restore - rather than a more careful reading of the same status message.