A junior engineer's plan for handling transformation failures is to rerun the job repeatedly until it succeeds. What is wrong with that plan?
Select an answer to reveal the explanation.
Short Explanation
Rerunning blindly just replays the same failure. If you don't read the diagnostics and fix the cause, the same failure is likely to recur.
Full Explanation
The correct answer is the one saying that retrying without reading the diagnostics does not address the underlying cause, so the same failure is likely to recur. Whatever caused the first failure, such as a bad input, a missing dependency, an unsupported construct, or an unhealthy component, is still there on the second attempt unless something is actually diagnosed and fixed. The option saying rerunning a failed transformation job is not permitted by the tooling is incorrect because retrying a job is technically possible; the objection is that it is unproductive, not that it is blocked. The option saying each retry consumes a COBOL section, so eventually there is no source left to transform, is incorrect because retrying does not consume or remove source code. The option saying retrying is fine, but only if it is done exactly three times before escalating, is incorrect because there is no meaningful retry count that fixes an undiagnosed problem; the number of attempts is not the issue. For the exam, retrying without diagnosis is the specific mistake this checklist is meant to prevent.