GH-200 Question #34: Real Exam Question with Answer & Explanation

Question

A workflow that had been working now stalls in a waiting state until failing. The workflow file process-ml.yaml has not changed and contains jobs specifying runs-on: [gpu ]. Which of the following steps would troubleshoot the issue? (Each answer presents a complete solution. Choose two.)

Options

• AReview the contents of the Runner_*.log files in the _diag folder.
• BIncrease the usage limits for the GitHub-hosted runners.
• CCheck the "Set up job" step for the logs of the last successful run to determine the runner.
• DUpdate the org settings to enable GPU-based GitHub-hosted runners.
• ERotate the GITHUB_TOKEN secret for the appropriate runners.

Explanation

{"question": 1, "explanation": "The workflow uses 'runs-on: [gpu]', indicating it targets GPU-based runners. Since the workflow file is unchanged but the workflow now stalls and fails, the issue is likely with runner availability or configuration. (A) Reviewing Runner_*.log files in the _diag folder is the standard troubleshooting step for self-hosted runner issues - these logs capture detailed runner diagnostics and error messages. (D) GitHub-hosted GPU runners are a newer, premium feature that must be explicitly enabled in organization settings; if they were never enabled or if access lapsed, jobs targeting them would queue indefinitely and then fail. (B) is wrong because GPU runners are not a standard GitHub-hosted tier subject to usage limits in the same way. (C) checking the last successful run's 'Set up job' logs could provide runner info but does not resolve the issue. (E) rotating GITHUB_TOKEN is unrelated to runner availability.", "generated_by": "claude-sonnet", "llm_judge_score": 3}

No community discussion yet for this question.