Run a benchmark and refuse to average away the errors

short · 55 min · Objective 3.2

Task

Evaluate a published benchmark against a lab host, then compute compliance two ways -- one that hides the unmeasured rules and one that does not.

Steps

  1. Run the benchmark with an appropriate profile and save the full results, including per-rule outcomes, to lab/scap-results.txt.
  2. Count the outcomes into lab/scap-outcomes.csv as outcome,count, keeping pass, fail, error, unknown and notapplicable SEPARATE.
  3. Compute two figures in lab/scap-compliance.md: passes over total rules, and passes over rules actually EVALUATED, stating the difference.
  4. Tailor the profile to deselect two rules that genuinely do not apply, recording each as an exception in lab/exceptions.csv with a justification, an owner and an expiry.
  5. Report BOTH figures -- against the published benchmark and against your tailored profile -- plus the count of deviations, in lab/scap-report.md.

Verify

awk -F, 'NR>1 {print $1}' lab/scap-outcomes.csv | sort | tr '\n' ' '
awk -F, 'NR>1 && $1=="error" {print $2}' lab/scap-outcomes.csv
grep -Ec '[0-9]+(\.[0-9]+)?%' lab/scap-compliance.md
grep -Eci 'published benchmark|tailored|deviation' lab/scap-report.md

At least four distinct outcomes recorded, both percentages computed, and the report carrying both numbers plus a deviation count. A percentage that counts errors as passes is the category error this course treats as foundational.

This is an independent study companion for CompTIA SecurityX CAS-005 and is not produced by or endorsed by CompTIA.