GCP Professional Cloud DevOps Engineer Practice Question 3: Toil Identification Reduction
Google Cloud Professional Cloud DevOps Engineer · Test 1 · question 3 of 10
A review of the last quarter's on-call shifts shows that engineers spend roughly 60% of their time on the same handful of tasks: rotating TLS certificates by hand, filing and applying quota increases, and restarting a batch job that wedges most nights. Each task has a clear, well-written runbook, takes 10-30 minutes, and its frequency has grown in step with the number of customers onboarded.
How should the team classify and address this work?
Show answer and explanation
Answer:
Toil has a specific definition in SRE: work tied to running a production service that is manual, repetitive, automatable, tactical (interrupt-driven rather than strategy-driven), devoid of enduring value, and that grows at least linearly with service size. Every task listed matches all six characteristics; most tellingly, the volume grows with customer count, which is the signal that the work will eventually consume the whole team. Google's guideline caps toil at 50% of an SRE's time, with the remainder spent on engineering that reduces future toil: certificate rotation becomes an automated renewal, quota bumps become a monitored automated request or a capacity plan, and the nightly wedge gets a real fix rather than a nightly restart.
Option B misreads what a runbook signifies. A crisp runbook is evidence that a task is *automatable* (it is already an algorithm a human is executing), so it strengthens the case that the work is toil, not the reverse.
Option C treats the symptom. Adding people lowers the per-person percentage temporarily, but the work still scales with the service, so the team is back where it started after the next growth step, and it has now committed more engineers to work that produces no enduring value.
Option D just relocates the toil. Handing repetitive operational work to a separate team recreates the traditional dev/ops split that SRE was designed to dissolve, removes the incentive for the people who can fix the underlying causes to do so, and leaves the linear growth problem intact.