DevOps Phrases: Infra, Pipelines & On-Call Language
5 exercises on pipeline, deployment, and on-call phrases. Choose the most natural and professional option.
0 / 5 completed
1 / 5
How do you report that CI/CD passed and the build is healthy?
KEY PHRASE: "The pipeline is green — we're good to merge" "The pipeline is green" is the idiomatic shorthand in DevOps culture — green means all checks passed, all tests passed, and the build artifact is healthy. It's universally understood in CI/CD contexts without requiring explanation. "Good to merge" adds the actionable recommendation, completing the communication. Real examples: "Pipeline's green — deploying to staging now"; "Checked the pipeline — it's green, feel free to merge your PR." Options A and C are technically accurate but miss the idiomatic fluency. Option B adds uncertainty ("seems to be") which undermines confidence in the status.
2 / 5
During a deployment, error rates spike. How do you alert the team?
KEY PHRASE: "We're seeing elevated 5xx rates — pausing the rollout..." "Elevated 5xx rates" is the precise DevOps/SRE phrasing: it names the HTTP error class (5xx = server errors), uses "elevated" rather than "high" to imply comparison against a baseline, and pairs with a concrete action (pausing). This tells engineers exactly where to look. Real examples: "We're seeing elevated 5xx rates from the auth service — rolling back"; "Elevated 4xx rates on /api/v2 — investigating the request validation change." Options B and C are vague. Option D focuses on user impact rather than the technical signal, which slows down the investigation.
3 / 5
A deployment is failing and needs to be reversed immediately. What do you say?
KEY PHRASE: "I'm going to manually trigger a rollback — stand by" "Trigger a rollback" is the correct DevOps action phrase — "trigger" reflects that rollback is a deliberate pipeline action you initiate, not a passive undo. "Manually" clarifies you're bypassing automation. "Stand by" signals urgency and coordination, telling the team to hold other actions. Real examples: "I'm going to manually trigger a rollback — ETA 3 minutes"; "Triggering a rollback on prod now — please don't merge anything until I confirm." Options A, C, and D are vaguer and don't signal the mechanism, the timing, or the coordination protocol.
4 / 5
You're running a gradual traffic shift to a new version. What's the correct status update?
KEY PHRASE: "The canary looks healthy at X% — error rate is nominal, proceeding to Y%" "The canary looks healthy at X%" is standard canary deployment language used by SREs and platform engineers worldwide. It names the deployment strategy (canary), the current traffic percentage (5%), a health signal (error rate is nominal), and the next step (proceeding to 25%). This is exactly what you'd see in a deployment runbook or incident channel update. Real examples: "Canary at 10% — latency nominal, no 5xx, promoting to 50%"; "Canary healthy after 20 minutes at 5% — proceeding with rollout." Options A, B, D avoid the correct terminology entirely.
5 / 5
You receive an on-call alert. How do you acknowledge it to the team?
KEY PHRASE: "PagerDuty fired — I'm on it and starting investigation now" "PagerDuty fired" is the idiomatic on-call shorthand — PagerDuty is the de facto incident-alerting tool, and "fired" means it triggered an alert. "I'm on it" signals immediate engagement, and "starting investigation" tells the team the response has begun. This combination — acknowledge, name the alert source, commit to action — is the standard on-call pickup phrase in engineering teams. Real examples: "PagerDuty fired on high error rate — picking it up"; "Got the PD alert — I'm on it, will update in 5." Option D is especially wrong: "when I can" implies delay, which is unacceptable for on-call incidents.