Cloud Deploy CD Pipeline with Manual QA

PCDOE By islever Editorial Last updated

Designing pipelines.

Short answer

D Create two pipeline stages, and use a standard deployment strategy.

Question

Your company wants to implement a CD pipeline in Cloud Deploy for a web service deployed to GKE. The web service currently does not have any automated testing. The Quality Assurance team must manually verify any new releases of the web service before any production traffic is processed. You need to design the CD pipeline. What should you do?
  1. Create a single pipeline stage, and use a standard deployment strategy.
  2. Create a single pipeline stage, and use a canary deployment strategy.
  3. Create two pipeline stages, and use a canary deployment strategy.
  4. Create two pipeline stages, and use a standard deployment strategy. Correct Answer

Key Takeaways

The exam tests the ability to structure a pipeline around human-in-the-loop verification; the common trap is assuming canary deployment is always superior for risk reduction, ignoring that canary relies on automated metrics which don't exist here.

This question tests the design of a Cloud Deploy pipeline for GKE when automated testing is absent but manual QA verification is required. It establishes that two stages are needed to separate deployment from validation, and a standard strategy is safer after manual approval than a canary.

Common mistake: Many learners choose C (Canary) because they associate 'canary' with safety. However, canary deployments typically rely on automated traffic shifting and monitoring metrics to decide whether to promote or roll back. Without automated testing/monitoring, a manual check on a full rollout (Standard) after a staging phase is the more robust manual workflow.

Expert Analysis

Why the Answer Is Correct

The correct approach is to create two pipeline stages: one for initial deployment (e.g., to a staging environment) and a second for production promotion. Since there is no automated testing, the Quality Assurance team needs a stable environment to perform manual verification before any production traffic is touched. Using a 'standard deployment strategy' in the final stage means that once QA approves the release in the previous stage, the new version is deployed to production (often via a rolling update or blue/green switch) to serve all traffic. This separates the 'build/deploy' action from the 'verify/promote' action, ensuring manual QA gates are respected.

Why the Other Options Are Wrong

Option A fails because a single stage cannot accommodate the mandatory manual verification step between deployment and production traffic processing. Option B uses a canary strategy, which gradually shifts traffic. While canary is good for automated risk mitigation, it assumes you can monitor health automatically. If QA must manually verify before any production traffic, a canary (which sends some traffic immediately upon deploy) violates the requirement. Option D suggests a standard strategy, but without the separation of stages, it doesn't provide the necessary gate for the QA team to intervene.

Community Comment Notes

The community is split, but several users like kleranti, RushinthJohn, kitaharazyl, BlackCoffee, and a36df5b argue for D. As RushinthJohn noted, "Two pipeline stages... allows for a dedicated QA stage before production." Others like hproms and odinson prefer C, stating "Canary will be better," likely confusing general best practices with this specific constraint of manual-only QA. The consensus among those who understand the 'manual verification before any traffic' constraint leans toward the staged approach with standard promotion.

Exam Strategy

When a question specifies 'manual verification' and 'no automated testing', look for pipeline structures that explicitly separate deployment from promotion/validation. Avoid canary/rolling strategies if they imply automatic traffic shifting based on metrics that don't exist; instead, favor a 'staging then production' stage model where the final stage acts as the controlled rollout after human approval.

Frequently Asked Questions

Why not use Canary deployment for manual QA?

Canary deployments shift a small percentage of traffic automatically. If QA must verify BEFORE any production traffic is processed, you cannot start a canary. You must deploy to a non-production stage first for manual checks.

What does 'standard deployment strategy' mean in Cloud Deploy?

It refers to deploying the new version to replace the old one (like a rolling update or blue/green) in the target environment, rather than gradually shifting traffic percentages as in a canary.

More PCDOE FAQ →

Discussion 6 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 6

  1. RushinthJohn 👍 3 Answered C
    Two pipeline stages: Similar to C, this allows for a dedicated QA stage before production. Standard deployment strategy (in the Production stage): After QA, if a "standard deployment" (like a rolling update or blue/green switch) is used, it means the new version gets 100% of production traffic fairly quickly. Problem: While this allows for manual QA, without automated testing, jumping from manual QA directly to 100% production traffic is riskier than a canary strategy. A standard deployment doesn't offer the gradual rollout and risk mitigation that a canary does, especially critical when automated testing is absent.
  2. kleranti 👍 3 Answered D
    D is better
  3. hproms 👍 2 Answered C
    I think C is better when deploying with manual inspection without QA.
  4. odinson 👍 2 Answered C
    Canary will be better
  5. BlackCoffee 👍 1 Answered D
    D - Create two pipeline stages, and use a standard deployment strategy.
Show 1 more comment
  1. a36df5b 👍 1 Answered D
    D. Create two pipeline stages, and use a standard deployment strategy.

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Submitted for moderation before publishing. Keep it helpful and respectful.

Binary Authorization Attestor Implementation

S-grade

GKE Cluster and Namespace Design for Data Isolation

S-grade

Granting Log Export Permissions in Google Cloud

S-grade

Optimizing SLO Balance Between Reliability and Velocity

S-grade

Practice All PCDOE Questions

Access 75 questions with complete answers and detailed explanations.

View Full PCDOE Practice Test →

← Back to PCDOE Study Guide