Prometheus Certified Associate (PCA) Practice Exam

What the Prometheus Certified Associate (PCA) exam covers

• Observability Concepts30 questions
• Prometheus Fundamentals74 questions
• PromQL70 questions
• Instrumentation and Exporters67 questions
• Alerting and Dashboards57 questions

Free Prometheus Certified Associate (PCA) sample questions

A sample of 10 questions with answers and explanations. Sign up free to practice all 298.

1. Question 1Observability Concepts

   What type of monitoring system is Prometheus?

   • AA metrics-based, time-series monitoring and alerting system that pulls (scrapes) metricsCorrect
   • BA container runtime
   • CA log aggregation database only
   • DA DNS server
   ✓ Correct answer: A

   Prometheus is a metrics-based, time-series monitoring system that pulls metrics via HTTP scraping. Prometheus provides foundational observability for modern cloud infrastructure. Understanding these concepts is essential for effective monitoring strategy.

   Why the other options are wrong

   • BA container runtime is incorrect because Container runtime is separate from metrics monitoring.
   • CA log aggregation database only is incorrect because this does not provide the specific technical capability required by the question.
   • DA DNS server is incorrect because DNS is for name resolution, not metrics collection.
2. Question 2Prometheus FundamentalsSelect all that apply

   Which TWO metric types summarize distributions (e.g., request latencies)? (Choose TWO)

   • AHistogramCorrect
   • BSummaryCorrect
   • CDNS record
   • DCounter only
   ✓ Correct answer: A, B

   Summary Both Histogram and Summary are essential components within the Prometheus Fundamentals domain. These features work together to provide comprehensive functionality required for effective platform implementation. Understanding how these capabilities integrate is critical for certification-level knowledge.

   Why the other options are wrong

   • CDNS record is incorrect because DNS is a name resolution service unrelated to the stated requirement.
   • DCounter only is incorrect because this does not provide the specific technical capability required by the question.
3. Question 3Prometheus Fundamentals

   What is the effect of setting a very short --storage.tsdb.retention.time relative to a query asking for data over a longer range?

   • AThe query fails with a fatal error
   • BRetention is automatically extended to cover the query range
   • CSamples older than the retention window are deleted from local TSDB and are simply unavailable to the queryCorrect
   • Dprometheus re-scrapes historical data to satisfy the query
   ✓ Correct answer: C

   How long Prometheus stores metrics before deletion. Time-series storage optimizes for high-volume metric ingestion and efficient point-in-time queries. Retention policies manage storage growth while preserving analysis windows.

   Why the other options are wrong

   • AThe query fails with a fatal error is incorrect because this does not provide the specific technical capability required by the question.
   • BRetention is automatically extended to cover the query range is incorrect because this does not provide the specific technical capability required by the question.
   • Dprometheus re-scrapes historical data to satisfy the query is incorrect because this does not provide the specific technical capability required by the question.
4. Question 4PromQL

   Which PromQL operator and modifier match two metrics one-to-many, keeping the left side's many series (e.g., joining a per-pod metric with a per-node metric)?

   • A* join (node) ...
   • B* and (node) ...
   • C* unless (node) ...
   • D... * on (node) group_left(extra_label) ... (one-to-many keeping the left side)Correct
   ✓ Correct answer: D

   Quantitative measurements of system behavior collected and analyzed over time. Prometheus provides foundational observability for modern cloud infrastructure. Understanding these concepts is essential for effective monitoring strategy.

   Why the other options are wrong

   • A* join (node) ... is incorrect because this does not provide the specific technical capability required by the question.
   • B* and (node) ... is incorrect because this does not provide the specific technical capability required by the question.
   • C* unless (node) ... is incorrect because this does not provide the specific technical capability required by the question.
5. Question 5PromQL

   After migrating an app, both an old metric http_requests_total and a renamed metric requests_http_total exist for an overlap period. To build one continuous rate series across the rename, which PromQL handles the union of the two counters?

   • Asum(rate(http_requests_total[5m])) or sum(rate(requests_http_total[5m]))Correct
   • Bhttp_requests_total and requests_http_total
   • Cabsent(http_requests_total)
   • Drate(http_requests_total + requests_http_total)[5m]
   ✓ Correct answer: A

   Quantitative measurements of system behavior collected and analyzed over time. Counters track cumulative values like request counts and error totals. The monotonic nature makes them ideal for rate calculations showing request frequency or error rates.

   Why the other options are wrong

   • Bhttp_requests_total and requests_http_total is incorrect because this does not provide the specific technical capability required by the question.
   • Cabsent(http_requests_total) is incorrect because this does not provide the specific technical capability required by the question.
   • Drate(http_requests_total + requests_http_total)[5m] is incorrect because this does not provide the specific technical capability required by the question.
6. Question 6Instrumentation and ExportersSelect all that apply

   A team adds a label whose value is the full request URL including unique query strings. Which TWO are likely operational consequences? (Choose TWO)

   • AThe metric is automatically converted into a histogram
   • BAutomatic aggregation of the URLs into a single low-cardinality series
   • CA cardinality explosion that drives up Prometheus memory and TSDB series counts, potentially causing OOMCorrect
   • DSlower queries and rule evaluations as the index and head block grow with the number of active seriesCorrect
   ✓ Correct answer: C, D

   Slower queries and rule evaluations as the index and head block grow with the number of active series Both A cardinality explosion that drives up Prometheus memory and TSDB series counts, potentially causing OOM and Slower queries and rule evaluations as the index and head block grow with the number of active series are essential components within the Instrumentation and Exporters domain. These features work together to provide comprehensive functionality required for effective platform implementation. Understanding how these capabilities integrate is critical for certification-level knowledge.

   Why the other options are wrong

   • AThe metric is automatically converted into a histogram is incorrect because this does not provide the specific technical capability required by the question.
   • BAutomatic aggregation of the URLs into a single low-cardinality series is incorrect because this does not provide the specific technical capability required by the question.
7. Question 7Alerting and Dashboards

   Which Alertmanager concept groups related alerts into a single notification?

   • AScraping
   • BRecording rules
   • CGrouping (group_by)Correct
   • DRelabeling
   ✓ Correct answer: C

   Processes alerts from Prometheus and handles routing, grouping, and notifications. Prometheus provides foundational observability for modern cloud infrastructure. Understanding these concepts is essential for effective monitoring strategy.

   Why the other options are wrong

   • AScraping is incorrect because this does not provide the specific technical capability required by the question.
   • BRecording rules is incorrect because this does not provide the specific technical capability required by the question.
   • DRelabeling is incorrect because this does not provide the specific technical capability required by the question.
8. Question 8PromQL

   A Grafana dashboard panel runs sum(rate(http_requests_total[5m])) by (path) across 200 services and is slow on every refresh. Which optimization reduces query cost the most without losing the displayed value?

   • APrecompute the expression with a recording rule and have the panel query the recorded seriesCorrect
   • BWrap the query in absent() to skip empty series
   • Cincrease the range from [5m] to [1h] to read fewer samples
   • DAdd a histogram_quantile() around the sum
   ✓ Correct answer: A

   Service Level Objective - target metric like 99.9% availability. SLIs are the measured indicators tracked through metrics while SLOs are the targets we commit to. Error budgets quantify how much service can degrade before violating SLOs.

   Why the other options are wrong

   • BWrap the query in absent() to skip empty series is incorrect because this does not provide the specific technical capability required by the question.
   • Cincrease the range from [5m] to [1h] to read fewer samples is incorrect because this does not provide the specific technical capability required by the question.
   • DAdd a histogram_quantile() around the sum is incorrect because this does not provide the specific technical capability required by the question.
9. Question 9Prometheus Fundamentals

   A target's /metrics page loads fine in your browser, but in Prometheus the target shows DOWN with the error 'context deadline exceeded'. The endpoint takes about 12 seconds to render under load. Which change most directly fixes the scrape failure?

   • AAdd the target to the Pushgateway instead
   • BLower scrape_interval to 1s so Prometheus retries faster
   • CDisable TLS verification on the job
   • DIncrease scrape_timeout (and if needed scrape_interval) for that job so the slow endpoint can finish renderingCorrect
   ✓ Correct answer: D

   Prometheus is a metrics-based, time-series monitoring system that pulls metrics via HTTP scraping. This pull-based approach enables monitoring without requiring targets to be configured with collector addresses. Targets simply expose a /metrics endpoint that Prometheus periodically requests.

   Why the other options are wrong

   • AAdd the target to the Pushgateway instead is incorrect because this does not provide the specific technical capability required by the question.
   • BLower scrape_interval to 1s so Prometheus retries faster is incorrect because this does not provide the specific technical capability required by the question.
   • CDisable TLS verification on the job is incorrect because this does not provide the specific technical capability required by the question.
10. Question 10Prometheus Fundamentals

    To satisfy a compliance requirement that the Prometheus web UI and API be served only over encrypted connections, which configuration enables HTTPS natively in Prometheus 2.x without a separate reverse proxy?

    • Aenable --web.enable-lifecycle
    • BSet scheme: https in the global scrape_config block
    • Cpass --web.config.file pointing to a web config that defines tls_server_config with cert_file and key_fileCorrect
    • DAdd tls_config to each alerting rule
    ✓ Correct answer: C

    Prometheus is a metrics-based, time-series monitoring system that pulls metrics via HTTP scraping. Prometheus provides foundational observability for modern cloud infrastructure. Understanding these concepts is essential for effective monitoring strategy.

    Why the other options are wrong

    • Aenable --web.enable-lifecycle is incorrect because this does not provide the specific technical capability required by the question.
    • BSet scheme: https in the global scrape_config block is incorrect because this does not provide the specific technical capability required by the question.
    • DAdd tls_config to each alerting rule is incorrect because this does not provide the specific technical capability required by the question.

Prometheus Certified Associate (PCA) practice exam FAQ