100+ Free PCA Practice Questions

Pass your Prometheus Certified Associate (PCA) exam on the first try — instant access, no signup required.

✓ No registration✓ No credit card✓ No hidden fees✓ Start practicing immediately

100+ Questions

100% Free

1 / 10

Question 1

Score: 0/0

What kind of database does Prometheus use to store metrics?

Relational (SQL)

Time-series

Document (JSON)

Graph

to track

2026 Statistics

Key Facts: PCA Exam

Questions

Linux Foundation

90 min

Time Limit

Linux Foundation

75%

Passing Score

Linux Foundation

$250

Exam Fee

Linux Foundation

1 retake

Included

PCA product page

12 months

Eligibility Window

Candidate handbook

As of April 15, 2026, the Linux Foundation lists PCA as a 60-question, 90-minute online proctored multiple-choice exam with a 75% passing score and a $250 exam-only fee. The current curriculum weights PromQL at 28%, Prometheus Fundamentals at 20%, Observability Concepts at 18%, Alerting & Dashboarding at 18%, and Instrumentation & Exporters at 16%. One retake is included and certification validity is 2 years.

Sample PCA Practice Questions

Try these sample questions to test your PCA exam readiness. Each question includes a detailed explanation. Start the interactive quiz above for the full 100+ question experience with AI tutoring.

1What kind of database does Prometheus use to store metrics?

A.Relational (SQL)

B.Time-series

C.Document (JSON)

D.Graph

Explanation: Prometheus is a time-series database (TSDB) optimized for storing numeric samples indexed by metric name and label set over time.

2By default, how does Prometheus collect metrics from targets?

A.Pushes from targets via UDP

B.Pulls (scrapes) HTTP endpoints on targets

C.Polls syslog daemons

D.Reads from an Amazon SQS queue

Explanation: Prometheus uses a pull model: it scrapes an HTTP endpoint (typically /metrics) on each target at the configured interval and parses the exposition format.

3Which of the following is NOT a standard Prometheus metric type?

A.counter

B.gauge

C.histogram

D.queue

Explanation: Prometheus defines four metric types: counter, gauge, histogram, and summary. There is no 'queue' metric type.

4A metric that represents the number of HTTP requests served since process start is best modeled as which type?

A.counter

B.gauge

C.summary

D.histogram

Explanation: Counters monotonically increase until the process restarts. Total HTTP requests served fits this exactly; use rate() in PromQL to derive per-second throughput.

5Where is scrape target configuration typically defined in Prometheus?

A.prometheus.yml scrape_configs

B.alertmanager.yml receivers

C.grafana.ini datasources

D.docker-compose environment variables only

Explanation: prometheus.yml contains scrape_configs with job_name, static_configs or service discovery entries, and optional relabeling rules.

6Which component pairs with Prometheus to handle alert routing and notifications?

A.Grafana Loki

B.Alertmanager

C.Thanos Sidecar

D.node_exporter

Explanation: Alertmanager receives alerts from Prometheus and handles deduplication, grouping, silencing, inhibition, and routing to notification integrations.

7Which of the following is the correct Prometheus exposition format sample?

A.{ "name": "http_requests_total", "value": 42 }

B.http_requests_total{method="GET",code="200"} 42

C.http_requests_total = 42 // method=GET

D.<metric name="http_requests_total">42</metric>

Explanation: The Prometheus text exposition format is: metric_name{label="value",...} sample_value [optional_timestamp]. That is the canonical line-oriented plain-text format.

8Prometheus service discovery mechanisms include all of the following EXCEPT:

A.Kubernetes

B.Consul

C.File-based (file_sd)

D.SMTP

Explanation: Prometheus supports dozens of SD mechanisms including Kubernetes, Consul, EC2, Azure, DNS, and file_sd. SMTP is an email protocol and has no role in service discovery.

9Which component of Prometheus handles long-term storage beyond local disk retention?

A.No built-in solution; use remote_write to a compatible backend (Thanos, Cortex, Mimir)

B.Prometheus itself retains data for 10 years by default

C.Alertmanager

D.Grafana

Explanation: Core Prometheus keeps data on local disk (default retention 15 days). For longer horizons, use remote_write to projects like Thanos, Cortex, or Grafana Mimir.

10What does a histogram metric expose in the exposition format?

A.Only a single _bucket time series

B._bucket (cumulative buckets), _sum, _count, plus +Inf bucket

C.Pre-calculated quantiles only

D.No time series; histograms are client-only

Explanation: A histogram exposes cumulative bucket series (..._bucket{le="..."}), a _sum of observations, a _count of observations, and an implicit +Inf bucket. PromQL can then compute quantiles server-side with histogram_quantile.

About the PCA Exam

The Prometheus Certified Associate (PCA) validates foundational observability and monitoring skills using Prometheus, the CNCF-graduated systems monitoring and alerting toolkit. It is a remotely proctored multiple-choice exam that covers PromQL, instrumentation, exporters, alerting, dashboarding, and core observability concepts.

Assessment

60 multiple-choice questions

Time Limit

90 minutes

Passing Score

75%

Exam Fee

$250 (Cloud Native Computing Foundation (CNCF) / Linux Foundation)

PCA Exam Content Outline

28%

PromQL

Selectors, matchers, aggregation operators, rate, irate, increase, histogram_quantile, subqueries, and writing correct query expressions.

20%

Prometheus Fundamentals

Architecture, pull model, scrape configuration, service discovery, metric types (counter, gauge, histogram, summary), and storage.

18%

Observability Concepts

Metrics vs logs vs traces, the RED and USE methods, SLOs, SLIs, error budgets, and cardinality.

18%

Alerting & Dashboarding

Alerting rules, recording rules, Alertmanager routing and silences, notification integrations, and Grafana dashboards.

16%

Instrumentation and Exporters

Client libraries, exposition formats, node_exporter and other common exporters, and instrumenting applications.

How to Pass the PCA Exam

What You Need to Know

Passing score: 75%
Assessment: 60 multiple-choice questions
Time limit: 90 minutes
Exam fee: $250

Keys to Passing

Complete 500+ practice questions
Score 80%+ consistently before scheduling
Focus on highest-weighted sections
Use our AI tutor for tough concepts

PCA Study Tips from Top Performers

1Spend the most time on PromQL because it is 28% of the exam.

2Memorize the four core metric types: counter, gauge, histogram, summary.

3Know when to use rate() vs irate() vs increase().

4Understand histogram_quantile() and how bucket boundaries affect accuracy.

5Practice writing alerting rules and recognize recording rule syntax.

6Study Alertmanager routing trees, grouping, inhibition, and silences.

7Review USE method, RED method, SLOs, SLIs, and error budgets at a conceptual level.

Frequently Asked Questions

How many questions are on the PCA exam?

PCA is a 60-question multiple-choice exam delivered online through a PSI proctored session. Candidates have 90 minutes to finish and results are emailed within 24 hours of the session.

What score do you need to pass PCA?

PCA requires a score of 75% or higher to pass, consistent with other Linux Foundation multiple-choice certifications. Most successful candidates aim for comfortably above 75% on timed practice sets before scheduling.

What are the official PCA domain weights?

The current PCA curriculum weights PromQL at 28%, Prometheus Fundamentals at 20%, Observability Concepts at 18%, Alerting and Dashboarding at 18%, and Instrumentation and Exporters at 16%. PromQL is the single heaviest domain, so fluency with rate, aggregation, histogram_quantile, and label matching is critical.

Do I need to know Grafana for PCA?

Yes, at a concept and usage level. The Alerting and Dashboarding domain includes dashboard creation concepts, and Grafana is the canonical visualization tool paired with Prometheus. You should recognize variables, panels, and how Grafana queries Prometheus, but deep dashboard engineering is out of scope.

How much does the PCA exam cost?

The Linux Foundation lists PCA at $250 for the exam only. Bundles with the LFS241 Monitoring Systems and Services with Prometheus course are sometimes offered at $299, and a THRIVE-ONE subscription bundle is $495.

How long should I study for PCA?

Most candidates with prior monitoring experience prepare in 3 to 5 weeks with 25 to 40 focused study hours. Spend the majority of that time on PromQL because it is the largest domain and the area where candidates lose the most points.