Practice Exam PCA: Prometheus Certified Associate

Question bank: 110 · timed mock exam: 90 min

Official exam page on the vendor's site (opens in a new tab) Official exam guide (opens in a new tab)

Compare: PCA vs CGOA

  • PromQL 31 questions
  • Prometheus Fundamentals 22 questions
  • Observability Concepts 21 questions
  • Alerting & Dashboarding 19 questions
  • Instrumentation and Exporters 17 questions
Study mode (free) Exam mode (subscribers)

Free: 25 questions in study mode, all domains, no account.

Subscription: the 110 questions of this certification, the timed exam mode, and all other certifications too.

Subscribe · €9.99 / month

Secure card payment (Stripe) · No commitment · Cancel online

How to prepare for Prometheus Certified Associate with InfraPrep

The 110 questions cover 5 domains: PromQL, Prometheus Fundamentals, Observability Concepts, Alerting & Dashboarding, Instrumentation and Exporters. 25 of them are free, without an account.

Study mode

Answer at your pace: each answer is corrected straight away, with an explanation and the link to the documentation it relies on.

Exam mode

A timed exam of 90 minutes, with a question map to flag and come back to questions before you submit.

Results by domain

Your score is broken down by domain, so you know what to review first.

Original questions

Every question is written from the public documentation, never copied from the exam. You can contest an answer from study mode.

10 free sample questions

Question 1 Observability Concepts

A new engineer joins the SRE team and asks how the Prometheus documentation itself defines the word "metric". Which statement matches the official definition?

  1. Metrics are structured log lines that Prometheus indexes and stores for full-text search
  2. Metrics are the alert notifications generated by Alertmanager when a threshold is crossed
  3. Metrics are the dashboard panels rendered by a visualization tool from raw event streams
  4. Metrics are numerical measurements
Show the answer and the explanation

Correct answer: D

The documentation defines metrics as numerical measurements, with the term "time series" describing how changes in that value are recorded over time.

Source: official documentation

Question 2 Prometheus Fundamentals

Which statement best summarizes the core architecture of a Prometheus server, as described on the official overview page?

  1. Prometheus scrapes metrics from instrumented jobs, stores all scraped samples locally
  2. Prometheus writes every sample directly to Alertmanager, which is responsible for local storage
  3. Prometheus requires every target to push its metrics into a shared distributed database before they can be queried
  4. Prometheus only stores data in memory and discards it once a dashboard has rendered it
Show the answer and the explanation

Correct answer: A

The overview's architecture description states that Prometheus scrapes metrics from instrumented jobs, stores all scraped samples locally, and runs rules over that data to aggregate/record new series or generate alerts.

Source: official documentation

Question 3 PromQL

A PromQL expression `http_requests_total{job="api"}` (with no range selector) is evaluated at a single point in time. Per the querying basics documentation, which expression type does this query return?

  1. A range vector, containing a range of data points over time for each series
  2. An instant vector: time series with a single sample each, sharing the same timestamp
  3. A scalar, a simple numeric floating point value
  4. A string, currently unused for this kind of selector
Show the answer and the explanation

Correct answer: B

The querying basics page defines an instant vector as a set of time series containing a single sample for each time series, all sharing the same timestamp, which is exactly what a bare metric selector without a range returns.

Source: official documentation

Question 4 Instrumentation and Exporters

A developer's application is written in a language that has no official Prometheus client library. According to the client libraries documentation, what is the recommended way for that application to expose metrics?

  1. The application cannot expose Prometheus metrics unless it is rewritten in Go, Java/Scala, Node.js, Python, Ruby, or Rust
  2. It must run a separate exporter written by the Prometheus team for every unofficial language
  3. If no client library exists for the language, implement one of the supported exposition formats directly
  4. It must send its metrics to Alertmanager, which will convert them into Prometheus format
Show the answer and the explanation

Correct answer: C

The client libraries documentation states that if no client library is available for your language, or you want to avoid dependencies, you may implement one of the supported exposition formats yourself to expose metrics.

Source: official documentation

Question 5 Alerting & Dashboarding

A team upgrading to Prometheus 3.0 notices that console templates, which they used for simple dashboards, no longer render out of the box. According to the console templates documentation, what changed?

  1. Console templates were renamed to Perses templates but otherwise work identically with no configuration changes
  2. Console templates now require a separate paid license from the Prometheus team
  3. Console templates were merged into PromQL and are now written as PromQL expressions instead of HTML
  4. Starting with Prometheus 3.0, console templates and libraries are no longer bundled
Show the answer and the explanation

Correct answer: D

The documentation explicitly warns that starting with Prometheus 3.0, console templates and libraries are no longer bundled with Prometheus, and that using them now requires providing your own templates and libraries via the `--web.console.templates` and `--web.console.libraries` command-line flags.

Source: official documentation

Question 6 Observability Concepts

A colleague argues that, from an observability-concepts point of view, logs/events and metrics are basically the same kind of data and can be swapped for one another. Which statement from the OpenMetrics specification (an observability data-model reference linked from the Prometheus documentation) best corrects this claim?

  1. Logs and events are always derived from metrics, so they never need to be collected separately
  2. Contrary to metrics, singular events occur at a specific time, while metrics tend to aggregate data temporally
  3. Metrics are only a visualization layer built on top of logs
  4. Events and metrics use the exact same storage engine in Prometheus, so the distinction is purely cosmetic
Show the answer and the explanation

Correct answer: B

The specification explicitly contrasts the two: events happen at a specific point in time, whereas metrics aggregate data over time.

Source: official documentation

Question 7 Prometheus Fundamentals

An operator writes a scrape configuration for a new job but does not set a job-level `scrape_interval`. According to the configuration reference, what value will Prometheus use for that job?

  1. The job falls back to the value configured in the top-level `global.scrape_interval`
  2. The job is scraped exactly once and never again
  3. The job uses a hard-coded interval of 5 seconds regardless of any configuration
  4. Prometheus refuses to start until a job-level scrape_interval is explicitly set
Show the answer and the explanation

Correct answer: A

The configuration reference shows that the job-level `scrape_interval` field defaults to the value of `global_config.scrape_interval`, i.e. it falls back to the global setting when not set per job.

Source: official documentation

Question 8 PromQL

An analyst runs `deriv(cpu_temperature_celsius[10m])` against a gauge that only has one float sample inside the 10-minute window. Based on the functions documentation, what should they expect?

  1. The function calculates the per-second derivative using simple linear regression
  2. `deriv()` always returns zero when fewer than three samples are present
  3. `deriv()` is only defined for counters, never for gauges
  4. The query fails immediately with a parse error because the range is too short
Show the answer and the explanation

Correct answer: A

The documentation states that `deriv()` calculates the per-second derivative of each float time series using simple linear regression, and that the range vector must have at least two float samples to perform the calculation — with only one sample, no derivative can be computed.

Source: official documentation

Question 9 Instrumentation and Exporters

An instrumentation guide describes a metric whose value should only ever increase, or reset to zero when the process restarts, and that is useful for accumulating the number of HTTP requests handled. Which metric type does this describe, per the instrumentation best practices?

  1. Gauge
  2. Summary
  3. Counter
  4. Histogram
Show the answer and the explanation

Correct answer: C

The best-practices documentation states that counters can only go up (and reset, such as when a process restarts), and that they are useful for accumulating the number of events, such as the total number of HTTP requests.

Source: official documentation

Question 10 Alerting & Dashboarding

An alerting rule uses `expr: up == 0` with `for: 10m`. At minute 3 after the target first goes down, the alert condition is already true. According to the alerting rules documentation, what is required before Prometheus counts this alert as firing?

  1. Nothing more: the alert fires immediately as soon as the expression is true at any point in time
  2. Prometheus must see the alert expression stay true continuously for the configured `for` duration before firing
  3. The rule must also define a recording rule with the same name before it can fire
  4. The alert can only fire once Alertmanager has been restarted
Show the answer and the explanation

Correct answer: B

The documentation explains that the optional `for` clause causes Prometheus to wait for a certain duration between first encountering a new expression output vector element and counting the alert as firing for that element, so `for: 10m` means it only fires after remaining active for 10 minutes.

Source: official documentation

Frequently asked questions

Are there free questions, and do I need an account?

Yes. 25 questions of this certification are free in study mode, across all its domains, without a subscription and without registration. The account is created when you subscribe, at checkout.

How does the subscription work, and can I cancel?

One monthly subscription covers all certifications and the timed practice exams. There is no minimum period. You cancel online, at any time, with the "Cancel your contract" function in My account; cancellation takes effect at the end of the month already paid, and no further payment is taken.

Where do the questions come from?

Every question is original, written from the vendor's public documentation. Each answer comes with an explanation and a link to the documentation page it relies on. InfraPrep is not affiliated with the vendors, and certification names belong to their owners.

In which language are the questions?

The questions are written in English, also on the French version of the site.

Is the catalog updated?

Yes, the catalog changes over time: new certifications and new questions are added.