PCA Cheat Sheet 2026
The 30 highest-yield PCA facts, distilled from real exam questions. Print it, save it as a PDF, or study it here — free, no sign-up.
60 questions
90 min time limit
75.00% to pass
- What is the Prometheus exposition format, and which content type does it use? → A text-based line format exposed over HTTP with Content-Type: text/plain; version=0.0.4
- What content type must a Prometheus metrics endpoint return for the text exposition format? → text/plain; version=0.0.4
- What does the `up` metric that Prometheus automatically generates represent? → Whether a scrape target was successfully reached (1) or failed (0)
- What is the purpose of the `resets()` function in PromQL? → Returns the number of counter resets (value drops) within the range window
- What happens to 'job' and 'instance' labels when metrics are federated from one Prometheus server to another? → They are preserved as-is from the originating Prometheus server
- In the Prometheus data model, what uniquely identifies a time series? → The metric name plus its set of label name-value pairs
- In Prometheus web configuration, what does the `http2` field control? → Whether Prometheus serves its own API over HTTP/2
- Which flag controls how long Prometheus retains data when using time-based retention? → --storage.tsdb.retention.time
- What is the role of Grafana in Prometheus? → To visualize data in dashboards
- Which HTTP endpoint should you check on a Prometheus server to verify that a specific target is being scraped successfully? → /api/v1/targets
- How are Prometheus metrics structured? → As logs and alerts
- Which Prometheus configuration option controls how often a target is scraped if not overridden at the job level? → `scrape_interval`
- Which exporter is the standard choice for exposing Linux host-level metrics such as CPU, memory, and disk? → node_exporter
- What does `absent(up{job="myapp"})` return when the time series EXISTS? → An empty vector (no data)
- What does the `remote_read` configuration in Prometheus enable? → Querying historical data from a remote storage backend when local data is unavailable
- How does the `offset` modifier affect a PromQL query like `http_requests_total offset 1h`? → It retrieves the metric value as it was 1 hour ago
- How does Prometheus handle a bearer token stored in a file for scrape authentication? → It reads the token file on each scrape, automatically picking up rotated tokens
- When troubleshooting a failing alerting rule, which Prometheus API endpoint provides the current evaluation state and any errors for alerting rules? → /api/v1/rules
- Which Prometheus configuration option allows scraping a target that uses a self-signed TLS certificate without verification? → insecure_skip_verify: true
- A Prometheus scrape target returns a metric with timestamp older than the staleness window. What happens? → Prometheus stores the stale sample with a special NaN value
- What OAuth2 configuration capability does Prometheus support natively in scrape_configs? → Client credentials flow to obtain a bearer token before scraping
- What is the significance of the `__name__` label in Prometheus? → It holds the metric name and can be used in label matchers
- When configuring a Prometheus scrape job, what does `honor_timestamps: true` do? → Uses the timestamps provided by the target's exposition format instead of the scrape time
- What is a Prometheus 'target' in the context of its architecture? → An HTTP endpoint Prometheus scrapes for metrics
- What does the `quantile` label represent in a Prometheus Summary metric? → A pre-calculated percentile value over a sliding time window
- What is the main purpose of metrics in Prometheus? → To provide quantitative data on system performance
- Which exporter is the standard choice for exposing Linux/Unix OS-level metrics like CPU, memory, and disk to Prometheus? → node_exporter
- What does the Prometheus remote_read feature enable? → Querying historical data from a remote storage backend via Prometheus's query API
- Which PromQL function would you use to predict a gauge's value 1 hour from now based on recent trends? → `predict_linear()`
- What is the primary difference between `rate()` and `irate()` in Prometheus? → `rate()` averages over the full range while `irate()` uses only the last two data points
Turn these facts into recall:
Was this helpful?