Question 1

What function calculates the tp-quantile from a histogram?

Accepted Answer

A

Explanation: The histogramquantile() function is the specific PromQL function used to calculate the φ-quantile from a histogram. It takes two arguments: the desired quantile (a float between 0 and 1) and an instant vector containing the histogram's bucket counts (the time series with the bucket suffix). It then computes the estimated value at that quantile, allowing users to calculate metrics like the 95th percentile request latency from a histogram metric.

Question 2

How can you select all the up metrics whose instance label matches the regex fe-.*?

Accepted Answer

D

Explanation: In Prometheus Query Language (PromQL), time series are filtered using label selectors. To filter based on a regular expression match, the =~ operator is used. The query up{instance=~"fe-."} correctly selects all time series with the metric name up where the instance label value matches the regular expression fe-.. This expression matches any instance label that starts with the string "fe-".

Question 3

Which exporter would be best suited for basic HTTP probing?

Accepted Answer

B

Explanation: The Blackbox exporter is specifically designed for "blackbox" probing of endpoints. It allows Prometheus to monitor external-facing services over various protocols, including HTTP, HTTPS, DNS, TCP, and ICMP. For basic HTTP probing, it can be configured to check an endpoint's availability, measure response time, verify SSL certificates, and validate HTTP status codes or response body content. This makes it the ideal tool for assessing service health from a user's perspective, which is the core of HTTP probing. The other exporters are designed for "whitebox" monitoring of specific technologies.

Question 4

What Prometheus component would you use if targets are running behind a Firewall/NAT?

Accepted Answer

D

Explanation: PushProx is a specialized component designed to allow Prometheus to scrape targets that are not directly reachable, such as those behind a firewall or NAT. It works by having a pushprox-client on the target's network initiate a long-lived outbound connection to a pushprox-proxy. The Prometheus server, which can reach the proxy, sends its scrape request to the proxy. The proxy then forwards this request over the pre-established connection to the client, which collects the metrics from the target and sends them back. This effectively reverses the connection initiation direction, enabling the standard pull model across restrictive network boundaries.

Question 5

What is the name of the official *nix OS kernel metrics exporter?

Accepted Answer

B

Explanation: The nodeexporter is the official and most widely used Prometheus exporter for gathering hardware and operating system metrics from NIX kernels, including Linux. It is designed to expose a wide variety of system-level statistics, such as CPU utilization, memory usage, disk I/O, and network traffic. In the Prometheus ecosystem, the term "node" is a synonym for a host or machine, hence the name nodeexporter. It is a standalone binary that runs on the target machine to collect and expose these metrics over HTTP for a Prometheus server to scrape.

Question 6

Which function would you use to calculate the 95th percentile latency from histogram data?

Accepted Answer

B

Explanation: The histogramquantile() function is specifically designed in Prometheus Query Language (PromQL) to calculate quantiles from histogram buckets. The query histogramquantile(0.95, sum(rate(httprequestdurationsecondsbucket[5m])) by (le)) correctly follows the required pattern. It first calculates the per-second rate of observations in each bucket over a 5-minute window using rate(). Then, sum(...) by (le) aggregates these rates across different instances while preserving the bucket boundary (le) label, which is essential for the quantile calculation. Finally, histogramquantile() takes this aggregated data and the desired quantile (0.95) to compute the 95th percentile value.

Question 7

Which PromQL expression computes the rate of API Server requests across the different cloud
providers from the following metrics?
apiserver_request_total{job="kube-apiserver", instance="192.168.1.220:6443", cloud="aws"} 1
apiserver_request_total{job="kube-apiserver", instance="192.168.1.121:6443", cloud="gcloud"} 5

Accepted Answer

C

Explanation: The goal is to compute the rate of requests aggregated by the cloud label. The correct approach is to first calculate the per-second rate for each individual time series using rate(apiserverrequesttotal[5m]). This function operates on the apiserverrequesttotal counter over a 5-minute window. The result is an instant vector of rates. Subsequently, the sum() by (cloud) aggregation operator is applied to this vector of rates. This sums up the rates for all instances belonging to the same cloud provider, effectively giving the total request rate per cloud.

Question 8

Which of the following is an invalid @ modifier expression?

Accepted Answer

D

Explanation: The @ modifier in PromQL is used to change the evaluation time for an expression. When used with aggregation operators such as sum(), the canonical and syntactically correct method is to apply the modifier to the entire aggregation expression. The expression sum(httprequeststotal{method="GET"}) @ 1609746000 correctly evaluates the sum() function as if the current time were the specified timestamp. The syntax in option D, sum(httprequeststotal{method="GET"} @ 1609746000), which places the @ modifier on the instant vector selector inside the aggregation function, is not the standard syntax for this operation and is considered invalid.

Question 9

Which metric type uses the delta() function?

Accepted Answer

C

Explanation: The delta() function in PromQL calculates the difference between the first and last value of each time series in a specified time range. This function is specifically designed for the Gauge metric type, which represents a numerical value that can arbitrarily go up and down, such as temperature or current memory usage. For example, delta(cputempcelsius[2h]) calculates how much the CPU temperature has changed over the last two hours. For Counter metrics, which only increase, the increase() function is the appropriate choice as it correctly handles counter resets.

Question 10

What's "wrong" with the myapp_filG_uploads_total{userid=,,5123",status="failed"} metric?

Accepted Answer

A

Explanation: The primary issue with the metric myappfilGuploadstotal{userid=,,5123",status="failed"} is the use of userid as a label. User IDs are considered high-cardinality labels because they can have a vast and unbounded number of unique values (one for each user). In Prometheus, every unique combination of a metric name and its labels creates a new time series. Using high-cardinality labels like userid leads to a "cardinality explosion," which drastically increases memory consumption, storage requirements, and query latency on the Prometheus server, potentially making it unstable. Best practices strongly advise against using identifiers with high or unbounded cardinality as labels.

Question 11

What is considered the best practice when working with alerting notifications?

Accepted Answer

B

Explanation: The best practice in modern monitoring is to create a high-signal, low-noise alerting system. This is achieved by alerting on symptoms that are, or are about to become, externally visible and impact users (e.g., high error rates, increased latency). This approach ensures that every alert is meaningful, actionable, and warrants human intervention. Focusing on symptoms rather than every possible underlying cause prevents "alert fatigue," where operators become desensitized to frequent, low-impact notifications. This philosophy prioritizes system reliability from the end-user's perspective and makes the on-call process more effective and sustainable.

Question 12

What should you do with counters that have labels?

Accepted Answer

C

Explanation: Initializing a counter with all its potential label values to zero is a crucial best practice. When a metric with a specific label set has not yet been incremented, it is not exposed by the client library. This can cause a time series to appear suddenly, leading to incorrect calculations by functions like rate() and increase(), which rely on the continuous existence of a series to work correctly. By instantiating all possible label combinations with a zero value upon application startup, you ensure that the time series exists from the beginning, which leads to more predictable and accurate alerting and graphing.

Question 13

Which PromQL statement returns the average free bytes of the filesystems over the last hour?

Accepted Answer

A

Explanation: The PromQL function avgovertime() is specifically designed to calculate the average value of all data points within a specified time interval for each time series. The query avgovertime(nodefilesystemavailbytes[1h]) correctly applies this function to the nodefilesystemavailbytes metric, which is a gauge representing available disk space. The [1h] is a range vector selector that specifies the time duration—one hour—over which the average should be calculated. This directly answers the question of finding the average free bytes over the last hour.

Question 14

How do you configure the rule evaluation interval in Prometheus?

Accepted Answer

A

Explanation: The Prometheus rule evaluation interval is configured in two primary locations. A default interval for all rule groups is set globally in the main Prometheus configuration file (prometheus.yml) under the global block using the evaluationinterval parameter. This global default can be overridden for specific groups of rules by defining an interval key within a rulegroup in a separate rule configuration file (e.g., rules.yml). This hierarchical approach allows for a general setting while providing granular control over how frequently specific rule sets are evaluated.

Question 15

What is the difference between client libraries and exporters?

Accepted Answer

C

Explanation: The fundamental difference lies in their application. Client libraries are used to add instrumentation directly into the source code of an application you control, allowing it to expose its own internal state as Prometheus metrics. In contrast, an exporter is a standalone application that runs alongside a target service that cannot be directly instrumented (e.g., a database, OS kernel, or third-party software). The exporter collects metrics from the target service and then uses a Prometheus client library internally to translate and expose these metrics in the Prometheus format for scraping.

Free Linux Foundation PCA Actual Exam Questions