I85PREP
⌘ K
New Relic APM Practitioner study materials
New Relic APM Practitioner study resources

New Relic APM Practitioner practice questions

One question at a time. Build your New Relic APM Practitioner understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Application Performance Monitoring (APM 360)

In New Relic APM, what does the Apdex score measure for a monitored application?

  1. The ratio of error responses to total throughput across all transactions
  2. The average CPU utilization of all hosts reporting to the application
  3. The total number of unique end users who loaded a browser page
  4. The percentage of requests whose response time is at or below a configured satisfactory threshold, with tolerating and frustrating bands Correct answer.

Explanation

Apdex is a user-satisfaction metric derived from a target response time T. Requests at or below T are satisfied, requests between T and 4T are tolerating, and slower requests are frustrating; the score is (satisfied + tolerating/2) / total. It is not a resource or error metric.

Question 2Application Performance Monitoring (APM 360)

A transaction in New Relic APM shows a high total time but very low application (self) time, with most time attributed to external service calls. What does this most directly indicate?

  1. The application code itself is spending most of its execution time in local method calls
  2. The host running the agent is under-provisioned on memory
  3. The Apdex threshold T is configured too low for this transaction
  4. The transaction is waiting on downstream dependencies rather than doing its own work Correct answer.

Explanation

In the transaction breakdown, application time excludes time spent in external services and datastores. When self time is small and external time dominates, the latency originates in a downstream dependency, so investigation should focus there rather than on local code.

Question 3Application Performance Monitoring (APM 360)

Which New Relic APM feature lets you follow a single request across multiple services and see the timing of each downstream call?

  1. Alert policies
  2. Distributed tracing Correct answer.
  3. Dashboards
  4. Synthetic monitoring

Explanation

Distributed tracing propagates trace context across service boundaries so one request can be reconstructed as spans in different services. This is how cross-service latency and call chains are visualized in APM.

Question 4Application Performance Monitoring (APM 360)

An engineer wants New Relic APM to flag slow database queries as separate trace segments with their SQL statement. Which capability provides this?

  1. Synthetic scripted browser steps
  2. Infrastructure agent process sampling
  3. Slow query traces captured by the database instrumentation in the agent Correct answer.
  4. Browser (RUM) session traces

Explanation

The APM agent's datastore instrumentation records query segments and can capture slow query traces including the obfuscated SQL and stack trace. These appear in the datastore and slow query views.

Question 5Application Performance Monitoring (APM 360)

What is the purpose of the error rate metric in New Relic APM?

  1. It reports the number of distinct exception classes thrown by the JVM
  2. It reports the percentage of hosts that failed a health check
  3. It reports the fraction of requests that returned an error response during a time window Correct answer.
  4. It reports the count of alert violations opened in the period

Explanation

Error rate expresses errored transactions as a proportion of total transactions over the selected window. It is distinct from raw exception counts and from alert incident counts.

Question 6Application Performance Monitoring (APM 360)

A team notices that a specific transaction's error rate spiked after a deployment, but overall application error rate is unchanged. What is the most efficient next step in New Relic APM?

  1. Raise the Apdex threshold so the transaction is no longer scored
  2. Disable the agent on one host to compare behavior
  3. Delete the alert policy and recreate it
  4. Open the transaction's Errors page and group errors by error class and message Correct answer.

Explanation

Because the spike is localized to one transaction, drilling into that transaction's error traces and grouping by class or message isolates the failing code path quickly. Broad configuration changes would obscure the signal.

Question 7Application Performance Monitoring (APM 360)

Which metric best represents the time the application code itself spends executing, excluding time in external calls and datastores?

  1. Total response time
  2. Throughput
  3. Apdex score
  4. Application (self) time Correct answer.

Explanation

Application time, also called self time, is total transaction time minus time attributed to external services and datastore segments. It isolates the code the team owns.

Question 8Application Performance Monitoring (APM 360)

An APM user needs to know how many requests per minute a web transaction handles. Which metric should they inspect?

  1. Memory usage
  2. Error rate
  3. Throughput (requests per minute) Correct answer.
  4. Apdex

Explanation

Throughput in APM is expressed as requests per minute for the selected transaction or application. It measures volume rather than latency or satisfaction.

Question 9Application Performance Monitoring (APM 360)

What does a New Relic APM key transaction allow you to do?

  1. Capture browser session replays for one endpoint
  2. Route traffic for one transaction to a dedicated host pool
  3. Track a business-critical transaction with its own Apdex threshold and dedicated alerting Correct answer.
  4. Encrypt the payload of a specific request at the agent level

Explanation

Key transactions let you single out important flows and set a separate Apdex target T plus dedicated alert conditions. This gives focused SLO-style visibility beyond the application-wide score.

Question 10Application Performance Monitoring (APM 360)

In the New Relic APM UI, where would you look to see how much of a transaction's time is spent in a particular external HTTP call?

  1. The alert policy's condition thresholds
  2. The transaction breakdown and externals section of the transaction's summary Correct answer.
  3. The host's network interface metrics
  4. The synthetic monitor's step timings

Explanation

The transaction summary includes a time breakdown chart plus lists of segments such as externals and datastores, showing each call's contribution. This is the standard place to attribute latency.

Question 11Application Performance Monitoring (APM 360)

Which statement about New Relic APM's browser (RUM) monitoring is accurate?

  1. It captures only synthetic traffic generated from selected locations
  2. It replaces the APM agent for server-side transaction tracing
  3. It instruments server-side SQL queries for slow-query analysis
  4. It measures page-load and AJAX timing in the end user's browser and reports to New Relic Correct answer.

Explanation

Browser monitoring uses a JavaScript agent to time page loads, AJAX calls, and JavaScript errors in real user browsers. It complements rather than replaces server-side APM instrumentation.

Question 12Application Performance Monitoring (APM 360)

A service's Apdex dropped sharply while throughput stayed flat and error rate stayed near zero. What is the most likely explanation?

  1. Response times increased for the same volume of successful requests Correct answer.
  2. The agent stopped reporting throughput data
  3. The number of hosts reporting dropped unexpectedly
  4. The application started returning many HTTP 500 responses

Explanation

With stable throughput and negligible errors, the Apdex decline must come from slower responses pushing satisfied requests into tolerating or frustrating bands. Errors or missing data would change other metrics too.

Question 13Application Performance Monitoring (APM 360)

Which New Relic capability groups related transactions under a single named logical service so they can be analyzed together?

  1. APM application grouping of transaction names Correct answer.
  2. Log parsing rules
  3. Synthetic monitor scripts
  4. Alert conditions

Explanation

The agent groups transactions into an APM application, and transaction naming determines how requests roll up within it. This grouping is what the application-level overview aggregates.

Question 14Application Performance Monitoring (APM 360)

An engineer wants to reduce noise from health-check endpoints in APM data. Which approach is most appropriate?

  1. Lower the alert threshold for throughput
  2. Exclude or ignore the health-check transaction names via agent configuration so they are not reported Correct answer.
  3. Disable error collection entirely
  4. Raise the Apdex threshold for the whole application

Explanation

Agent configuration can ignore specific transaction names or URLs, removing low-value endpoints from metrics without affecting other transactions. This is the targeted fix for noise from probes.

Question 15Application Performance Monitoring (APM 360)

Which statement best describes how New Relic APM handles errors by default?

  1. It stores errors only when distributed tracing is enabled
  2. It only records errors that occur in browser JavaScript
  3. It suppresses all errors until an alert policy fires
  4. It captures error traces with class, message, and stack trace for errored transactions Correct answer.

Explanation

The APM agent records error traces including the error class, message, and stack trace, and rolls them up into error rate metrics. Alerting is a separate layer built on top of this data.

Question 16Application Performance Monitoring (APM 360)

A transaction's response time is high, and the breakdown shows most time in a datastore segment. Which investigation is most direct?

  1. Review slow query traces and the datastore call details for that segment Correct answer.
  2. Inspect the alert policy's incident history
  3. Review synthetic monitor location results
  4. Check the browser page-load waterfall

Explanation

When the datastore segment dominates, slow query traces and datastore call details reveal which queries are slow and their obfuscated SQL. That points directly at indexes or query plans.

Question 17Application Performance Monitoring (APM 360)

What is the role of the Apdex T value in APM?

  1. It is the maximum number of errors allowed per minute
  2. It is the retention period for metric data
  3. It is the target response time that separates satisfied from tolerating requests Correct answer.
  4. It is the sampling rate for distributed traces

Explanation

T is the threshold at which a request is considered satisfied; requests up to 4T are tolerating. Choosing T appropriately is essential because it defines the whole Apdex scale for that application or key transaction.

Question 18Application Performance Monitoring (APM 360)

Which APM view helps identify which deployment introduced a regression in response time?

  1. Deployments markers overlaid on the application's response time chart Correct answer.
  2. The synthetic monitor script editor
  3. The browser session trace list
  4. The infrastructure process list

Explanation

Deployment markers record releases on the APM timeline, so a step change in response time can be correlated with a specific deploy. This shortens root-cause time after releases.

Question 19Application Performance Monitoring (APM 360)

In distributed tracing, what does a span represent?

  1. The entire application's monthly error budget
  2. A single timed operation within a trace, such as a service call or database query Correct answer.
  3. A synthetic monitor's scheduling interval
  4. A host's CPU utilization sample

Explanation

A trace is composed of spans, each representing one timed unit of work with attributes such as service, duration, and parent. Together the spans form the call tree for a request.

Question 20Application Performance Monitoring (APM 360)

An application has high throughput but a low Apdex score. Which combination of changes would most directly improve the score?

  1. Adding more synthetic monitors from additional locations
  2. Lowering the error rate threshold in the alert policy
  3. Increasing the number of hosts reporting to the application
  4. Reducing response times for the slowest requests so more fall within the target threshold Correct answer.

Explanation

Apdex weights satisfied and tolerating requests, so shifting slow requests under T (or under 4T) raises the score. Adding hosts or monitors does not change per-request latency.

Question 21Application Performance Monitoring (APM 360)

Which New Relic APM metric would you use to detect that an application is serving fewer requests than expected during a traffic surge?

  1. Throughput compared against baseline request rates Correct answer.
  2. Memory usage compared against host capacity
  3. Apdex compared against its threshold
  4. Error rate compared against zero

Explanation

Throughput measures requests per minute, so a shortfall against an expected baseline reveals reduced serving capacity. Apdex and error rate describe quality, not volume.

Question 22Application Performance Monitoring (APM 360)

What happens when a transaction name is not explicitly set in the New Relic APM agent?

  1. The agent raises an error and stops instrumentation
  2. The transaction is always grouped under a single default name
  3. The transaction is dropped and never reported
  4. The agent derives a name from the framework routing or the request path Correct answer.

Explanation

Agents automatically name transactions based on framework routing or URL patterns when no custom naming is applied. Explicit naming is used to control grouping when defaults are too granular or too coarse.

Question 23Application Performance Monitoring (APM 360)

Which statement about alerting on APM metrics is correct?

  1. Alerts require browser monitoring to be enabled
  2. Alert policies automatically modify the Apdex T value
  3. Alert conditions can only be based on infrastructure CPU metrics
  4. Alert conditions evaluate a chosen metric such as Apdex or error rate over a defined threshold and window Correct answer.

Explanation

APM alert conditions target metrics like Apdex, error rate, or response time and evaluate them against thresholds over a time window. Policies then group conditions and route notifications.

Question 24Application Performance Monitoring (APM 360)

A team wants to see the slowest individual requests with their full call trees for debugging. Which APM feature should they use?

  1. Log parsing rules
  2. Transaction traces Correct answer.
  3. Synthetic location results
  4. Host process samples

Explanation

Transaction traces capture detailed timing and the call tree for individual slow requests, showing where time is spent. They are the standard tool for deep per-request debugging.

Question 25Application Performance Monitoring (APM 360)

What is the primary benefit of enabling distributed tracing in New Relic APM?

  1. Removal of the need for any APM agent
  2. Automatic generation of synthetic test scripts
  3. Automatic reduction of application response time
  4. End-to-end visibility of a request across services with per-span timing Correct answer.

Explanation

Distributed tracing links spans across service boundaries so a request's full journey and per-service timing are visible. It is an observability enhancement, not a performance fix or replacement for agents.

Question 26Application Performance Monitoring (APM 360)

In New Relic APM, which component is responsible for instrumenting application code and reporting transaction data to New Relic?

  1. The browser monitoring JavaScript snippet
  2. The infrastructure monitoring daemon
  3. The New Relic REST API v2 endpoint
  4. The APM agent installed in the application runtime Correct answer.

Explanation

The APM agent runs inside the application runtime, hooks into framework and library calls, and reports transaction, error, and metric data to New Relic. The REST API and browser snippet consume or collect different data types and do not instrument server-side code.

Question 27Application Performance Monitoring (APM 360)

Which APM metric represents the percentage of transactions that produced errors during a given period?

  1. Error rate Correct answer.
  2. Throughput
  3. Apdex score
  4. Response time

Explanation

Error rate is the share of transactions that produced errors relative to total transactions, which is the APM signal for reliability. Apdex, throughput, and response time describe satisfaction, volume, and latency respectively.

Question 28Application Performance Monitoring (APM 360)

A developer wants the APM agent to record a custom timing for a business operation that is not a web request. Which agent API call is designed for this?

  1. noticeError for the operation
  2. setTransactionName for the operation
  3. startWebTransaction for the operation
  4. recordMetric for a custom metric name Correct answer.

Explanation

Custom metrics are created with the agent's recordMetric API, which lets you capture arbitrary business timings and counters. Web transaction APIs, error notification, and transaction naming serve different purposes and do not record standalone custom timings.

Question 29Application Performance Monitoring (APM 360)

What does the New Relic APM transaction trace capture when a slow transaction exceeds the trace threshold?

  1. A heap dump of the JVM at the moment of the request
  2. Only the total duration of the transaction
  3. The full HTTP request and response bodies
  4. A timed breakdown of segments across the transaction's call tree Correct answer.

Explanation

Transaction traces record segment-level timing so you can see where time was spent across the call tree. They do not capture heap dumps or full payload bodies by default, and they contain far more than a total duration.

Question 30Application Performance Monitoring (APM 360)

Which New Relic APM feature lets you compare metric behavior across two time windows side by side?

  1. The compare view in the APM chart Correct answer.
  2. The deployment timeline
  3. The service level objective editor
  4. The entity map

Explanation

The compare view overlays a second time window on APM charts so you can contrast current behavior with a baseline period. The entity map shows topology, deployment markers annotate changes, and SLOs define targets rather than comparisons.

Question 31Application Performance Monitoring (APM 360)

Which APM data type stores aggregated metric values over time and is the basis for charts and alert conditions?

  1. Log events
  2. Span events
  3. Deployment markers
  4. Metric timeslice data Correct answer.

Explanation

Metric timeslice data aggregates values such as response time and throughput into time buckets that power charts and NRQL alert conditions. Logs, spans, and deployment markers are separate event types with different schemas.

Question 32Application Performance Monitoring (APM 360)

Which agent configuration setting controls how long a transaction must run before the agent generates a transaction trace?

  1. The slow SQL threshold
  2. The error collector threshold
  3. The transaction trace threshold Correct answer.
  4. The Apdex T threshold

Explanation

The transaction trace threshold (often expressed in seconds or as a multiple of the Apdex T) determines when a slow transaction qualifies for tracing. Apdex T defines satisfaction targets, and the error and slow SQL settings govern different capture rules.

Question 33Application Performance Monitoring (APM 360)

Which New Relic APM concept groups transactions that share a common naming pattern so they can be tracked as one unit?

  1. Synthetic monitoring
  2. Metric grouping via transaction naming Correct answer.
  3. Log forwarding
  4. Span sampling

Explanation

Transaction naming determines how requests are grouped into the same metric, and metric grouping rules can normalize names to avoid explosion. Span sampling, log forwarding, and synthetics address different concerns.

Question 34Application Performance Monitoring (APM 360)

An APM chart shows response time rising steadily while throughput is unchanged and error rate is flat. Which APM view should you open first to localize the slowdown?

  1. The deployment history list
  2. The alert policy summary
  3. The transaction breakdown table sorted by total time Correct answer.
  4. The service level objective report

Explanation

The transaction breakdown table ranks transactions by time consumption, quickly revealing which transaction accounts for the increase. Deployment history, alert summaries, and SLO reports provide context but do not localize where time is spent.

Question 35Application Performance Monitoring (APM 360)

A team wants APM alert conditions to fire only when the same metric breaches a threshold for several consecutive minutes. Which configuration accomplishes this?

  1. Set the condition's threshold duration to the required window Correct answer.
  2. Enable distributed tracing on the agent
  3. Set the Apdex T value higher
  4. Add a deployment marker before each release

Explanation

The threshold duration on an alert condition requires the metric to violate the threshold for a sustained period before the condition opens. Apdex T, distributed tracing, and deployment markers do not control alert evaluation windows.

Question 36Application Performance Monitoring (APM 360)

Which agent API lets you attach custom attributes to the current transaction so they appear on trace and error data?

  1. setUserId on the browser agent
  2. ignoreTransaction on the current request
  3. addCustomAttribute on the transaction Correct answer.
  4. recordCustomEvent on the incident

Explanation

Adding custom attributes to a transaction makes them available on transaction traces and error events for filtering and analysis. The other calls set browser user identity, record a custom event, or exclude a transaction from reporting.

Question 37Application Performance Monitoring (APM 360)

A team needs to exclude a noisy health-check endpoint from Apdex calculations while still tracking its availability. Which approach fits best?

  1. Delete the endpoint from the application
  2. Configure the agent to ignore that transaction for Apdex Correct answer.
  3. Disable the agent for the entire application
  4. Raise the Apdex T value for all transactions

Explanation

The agent supports excluding specific transactions from Apdex scoring so they still appear in throughput and error data without skewing satisfaction metrics. Disabling the agent, removing the endpoint, or globally raising T would harm broader observability.

Question 38Application Performance Monitoring (APM 360)

Which NRQL query returns average response time for a specific APM transaction over the last hour?

  1. SELECT average(duration) FROM Span WHERE name = 'MyRoute'
  2. SELECT average(duration) FROM Transaction WHERE name = 'WebTransaction/MyRoute' SINCE 1 hour ago Correct answer.
  3. SELECT count(*) FROM Transaction SINCE 1 hour ago
  4. SELECT sum(duration) FROM Transaction FACET name

Explanation

Transaction event data exposes duration, and averaging it filtered by transaction name yields the average response time for that route. Counting returns volume, querying Span targets trace data, and summing without averaging answers a different question.

Question 39Application Performance Monitoring (APM 360)

A service calls a downstream HTTP API that has become slow. Where in the APM transaction trace would you look to confirm the downstream call is the bottleneck?

  1. The garbage collection metrics panel
  2. The external segment entry for the downstream host Correct answer.
  3. The browser page load waterfall
  4. The application code segment for the entry point

Explanation

External segments in a transaction trace record time spent calling out to other services and are labeled by host. The entry-point segment, JVM GC metrics, and browser waterfall describe other layers of the request path.

Question 40Application Performance Monitoring (APM 360)

Which statement describes how the APM agent handles errors that occur inside an asynchronous background job?

  1. Errors in background jobs are never reported to New Relic
  2. Background job errors are automatically converted into browser errors
  3. The agent reports background errors only when distributed tracing is disabled
  4. Errors are captured only if the agent instruments the job framework or the error is explicitly reported Correct answer.

Explanation

The agent captures errors when its framework instrumentation wraps the job or when code explicitly reports the error through the agent API. Without instrumentation or explicit reporting, the agent has no visibility into the failure.

Question 41Application Performance Monitoring (APM 360)

An APM user wants to see which transaction names contribute the most total time to the application during peak hours. Which analysis is most appropriate?

  1. Average Apdex across the whole day
  2. Facet transaction duration by name and sum total time for the peak window Correct answer.
  3. List all error messages by frequency
  4. Count deployments during the peak window

Explanation

Summing duration grouped by transaction name over the peak window reveals which routes consume the most aggregate time. Whole-day Apdex, deployment counts, and error frequency answer different questions.

Question 42Application Performance Monitoring (APM 360)

What is the effect of setting the Apdex T value to a higher number for a transaction?

  1. Throughput increases because more requests are recorded
  2. More requests count as satisfied, raising the Apdex score Correct answer.
  3. Error rate decreases because failures are reclassified
  4. Fewer requests count as satisfied, lowering the Apdex score

Explanation

Apdex T is the satisfied threshold; raising it means more response times fall within the satisfied band, which increases the score. It does not change throughput or reclassify errors.

Question 43Application Performance Monitoring (APM 360)

Which APM capability should you enable to follow a single request as it moves from a web front end through multiple backend services?

  1. Distributed tracing Correct answer.
  2. Browser monitoring
  3. Infrastructure monitoring
  4. Synthetic monitoring

Explanation

Distributed tracing propagates trace context across services so a single request can be followed end to end. Browser, synthetic, and infrastructure monitoring observe different layers and do not stitch cross-service traces.

Question 44Application Performance Monitoring (APM 360)

A team wants to be notified when the error rate of a specific APM transaction exceeds a percentage for five minutes. Which alert configuration element ties the condition to that transaction?

  1. A dashboard widget showing the transaction
  2. A NRQL condition filtered by the transaction name Correct answer.
  3. A static threshold on host CPU
  4. A synthetic monitor for the route

Explanation

NRQL alert conditions can filter on transaction name so the threshold applies only to that route. Host CPU thresholds, synthetic monitors, and dashboard widgets do not create transaction-scoped alert conditions.

Question 45Application Performance Monitoring (APM 360)

An engineer needs to reduce the volume of transaction traces stored while keeping visibility into the slowest requests. Which agent setting is most relevant?

  1. The log level
  2. The transaction trace threshold Correct answer.
  3. The Apdex T value
  4. The browser sampling rate

Explanation

Raising the transaction trace threshold means only the slowest transactions qualify for tracing, reducing stored trace volume. Apdex T, log level, and browser sampling affect other data streams.

Question 46Application Performance Monitoring (APM 360)

A team observes that a single transaction's 95th percentile response time is far worse than its average. What does this indicate about the transaction?

  1. Throughput is declining over the period
  2. A subset of requests is much slower than the typical request Correct answer.
  3. The error rate is necessarily elevated
  4. The average is being skewed upward by fast requests

Explanation

A high 95th percentile relative to the average signals a slow tail: a minority of requests take much longer than the typical request. It does not imply skew from fast requests, declining volume, or elevated errors.

Question 47Application Performance Monitoring (APM 360)

A service shows low Apdex despite acceptable average response time. Which explanation is most consistent with this pattern?

  1. Errors are being excluded from the Apdex calculation
  2. Throughput has dropped below the expected baseline
  3. The Apdex T value is set too high for the workload
  4. A meaningful share of requests exceeds the tolerating threshold even though most are fast Correct answer.

Explanation

Apdex penalizes requests that exceed the tolerating threshold, so a heavy slow tail can depress the score even when the average looks fine. A higher T would raise the score, low throughput is unrelated, and errors are counted as frustrated.

Question 48Application Performance Monitoring (APM 360)

An APM agent reports high time in a datastore segment, but database monitoring shows healthy query latency. Which explanation deserves investigation first?

  1. Connection pool contention or slow connection acquisition in the application Correct answer.
  2. The browser agent not being installed
  3. The deployment marker being missing
  4. The Apdex T value being too low

Explanation

Datastore segment time includes time waiting for a connection from the pool, so contention can inflate it even when the database answers queries quickly. Apdex T, browser instrumentation, and deployment markers do not explain datastore segment time.

Question 49Application Performance Monitoring (APM 360)

A team wants to correlate a latency regression with a specific code change and confirm which transactions were affected. Which combination of APM capabilities supports this analysis?

  1. Synthetic monitor uptime checks alongside log forwarding
  2. Infrastructure host metrics alongside browser page views
  3. Deployment markers alongside transaction-level response time charts Correct answer.
  4. Apdex T adjustments alongside alert policy summaries

Explanation

Deployment markers annotate the timeline with change events, and transaction-level charts show which routes shifted at that point. Uptime checks, host metrics, and alert summaries do not tie a code change to per-transaction latency shifts.

Question 50Application Performance Monitoring (APM 360)

An application reports a large number of distinct transaction names that fragment its metrics. Which remediation is most appropriate?

  1. Increase the transaction trace threshold
  2. Switch from Apdex to error rate as the primary metric
  3. Apply metric grouping rules to normalize dynamic parts of the transaction name Correct answer.
  4. Disable the agent's transaction naming entirely

Explanation

Dynamic identifiers in transaction names cause metric fragmentation, and grouping rules normalize those names into a manageable set. Disabling naming, changing primary metrics, or raising the trace threshold do not address name explosion.