OpenTelemetry Quiz
Quiz
Question 1 of 31
(0 answered)
Question 1
What is the primary purpose of OpenTelemetry?
✓
Correct!
OpenTelemetry (OTel) is an open-source observability framework that provides a standardized way to collect, process, and export telemetry data (traces, metrics, and logs) from applications and infrastructure.
✗
Incorrect
OpenTelemetry (OTel) is an open-source observability framework that provides a standardized way to collect, process, and export telemetry data (traces, metrics, and logs) from applications and infrastructure.
Think about what makes OTel vendor-agnostic.
Question 2
Which of the following are the four telemetry signals supported by OpenTelemetry?
✓
Correct!
OpenTelemetry supports four types of telemetry signals: Traces (distributed request tracking), Metrics (numerical measurements), Logs (timestamped event records), and Baggage (key-value pairs propagated across services).
✗
Incorrect
OpenTelemetry supports four types of telemetry signals: Traces (distributed request tracking), Metrics (numerical measurements), Logs (timestamped event records), and Baggage (key-value pairs propagated across services).
Events is not listed as a separate signal type in OTel.
Question 3
In OpenTelemetry, the trace_id changes at each service hop while span_id remains constant throughout the request flow.
✓
Correct!
It’s the opposite: trace_id NEVER changes as the request flows through services (it uniquely identifies the entire trace), while span_id CHANGES at each service hop to identify each operation.
✗
Incorrect
It’s the opposite: trace_id NEVER changes as the request flows through services (it uniquely identifies the entire trace), while span_id CHANGES at each service hop to identify each operation.
Think about what uniquely identifies the entire distributed request.
Question 4
What is the correct order of components in an OpenTelemetry Collector pipeline?
✓
Correct!
The OTel Collector processes data in this order: Receivers (ingest data) → Processors (transform/filter) → Exporters (send to backends).
✗
Incorrect
The OTel Collector processes data in this order: Receivers (ingest data) → Processors (transform/filter) → Exporters (send to backends).
Data must be received before it can be processed or exported.
Question 5
The default OTLP gRPC port for the OpenTelemetry Collector is ____.
✓
Correct!
OTLP endpoints use port 4317 for gRPC and port 4318 for HTTP.
✗
Incorrect
OTLP endpoints use port 4317 for gRPC and port 4318 for HTTP.
It’s a 4-digit number starting with 43.
Question 6
Given this traceparent header, what is the trace_id?
traceparent: 00-0af7651916cd43dd8448eb211c80319c-b7ad6b7169203331-01What will this code output?
✓
Correct!
The traceparent format is: VERSION-TRACE_ID-PARENT_SPAN_ID-TRACE_FLAGS. The trace_id is the 32-character hex string (128 bits) that uniquely identifies the entire trace.
✗
Incorrect
The traceparent format is: VERSION-TRACE_ID-PARENT_SPAN_ID-TRACE_FLAGS. The trace_id is the 32-character hex string (128 bits) that uniquely identifies the entire trace.
The trace_id is the longest component in the header.
Question 7
What are the three types of metric instrument types in OpenTelemetry and when would you use each?
What are the three types of metric instrument types in OpenTelemetry and when would you use each?
Counter: Cumulative value that only increases (e.g., total requests, errors)
Gauge: Point-in-time value that can go up or down (e.g., CPU usage, memory, queue depth)
Histogram: Distribution of values with configurable buckets (e.g., request durations, response sizes)
Did you get it right?
✓
Correct!
✗
Incorrect
Question 8
Arrange these OpenTelemetry Collector processors in the recommended order for a production pipeline:
Drag to arrange in the correct order
⋮⋮
memory_limiter
⋮⋮
batch
⋮⋮
attributes
⋮⋮
filter
✓
Correct!
Best practice order: memory_limiter (prevent OOM first), filter (drop unwanted data early), attributes (enrich/modify), batch (optimize export last).
✗
Incorrect
Best practice order: memory_limiter (prevent OOM first), filter (drop unwanted data early), attributes (enrich/modify), batch (optimize export last).
Question 9
What is the key difference between head-based and tail-based sampling?
✓
Correct!
Head-based sampling makes the decision at trace start (root span), while tail-based sampling waits until the trace is complete to make a decision based on criteria like errors, latency, or attributes.
✗
Incorrect
Head-based sampling makes the decision at trace start (root span), while tail-based sampling waits until the trace is complete to make a decision based on criteria like errors, latency, or attributes.
Think about when the sampling decision is made.
Question 10
Which are valid use cases for the transform processor?
✓
Correct!
The transform processor uses OTTL to modify telemetry data: redacting PII, sanitizing URLs, removing credentials, and normalizing formats. It cannot create collector instances.
✗
Incorrect
The transform processor uses OTTL to modify telemetry data: redacting PII, sanitizing URLs, removing credentials, and normalizing formats. It cannot create collector instances.
The transform processor modifies data, not infrastructure.
Question 11
Zero-code instrumentation requires modifying application source code to add OpenTelemetry support.
✓
Correct!
Zero-code (automatic) instrumentation attaches an agent at runtime without modifying application code. For example, Java uses
-javaagent:opentelemetry-javaagent.jar.✗
Incorrect
Zero-code (automatic) instrumentation attaches an agent at runtime without modifying application code. For example, Java uses
-javaagent:opentelemetry-javaagent.jar.The name ‘zero-code’ is descriptive of its approach.
Question 12
Complete the Python code to create a new span:
Fill in the missing method call
from opentelemetry import trace
tracer = trace.get_tracer(__name__)
with tracer._____("process_order"):
process_payment()
update_inventory()✓
Correct!
The
start_as_current_span method creates a new span and sets it as the current span within the context manager.✗
Incorrect
The
start_as_current_span method creates a new span and sets it as the current span within the context manager.Question 13
What is the purpose of exemplars in OpenTelemetry?
✓
Correct!
Exemplars link specific trace examples to aggregated metrics, enabling direct navigation from a metric alert to the actual trace that caused it. This enables faster root cause analysis.
✗
Incorrect
Exemplars link specific trace examples to aggregated metrics, enabling direct navigation from a metric alert to the actual trace that caused it. This enables faster root cause analysis.
Think about bridging metrics and traces.
Question 14
The W3C Trace Context standard uses the ____ header to transmit trace context between services.
✓
Correct!
The traceparent header is the required W3C Trace Context header containing VERSION-TRACE_ID-PARENT_SPAN_ID-TRACE_FLAGS. The optional tracestate header allows vendor-specific data.
✗
Incorrect
The traceparent header is the required W3C Trace Context header containing VERSION-TRACE_ID-PARENT_SPAN_ID-TRACE_FLAGS. The optional tracestate header allows vendor-specific data.
It combines ’trace’ with the relationship to the caller.
Question 15
Which deployment patterns are valid for OpenTelemetry Collector?
✓
Correct!
The three documented deployment patterns are: Agent (collector alongside app), Gateway (centralized collector cluster), and Hybrid (agents for basic batching, gateway for heavy processing).
✗
Incorrect
The three documented deployment patterns are: Agent (collector alongside app), Gateway (centralized collector cluster), and Hybrid (agents for basic batching, gateway for heavy processing).
Three patterns are explicitly documented.
Question 16
What is the spanmetrics connector and what RED metrics does it generate?
What is the spanmetrics connector and what RED metrics does it generate?
Spanmetrics Connector: Generates metrics from trace spans (replaces deprecated spanmetrics processor).
RED Metrics Generated:
- Rate:
calls_total(total call count) - Errors: Error counts by status code
- Duration:
duration_milliseconds_sum/count
Dimensions include: service, operation, status_code
Did you get it right?
✓
Correct!
✗
Incorrect
Question 17
What happens when
otel.bsp.max.export.batch.size is larger than otel.bsp.max.queue.size?✓
Correct!
If max.export.batch.size is larger than the queue size, the batch span processor won’t be able to form a batch of that size because there aren’t enough spans in the queue.
✗
Incorrect
If max.export.batch.size is larger than the queue size, the batch span processor won’t be able to form a batch of that size because there aren’t enough spans in the queue.
Think about the relationship between queue capacity and batch size.
Question 18
The tracestate header is required for W3C Trace Context propagation.
✓
Correct!
Only the traceparent header is required. The tracestate header is optional and allows vendors to add proprietary information without breaking the standard.
✗
Incorrect
Only the traceparent header is required. The tracestate header is optional and allows vendors to add proprietary information without breaking the standard.
One header is mandatory, the other is optional.
Question 19
Arrange these components in the order data flows through the OpenTelemetry architecture:
Drag to arrange in the correct data flow order
⋮⋮
Observability Backend
⋮⋮
Application (Traces/Metrics/Logs)
⋮⋮
OTLP Protocol
⋮⋮
OTel Collector
✓
Correct!
Data flows: Application generates telemetry → OTLP transmits it → Collector processes it → Backend stores/visualizes it.
✗
Incorrect
Data flows: Application generates telemetry → OTLP transmits it → Collector processes it → Backend stores/visualizes it.
Question 20
Which processor should you use to prevent memory overload in the OTel Collector?
✓
Correct!
The memory_limiter processor prevents memory overload by dropping data when memory usage exceeds configured limits. It should be placed first in the processor chain.
✗
Incorrect
The memory_limiter processor prevents memory overload by dropping data when memory usage exceeds configured limits. It should be placed first in the processor chain.
The processor name directly describes its function.
Question 21
What does trace_flags value ‘01’ indicate in a traceparent header?
traceparent: 00-4bf92f3577b34da6a3ce929d0e0e4736-00f067aa0ba902b7-01What will this code output?
✓
Correct!
Trace flags ‘01’ means the trace is sampled (being recorded). ‘00’ would mean not sampled. This is an 8-bit flag field.
✗
Incorrect
Trace flags ‘01’ means the trace is sampled (being recorded). ‘00’ would mean not sampled. This is an 8-bit flag field.
The flag indicates the sampling decision.
Question 22
What are the benefits of using the OpenTelemetry Collector over direct export from applications?
✓
Correct!
The Collector provides decoupling, centralized processing, reduced app load, multi-backend export, and buffering/retry. It does NOT eliminate the need for instrumentation.
✗
Incorrect
The Collector provides decoupling, centralized processing, reduced app load, multi-backend export, and buffering/retry. It does NOT eliminate the need for instrumentation.
Four of these are actual benefits mentioned in the documentation.
Question 23
Explain the difference between the attributes processor and the transform processor.
Explain the difference between the attributes processor and the transform processor.
Attributes Processor: Simple operations on attributes
- insert, update, delete, hash
- Extract values with patterns
- Good for basic attribute management
Transform Processor: Advanced transformations using OTTL
- Complex regex replacements
- Conditional logic with ‘where’ clauses
- Cross-field operations
- Truncation, case changes
- More powerful but more complex syntax
Did you get it right?
✓
Correct!
✗
Incorrect
Question 24
The Target Allocator in the OpenTelemetry Operator distributes ____ scrape targets across multiple collector instances.
✓
Correct!
The Target Allocator discovers Prometheus scrape targets (ServiceMonitors, PodMonitors) and distributes them evenly across collector instances for load balancing.
✗
Incorrect
The Target Allocator discovers Prometheus scrape targets (ServiceMonitors, PodMonitors) and distributes them evenly across collector instances for load balancing.
It relates to a popular metrics collection system.
Question 25
In a hybrid deployment pattern, agents handle heavy processing while gateways perform basic batching.
✓
Correct!
It’s the opposite: In a hybrid pattern, agents handle basic batching (lightweight) while the gateway performs expensive processing like tail sampling and enrichment.
✗
Incorrect
It’s the opposite: In a hybrid pattern, agents handle basic batching (lightweight) while the gateway performs expensive processing like tail sampling and enrichment.
Think about where you’d want to concentrate compute-intensive tasks.
Question 26
Which scrape job collects container-level resource usage metrics and is embedded in the Kubelet?
✓
Correct!
cAdvisor (Container Advisor) is embedded in the Kubelet and collects container-level resource usage metrics. It’s scraped from
https://<node>:10250/metrics/cadvisor.✗
Incorrect
cAdvisor (Container Advisor) is embedded in the Kubelet and collects container-level resource usage metrics. It’s scraped from
https://<node>:10250/metrics/cadvisor.Its name suggests container-level monitoring.
Question 27
Complete the YAML to configure a filter processor that excludes health check endpoints:
Fill in the attribute name to filter on
processors:
filter:
traces:
span:
- 'attributes["_____"] == "/health"'✓
Correct!
The http.url attribute contains the URL path being accessed. Filtering on this attribute allows excluding health check endpoints from traces.
✗
Incorrect
The http.url attribute contains the URL path being accessed. Filtering on this attribute allows excluding health check endpoints from traces.
Question 28
Which of these are valid resource detectors in OpenTelemetry?
✓
Correct!
The documented resource detectors include env, kubernetes, ec2, gcp, and docker. Azure is not mentioned in the provided documentation.
✗
Incorrect
The documented resource detectors include env, kubernetes, ec2, gcp, and docker. Azure is not mentioned in the provided documentation.
Four detectors are explicitly listed in the documentation.
Question 29
What is context propagation and why is it essential for distributed tracing?
What is context propagation and why is it essential for distributed tracing?
Context Propagation: The mechanism that passes trace metadata (trace_id, span_id) between services via HTTP headers.
Why Essential:
- Without it, each service creates disconnected, independent spans
- Enables linking all spans from a single request into one trace
- Provides end-to-end visibility across distributed systems
- Allows reconstruction of the complete request journey
Transmitted via: W3C traceparent header
Did you get it right?
✓
Correct!
✗
Incorrect
Question 30
What is the default value for
otel.bsp.max.export.batch.size?✓
Correct!
The default max export batch size is 512 spans. Recommended range is 512-2048 depending on span size.
✗
Incorrect
The default max export batch size is 512 spans. Recommended range is 512-2048 depending on span size.
It’s a power of 2 between 256 and 1024.
Question 31
Histograms perform server-side quantile calculation while Summaries perform client-side quantile calculation.
✓
Correct!
Histograms use configurable buckets for server-side quantile calculation and can be aggregated across dimensions. Summaries pre-calculate quantiles (p50, p90, p99) on the client side and cannot be aggregated.
✗
Incorrect
Histograms use configurable buckets for server-side quantile calculation and can be aggregated across dimensions. Summaries pre-calculate quantiles (p50, p90, p99) on the client side and cannot be aggregated.
One can be aggregated, the other cannot.
Quiz Results
Score
0/0
Accuracy
0%
Right
0
Wrong
Skipped
0
Last updated on