OpenTelemetry Interview Questions Hindi mein

OpenTelemetry कोर्स के 25 सवाल, हर सवाल के सही जवाब और explanation के साथ। जवाब देखने से पहले ख़ुद सोचें।

OpenTelemetry Foundations

1. What does OpenTelemetry NOT provide?

  1. The OTLP protocol
  2. APIs and SDKs for instrumentation
  3. A storage and visualisation backend
  4. Semantic conventions
जवाब देखें

जवाब: A storage and visualisation backend

You bring a backend such as Jaeger, Tempo or a vendor.

पाठ: Why OpenTelemetry Exists

2. Which signal shows how long each step of a single request took across services?

  1. Traces
  2. Metrics
  3. Resource attributes
  4. Configuration files
जवाब देखें

जवाब: Traces

Spans record timings per step.

पाठ: Signals: Traces, Metrics and Logs

Distributed Tracing Concepts

4. Which span kind describes a service making an outgoing HTTP call?

  1. SERVER
  2. CLIENT
  3. INTERNAL
  4. CONSUMER
जवाब देखें

जवाब: CLIENT

The receiving side records a SERVER span.

पाठ: Traces and Spans

5. How is an exception usually recorded on a span?

  1. As a new trace
  2. As a span event with exception details, plus an Error status
  3. As a metric only
  4. By deleting the span
जवाब देखें

जवाब: As a span event with exception details, plus an Error status

record_exception adds an event.

पाठ: Attributes, Events and Status

6. Which HTTP header carries W3C trace context?

  1. content-type
  2. x-request-id
  3. authorization
  4. traceparent
जवाब देखें

जवाब: traceparent

tracestate and baggage are optional companions.

पाठ: Context Propagation

Instrumenting Applications

7. Which environment variable sets the service name for OpenTelemetry SDKs?

  1. OTEL_SERVICE_NAME
  2. SERVICE
  3. OTEL_NAME_OF_APP
  4. APP_TRACE_ID
जवाब देखें

जवाब: OTEL_SERVICE_NAME

A standard SDK environment variable.

पाठ: Zero-Code Instrumentation

8. What does start_as_current_span do?

  1. Sends a metric
  2. Creates a new trace every time
  3. Starts a span, makes it the current (parent) context, and ends it when the block exits
  4. Deletes the parent span
जवाब देखें

जवाब: Starts a span, makes it the current (parent) context, and ends it when the block exits

Children nest automatically.

पाठ: Manual Spans and Attributes

9. Which attribute belongs on the resource rather than individual spans?

  1. app.cart.items
  2. http.response.status_code
  3. service.name
  4. exception.message
जवाब देखें

जवाब: service.name

Resources describe the source.

पाठ: Resources and Semantic Conventions

Metrics and Logs With OpenTelemetry

10. Which instrument fits checkout latency?

  1. Histogram
  2. Counter
  3. UpDownCounter
  4. Resource
जवाब देखें

जवाब: Histogram

Distributions need histograms.

पाठ: Metric Instruments

11. What does correlation add to a log record?

  1. A random UUID unrelated to traces
  2. A screenshot
  3. The full source code
  4. The trace ID and span ID of the active span
जवाब देखें

जवाब: The trace ID and span ID of the active span

Jump from logs to traces.

पाठ: Logs and Trace Correlation

12. What is an exemplar?

  1. A metric type
  2. A sample trace ID attached to a metric data point
  3. A log level
  4. A sampling policy
जवाब देखें

जवाब: A sample trace ID attached to a metric data point

Links metrics to traces.

पाठ: Linking Metrics to Traces

The OpenTelemetry Collector

13. Which Collector component type transforms or filters data?

  1. Receiver
  2. Processor
  3. Exporter
  4. Extension
जवाब देखें

जवाब: Processor

Receivers in, processors transform, exporters out.

पाठ: Receivers, Processors, Exporters

14. What happens to a processor defined in the configuration but not listed in any pipeline?

  1. It becomes a receiver
  2. It applies to all pipelines
  3. The Collector deletes it
  4. It is not used
जवाब देखें

जवाब: It is not used

Pipelines enable components.

पाठ: Collector Configuration

15. Where does an "agent" Collector usually run?

  1. Close to the application, as a sidecar or one per node
  2. Only inside the backend database
  3. In the browser
  4. On the developer laptop only
जवाब देखें

जवाब: Close to the application, as a sidecar or one per node

Gateways are central.

पाठ: Deployment Patterns

Sampling and Data Processing

16. What is a weakness of head sampling?

  1. It requires a database
  2. It is expensive
  3. It may drop rare errors or slow traces because it decides before the outcome is known
  4. It cannot be configured
जवाब देखें

जवाब: It may drop rare errors or slow traces because it decides before the outcome is known

Decisions are made up front.

पाठ: Head Sampling

17. Why must tail sampling run where all spans of a trace arrive together?

  1. Spans are encrypted otherwise
  2. It only works on metrics
  3. The decision needs the complete trace
  4. Because exporters require it
जवाब देखें

जवाब: The decision needs the complete trace

Route by trace ID to one instance.

पाठ: Tail Sampling

18. Which processor adds Kubernetes pod and namespace names to telemetry?

  1. k8sattributes
  2. batch
  3. debug
  4. memory_limiter
जवाब देखें

जवाब: k8sattributes

Enrichment from the Kubernetes API.

पाठ: Enrichment, Filtering and Redaction

Backends and Debugging With Traces

19. Why is switching observability vendors easier with OpenTelemetry?

  1. Applications must be rewritten anyway
  2. Vendors share one database
  3. Traces are not stored
  4. Applications emit OTLP, so only Collector exporter configuration changes
जवाब देखें

जवाब: Applications emit OTLP, so only Collector exporter configuration changes

Decoupled instrumentation.

पाठ: Choosing Backends

20. What does a CLIENT span that is much longer than its matching SERVER span suggest?

  1. The server is slow
  2. Time is lost between services, such as network, queueing or connection waits
  3. The trace is complete and healthy
  4. Sampling is disabled
जवाब देखें

जवाब: Time is lost between services, such as network, queueing or connection waits

Compare both sides of a call.

पाठ: Debugging Latency and Errors With Traces

21. What can the OpenTelemetry Operator inject into pods?

  1. Database schemas
  2. Zero-code instrumentation, enabled by annotations
  3. TLS certificates for users
  4. Grafana dashboards
जवाब देखें

जवाब: Zero-code instrumentation, enabled by annotations

Via the Instrumentation resource.

पाठ: OpenTelemetry on Kubernetes

OpenTelemetry in Production

22. Which is a good way to cut trace volume without losing error visibility?

  1. Sampling 0% of everything
  2. Disabling tracing for failing services
  3. Removing trace IDs from logs
  4. Tail sampling that keeps all error traces and a fraction of the rest
जवाब देखें

जवाब: Tail sampling that keeps all error traces and a fraction of the rest

Keep the valuable traces.

पाठ: Overhead, Volume and Cost

23. What should you record instead of a full URL with query parameters?

  1. The route template, such as /orders/{id}
  2. The user's password
  3. The entire request body
  4. Cookies
जवाब देखें

जवाब: The route template, such as /orders/{id}

Avoid sensitive values and high cardinality.

पाठ: Security and Privacy

24. What is a sensible first step when adopting OpenTelemetry?

  1. Choose sampling after a year
  2. Instrument every service manually at once
  3. Instrument one critical user journey end to end and verify propagation
  4. Skip the Collector and resource attributes
जवाब देखें

जवाब: Instrument one critical user journey end to end and verify propagation

Prove value on a slice first.

पाठ: Rolling Out OpenTelemetry

25. Which belongs on an OpenTelemetry checklist?

  1. Disable context propagation for speed
  2. Put user emails in span names
  3. Every service sets service.name and service.version
  4. Export without any batching
जवाब देखें

जवाब: Every service sets service.name and service.version

Consistent identity enables everything else.

पाठ: An OpenTelemetry Checklist

← कोर्स पर वापस जाएँ