KCNA ยท Observability
Logging, tracing and OpenTelemetry
Node-level logging agents (Fluentd, Fluent Bit), distributed tracing, Jaeger, OpenTelemetry signals and collector.
Key points
- Containers log to stdout and stderr; the runtime stores logs on the node, the kubelet serves them to kubectl logs.
- These files vanish with the node or pod, so a node-level agent (Fluentd, Fluent Bit) ships them to a central backend.
- Jaeger: CNCF graduated distributed tracing system.
- OpenTelemetry: vendor-neutral standard and toolset (APIs, SDKs, Collector) to generate and export metrics, logs and traces.
Source: https://kubernetes.io/docs/concepts/cluster-administration/logging/
Practice questions
What is the main purpose of the OpenTelemetry Collector?
- Receive, process and export telemetry data to backends
- Schedule pods onto nodes
- Store all logs permanently on the control plane
- Run containers on each node
Show answer
Answer: A. Receive, process and export telemetry data to backends
The Collector receives, processes and exports telemetry in a vendor-neutral way. It is not a storage database, a scheduler or a container runtime.
Which of these are signals that OpenTelemetry defines?
- Traces
- Container images
- Metrics
- Logs
- Network policies
Show answer
Answer: A. Traces, C. Metrics, D. Logs
OpenTelemetry covers traces, metrics and logs. Container images and network policies are unrelated to telemetry signals.
True or false: OpenTelemetry is a backend that stores and visualizes your traces.
Show answer
Answer: False
False. OpenTelemetry generates, collects and exports telemetry; you send it to a backend such as Jaeger or Prometheus for storage and analysis.
Practice 8 questions on this topic in Bitzu
Short lessons, instant explanations and spaced review of what you get wrong. Free on iOS and Android.
