Source: https://www.youtube.com/watch?v=dfpGU9al_i4

Merger of 2 open source standards:

• Open Tracing
• Open Census

Types of data

Tracing

• Made of spans.
• Spans represent the work being done by a sub-component of a service.
  • Name
  • Start timestamp
  • End timestamp
  • Attributes
    • These are predefined depending on the technology
      • Http:
        • http.method
        • http.status_code
      • Database
        • db.type
        • db.instance
        • db.statement
      • Messaging
        • messaging.system
        • messaging.destination

Metrics

• A measurement of something
• Name
• Description
• Unit
• Kind
  • There are 3 “kinds” of metrics
    • counter: Increments over time
    • metric: aggregation of a value over time
    • observer: set of values at a given time
• Label
• Aggregation
• Timestamp

Logs

• Video is old. Logging is available now.

OTel Components

Collectors

• May not be required
• Can make deployment faster/easier
• Deployment Options
  • Agent
  • Containerized
  • Daemon set
  • Gateway
    • Multiple collectors running as a service
    • Point your collectors to this then export to a common source.

Collector Components

• Receiver
  • Get the data to the collector
  • Push or Pull
  • otlp protocols
    • grpc
      • Specify endpoint:
      • endpoint: 0.0.0.0.:55680
    • http (otlphttp = protobuf)
      • Specify endpoint
      • endpoint: 0.0.0.0:55681
• Processor
  • Transform the data you’ve received
• Exporter
  • Where to send the data
    • Can send to multiple locations

Collector Pipelines

• Really just the YAML file

Extensions

• Trace Extensions
  • Receiver . Exporter
    • OTLP
    • Jaeger
    • Zipkin
  • Processor
    • Attributes
    • batch
    • queued retry
    • resource
    • sampling
    • span
• Metric Extensions
  • Receiver
    • OTLP
    • Host
    • Promethius
  • Exporter
    • OTLP
    • Prometheus
• Community Based
  • Build Your Own extensions
  • Enhance the collector pipeline
    • Perhaps build a dedicated exporter

Client Library (Instrumentation Library)

• Applications require an instrumentation library (manual or automatic)
• Automatic
  • Various languages supported
  • Natively supported language list
• Manual
  • You have to do all of this.