Jaeger

Overview

Jaeger is an open source implementation of Zipkin that can be used to collect and visualize traces.

Big Bang Touch Points

graph TB
  subgraph "jaeger"
  jaegerpods("Jaeger-AllInOne")
  elasticcredentials --> jaegerpods("Jaeger-AllInOne")
  end      

  subgraph "ingress"
    ingressgateway --> jaegerpods("Jaeger-AllInOne")
  end

  subgraph "logging"
    subgraph "elasticsearch"

    credentials --> elasticcredentials
    jaegerpods("Jaeger-AllInOne") --> logging-ek-es-http
    logging-ek-es-http --> LoggingElastic(Elasticsearch Storage )
    end
  end

  subgraph "workloads"
    sidecar --> jaegerpods("Jaeger-AllInOne")
  end

Storage

When Jaeger receives traces, it needs a location to store them. The default configuration in the Helm Chart is to use in memory storage. This, of course, doesn’t provide High Availability. To provide storage, the chart uses the deployed Elasticsearch instance deployed in the logging namespace.

Istio Configuration

Istio is configured with knowledge of the jaeger ingest service so istio sidecars attached to workloads can send trace data. This is done via the meshconfig:

  meshConfig:
    accessLogFile: /dev/stdout
    defaultConfig:
      tracing:
        sampling: 100
      zipkinAddress: jaeger-jaeger-operator-jaeger-collector.istio-system.svc:9411
    enableTracing: false

High Availability

Jaeger is deployed with HorizonalPodAutoscalers for the collector and the querying pods. Use the below yaml to update the maxReplicas on the HPA:

jaeger:
  values:
    jaeger:
      spec:
        query:
          maxReplicas: 5
        collector:
          maxReplicas: 5

Single Sign on (SSO)

Jaeger does not have built in SSO. In order to provide SSO, this deployment leverages Authservice.

flowchart LR

A --> K[(Keycloak)]

subgraph external
K
end

subgraph auth["authservice namespace"]
    A(authservice) --> K
end



ingress --> IP


subgraph "jaeger namespace"
    subgraph "jaeger pod"
        J["jaeger"]
        IP["istio proxy"] --> A
        IP --> J
    end
end    

Licencing

Jaeger has no licensing options nor requirements.

For production workloads, Jaeger uses Elasticsearch to store and query for traces.

Dependencies

Jaeger can be run without dependencies, but to ensure resiliency of data, it uses Elasticsearch for its span and trace database.

---

Last update: 2022-09-14 by Micah Nagel