jaeger.md

Deploy Jaeger to collect & visualize trace data

If you have an application running in MicroShift VM exporting OTLP trace spans, you can collect and visualize the traces by deploying a jaeger container on the host system via podman.

Run jaeger in container on local system

# Run jaeger image on local system
podman run -d --name jaeger   -p 16686:16686   -p 14250:14250   jaegertracing/all-in-one:latest

Forward port 14250 on local system to MicroShift VM

ssh -R 14250:localhost:14250 redhat@<MicroShift VM IP> -N

If you don't have an application instrumented to export traces, you can configure CRI-O to export traces!

Enable CRI-O tracing by adding a crio.conf.d file

sudo su
mkdir /etc/crio/crio.conf.d
cat <<EOF > /etc/crio/crio.conf.d/otel.conf
[crio.tracing]
tracing_sampling_rate_per_million=999999
enable_tracing=true
EOF

Restart cri-o service

systemctl daemon-reload
systemctl restart crio # or systemctl enable crio --now
systemctl status crio # should be running

You might also deploy the sample application. The sample application is instrumented to generate OTLP trace data. CRI-O and the sample-app are now exporting trace data to Jaeger running on the local machine. You should now see spans in the jaeger dashboard at localhost:16686, provided you have deployed and configured OpenTelemetry Collector to receive OTLP data from the application service (in this example I have also configured CRI-O to export trace data, and OpenTelemetry Collector is running with hostNetwork:true) and provided the OpenTelemetry Collector is configured to export to Jaeger.