3. Tracing Monitoring: Spring Boot 3 -- OpenTelemetry -- Jaeger -- Zipkin

In this video, we're going to extend a bit the diagram in the previous video with the presence of Jaeger and Zipkin. By this you can get some foundation knowledge about Spring Boot 3 Actuator, Micrometer work together, and have some playground to play with OpenTelemetry Connector, Jaeger and Zipkin
Github: github.com/nlinhvu/hello-service
Spring Boot 3 Observability Series: • Spring Boot 3 Observab...
References:
Spring Boot Actuator: docs.spring.io/spring-boot/do...
Tracing: docs.spring.io/spring-boot/do...

[@berlisrodriguez5029]

you rock man, this video series is amazing!

[@linhvudev]

thank you very much! glad you enjoy it

[@user-oh6ph6oh3k]

I was really happy to hear your voice because it remembers me of a few vietnamese friends and as expected, I've learned a lot so far. Thank you very much :). I'm not sure if it is an issue but since spring boot version 3.2 the OtlpHttpSpanExporter ist not part of the OtlpAutoConfiguration class anymore.

[@linhvudev]

🥰 thank you so much, even though I don't think my voice as fluent as your friends 😅. In Spring Boot 3.2, just a quick check, OtlpHttpSpanExporter is moved out of OtlpAutoConfiguration, but its bean is also provided to OtlpAutoConfiguration through @Import(OtlpTracingConfigurations.Exporters.class) annotation. So if you put a debug point at otlpHttpSpanExporter method in OtlpTracingConfigurations.Exporters class, you will see the it will be called as well.

[@linhvudev]

For Otel Collector >= 0.85.0, now Jaeger supports OTLP directly. The default port for OTLP/gRPC is 4317. By that exporters.jaeger.endpoint=jaeger-all-in-one:14250 is deprecated. Please use exporters.otlp.endpoint=jaeger-all-in-one:4317 instead, update service.pipelines.metrics.exporters from 'jaeger' to 'otlp' (in otel-collector.yml) and expose port 4317 for Jaeger in docker compose file (opentelemetry.io/blog/2023/jaeger-exporter-collector-migration/)

[@vtvvnha2230]

Thank you for your videos

[@linhvudev]

glad you found it useful!

[@ajeethkumarr6744]

It's a Good KT, I have an issue with exporting "Actuator" traces to zipkin and jaeger somehow I don't see any info in zipkin & jaeger UI, I would really great to hear from you

[@user-hr4bn1zl6t]

Your video has answered some of my doubts about OpenTelemetry, thanks! but i still have question: can OTLP trace database operations?

[@linhvudev]

Thanks, glad that it could help. As long as your application has these database operations in terms of traceId and spanId, OTLP can obviously transfer them. But basically from the application perspective (using micrometer tracing) you know what kind of statement you sent to Database and how much did it take.

[@andrejflieger4182]

Nice Video and very good to get an idea about tracing. However how does that integrate with kubernetes / Openshift with the tracing operator in place.

[@linhvudev]

Thank you so much for your nice words! Production grade is totally a different story, there are always trade-offs based on the resources, and decisions of each team, and organization. If you're interested in that, depending on your use case, you can refer to these Deployment Patterns (opentelemetry.io/docs/collector/deployment/) to find the one that is most suitable for you. This link comes along with a KZbin video and a GitHub repository, they're worth taking a look to them as well

[@Useralammar2233]

Hi, thanks for this informative video, but i have question why we need both zipkin and jeager ? since they both show the distrbuted tracing info

[@linhvudev]

thanks for your interest, yes, you don't have to, I also said it in the video, just simply like to give the configuration for both, so if you're interested in either Jaeger or Zipkin, you can refer to one of them and make it run by yourself.