Hi everyone! I am new to this space and have been going through various distributed tracing solutions since last few days. Have a decent idea on opentelemetry stuff and explored some solutions with lightstep, sentry, datadog as well as using aws distro. Not really able to make the decision here since our current stack already has prometheus for metrics and a hacky trace id propagation for call graphs. We are looking for a solution which gives us deeper view in the system around which microservice, operation is becoming bottlleneck in thousands of calls in a single request. Also a bit confused on signoz, do we still use otel go sdk to instrument ourselves and just use signoz as a analysis backend? or do we have to use signoz as a solution itself? Also can someone help me with self hosted maintenance pros and cons of this solutions vs managed services like lightstep/xray. Thanks!

Hey @User Welcome to the community 👋 To your question regarding SigNoz - you have to instrument your go app using otel sdks. If you are using popular frameworks like gin, gorillamux it should be very easy to instrument your http requests - https://signoz.io/docs/instrumentation/golang/#automatically-create-tracesspans-on-http-requests SigNoz has otel collector built it - so you just have to point your app to SigNoz installation and visualise data in SigNoz dashboard. Based on what we are seeing on managed vs self hosted - we have seen people prefer self hosted to have more control and also not sending data outside your AWS/GCP vpc. Fintech companies are also generally more cautious here. if you are on k8s, etc - scaling SigNoz is fairly easy - but you would still need some devops/sre capabilities in your team to set things up

very helpful answer, Pranay! It would be great, if we could have a FAQ hosting such clear answers, as many developers or sysadmins aren’t used with APM implementations. It saved me a lot of effort on going out and looking things up. All the best!

Sure @User thats a great suggestion

Hi Pranay, Thanks for the detailed info! We use GRPC internally and I am a bit confused on automatic instrumentation and manual instrumentation for grpc communications. Do we have any examples or sample instrumentations which I can refer to?

@User Which framework do you use? @User Can you provide some sample example on auto+manual instrumentation for Hardik

We use golang with grpc and protobuf. We have a microservice architecture internally.

Hey @User We have a blog on instrumenting golang app (http/grpc) https://signoz.io/docs/instrumentation/golang/ Let me know if this helps.

Hey Yes, followed this and added interceptors and instrumented some calls. Working crudely so needs a bit more work. Thanks!

nice. @User I think we have a demo call scheduled tomorrow. So we can chat more then.

Yes

Precisely, Hardik asked the same things I was planning to ask today hahaha. Thanks for that.