so, how do you handle tracing in microservices.

Hey everyone, I've been wrestling with getting a solid distributed tracing setup in place for our growing microservices architecture. We're seeing more latency issues and it's getting hard to pinpoint the root cause when requests hop between several services. We've looked at Jaeger and Zipkin, but I'm curious about what real-world experiences people have had. Are you using OpenTelemetry? How do you handle sampling strategies to avoid overwhelming your tracing backend? Any gotchas or best practices you've learned the hard way? Let's discuss!