Welcome to otel.guru
Hello, World. I’m Vaishak.
This blog is dedicated to my journey with OpenTelemetry (OTel).
When I started experimenting with OTel a few years ago, it was a much smaller project—promising, but raw. Documentation was scarce, APIs were in flux, and finding answers often meant digging through GitHub issues or reading the source code.
Fast forward to today: OpenTelemetry has exploded. It now boasts over 20,000 contributors and has become the de facto standard for observability in cloud-native environments.
But with growth comes complexity.
The Gap: Lab vs. Production
There are plenty of “Getting Started” guides out there. You can spin up the OTel Demo in Docker, see some pretty traces in Jaeger, and call it a day. That’s the Lab.
But the Production reality is different.
- In the Lab, sampling is 100%. In Production, 100% sampling will bankrupt you.
- In the Lab, you control the code. In Production, you have to convince 50 teams to instrument their services correctly.
- In the Lab, errors are fun to debug. In Production, missing spans mean lost revenue and 3 AM pager alerts.
What to Expect Here
This blog isn’t just about how to install the collector. It’s about:
- Real-World Scale: Lessons learned from deploying OTel in large, messy enterprise environments.
- The “Why”, not just the “How”: Understanding the philosophy behind tracing, metrics, and logs.
- Experimentation: Deep dives into new features, eBPF auto-instrumentation, and edge cases.
- The Human Element: How to build an observability culture and get buy-in from developers and management.
I’m here to share my learnings—both the successes and the painful failures—so you don’t have to repeat them. Let’s explore the “dark corners” of distributed systems together.
If you’re ready to move beyond the demo app, stay tuned.