Attending @HumanXCo? We're hosting an intimate fireside chat with @honeycombio's @cyen before she takes the HumanX stage. Moderated by @conikeec, they're going to be discussing how teams can design observability for an AI-driven world. Space is limited! luma.com/agentic-reliability
We've been building an internal Claude Code plugin system at Intercom with 13 plugins, 100 skills, and hooks that turn Claude into a full-stack engineering platform. Lots done, more to do. Here's a thread of some highlights.
If you're a team of 5 and you don't have formal, tracked SLOs around:
* Request queue time (web server)
* Background job queue time (per queue)
... you should, because you already have an informal one (your customers complaining).
There have been some good posts on what "wide events" instrumentation is but not as many on how to go about it, what attributes you should add, or how to work with OpenTelemetry
I put everything I've learned in the last few years into one guide jeremymorrell.dev/blog/a-pra…
Service level objectives aren’t one-size-fits-all. @cyen, CEO of @honeycombio, recommends teams define their own SLOs to meet their unique needs! 🎯
Tune in for Christine’s take on shifting from tool-centric discussions to a customer-focused approach: circle.ci/4ds0Qyq
These three offerings respectively solve the three biggest Observability challenges:
1. Everything’s logs already; I don’t want to instrument it all again.
2. Scale is hard.
3. My engineers are morons.
Today we are launching the most significant expansion to @honeycombio's observability service, with 💥three💥 new offerings:
🪵Honeycomb for Log Analytics
🚰 Honeycomb Telemetry Pipeline
👷 Honeycomb Professional Services
honeycomb.io/blog/announcing…
How would you migrate your code from metrics to wide structured logs? 💥LIKE THIS💥 (courtesy @JeremyMorrell)
Also see this terrific how-to blog from @jessitron, "How to Get Infinite Custom Metrics for Free": honeycomb.io/blog/get-infini…
I'm not Charity, but I've toyed with the idea of recommending this approach to people for web services: gist.github.com/jmorrell/76a…
Essentially:
- add a new middleware that builds up a log line
- add more attributes during a request
- send that off to Honeycomb
ALT A screenshot of the code available in the gist link
It's been five years and Datadog still can't get close to @honeycombio in terms of experience of slicing and dicing tracing data.
Just added a new field to my traces. I want to group by that field to compare old vs. new. Datadog is not capable of that. At least not automatically
Trust us, the future of DevOps is bright! 😎
Check out @cyen, CEO of @honeycombio, and @z00b as they discuss the evolving role of observability in modern software development: circle.ci/4ds0Qyq
(or, as I sometimes think of it: the hard-won state of my/@mipsytipsy 's relationship, circa 2015, as a dev and an ops coming from opposite ends of the engineering org chart 💛)