<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Otel on Neil Hanlon</title><link>https://thepotato.tech/tags/otel/</link><description>Recent content in Otel on Neil Hanlon</description><generator>Hugo</generator><language>en-US</language><lastBuildDate>Wed, 30 Sep 2026 12:56:00 -0400</lastBuildDate><atom:link href="https://thepotato.tech/tags/otel/index.xml" rel="self" type="application/rss+xml"/><item><title>Wiring swamp's Traces Into My Own LGTM Stack</title><link>https://thepotato.tech/posts/configuring-swamp-otel/</link><pubDate>Wed, 30 Sep 2026 12:56:00 -0400</pubDate><guid>https://thepotato.tech/posts/configuring-swamp-otel/</guid><description>&lt;p>This one is mostly a how-to. If you are running swamp and already have somewhere to put OpenTelemetry traces, this is how I have mine wired into the LGTM stack at home. I will try to keep the editorializing under control, but I make no promises.&lt;/p>
&lt;p>Two things are worth establishing up front. swamp emits OTel traces and structured logs natively, both straight out of Deno&amp;rsquo;s telemetry runtime; this post is about the traces, which go out to an ingest gateway and into Tempo, and that is where the dashboard payoff is. The logs ride a separate OTLP signal that I point at a local Alloy, which forwards them to Loki, and I will mention that once more and then leave it alone. The second thing is that these are traces of swamp execution. They are not LLM traces, agent traces, token accounting, prompt telemetry, or any of the other things now living under the increasingly enormous &amp;ldquo;AI observability&amp;rdquo; umbrella, and that distinction matters more than it might sound.&lt;/p></description></item></channel></rss>