https://company-engineering.pages.dev/tags/observability/Recent content in Observability on Company EngineeringHugoenWed, 10 Jun 2026 00:00:00 +0000https://company-engineering.pages.dev/posts/observability-patterns-for-distributed-systems/Wed, 10 Jun 2026 00:00:00 +0000https://company-engineering.pages.dev/posts/observability-patterns-for-distributed-systems/<p>When a request crosses a dozen services, &ldquo;check the logs&rdquo; stops being useful advice. Observability is what lets us ask new questions about a running system without shipping new code to answer them. These are the patterns we rely on.</p> <h2 id="the-three-signals-and-what-each-is-for">The three signals, and what each is for</h2> <p>We treat metrics, logs, and traces as complementary rather than interchangeable:</p> <ul> <li><strong>Metrics</strong> answer <em>how much</em> and <em>how often</em> — and they&rsquo;re cheap enough to keep for everything. They&rsquo;re how we notice a problem.</li> <li><strong>Traces</strong> answer <em>where</em> — they show the path of a single request across services and where the time went. They&rsquo;re how we localize a problem.</li> <li><strong>Logs</strong> answer <em>why</em> — the detailed, contextual record of what a specific component did. They&rsquo;re how we explain a problem.</li> </ul> <p>You reach for them in that order: a metric alerts, a trace narrows it to a service, logs explain the failure.</p>