Observability Glossary

Browse 46 observability terms defined in plain English, from the cultural dictionary of computing.

46 Observability Terms

AI Debugging
The process of investigating and fixing problems in AI systems, such as bad outputs, retrieval errors, prompt issues, latency spikes, or evaluation failures....
AI Logging
The recording of AI-related events such as prompts, responses, model versions, latency, token use, tool calls, and safety decisions for debugging or...
AI Trace
A trace or record showing the sequence of steps an AI request took through prompts, retrieval, tool calls, model invocations, and post-processing. AI traces...
Blinking Light
A noisy alert, dashboard signal, or flashy metric that attracts attention without necessarily indicating the most important problem. In ops slang, blinking...
BPF Security
Security monitoring, enforcement, or isolation techniques built on Berkeley Packet Filter or extended BPF capabilities in the kernel. BPF security is powerful...
Cloud Logging
The collection, storage, querying, and routing of logs through cloud-native logging services and pipelines. Cloud logging often combines infrastructure,...
Cloud Monitoring
The observation of cloud resources, services, and workloads through metrics, logs, traces, and health signals provided by cloud-native or integrated monitoring...
Code Phantom
A problem or behavior that appears briefly and mysteriously, then disappears before it can be fully captured. In engineering slang, code phantoms are the...
Code Shadow
A hidden dependency, side effect, or secondary behavior that follows visible code without being obvious. In engineering slang, code shadows are why systems can...
Continuous Profiling
The ongoing collection of profiling data from running applications in production or near-production environments to understand where CPU time, memory, or other...
Correlation ID
A unique identifier generated at the entry point of a request and propagated through all downstream services, appearing in every log line and trace span....
Custom Logger
A logging implementation or wrapper tailored to a team's formatting, routing, metadata, or transport requirements. Custom loggers often add context such as...
Data Lineage
The tracking of data's journey from source through transformations to final consumption — which systems produced it, what transformations were applied, and who...
distributed tracing
A method of tracking requests as they propagate through distributed systems by assigning each request a unique trace ID that follows it across service...
Elastic Stack
The ecosystem around Elasticsearch, Logstash, Kibana, and related tools for search, log analysis, and observability.
Fluentd
An open-source data collector for unifying logs and routing them to storage, search, or analytics backends.
Flying Blind
Operating without enough visibility, instrumentation, or reliable feedback to know what is really happening. In engineering slang, flying blind is one of the...
Grafana
An open-source platform for dashboards, visualization, and observability data, widely used for monitoring systems and business metrics. In operations culture,...
Grafana Community
The ecosystem of users, maintainers, plugin authors, and operators working with Grafana and related tooling. In observability culture, the Grafana community is...
Jaeger
An open source distributed tracing system used to collect, store, and visualize traces across microservices. Engineers use Jaeger to follow a request through...
LLM Observability
The practice of instrumenting and analyzing large language model systems so teams can understand their behavior, performance, cost, and failure modes. LLM...
Log Aggregation
The practice of collecting, centralizing, and indexing log output from many distributed services and hosts into a single searchable system. Tools like the ELK...
Log Everything
A blunt operational instinct to emit as much diagnostic information as possible so incidents become easier to reconstruct. It reflects a real need for...
Logger
An object or module responsible for recording application events to an output destination — console, file, or remote service — with metadata such as timestamp,...
Logging
Logging is the practice of recording timestamped events, errors, warnings, and diagnostic messages from a running application to structured or unstructured...
Logging Framework
A library that provides a complete logging infrastructure — including logger creation, severity levels, output formatting (text, JSON), handler/appender...
Log Level
A severity classification assigned to each log message — typically DEBUG, INFO, WARN, ERROR, and FATAL (in ascending severity) — allowing operators to filter...
Log Parser
A tool or library that extracts structured data from raw log lines — parsing timestamps, severity levels, request IDs, and other fields from unstructured text...
Metrics Collection
The practice of instrumenting applications and infrastructure to emit quantitative measurements — such as request latency, error rates, CPU usage, and queue...
Monitoring
Monitoring is the practice of continuously collecting, aggregating, and analyzing metrics, logs, and health signals from applications and infrastructure to...
More Logs
A common reflexive answer to uncertainty in debugging, often sensible and sometimes overused. In engineering slang, 'more logs' is both a practical move and a...
Observability Security
The protection of logs, traces, metrics, dashboards, and observability pipelines so they do not leak secrets, accept tampered data, or become a high-value...
OpenTelemetry
An open-source observability framework that provides a single set of APIs, SDKs, and tools for generating and collecting traces, metrics, and logs....
Performance Monitoring
The ongoing collection, aggregation, and alerting on system and application performance metrics — such as response latency, throughput, error rates, and...
Prometheus
An open-source systems monitoring and alerting toolkit, originally built at SoundCloud. Uses a pull-based model to scrape metrics endpoints, stores time-series...
Prometheus Alertmanager
The alert-routing component of the Prometheus ecosystem, used to group, deduplicate, and send alerts to notification channels.
Prometheus Community
The ecosystem of users, maintainers, and contributors around Prometheus and related monitoring tooling. In infrastructure culture, the Prometheus community is...
Pulse
A periodic signal or lightweight health check emitted by a service or process to indicate it is alive and functioning. Unlike a full health check, a pulse (or...
Sentry
An error tracking and application monitoring platform used to capture exceptions, crashes, and performance issues.
Seq
A centralized structured logging server that ingests, indexes, and lets you query application logs with a rich filter syntax, popular in the .NET ecosystem.
Silent Error
A problem that happens without obvious alerts, logs, or visible breakage until later damage surfaces. In engineering slang, silent errors are dangerous because...
Stack Trace
A sequential list of the active function calls (stack frames) at the point an exception was thrown or a breakpoint was hit, showing the chain of callers from...
Structured Logging
Logging in a machine-parseable format (JSON, logfmt) with consistent key-value fields instead of free-text strings. Enables filtering, aggregation, and...
Trace
A record of the execution path through a system, capturing the sequence of function calls, service hops, or operations. In distributed systems, a trace follows...
Trace
A record of a request's journey through a distributed system, composed of spans (timed operations) connected by parent-child relationships. Each span captures...
Tracing
Tracing is an observability technique that records the end-to-end path of a request as it traverses services in a distributed system. A trace is composed of...

Related Topics