Exploring Expression Browser

This guide shows how to open and use the Prometheus expression browser (the built-in web UI) to run PromQL queries, inspect time series, evaluate expressions at specific times, and create simple visualizations. It’s ideal for ad-hoc troubleshooting and interactive exploration of your Prometheus metrics.

Accessing the expression browser

Open your web browser and navigate to your Prometheus server. For a local instance the default is: http://localhost:9090 The expression browser UI includes:

An “Expression” input for PromQL queries.
Options to view raw query results or switch to the Graph tab for time-series visualization.
An “Evaluation time” field to evaluate queries at specific timestamps.

Quick example: the `up` metric

The up metric is the simplest diagnostic query — 1 indicates a reachable target, 0 indicates unreachable.

Type up into the Expression field.
Click Execute.

Example output:

up{instance="192.168.1.168:9100", job="node"} 1
up{instance="192.168.21.43:80", job="ec2"} 0
up{instance="192.168.40.248:80", job="ec2"} 0
up{instance="localhost:9090", job="prometheus"} 1

Each line shows the metric name, its label set, and the current value (1 = up, 0 = down).

Inspecting multi-label metrics

Many metrics include multiple labels. For example, node_cpu_seconds_total returns per-CPU and per-mode counters:

node_cpu_seconds_total{cpu="0", instance="192.168.1.168:9100", job="node", mode="idle"} 1434.34
node_cpu_seconds_total{cpu="0", instance="192.168.1.168:9100", job="node", mode="iowait"} 7.67
node_cpu_seconds_total{cpu="0", instance="192.168.1.168:9100", job="node", mode="irq"} 0
node_cpu_seconds_total{cpu="0", instance="192.168.1.168:9100", job="node", mode="nice"} 24.7
node_cpu_seconds_total{cpu="0", instance="192.168.1.168:9100", job="node", mode="softirq"} 0.51
node_cpu_seconds_total{cpu="0", instance="192.168.1.168:9100", job="node", mode="steal"} 0
node_cpu_seconds_total{cpu="0", instance="192.168.1.168:9100", job="node", mode="system"} 21.6
node_cpu_seconds_total{cpu="0", instance="192.168.1.168:9100", job="node", mode="user"} 60.38

node_cpu_seconds_total is a cumulative (monotonic) counter. Plotting it directly shows total accumulated CPU-seconds. To visualize CPU usage rate over time, use rate(node_cpu_seconds_total[5m]), irate(...), or another rate function appropriate to your resolution.

Evaluating queries at a specific time

Use the “Evaluation time” field to run a query as of a historical timestamp. This returns the values that were valid at that time — helpful for debugging incidents or validating historical alert firings.

Graphing queries

Switch to the Graph tab to visualize query results: Example query to graph a particular series:

node_cpu_seconds_total{cpu="0", instance="192.168.1.168:9100", job="node", mode="idle"}

Click Execute (or the graph button) to render a time-series chart.
Adjust the time range (e.g., last 1 hour, last 30 minutes), change chart type, or zoom into a sub-interval.
If you omit label filters, one series is drawn for each matching time series; this can make the chart crowded when many series exist.

Graphing tip: use aggregation and rate functions to produce clearer graphs when many series are present, for example:

sum by(instance) (rate(node_cpu_seconds_total[5m]))

Editor features and ergonomics

The expression input supports:

Auto-completion for metric names and label keys.
Syntax highlighting.
Basic inline help.

You can enable/disable those editor features in the UI settings.

Useful server pages

The expression browser exposes several server pages that are useful when troubleshooting or inspecting Prometheus state. Use the menu at the top of the UI to navigate.

Page	Purpose
Configuration	View the current running Prometheus configuration and scrape settings.
Targets	See all scrape targets, discovered labels, last scrape time, and current health (UP/DOWN).
Rules	List recording and alerting rules and their evaluation state.
Alerts	View active alerts, labels, state, and links to alert history.

Example (truncated) running configuration:

global:
  scrape_interval: 15s
  scrape_timeout: 10s
  evaluation_interval: 15s
alerting:
  alertmanagers:
  - follow_redirects: true
    enable_http2: true
    scheme: http
    timeout: 10s
    api_version: v2
    static_configs:
    - targets: []
scrape_configs:
  - job_name: prometheus
    honor_timestamps: true
    scrape_interval: 15s
    scrape_timeout: 10s
    metrics_path: /metrics
    scheme: http
    follow_redirects: true
    enable_http2: true
    static_configs:
    - targets:
      - localhost:9090
  - job_name: node
    honor_timestamps: true
    scrape_interval: 15s
    scrape_timeout: 10s
    metrics_path: /metrics
    scheme: http
    follow_redirects: true
    enable_http2: true
    static_configs:
    - targets:
      - 192.168.1.168:9100
  - job_name: ec2
    honor_timestamps: true
    scrape_interval: 15s
    scrape_timeout: 10s
    metrics_path: /metrics
    scheme: http
    follow_redirects: true
    enable_http2: true

Targets page (example)

The Targets page is especially useful for tracking down unreachable endpoints: it shows each configured scrape target, discovered labels, the last scrape result, and the scrape health. When a target is DOWN, the page provides an error message (for example, “context deadline exceeded”) that helps diagnose network or service-side issues.

A screenshot of the Prometheus "Targets" page listing monitored endpoints. Two EC2 endpoints are shown as DOWN with "context deadline exceeded" errors, while the node and prometheus targets are UP.

Rules and Alerts

Rules: recording rules create new time series to simplify repeated PromQL expressions; alerting rules generate alerts when their conditions are met.
Alerts: lists currently active alerts, their labels, state (firing/pending), and links to alert history/details.

Quick reference — common PromQL patterns

Goal	Example
Check target health	`up`
CPU usage rate per instance	`sum by(instance) (rate(node_cpu_seconds_total[5m]))`
Single series graph	`node_cpu_seconds_total{cpu="0", instance="192.168.1.168:9100", job="node", mode="idle"}`
Aggregate across jobs	`sum by(job) (rate(node_cpu_seconds_total[5m]))`

Conclusion

The Prometheus expression browser is a lightweight, built-in tool for interactive querying, debugging, and quick visualizations. For production dashboards, team-facing views, and richer visualizations, integrate Prometheus with a dedicated dashboarding tool such as Grafana.

The expression browser is perfect for quick exploration and troubleshooting. For long-term dashboards and collaborative views, connect Prometheus to Grafana or another visualization tool.

Introduction

Observability Fundamentals

Prometheus Fundamentals

PromQL

Dashboarding & Visualization

Application Instrumentation

Service Discovery

Push Gateway

Alerting

Monitoring Kubernetes

Conclusion

Scaling & Long Term Storage

Exploring Expression Browser

Accessing the expression browser

Quick example: the `up` metric

Inspecting multi-label metrics

Evaluating queries at a specific time

Graphing queries

Editor features and ergonomics

Useful server pages

Targets page (example)

Rules and Alerts

Quick reference — common PromQL patterns

Conclusion

Links and references

Watch Video

​Accessing the expression browser

​Quick example: the up metric

​Inspecting multi-label metrics

​Evaluating queries at a specific time

​Graphing queries

​Editor features and ergonomics

​Useful server pages

​Targets page (example)

​Rules and Alerts

​Quick reference — common PromQL patterns

​Conclusion

​Links and references

Watch Video

Accessing the expression browser

Quick example: the `up` metric

Inspecting multi-label metrics

Evaluating queries at a specific time

Graphing queries

Editor features and ergonomics

Useful server pages

Targets page (example)

Rules and Alerts

Quick reference — common PromQL patterns

Conclusion

Links and references