Gauge

In this lesson we wrap up with Prometheus’ final core metric type: the Gauge. A Gauge records a single numerical value that can go up and down over time, which makes it ideal for tracking quantities that fluctuate — for example, the number of active requests currently being processed by an application. Key use cases:

Current number of in-progress requests
Free memory, queue length, or number of running jobs
Resource values that both increase and decrease

See Prometheus docs: Gauge metric type. Example: tracking in-progress requests with the Python prometheus_client

Important: the order of labelnames must match the order of values passed to .labels(...). In this example we define ['method', 'path'] and call .labels(request.method, request.path).

from prometheus_client import Gauge
import time
from flask import request

IN_PROGRESS = Gauge(
    'inprogress_requests',
    'Total number of requests in progress',
    labelnames=['method', 'path']
)

def before_request():
    # increment when a request starts
    IN_PROGRESS.labels(request.method, request.path).inc()
    request.start_time = time.time()

def after_request(response):
    # decrement when a request finishes
    IN_PROGRESS.labels(request.method, request.path).dec()
    return response

The Gauge API highlights:

inc() — increment the gauge by 1 (or a specified amount)
dec() — decrement the gauge by 1 (or a specified amount)
set(value) — explicitly set the gauge to value

Ensure every inc() is paired with a corresponding dec() (even when errors occur). In frameworks like Flask, prefer teardown_request, middleware, or a try/finally block around request handling so the gauge is always decremented when a request finishes.

Robust pattern using try/finally inside a handler:

def handle_request():
    IN_PROGRESS.labels(request.method, request.path).inc()
    try:
        # process the request...
        pass
    finally:
        IN_PROGRESS.labels(request.method, request.path).dec()

When exposed by your metrics endpoint, the gauge shows the current value per label combination. Example output:

inprogress_requests{method="GET",path="/cars"} 3.0
inprogress_requests{method="POST",path="/cars"} 0.0
inprogress_requests{method="POST",path="/boats"} 18.0
inprogress_requests{method="GET",path="/boats"} 7.0

Gauge vs Counter — quick comparison

Metric Type	Use Case	Behavior	Example
Gauge	Current value that can go up and down	Supports `inc()`, `dec()`, and `set()`	Number of active requests
Counter	Cumulative values that only increase	Only `inc()` (and `reset` on process restart)	Total requests served

Additional guidance

Use a Gauge when you need an up-and-down metric (active jobs, queue length, free resources).
For request duration or size distributions, prefer Histogram or Summary.
Label cardinality: avoid high-cardinality label values (e.g., raw user IDs or long paths) to prevent memory blowups in the Prometheus server.

Links and references

Never allow unpaired increments. Missing decrements will inflate your gauge and produce misleading monitoring alerts. Implement error paths and teardown hooks to guarantee balanced inc()/dec() calls.

Watch Video

HistogramSummary

Best Practice

Introduction

Observability Fundamentals

Prometheus Fundamentals

PromQL

Dashboarding & Visualization

Application Instrumentation

Service Discovery

Push Gateway

Alerting

Monitoring Kubernetes

Conclusion

Scaling & Long Term Storage

Watch Video