Response time and custom metrics. See the slowdown before the outage.

Every uptime check records how long your server took to answer. Add any number from your code, like queue depth or failed payments, with one line.

No signup · No credit card · See the live demo

CPU usage Last 60 s Live
Now 62%
80%
60 s ago Now

The number moves first. The outage follows.

Queue depth creeps up before the worker stalls. Failed payments pile up before anyone writes in. Send those numbers from your code and watch them move.

  • Response time on every check

    Every HTTP check records how long your endpoint took to answer. No answer within 10 seconds counts as down and sends the alert.

  • One line per metric

    Call setMetric or mutateMetric where the number changes. A new name shows up in your dashboard the first time you send it.

  • Counters and gauges

    Set a gauge like active users to an exact value. Mutate a counter like sign-ups by any amount, up or down.

  • Live as it happens

    Values stream into the dashboard while you watch. Each metric keeps its current value on a tile, with no refresh needed.

  • Minute, hour and day charts

    Zoom from per-minute detail for the last hour to daily points. Your plan sets how far back you can look, up to 30 days.

  • Next to your logs

    Metric tiles sit on the same screen as your logs, so a spike and the error behind it are side by side.

Three steps to your first chart.

No agent to run, no dashboard to build. Charts appear with the first value.

  1. Copy your key

    Sign in with GitHub or Google and copy your domain's ingest key from its settings. Every app on the domain sends with that one key and gets its own set of metrics.

  2. Install the SDK

    Install the SDK for your language and create the client with your API key. Or copy the setup prompt and let your AI assistant wire it in.

  3. Send a number

    Call setMetric or mutateMetric where the value changes. The metric gets its own tile and charts within seconds of the first value.

Track metrics in Node.js.

Install @logdash/node, create the client with your API key, then set or mutate any metric in one line. Python, Go, .NET, Java, Rust, Ruby and PHP work the same way.

Browse all SDKs
server.ts 
import { Logdash } from '@logdash/node';

const logdash = new Logdash(process.env.LOGDASH_API_KEY);

// to set absolute value
logdash.setMetric('users', 0);

// to modify existing metric
logdash.mutateMetric('users', 1);

Works with your stack

  • Node.js
  • Python
  • Go
  • logo_NET .NET
  • Java
  • file_type_rust Rust
  • Ruby
  • PHP
  • Curl

Response time and metrics FAQ

Have a different question? Reach out on Discord

Can Logdash monitor API response time?
Yes. Every HTTP check records how long your endpoint took to answer, and no answer within 10 seconds counts as down. For timings inside your code, like a slow query, send the duration as a metric with setMetric.
How do I track custom metrics in my app?
Install the Logdash SDK for your language and create the client with your API key. Then call setMetric or mutateMetric in Node.js, or set and mutate on the metrics client in the other SDKs. The metric appears in your dashboard the first time a value arrives.
What is the difference between set and mutate?
Set replaces the current value, which suits gauges like active users or queue depth. Mutate adds a positive or negative amount to the current value, which suits counters like sign-ups or processed jobs.
Which languages are supported?
There are official SDKs for Node.js, Python, Go, .NET, Java, Rust, Ruby and PHP. Anything else can send a metric with a single HTTP PUT request and your API key.
How long is metric history kept?
24 hours on the free Hobby plan, 7 days on Builder and 30 days on Pro. Recent values are kept per minute, older ones per hour and per day.
How many metrics can I track?
Up to 5 metrics per app on Hobby, 10 on Builder and 30 on Pro. Past the limit, new metric names are not recorded while your existing metrics keep working. Delete one you no longer need to free its slot.
Is there a free plan?
Yes. Hobby is free and includes 5 metrics per app with 24 hours of history, plus logs and uptime monitoring. Builder is $9 a month and Pro is $15 a month.
Can I get an alert when a metric changes?
Not yet. Alerts cover uptime today: when a monitor goes down, or takes longer than 10 seconds to answer, Logdash tells you on Telegram or a webhook. Response times and metrics are for spotting the trend on the dashboard.

Know before they do.

Paste your URL and we start watching it right now.

No signup · No credit card · Claim later with GitHub or Google

Or watch the live demo
Your first 30 days
  1. Day 1
    The alert beats the tweet Checks start 30 seconds after you paste a URL. When it goes down, you hear it first.
  2. Day 7
    The refresh reflex is gone Logs and metrics sit next to your uptime, so you stop checking by hand.
  3. Day 30
    Users hear it from you first A status page on your domain answers before they ask. You ship, Logdash watches.