Logs for $99. Once.

Self-hosted logging. Alerts, errors, and AI answers, all read from your logs. No per-GB meter.

4.8k+ Docker pulls

Setup


              $ curl -fsSL lognorth.com/install | bash

app.ts

import LogNorth from '@karloscodes/lognorth-sdk'

LogNorth.config('https://logs.yoursite.com', 'your-api-key')

// two calls. that's it.
LogNorth.log('Order placed', { total: 99 })
LogNorth.error('Payment failed', err)

↓

Feb 26 14:32:07 Order placed total=99

Feb 26 14:32:08 Payment failed checkout.ts:42 StripeError

The Problem

Stop paying more to watch your app than to run it.

Hosted log platforms solve a brutal multi-tenant problem: storing and querying billions of lines from thousands of customers at once. That scale is the hard part. It's also the part you don't have.

But you pay for it anyway. Metered by the gigabyte, thresholds to tune, on-call rotations to staff. You end up running observability instead of running your app.

Host your own logs. One tenant, not thousands. The hard part disappears, and logging and monitoring get simple. That's LogNorth.

The Answer

Logs do all of it.

Once it's just your logs, that's all you need. We only do logging. The alerts, errors, uptime, and AI answers are read out of one stream, no second system.

Logs alert

An error log pages you. Smarter than a flat 5%. It learns your baseline and fires when it's actually wrong.

Logs monitor

An outage shows up as silence. LogNorth learns your normal log volume and alerts the moment it drops to zero.

Logs debug

Logs and errors in one stream, grouped into 1 issue, not 1,000 duplicates. No separate tracker.

Logs feed AI

Point Claude Code or Cursor at your logs. It reads production, finds the failing request, and hands you the fix.

Logs stay readable

Plain lines a human scans in seconds. Errors in red, context attached. No grep, no parser.

Logs stay cheap

No per-GB meter, so you log everything, not just what you can afford. One SQLite file on a $5 box. $99 once.

SDKs: TypeScript Go Rails

Your AI agent reads all of it and investigates on its own. See how agents use LogNorth →

A closer look · Alerting

Percentages lie.

Take alerting. Everyone else pages you on a flat 5%: noise at 3am over nothing, dead silent at noon during a real fire. Tuning that threshold is the second full-time job.

3am

requests

1 error = 5%

Noise.

Noon

10,000

requests

500 errors = 5%

Fire.

We learn your baseline. Alert when it's actually wrong.

Alerts

Pinged where you'll see it.

When the logs say something's wrong, LogNorth reaches you on the channel you already check.

Telegram Bot token + chat ID

ntfy.sh Free push, no account

Email Your own SMTP

I got tired of paying monthly for logs.
So I built this.

Carlos · @karloscodes

Software developer. 15+ years shipping products.

Buy, don't rent

Buy once. The meter stops.

$99 for the software, plus a $5 box you own. Not a bill that runs forever.

Five years in: $399 vs $1,560. You keep $1,161.

Cancel a subscription and your logs are gone. You can't cancel what you own.

A small app at 12,000 requests a day. LogNorth is $99 once plus a small server, here Hetzner's cheapest box at $5/mo. The hosted side is a real entry plan at $26/mo with 90-day retention. That's the floor: send more and it climbs, and after 90 days the lookback ends. LogNorth keeps everything, as long as you want, on your own disk.

Pricing

No tiers. No enterprise. No sales calls.

$99 one-time

No subscriptions. 30-day money-back guarantee, no questions asked.

1 Docker image. No Kubernetes.
SQLite. No managed database.
Runs on a $5 VPS.
Your data never leaves your server.
Email support from me.

Buy LogNorth · $99

Installation

Logging in 3 minutes

Docker, TLS, automatic updates. Done.


              curl -fsSL lognorth.com/install | bash

FAQ

Questions

What do I get for $99?

Everything. Logs, errors, alerts, uptime checks, AI debugging, all SDKs. One price. No subscriptions. 30-day money-back guarantee.

Does it monitor uptime?

Yes. Paste your app URL when you create the app and LogNorth pings it every minute. 3 consecutive failures = down.

Why not use Sentry?

Sentry is a platform. LogNorth is a single Docker container. One file database. No dependencies. Runs on a $5 VPS. That's the trade-off: fewer features, zero complexity.

What frameworks are supported?

Go, Node/Bun, and Rails. Also accepts OpenTelemetry logs (OTLP/HTTP). We take what matters and discard the rest. No traces, no metrics, no complexity.

How does alerting work?

LogNorth groups errors into issues, learns your baseline traffic, and only alerts when something needs attention. 1 error in 1,000 requests? Normal. Sudden spike? That's a real problem. Push notifications via ntfy.sh, Telegram, or email with your own SMTP.

How long are logs stored?

You decide. Default is 90 days. Configure from 7 days to 1 year, or keep them forever. Shorter retention = smaller database.

Do I get updates?

Yes. Run lognorth update on your server and you're on the latest version. All updates within the current major version are included.