logpulse is a small command-line tool for watching Nginx access logs in real time.
Point it at a log file and it gives you a live view of request rates, error counts, busiest endpoints, and response times right in your terminal.
I built it in Go because I wanted something fast, portable, and easy to run on any server.
Sometimes you just want to SSH into a server and quickly understand what's happening.
Tools like Splunk and Datadog are powerful, but they can be overkill when all you need is a quick look at traffic and errors. On the other hand, tail -f shows raw logs but doesn't tell you much at a glance.
logpulse sits in the middle. It's lightweight, requires no setup, and runs entirely in your terminal.
- Watches a log file and processes new lines as they are written
- Shows requests per second, error rates, and average response times
- Tracks the top 5 most requested endpoints
- Groups errors by status code
- Displays an alert banner when errors exceed your threshold
- Works with Nginx Common Log Format
You need Go 1.20 or later.
git clone https://github.com/Jpeg-create/logpulse.git
cd logpulse
go build -o logpulse main.go parser.go tailer.go stats.go models.go dashboard.go cli.go
sudo mv logpulse /usr/local/bin/Start monitoring a log file:
logpulse --file /var/log/nginx/access.logWith additional options:
logpulse --file /var/log/nginx/access.log --alert-threshold 10 --errors-only| Flag | Default | Description |
|---|---|---|
--file |
sample.log |
Path to the log file to monitor |
--alert-threshold |
5 |
Number of errors that triggers the alert banner |
--errors-only |
false |
Only process requests with error status codes |
Press q to quit.
logpulse parses the standard Nginx Common Log Format:
192.168.1.1 - - [11/May/2026:13:45:01 +0000] "GET /api/users HTTP/1.1" 200 45
This is the default format used by Nginx and Apache, so it should work out of the box for most setups.
logpulse/
main.go # Application entry point
models.go # Data structures (LogEntry, Stats, etc.)
parser.go # Converts raw log lines into structs
tailer.go # Watches files for new log entries
stats.go # Aggregates metrics in a sliding window
dashboard.go # Terminal UI using Bubble Tea and Lip Gloss
cli.go # Command-line flags
The file watcher runs in a goroutine and sends new lines into a channel the moment they appear. The main goroutine sits on the other end of that channel, parses each line, updates the stats, and redraws the dashboard in place.
This is the part where Go genuinely shines. The watcher and the dashboard run concurrently without blocking each other, which is why the updates feel instant. No polling delay, no setup required, just a live view that stays in one spot.
- Bubble Tea for the terminal UI
- Lip Gloss for styling
MIT