logmera 0.1.1

Self-hosted AI observability backend

logmera

Self-hosted AI observability backend with FastAPI + PostgreSQL, plus a lightweight Python SDK logger.

Features

• Async FastAPI backend
• PostgreSQL storage via SQLAlchemy ORM
• Auto table creation for logs on startup
• Typer CLI with interactive DB URL setup
• Python SDK function: logmera.log(...)

Install

pip install logmera

Start Server

Use a browser-safe port like 8000 (avoid 6000 in browser).

logmera --host 127.0.0.1 --port 8000

Notes:

• The CLI is currently single-command mode, so use logmera --port 8000 (not logmera start ...).
• On first run, if DB is not configured, it prompts for DATABASE_URL.
• Entered DB URL is saved at ~/.logmera/config.env.

Database Configuration

Resolution order for DB URL:

1. --db-url
2. DATABASE_URL env var
3. ~/.logmera/config.env
4. Interactive prompt

Examples:

logmera --db-url "postgresql://postgres:postgres@localhost:5432/logmera"

logmera --no-prompt --db-url "postgresql://postgres:postgres@localhost:5432/logmera"

Optional DB pool env vars:

DB_POOL_SIZE=10
DB_MAX_OVERFLOW=20
DB_POOL_TIMEOUT=30

API Endpoints

• GET / -> welcome message
• GET /health -> {"status":"ok"}
• POST /logs -> create log
• GET /logs -> list logs ordered by created_at DESC

Example: Health

curl http://127.0.0.1:8000/health

Example: Create Log

curl -X POST http://127.0.0.1:8000/logs \
  -H "Content-Type: application/json" \
  -d '{
    "project_id":"demo",
    "prompt":"Hello",
    "response":"Hi there",
    "model":"gemini-2.5-flash",
    "latency_ms":123,
    "status":"success"
  }'

Example: List Logs

curl http://127.0.0.1:8000/logs

SDK Usage

logmera/sdk.py exposes logmera.log(...).

import logmera

ok = logmera.log(
    project_id="demo-chat",
    prompt="Hello",
    response="Hi",
    model="gemini-2.5-flash",
    latency_ms=95,
    status="success",
)

SDK env vars:

• LOGMERA_URL (default: http://127.0.0.1:8000)
• LOGMERA_TIMEOUT_SECONDS (default: 10)
• LOGMERA_RETRIES (default: 2)

Local PostgreSQL Quick Start (Docker)

docker run --name logmera-postgres \
  -e POSTGRES_USER=postgres \
  -e POSTGRES_PASSWORD=postgres \
  -e POSTGRES_DB=logmera \
  -p 5432:5432 -d postgres:16

Then start:

logmera --db-url "postgresql://postgres:postgres@localhost:5432/logmera"