Schedule Maigret Jobs Without Surprises

Automation keeps identity sweeps current, but needs guardrails to avoid hammering sites or leaking data.

Wrapper script

#!/usr/bin/env bash
set -euo pipefail

export ALL_PROXY="socks5://127.0.0.1:9050"
STAMP="$(date +%F)"
INPUT="/opt/maigret/usernames.txt"
OUT="/var/reports/maigret"
LOG="/var/log/maigret-${STAMP}.log"

maigret --input "$INPUT" \
  --html --json --folderoutput \
  --output "$OUT" \
  --timeout 45 --max-connections 30 \
  --print-not-found false \
  >>"$LOG" 2>&1

Cron entry

0 1 * * 1 torsocks /opt/maigret/run-maigret.sh

systemd timer (alternative)

# /etc/systemd/system/maigret.service
[Service]
Type=oneshot
ExecStart=/opt/maigret/run-maigret.sh

# /etc/systemd/system/maigret.timer
[Timer]
OnCalendar=Mon *-*-* 01:00:00
Persistent=true

sudo systemctl enable --now maigret.timer

Monitoring checklist

• Tail journalctl -u maigret.service for failures.
• Set up logrotate for /var/log/maigret-*.log.
• Run maigret --self-check weekly to disable broken sites.
• Use flock to prevent overlapping executions.

Alerting idea

  flowchart LR
    A[systemd timer] --> B[Maigret wrapper]
    B --> C[Reports directory]
    B --> D[Logfile]
    D --> E[Promtail]
    E --> F[Grafana Loki Alerts]

By combining torified networking, log rotation, and service timers you avoid both detection risks and redundant scans.