Astrill Watchdog

Monitor and auto-reconnect Astrill VPN on Ubuntu Linux (deb GUI package). Detects dropped connections via tun interface + ping, then reconnects using Astrill...

Monitors Astrill VPN on Ubuntu (deb GUI package) and automatically restarts it when the StealthVPN tunnel drops.

What it does

Watches tun0 + ping every 30 seconds. On failure, performs a full Astrill restart:

• pkill astrill kills the process tree (root-owned children asproxy/asovpnc die with the parent — no sudo needed)
• setsid /autostart relaunches Astrill with the full desktop environment (DISPLAY, DBUS, WAYLAND_DISPLAY) so it can initialize its GUI/Wayland stack from a systemd service context
• Astrill auto-connects to the last used server

On restart failure: logs a CRITICAL block, resumes checking next cycle. Never exits.

Requirements

• Ubuntu Linux, Astrill deb GUI package (/usr/local/Astrill/astrill)
• ping, ip, pgrep, pkill, setsid (Ubuntu defaults)
• Active desktop session (DISPLAY/DBUS/WAYLAND) — required for Astrill relaunch

Installation

bash

bash setup.sh

No sudo. Installs the watchdog, creates a systemd user unit, and starts the service. Enabled on login automatically.

Usage

bash

astrill-watchdog.sh start    # start watchdog (also done by systemd on login)
astrill-watchdog.sh stop     # stop watchdog
astrill-watchdog.sh status   # health summary + last 20 log lines
astrill-watchdog.sh once     # single health check + restart if needed, then exit

Files

Path	Purpose
~/.config/astrill-watchdog/astrill-watchdog.sh	Watchdog script
~/.config/systemd/user/astrill-watchdog.service	Systemd user unit
~/.local/state/astrill-watchdog/watchdog.log	Log file (rotates at 5000 lines)
~/.local/state/astrill-watchdog/watchdog.pid	PID file

Configuration

Edit the config block at the top of astrill-watchdog.sh:

bash

CHECK_INTERVAL=30      # seconds between health checks
RECONNECT_WAIT=60      # seconds to wait after restart before health check
PING_HOST="8.8.8.8"
PING_COUNT=3
PING_TIMEOUT=3
LOG_MAX_LINES=5000

After editing, restart: systemctl --user restart astrill-watchdog.service

Diagnostics

bash

# Live log tail
tail -f ~/.local/state/astrill-watchdog/watchdog.log

# Systemd journal
journalctl --user -u astrill-watchdog.service -n 30

# Full status summary
astrill-watchdog.sh status