https://github.com/lengordy/net-watchdog
Deterministic self-healing infrastructure watchdog for Linux VPS environments with modular health checks and structured recovery.
https://github.com/lengordy/net-watchdog
devops docker infrastructure linux monitoring self-healing systemd vps watchdog
Last synced: 2 months ago
JSON representation
Deterministic self-healing infrastructure watchdog for Linux VPS environments with modular health checks and structured recovery.
- Host: GitHub
- URL: https://github.com/lengordy/net-watchdog
- Owner: lengordy
- License: mit
- Created: 2026-02-18T03:39:48.000Z (2 months ago)
- Default Branch: main
- Last Pushed: 2026-02-23T02:05:14.000Z (2 months ago)
- Last Synced: 2026-02-23T13:14:10.288Z (2 months ago)
- Topics: devops, docker, infrastructure, linux, monitoring, self-healing, systemd, vps, watchdog
- Language: Shell
- Homepage:
- Size: 66.4 KB
- Stars: 2
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- License: LICENSE
Awesome Lists containing this project
README
# Net Watchdog
Self-healing watchdog for Docker-based VPS environments.
Detects network degradation and applies prioritized recovery with anti-flap protection.
---
## What's New (v1.1.0)
- Container uptime-based VPN grace logic (`VPN_GRACE_SEC`)
- Prevents false alerts during planned container restarts
- Explicit `NET OK` notification after recovery
- Improved Telegram state transitions
- Added `.env.example` configuration template
---
## Quick Start
```bash
git clone https://github.com/lengordy/net-watchdog.git
cd net-watchdog
chmod +x net-watchdog.sh
```
Create environment file:
```bash
cp .env.example .env
```
Or export variables manually:
```bash
export TG_TOKEN=your_bot_token
export TG_CHAT_ID=your_chat_id
```
Run once:
```bash
./net-watchdog.sh
```
Optional: enable systemd timer
```bash
sudo cp examples/systemd/net-watchdog.* /etc/systemd/system/
sudo systemctl daemon-reload
sudo systemctl enable --now net-watchdog.timer
```
---
## Health Checks
- Default route
- Multi-target ping
- DNS resolution
- HTTPS reachability
- Local TCP port
- Docker daemon
- Container state
- Docker internal DNS
- Memory pressure
- MTU viability
- Container uptime grace handling
---
## Recovery Strategy
Recovery is controlled and anti-flap protected.
Priority order:
1. Restart container
2. Restart Docker
3. Restart systemd-resolved
4. Restart systemd-networkd
5. Optional reboot (explicitly enabled only)
Actions are triggered only after consecutive failures and cooldown validation.
No aggressive restarts. No blind loops.
---
## Modes
Run diagnostic only:
```bash
DIAG=1 ./net-watchdog.sh
```
Run self-test:
```bash
SELFTEST=1 ./net-watchdog.sh
```
Force emergency reboot after threshold:
```bash
FORCE_REBOOT=1 ./net-watchdog.sh
```
---
## License
MIT