From 927f6878632162d4e0eae9edf61b559729e885a6 Mon Sep 17 00:00:00 2001
From: ergosteur <1992147+ergosteur@users.noreply.github.com>
Date: Thu, 6 Aug 2026 15:44:49 +0000
Subject: [PATCH] Add Proxmox VM and LXC watchdog scripts

Cron-driven liveness checks that ping each guest, restart it via qm/pct if
unreachable, and log plus email the outcome. Guest lists are configured in
the associative array at the top of each script.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
---
 lxc-watchdog.sh | 74 +++++++++++++++++++++++++++++++++++++++++++++++++
 vm-watchdog.sh  | 59 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 133 insertions(+)
 create mode 100755 lxc-watchdog.sh
 create mode 100755 vm-watchdog.sh

diff --git a/lxc-watchdog.sh b/lxc-watchdog.sh
new file mode 100755
index 0000000..2ce2e5e
--- /dev/null
+++ b/lxc-watchdog.sh
@@ -0,0 +1,74 @@
+#!/bin/bash
+#
+# WatchDog script for Proxmox LXC Containers
+# v2.0
+#
+# Install using `crontab -e` like this: * * * * * /root/lxc_watchdog.sh
+#
+
+# Configurable Settings
+declare -A Containers=(
+    ["AdGuardLXC"]="100 192.168.31.10"
+    ["MainLXC"]="101 192.168.31.100"
+)
+
+logFile="/var/log/proxmox_lxc_watchdog.log"
+email="your_email@example.com"
+
+# Log function
+log () {
+    output="[LXC WatchDog] $(date '+%Y-%m-%d %H:%M:%S') $1"
+    echo "$output" >> "$logFile"
+    logger "$output"
+}
+
+# Send email alert
+sendAlert () {
+    subject="[Proxmox LXC WatchDog] $1"
+    body="$2"
+    echo -e "$body" | mail -s "$subject" "$email"
+}
+
+# Check LXC Container function
+checkContainer () {
+    local containerName=$1
+    local containerId=$2
+    local containerIp=$3
+
+    if /usr/bin/ping -c 1 "$containerIp" &> /dev/null; then
+        local containerStatus
+        containerStatus=$(/usr/sbin/pct status "$containerId" | awk '{print $2}')
+        if [[ "$containerStatus" == "running" ]]; then
+            local containerUptime
+            containerUptime=$(cat /proc/uptime | awk '{print $1}' | cut -d. -f1)
+            log "$containerName ($containerIp) is alive. Uptime: $containerUptime seconds."
+        else
+            log "$containerName ($containerIp) is not running. Attempting to start..."
+            if /usr/sbin/pct start "$containerId"; then
+                log "$containerName started successfully."
+                sendAlert "$containerName Started" "The LXC container '$containerName' was started by the watchdog script."
+            else
+                log "Failed to start $containerName. Manual intervention required."
+                sendAlert "$containerName Start Failed" "The LXC container '$containerName' failed to start. Please check Proxmox immediately."
+            fi
+        fi
+    else
+        log "$containerName ($containerIp) is down (no ping). Attempting to restart..."
+        if /usr/sbin/pct stop "$containerId" && /usr/sbin/pct start "$containerId"; then
+            log "$containerName restarted successfully."
+            sendAlert "$containerName Restarted" "The LXC container '$containerName' was restarted by the watchdog script."
+        else
+            log "Failed to restart $containerName. Manual intervention required."
+            sendAlert "$containerName Restart Failed" "The LXC container '$containerName' failed to restart. Please check Proxmox immediately."
+        fi
+    fi
+}
+
+# Main Logic
+for containerName in "${!Containers[@]}"; do
+    read -r containerId containerIp <<< "${Containers[$containerName]}"
+    checkContainer "$containerName" "$containerId" "$containerIp" &
+done
+
+# Wait for all background checks to complete
+wait
diff --git a/vm-watchdog.sh b/vm-watchdog.sh
new file mode 100755
index 0000000..58d98f4
--- /dev/null
+++ b/vm-watchdog.sh
@@ -0,0 +1,59 @@
+#!/bin/bash
+#
+# WatchDog script for Proxmox VMs
+# v2.0
+#
+# Install using `crontab -e` like this: * * * * * /root/watchdog.sh
+#
+# Configurable Settings
+declare -A VMs=(
+    ["pfsense"]="2801 10.20.28.1"
+)
+
+logFile="/var/log/proxmox_watchdog.log"
+email="alerts@example.com"
+
+# Log function
+log () {
+    output="[WatchDog] $(date '+%Y-%m-%d %H:%M:%S') $1"
+    echo "$output" >> "$logFile"
+    logger "$output"
+}
+
+# Send email alert
+sendAlert () {
+    subject="[Proxmox WatchDog] $1"
+    body="$2"
+    echo -e "$body" | mail -s "$subject" "$email"
+}
+
+# Check VM function
+checkVM () {
+    local vmName=$1
+    local vmId=$2
+    local vmIp=$3
+
+    if /usr/bin/ping -c 1 "$vmIp" &> /dev/null; then
+        local vmUptime
+        vmUptime=$(/usr/sbin/qm status "$vmId" -verbose | grep uptime | cut -f2 -d' ')
+        log "$vmName ($vmIp) is alive. Uptime: $vmUptime seconds."
+    else
+        log "$vmName ($vmIp) is down. Attempting restart..."
+        if /usr/sbin/qm stop "$vmId" && /usr/sbin/qm start "$vmId"; then
+            log "$vmName restarted successfully."
+            sendAlert "$vmName Restarted" "The VM '$vmName' was restarted by the watchdog script."
+        else
+            log "Failed to restart $vmName. Manual intervention required."
+            sendAlert "$vmName Restart Failed" "The VM '$vmName' failed to restart. Please check Proxmox immediately."
+        fi
+    fi
+}
+
+# Main Logic
+for vmName in "${!VMs[@]}"; do
+    read -r vmId vmIp <<< "${VMs[$vmName]}"
+    checkVM "$vmName" "$vmId" "$vmIp" &
+done
+
+# Wait for all background checks to complete
+wait