Bash Watchdogs, Discord Alerts, and Dynamic DNS

SSH Watcher
Port Watcher
Discord Notifications
Reboot Notifications
DDNS Updater

#SSH Watcher

Make sure your server is set up to allow SSH logins with public keys and, at the server console or shell, as root, login to your local server (yes, recursively):

ssh <ipaddress you want to be monitored>

The first time you do this, you should get a message about remembering the fingerprint – say “yes”.

You will be logged into the server in an SSH shell from your current SSH shell. You can exit.
Next time you log in like this:

ssh <ipaddress you want to be monitored>

there should be no prompt.

Create a file /opt/watchdog/watchdog-ssh.sh

#!/bin/bash
SSH_USER="root"
SSH_HOST="<ipaddress you want to be monitored>"
SSH_PORT="<port of ssh>"

attempt_ssh_login() {
    local user="$1"
    local host="$2"
    local port="$3"
    echo "Attempting SSH login to $user@$host:$port..."
    ssh -o BatchMode=yes \
        -o ConnectTimeout=10 \
        -o StrictHostKeyChecking=no \
        -o PasswordAuthentication=no \
        -p "$port" "$user@$host" "exit" &> /dev/null
    return $?   # Return the exit status of the ssh command
}

# Perform the SSH login attempt
attempt_ssh_login "$SSH_USER" "$SSH_HOST" "$SSH_PORT"

# Check the exit status
SSH_STATUS=$?

if [ "$SSH_STATUS" -eq 0 ]; then
    # --- Actions on SUCCESS ---
    echo "SSH login to $SSH_USER@$SSH_HOST:$SSH_PORT successful!"
    echo "Nothing to do, exiting."
    # You can put a script here but consider this runs every 5 minutes.
    exit 0

elif [ "$SSH_STATUS" -ne 0 ]; then
    # --- Actions on FAILURE ---
    echo "SSH login to $SSH_USER@$SSH_HOST:$SSH_PORT FAILED (Exit Code: $SSH_STATUS)."
    echo "Restarting sshd..."
    systemctl restart ssh.service   # Restart the SSH service; your distro might vary.
    # /path/scriptToRun.sh  # you can run a script here; make sure it is +x and on PATH.
    exit 1

else
    # This block should not be reached if SSH_STATUS is always 0 or non-zero, but for completeness:
    echo "An unexpected error occurred."
    exit 2
fi

Make the file executable:

chmod +x /opt/watchdog/watchdog-ssh.sh

Open crontab and put the line there to run every 5 minutes:

crontab -e

Add this on a new line:

*/5 * * * * /opt/watchdog/watchdog-ssh.sh

You can manually run the script (be aware it restarts sshd) or wait a few minutes and check the crontab logs (may vary with distro):

grep CRON /var/log/syslog

Or you can use the script logic to write something meaningful or send a Discord notification.

Note: Some distros don’t like */5 in a cron job. If you have that trouble, try http://crontab.guru for tips.

Also, you don’t have to use root for this. You can use any user with permissions and a valid public key.

#Port Watcher

Similarly, a port can be checked with nc (netcat). In this case, create a file /opt/watcher/watcher-minecraft.sh

#!/bin/bash
TARGET_HOST="172.16.30.212"   # The IP address or hostname of the target server
TARGET_PORT="25565"           # The port number to check

check_port_status() {
    local host="$1"
    local port="$2"
    echo "Checking if port $port on $host is open..."
    if nc -z -w 1 "$host" "$port" &> /dev/null; then
        return 0   # Port is open
    else
        return 1   # Port is not open
    fi
}

# Perform the port check
check_port_status "$TARGET_HOST" "$TARGET_PORT"

# Check the exit status of the port check
PORT_STATUS=$?

if [ "$PORT_STATUS" -eq 0 ]; then
    # --- Actions on SUCCESS (Port is Open) ---
    echo "Port $TARGET_PORT on $TARGET_HOST is open. Nothing to do, exiting."
    exit 0

elif [ "$PORT_STATUS" -ne 0 ]; then
    # --- Actions on FAILURE (Port is NOT Open) ---
    echo "Port $TARGET_PORT on $TARGET_HOST is NOT open (Exit Code: $PORT_STATUS)."
    echo "Attempting to restart Minecraft, watch Discord for updates..."
    cd /opt/docker
    docker compose restart minecraft   # Restart the service or run a script

    if [ $? -eq 0 ]; then
        echo "Service restarted successfully."
    else
        echo "Failed to restart service. Check Discord for updates."
    fi
    exit 1

else
    # This block should not theoretically be reached
    echo "An unexpected error occurred during port check."
    exit 2
fi

This is the easiest way, I think, to keep track of things as the app is nice and the service is reliable.
I find sometimes messages might be 1–2 minutes after the fact, so this is not a real-time messaging solution.

Create a Discord account, server, and webhook.

You should have a string that looks like this:

https://discordapp.com/api/webhooks/115954877264954...

Make a file /opt/discord/discord.conf and put the string in, like this:

webhook="https://discordapp.com/api/webhooks/115954877264954..."

Lock the file down:

chmod 400 /opt/discord/discord.conf

Make a file /opt/discord/discord.sh:

#!/bin/bash
. /opt/discord/discord.conf
curl -H "Content-Type:application/json" \
     -d "{\"username\": \"$1\", \"content\": \"$2\"}" $webhook

Make this file executable:

chmod 775 /opt/discord/discord.sh

Now, if you want to send a notification to your Discord channel from the command line:

root@server12:~# /opt/discord/discord.sh "username" "message"

This will send a message with the fields to the Discord channel on the server.
Make sure the server is private.

Similarly, you can add that line to one of the actions above:

elif [ "$PORT_STATUS" -ne 0 ]; then
    # --- Actions on FAILURE (Port is NOT Open) ---
    echo "Port $TARGET_PORT on $TARGET_HOST is NOT open (Exit Code: $PORT_STATUS)."
    echo "Attempting to restart Minecraft, watch Discord for updates..."
    /opt/discord/discord.sh "server12" "Attempting to restart the Minecraft container..."
    cd /opt/docker
    docker compose restart minecraft

#Reboot Notifications

A little-known key of crontab is @reboot, and it can be used like this:

crontab -e

Add this on its own line:

@reboot /opt/discord/discord.sh "$(hostname)" "$(hostname) starting up..."

Now, whenever your Linux server starts up, you get a short message; this is good to get confirmation of a restart or a warning the power may have gone off.

#DDNS Updater

Often when self-hosting the problem is related to the ever-changing IP address at home. To get around this, I created a Cloudflare account, then set up the free DNS services and hosted a domain I own there. I then created a record called home.<mydomain> with IP address 1.2.3.4.

Make sure with nslookup the IP address is visible and resolves:

nslookup home.<mydomain>

Should return the IP 1.2.3.4.

In Cloudflare, click on your profile picture, then Preferences, then API Tokens and make a token that has EDIT rights to <mydomain> domain.

Install jq (varies by distro):

apt install jq

Get your zone ID from Cloudflare’s API:

curl -X GET "https://api.cloudflare.com/client/v4/zones?name=<mydomain>" \
     -H "Authorization: Bearer <cloudflare token>" \
     -H "Content-Type: application/json"

The output is 30- or 40-character alphanumerics; that is the <zoneid> for <mydomain>.

Create a script /opt/ddns/ddns.sh:

#!/bin/bash
# Don't forget to install the jq package

## keep these private
cloudflare_auth_key=<cloudflare token>

# Cloudflare zone is the zone which holds the record
dnsrecord=home.<mydomain>
zoneid=<zone id>

# Get the current external IP address
ip=$(curl -4 icanhazip.com)

#echo "Current IP is $ip"

if host $dnsrecord 1.1.1.1 | grep "has address" | grep "$ip"; then
  echo "$dnsrecord is currently set to $ip; no changes needed"
else
  /opt/discord/discord.sh "$(hostname)" "Cloudflare address for $dnsrecord should be $ip, I'll update the address..."

  # get the dns record id
  dnsrecordid=$(curl -s -X GET "https://api.cloudflare.com/client/v4/zones/$zoneid/dns_records?type=A&name=$dnsrecord" \
    -H "Authorization: Bearer $cloudflare_auth_key" \
    -H "Content-Type: application/json" | jq -r  '{"result"}[] | .[0] | .id')

  # update the record
  curl -s -X PUT "https://api.cloudflare.com/client/v4/zones/$zoneid/dns_records/$dnsrecordid" \
    -H "Authorization: Bearer $cloudflare_auth_key" \
    -H "Content-Type: application/json" \
    --data "{\"type\":\"A\",\"name\":\"$dnsrecord\",\"content\":\"$ip\",\"ttl\":1,\"proxied\":false}" | jq
fi

Make the file executable:

chmod 775 /opt/ddns/ddns.sh

The script checks the IPv4 address of the public interface against what is in Cloudflare and makes a change if needed. It also sends a Discord notice if a change was required.

I know my ISP resets the IP address every morning at 03:00 and I get a new one if the modem reboots for any reason, so this script can be added to crontab to run every 20 minutes:

crontab -e

Put this on a new line:

*/20 * * * * /opt/ddns/ddns.sh

You can test this by restarting your modem to get a new IP, then manually running the script. You should see nslookup returning your public IP after a few minutes and, of course, a Discord notification. You can also use the logic of the script to run another script in case some service needs to be restarted with an IP renewal, like a VPN service.

All the scripts above can be combined to make them monitor each other and provide some resilience. For example, you can use a modification of the port-monitoring script to check the DNS record and alert via Discord if a record does not return results.

Something else: I know not everybody is into AI, but for these short scripts most AIs are actually really good. Try this in Gemini:

I am a sysadmin looking after a server. I want a bash script that checks if a specific file has been changed in the last 10 minutes and return a 1 else 0.

URL: https://ib.bsb.br/bash-watchdogs-discord-alerts-and-dynamic-dns

#SSH Watcher

#Port Watcher

#Discord Notifications

#Reboot Notifications

#DDNS Updater