Tag Archives: security

Proxmox Time Sync

Posted on by
Reply

Syncing time is essential on Proxmox servers, especially in a cluster — time drift can cause serious issues with quorum, backups, and HA.

Here’s how to ensure accurate time sync on your Proxmox nodes:

✅ Recommended Method: Use systemd-timesyncd (built-in on Debian)

🔧 1. Enable and start time sync:

sudo timedatectl set-ntp true

Verify:

timedatectl status

Look for:

NTP service: active
System clock synchronized: yes

This uses systemd-timesyncd, which is light and reliable for Proxmox servers.

🧰 Alternative: Use chrony (Recommended for more control)

chrony is more robust, especially if your nodes might be offline/online often or you’re doing HA.

1. Install chrony:

sudo apt install chrony

2. Edit config (optional):

sudo nano /etc/chrony/chrony.conf

Add or edit NTP servers:

server time.google.com iburst
server 0.pool.ntp.org iburst
server 1.pool.ntp.org iburst

3. Restart and enable:

sudo systemctl enable chrony
sudo systemctl restart chrony

4. Check sync status:

chronyc tracking

🧪 Verify Time Sync on Proxmox

On each node:

timedatectl

All nodes should show:

  • NTP service: active
  • System clock synchronized: yes
  • Same or very close times

🧠 Bonus: Sync RTC (hardware clock)

To sync system time to the hardware clock:

sudo hwclock --systohc

Let me know if you’re using a cluster and want to designate one Proxmox node as an NTP server for others, or if you’d like to sync from your Pi-hole box or router!

Perfect — turning one of your Proxmox nodes into an NTP server is a great idea for local time sync, especially in a cluster or if internet access is limited.

Here’s how to do it using chrony, which is ideal for Proxmox nodes:

🧭 Overview

  • One Proxmox node will be the NTP server.
  • The others will sync from it as NTP clients.

🖥️ On the Proxmox NTP Server (e.g., pve1)

1. Install chrony if not already installed:

sudo apt update
sudo apt install chrony

2. Edit the config to allow LAN clients

sudo nano /etc/chrony/chrony.conf

Add or uncomment these lines:

# Allow LAN subnet (adjust to your LAN range)
allow 192.168.1.0/24

# Make this machine act as an NTP server
local stratum 10

Optional: Use a known upstream NTP source for initial sync:

server time.google.com iburst
server 0.pool.ntp.org iburst

3. Restart chrony and enable it

sudo systemctl restart chrony
sudo systemctl enable chrony

🖥️ On the Proxmox Clients (e.g., pve2, pve3)

1. Install chrony:

sudo apt install chrony

2. Configure it to sync from the Proxmox NTP server:

sudo nano /etc/chrony/chrony.conf

Comment out existing server lines, and add:

server 192.168.1.10 iburst   # Replace with the IP of your Proxmox NTP server

3. Restart and enable:

sudo systemctl restart chrony
sudo systemctl enable chrony

4. Verify sync status:

chronyc sources
chronyc tracking

You should see something like:

Name/IP address             Stratum Poll Reach LastRx Last sample
===============================================================================
192.168.1.10                     10   6   377    32    -34us[ -44us] +/-  200us

🧪 Bonus: Test if the server is responding

From another node:

chronyc sources

Or:

ntpq -p 192.168.1.10

Let me know if you want to also allow the Pi-hole box or other clients on your LAN to sync from this NTP server too — it’s as easy as expanding the subnet in allow.

The Hardened (Home) Data Center

Posted on by
Reply

What’s a good plan for remote access to my smart home during a power outage when I’m on an extended vacation?

Jump to What I Did.

Since you’ll be away for a while, you need a robust failover plan to ensure continuous remote access to your smart home and Proxmox cluster. Here’s a multi-layered approach using your existing equipment:

🌐 1️⃣ Internet Failover

Since you have:

  • Primary ISP: AT&T Fiber (Nokia BGW320-505)
  • Backup ISP: T-Mobile 5G Hotspot (30GB/month before throttling)
  • Spare ASUS RT-AC68U Router (WAN Failover Capable)

🛠 Steps to Set Up Failover

  1. Enable Dual WAN Failover on ASUS Router
    • Set AT&T as primary WAN and T-Mobile Hotspot as secondary WAN
    • Choose failover mode (not load balancing)
    • Adjust health check settings to detect AT&T failures
  2. Avoid Double NAT Issues
    • If AT&T’s Nokia router is in router mode, enable bridge mode on the ASUS router
    • Assign static routes if needed to prevent conflicts

🔌 2️⃣ Power Resilience

Since you have a UPS, ensure:

  • Proxmox nodes, routers, and modem are on the UPS
  • Use a smart plug on your modem & router (in case they lock up)

🛠 Automation Ideas

  • Smart Plug Auto-Reboot
    • Use a TP-Link Kasa or Tasmota plug
    • Set up a rule: If internet is down for 10+ min, cycle power to the router/modem

🔐 3️⃣ Remote Access Setup

🛠 VPN Options

  1. Tailscale VPN (Easiest & Most Reliable)
    • Run Tailscale on your Raspberry Pi (or Proxmox VM)
    • Enables access even if your IP changes
    • No port forwarding needed
  2. PiVPN (WireGuard or OpenVPN)
    • Run on a Raspberry Pi
    • Requires port forwarding, so it may fail if your ISP changes your IP
  3. Cloudflare Tunnel (Great Backup Plan)
    • Provides a free tunnel to access your smart home services
    • Works even if your public IP changes

🚨 4️⃣ Automatic Recovery from Failures

🛠 Proxmox & Server Auto-Recovery

  • Use Smart Plugs for Power Cycling
  • Watchdog Timer on Proxmox
    • If a Proxmox node stalls, it will auto-reboot

🛠 Network Watchdog on Raspberry Pi

Run a simple script to detect network failures and auto-reboot devices:

#!/bin/bash
if ! ping -c 5 8.8.8.8 &> /dev/null; then
  echo "Internet down! Rebooting router..."
  ssh user@smartplug "turn off && sleep 10 && turn on"
fi

(Schedule with crontab -e to run every 10 minutes)

📌 Summary: Best Remote Access Plan

✅ Failover Internet → ASUS RT-AC68U handles automatic switch to T-Mobile 5G
✅ UPS Protection → Keeps critical devices online during power outages
✅ Smart Plug Automation → Auto-reboots router/modem if unresponsive
✅ VPN Access (Tailscale) → Provides secure access even if ISP changes IP
✅ Proxmox Watchdog → Ensures servers auto-recover from crashes
✅ Cloudflare Tunnel (Optional) → Secondary remote access method

🚀 Next Steps

  • Do a test run before your trip
  • Monitor logs remotely (Tailscale + Home Assistant)
  • Set up alerts (Home Assistant + Pushover/Telegram)

Would you like help setting up any of these steps? 🚀

I implemented dual Wan on the ASUS router and used IP passthrough on the Nokia router. Failover mode works. Most of my automation works with Alexa. I have the UPS. I’ll implement the RPi watchdog. I think I’ll shutdown Proxmox during a power outage to preserver the UPS. I think I can get the temperature from my Ambient Weather station if it’s on the UPS. Two of my blink cameras work on batteries. I’ll need an Echo on UPS.

That sounds like a solid plan for maintaining remote access and automation during an outage. If your ASUS router supports custom scripts, you might consider adding automation to gracefully shut down Proxmox when the UPS battery level gets too low. Let me know if you want help setting that up!

how would you detect the battery level?

You can detect the UPS battery level using Network UPS Tools (NUT), which allows you to monitor your UPS from a Raspberry Pi, your Proxmox servers, or another system on your network. Here’s how you can set it up:

Step 1: Install NUT on Proxmox or a Raspberry Pi

On your Proxmox node or Raspberry Pi (whichever is connected to the UPS via USB):

sudo apt update
sudo apt install nut nut-client

Step 2: Configure NUT to Read UPS Data

Edit the NUT configuration file to recognize your UPS:

sudo nano /etc/nut/ups.conf

Add an entry for your UPS. For example, if it’s an APC UPS, it may look like this:

[myups]
    driver = usbhid-ups
    port = auto
    desc = "My UPS"

Save and exit.

Restart the NUT service:

sudo systemctl restart nut-server

Step 3: Check Battery Status

Run the following command to check UPS status:

upsc myups

Look for a value like:

battery.charge: 85

This tells you the current battery percentage.

Step 4: Automate Proxmox Shutdown on Low Battery

Create a script (/usr/local/bin/ups-shutdown.sh) that will shut down Proxmox when the battery is low:

#!/bin/bash
BATTERY_LEVEL=$(upsc myups battery.charge)

if [ "$BATTERY_LEVEL" -lt 20 ]; then
    echo "Battery low ($BATTERY_LEVEL%). Shutting down Proxmox..."
    /usr/sbin/poweroff
fi

Make it executable:

chmod +x /usr/local/bin/ups-shutdown.sh

Step 5: Run This Script Periodically

Add this to cron to check the battery level every minute:

sudo crontab -e

Add this line at the bottom:

* * * * * /usr/local/bin/ups-shutdown.sh

This setup will ensure Proxmox shuts down when the UPS battery gets too low.

Would you like help with setting up email or push notifications when the power goes out?

To get email or push notifications when the power goes out, you can integrate the UPS monitoring system with sendmail (for email notifications) or use Pushover or Telegram (for mobile push notifications).

1. Send Email Notification on Power Outage

You’ll need msmtp and mailutils to send emails from the Raspberry Pi or Proxmox server.

Install Required Packages

sudo apt update
sudo apt install msmtp mailutils

Configure msmtp

Edit the configuration file:

sudo nano /etc/msmtprc

Add the following, adjusting for your SMTP provider (Gmail, Outlook, etc.):

defaults
auth           on
tls            on
tls_trust_file /etc/ssl/certs/ca-certificates.crt
logfile        /var/log/msmtp.log

account default
host smtp.gmail.com
port 587
from your-email@gmail.com
user your-email@gmail.com
password your-email-password

Save and exit.

Test Email Sending

Try sending an email:

echo "Test email from Raspberry Pi" | mail -s "UPS Test Email" your-email@example.com

If you receive the email, the setup is working.

Modify the Shutdown Script to Send an Email

Edit the /usr/local/bin/ups-shutdown.sh script to send an email when the UPS is on battery:

#!/bin/bash
BATTERY_LEVEL=$(upsc myups battery.charge)
ONLINE_STATUS=$(upsc myups ups.status)

if [[ "$ONLINE_STATUS" != "OL" ]]; then
    echo "Power outage detected! UPS running on battery. Battery level: $BATTERY_LEVEL%" | mail -s "Power Outage Alert" your-email@example.com
fi

if [ "$BATTERY_LEVEL" -lt 20 ]; then
    echo "Battery low ($BATTERY_LEVEL%). Shutting down Proxmox..." | mail -s "Critical Battery Alert - Shutting Down" your-email@example.com
    /usr/sbin/poweroff
fi

2. Push Notifications Using Telegram

If you prefer mobile push notifications, you can use Telegram bots:

Create a Telegram Bot

  1. Open Telegram and search for BotFather.
  2. Start a chat and use the command /newbot.
  3. Give your bot a name and username.
  4. Copy the API token from BotFather.

Find Your Chat ID

  1. Open https://api.telegram.org/bot<YOUR_BOT_TOKEN>/getUpdates in a browser.
  2. Send your bot a message.
  3. Refresh the browser; your chat ID will appear.

Modify the Shutdown Script to Send Telegram Messages

Edit /usr/local/bin/ups-shutdown.sh:

#!/bin/bash
BATTERY_LEVEL=$(upsc myups battery.charge)
ONLINE_STATUS=$(upsc myups ups.status)
TELEGRAM_BOT_TOKEN="YOUR_BOT_TOKEN"
TELEGRAM_CHAT_ID="YOUR_CHAT_ID"

send_telegram() {
    curl -s -X POST "https://api.telegram.org/bot$TELEGRAM_BOT_TOKEN/sendMessage" \
        -d chat_id="$TELEGRAM_CHAT_ID" \
        -d text="$1"
}

if [[ "$ONLINE_STATUS" != "OL" ]]; then
    send_telegram "⚠️ Power outage detected! UPS running on battery. Battery level: $BATTERY_LEVEL%"
fi

if [ "$BATTERY_LEVEL" -lt 20 ]; then
    send_telegram "🔴 Critical Battery Alert! Battery level: $BATTERY_LEVEL%. Shutting down Proxmox."
    /usr/sbin/poweroff
fi

3. Running the Script on a Schedule

Edit cron:

sudo crontab -e

Add this line:

* * * * * /usr/local/bin/ups-shutdown.sh

This checks the UPS status every minute.

Would you like to add any additional notification methods, like Pushover or Home Assistant alerts?

What I Did

Sure! Here’s a quick and clear guide to set up SSH key-based login from your local machine to a remote Linux server, so you can skip password prompts.

🔑 Step-by-Step: SSH Key Setup

1. Check for existing SSH keys on your local machine

ls ~/.ssh/id_rsa.pub

If it exists, skip to step 3.

2. Generate a new SSH key pair (if needed)

ssh-keygen -t rsa -b 4096 -C "your_email@example.com"
  • Press Enter to accept defaults.
  • You can set a passphrase (or leave it blank for no prompt).

This creates:

  • Private key: ~/.ssh/id_rsa
  • Public key: ~/.ssh/id_rsa.pub

3. Copy your public key to the remote server

ssh-copy-id user@remote-host
  • Replace user with your username and remote-host with the IP or hostname.
  • It will prompt for your password once to copy the key.
Alternate method (if ssh-copy-id isn’t available):
cat ~/.ssh/id_rsa.pub | ssh user@remote-host 'mkdir -p ~/.ssh && cat >> ~/.ssh/authorized_keys && chmod 600 ~/.ssh/authorized_keys && chmod 700 ~/.ssh'

4. Test it

ssh user@remote-host

You should now log in without entering a password.

✅ Optional: Configure Passwordless sudo

If you want to run shutdown commands via SSH without typing the sudo password:

  1. On the remote server, run:
sudo visudo
  1. Add this line at the bottom (change user to your username):
user ALL=(ALL) NOPASSWD: /sbin/shutdown, /sbin/poweroff

Now your remote shutdown command works like this:

ssh user@remote-host 'sudo /sbin/shutdown -h now'

Crontab

  1. timedatectl # check time zone
  2. timedatectl set-timezone America/Chicago
  3. * * * * * date >> /home/pi/tz.txt #verify how crontab sees time

Automatic Internet Failover

Posted on by
Reply

I have a three node proxmox cluster running home assistant, jellyfin, file servers, pivpn, pihole for dhcp and a dozen other servers that I need to have remote access to. I have some spare raspberry pi’s and spare smart plugs. I have a T-Mobile hotspot with 30gb of monthly data before it throttles data. I have a backup UPS. I’m using a Nokia wireless router for connection to my ISP which is AT&T. I also have a spare ASUS RT-AC68U router that has a WAN failover feature. I think it may be necessary to reset the power to recover the proxmox servers. My servers will be unattended when I’m away. How can I make sure I don’t lose access to my home network.

To Do Regarding Following Responses

  1. Install Tailscale on RPi
  2. Setup ASUS router as access point with hotspot as secondary WAN
  3. Install smart plugs for AT&T router and Proxmox servers
  4. Install watchdog on RPi
  5. Install Pihole on RPi
  6. Install DDClient on RPi
  7. Simulate power outage
  8. Simulate ISP outage
  9. Simulate ISP failover
  10. Test hotspot for continuous operation
  11. Setup range extender

Solution

This is not a trivial solution. Using the AI responses from below, this is my implementation. My ASUS router supports dual WAN failover but my AT&T Nokia router connects to the internet. The best way to avoid double NAT is to put the AT&T router in bridge mode but that’s not available. However, it supports IP Passthrough.

To turn on IP Passthrough, the Nokia router needs have DHCP on the ASUS router turned on. Since I use Pihole for DHCP, I isolated my workstation and both routers to establish this with wifi off. The T-Mobile hotspot network was not configurable and conflicted with my long time 192.168 network, so I had to change my local network to a new 192.168 network. This was mostly trivial and tedious except for Proxmox. It broke my cluster and my Ceph shared storage. With some AI help, I repaired the broken cluster but not yet Ceph. I tethered my hotspot to the ASUS router, turned off the Nokia router and voilà it failed over to the hotspot. With the Nokia router back on, it failed back.

What I’m really trying to do is “harden my data center”. What’s the best combination of access and duration I can have to my “smart” home while I’m away during a power outage. Once I did a u-turn in Tucumcari when my home lost power in a cold spell. Failover to a battery backup 5G internet access is a big asset. Maybe a wifi thermometer and camera?

To avoid Double NAT when using the ASUS RT-AC68U as a failover router with your AT&T Nokia BGW320-505, you need to put the Nokia router into IP Passthrough mode.

🔧 Steps to Set Up IP Passthrough on AT&T Nokia BGW320-505

This will make the ASUS router handle all routing and NAT, preventing conflicts.

1️⃣ Connect Your ASUS Router to the Nokia Router

  • Plug an Ethernet cable from a LAN port on the Nokia BGW320-505 into the WAN port of the ASUS RT-AC68U.

2️⃣ Enable IP Passthrough on the Nokia BGW320-505

  1. Log in to the AT&T Router
    • Open a browser and go to: http://192.168.1.254
    • Sign in (default access code is on the router label)
  2. Navigate to IP Passthrough
    • Click Firewall → IP Passthrough
  3. Set Up Passthrough Mode
    • Allocation Mode: Passthrough
    • Passthrough Mode: DHCPS-Fixed
    • Passthrough MAC Address: Select the MAC address of the ASUS router’s WAN port
  4. Save & Apply Settings
    • Reboot the Nokia router and ASUS router.

3️⃣ Configure the ASUS Router

  1. Log in to ASUS RT-AC68U (http://192.168.1.1)
  2. Set WAN to Automatic IP (DHCP)
    • Go to WAN → Internet Connection
    • Set WAN Connection Type = Automatic IP
    • Disable DHCP Query Frequency
    • Click Apply
  3. Disable ASUS’s NAT Acceleration (Optional for stability)
    • Go to LAN → Switch Control
    • Set NAT Acceleration = Disabled
    • Click Apply

4️⃣ Confirm the Setup

  • On the ASUS Router, check the WAN IP Address (it should be the same as your public IP).
  • The Nokia router will no longer perform NAT, preventing Double NAT.
  • The ASUS router will now handle failover between AT&T and T-Mobile 5G.

✅ Now, all devices will use the ASUS router as their only gateway, preventing Double NAT! 🚀

ChatGPT Response

You have a solid setup with redundancy, but since your Proxmox servers will be unattended while you’re away, here’s a fail-safe strategy to ensure remote access and automatic recovery after a power outage or network failure.

1️⃣ Ensure Remote Access

(A) VPN Access via PiVPN

  • PiVPN is great, but if your primary ISP (AT&T) goes down or changes your external IP, you might lose access.
  • Solution: Set up a Dynamic DNS (DDNS) service like DuckDNS, No-IP, or Cloudflare to track your home IP address.
  • Alternative: Have a backup VPN (e.g., Tailscale or ZeroTier) running on a Raspberry Pi.

(B) T-Mobile Hotspot for Backup Internet

  • Use the ASUS RT-AC68U as a failover router:
    • Primary WAN: Nokia (AT&T ISP)
    • Secondary WAN: T-Mobile Hotspot (Connected via USB or Ethernet)
  • This ensures automatic failover if AT&T goes down.

2️⃣ Automatic Power Cycling for Recovery

(A) Smart Plugs for Remote Power Cycling

  • Use a smart plug (e.g., TP-Link Kasa, Tasmota, or Zigbee/Z-Wave) on:
    • Proxmox servers
    • Nokia router
    • ASUS router
  • Use a cloud-managed smart plug (like Kasa or Shelly) so you can remotely reboot devices if they become unresponsive.

(B) Raspberry Pi Watchdog for Network & Server Health

  • Set up a spare Raspberry Pi to:
    • Ping your key services (Proxmox, Pi-hole, VPN, Home Assistant).
    • If a service is down for X minutes, use a smart plug API to power cycle the Proxmox servers.
  • Example Python script for the Raspberry Pi: import os import requests import time # IPs of critical devices devices = { "Proxmox": "192.168.1.100", "Router": "192.168.1.1", "Home Assistant": "192.168.1.50" } # Smart plug API (TP-Link Kasa, Shelly, etc.) SMART_PLUG_URL = "http://smartplug-ip/power" def check_device(ip): return os.system(f"ping -c 3 {ip}") == 0 while True: for name, ip in devices.items(): if not check_device(ip): print(f"{name} is down! Restarting power...") requests.post(SMART_PLUG_URL, json={"state": "off"}) time.sleep(10) requests.post(SMART_PLUG_URL, json={"state": "on"}) time.sleep(300) # Check every 5 minutes

3️⃣ Prevent Data Overages on T-Mobile Hotspot

  • Limit backups & updates to run only on AT&T.
  • Block high-bandwidth services (like Jellyfin streaming) on the failover connection.

4️⃣ Test Before Leaving

  • Simulate a power outage: Unplug everything and check if it recovers automatically.
  • Simulate an ISP outage: Disconnect AT&T to ensure failover to T-Mobile works.

Final Setup Overview

✅ VPN Failover: PiVPN + DDNS + Tailscale/ZeroTier
✅ Internet Failover: ASUS RT-AC68U → Primary: AT&T, Secondary: T-Mobile
✅ Smart Power Cycling: Smart plugs controlled by a Raspberry Pi watchdog
✅ Data Protection: Limit traffic on T-Mobile failover

Would you like help setting up any of these components in more detail? 🚀

Let’s break this down into actionable steps. I’ll guide you through each part of the setup and you can let me know if you need more details or specific instructions.

1️⃣ Set Up VPN for Remote Access

Since you already have PiVPN, you should also add a backup access method to prevent getting locked out. Here’s how:

Option A: Enable Dynamic DNS (DDNS)

  • If AT&T changes your IP address, DDNS ensures you can always reach your home network.
  • Recommended services:
    • DuckDNS (Free)
    • No-IP (Free tier available)
    • Cloudflare DDNS (If you own a domain)

Steps to set up DuckDNS:

  1. Create an account at DuckDNS.
  2. Add a new domain (e.g., yourhome.duckdns.org).
  3. Set up a Raspberry Pi or Proxmox server to update your IP: sudo apt update && sudo apt install cron curl -y echo "*/5 * * * * root curl -s 'https://www.duckdns.org/update?domains=yourhome&token=YOURTOKEN&ip='" | sudo tee /etc/cron.d/duckdns sudo systemctl restart cron
  4. Configure PiVPN to use DuckDNS instead of an IP.
    Update /etc/wireguard/wg0.conf: Endpoint = yourhome.duckdns.org:51820

Option B: Backup VPN with Tailscale

  • Why? If AT&T goes down, Tailscale works even on your T-Mobile failover.
  • How? Install Tailscale on a Raspberry Pi and your remote devices: curl -fsSL https://tailscale.com/install.sh | sh sudo tailscale up
  • Use Tailscale as a secondary way to access your home servers.

2️⃣ Set Up WAN Failover (AT&T → T-Mobile)

Your ASUS RT-AC68U supports dual WAN failover.
Steps to configure it:

  1. Connect AT&T router to the ASUS RT-AC68U’s WAN port.
  2. Connect the T-Mobile hotspot to a LAN port on the ASUS router (or via USB if supported).
  3. Log into ASUS Router (192.168.1.1).
  4. Go to WAN → Dual WAN and enable Failover Mode.
  5. Set Primary WAN: AT&T
    Set Secondary WAN: T-Mobile hotspot
  6. Set Failover detection mode: “Network monitoring” (Ping 8.8.8.8).
  7. Save and test by unplugging AT&T.

✅ Now, if AT&T fails, T-Mobile takes over automatically.

3️⃣ Remote Power Cycling (Smart Plugs)

Use a cloud-based smart plug (like Kasa or Shelly) to remotely restart devices if needed.

Steps to Set Up TP-Link Kasa Smart Plug:

  1. Install Kasa app and connect the smart plug to your home WiFi.
  2. Plug in critical devices:
    • Proxmox servers
    • Nokia router
    • ASUS router
  3. Use the Kasa app to power cycle remotely if a device is unresponsive.

✅ Now you can manually reset devices remotely via the app.

4️⃣ Automatic Power Cycling with a Raspberry Pi

Use a Raspberry Pi to detect offline servers and power cycle them automatically.

Steps:

  1. Install Python on the Raspberry Pi: sudo apt update && sudo apt install python3-pip -y pip3 install requests
  2. Monitor devices using this Python script: import os import requests import time # Devices to monitor devices = { "Proxmox": "192.168.1.100", "Home Assistant": "192.168.1.50", "Router": "192.168.1.1" } # Smart plug API (TP-Link Kasa, Shelly, or Tasmota) SMART_PLUG_URL = "http://smartplug-ip/power" def check_device(ip): return os.system(f"ping -c 3 {ip}") == 0 while True: for name, ip in devices.items(): if not check_device(ip): print(f"{name} is down! Restarting power...") requests.post(SMART_PLUG_URL, json={"state": "off"}) time.sleep(10) requests.post(SMART_PLUG_URL, json={"state": "on"}) time.sleep(300) # Check every 5 minutes

✅ This automatically reboots Proxmox if it becomes unresponsive.

5️⃣ Limit T-Mobile Data Usage

T-Mobile has a 30GB cap before throttling, so block high-traffic services like Jellyfin.

Steps to limit data:

  1. Go to ASUS router settings (192.168.1.1).
  2. Go to Traffic Manager → Bandwidth Control.
  3. Set a bandwidth rule:
    • Allow only Home Assistant, PiVPN, and critical services on T-Mobile.
    • Block Jellyfin and non-essential traffic.

✅ This prevents unnecessary data usage on your failover connection.

6️⃣ Test Before Leaving

  1. Simulate power failure: Unplug everything and check if it restarts correctly.
  2. Simulate ISP failure: Disconnect AT&T and ensure the T-Mobile failover works.
  3. Test remote access: Ensure you can connect via PiVPN, Tailscale, and DDNS.

Final Setup Recap

✅ VPN Failover: PiVPN (AT&T) + Tailscale (T-Mobile) + DDNS
✅ Internet Failover: ASUS Router (AT&T → T-Mobile)
✅ Remote Power Cycling: Smart plugs + Raspberry Pi watchdog
✅ Data Protection: Block streaming services on T-Mobile
✅ Pre-Trip Testing: Power & ISP failure simulation

Do you need help with any of these steps? 🚀

How to Set Up Tailscale on Your Raspberry Pi for Remote Access

This guide will walk you through installing and configuring Tailscale on your Raspberry Pi to ensure you can securely access your home network remotely, even if your primary VPN fails.

📌 Step 1: Install Tailscale on Raspberry Pi

1️⃣ Update Your System

Before installing, update your system:

sudo apt update && sudo apt upgrade -y

2️⃣ Install Tailscale

Run the following command:

curl -fsSL https://tailscale.com/install.sh | sh

This script will automatically install the latest Tailscale package.

📌 Step 2: Start & Authenticate Tailscale

1️⃣ Start Tailscale

Enable and start the service:

sudo systemctl enable --now tailscaled

2️⃣ Authenticate Your Raspberry Pi

Run the command:

sudo tailscale up

It will provide a login link—open it in a web browser and sign in with Google, GitHub, or Microsoft.

Once authenticated, your Raspberry Pi is now part of your Tailscale network. 🎉

📌 Step 3: Access Your Home Network

  1. Get your Raspberry Pi’s Tailscale IP: tailscale ip -4 Example output: 100.99.88.77
  2. Connect from anywhere by using SSH: ssh pi@100.99.88.77 You can also access other devices on your network by enabling subnet routing.

📌 Step 4: Enable Subnet Routing (Access Entire Home Network)

By default, Tailscale only allows access to the Raspberry Pi. To access your entire home network, enable subnet routing:

1️⃣ Find Your Home Network Subnet

Run:

ip -4 addr show eth0

Example output:

inet 192.168.1.100/24 brd 192.168.1.255 scope global eth0

Here, 192.168.1.0/24 is your home network subnet.

2️⃣ Enable Subnet Routing

Run:

sudo tailscale up --advertise-routes=192.168.1.0/24

3️⃣ Approve the Route in Tailscale Admin Panel

  1. Go to https://login.tailscale.com/admin/machines
  2. Find your Raspberry Pi and approve the advertised route.

Now, you can access any device on your home network via Tailscale! 🔥

📌 Step 5: Ensure Tailscale Starts on Boot

Tailscale should already start automatically, but you can double-check with:

sudo systemctl enable tailscaled

📌 Step 6: (Optional) Use Tailscale Exit Node for Full Remote Access

If you want your Raspberry Pi to act as an exit node (allowing full internet access through your home network), enable it with:

sudo tailscale up --advertise-exit-node

Then approve it in the Tailscale Admin Panel.

🎯 Summary

✅ Tailscale is installed and running
✅ You can SSH into your Raspberry Pi from anywhere
✅ Your entire home network is accessible remotely
✅ Failover works even if your main ISP goes down

Would you like help testing the connection? 🚀

How to Test Your Tailscale Connection

Now that Tailscale is set up, let’s make sure everything is working properly. Follow these steps:

📌 Step 1: Check Tailscale Status

On your Raspberry Pi, run:

tailscale status

You should see output similar to this:

100.99.88.77   raspberrypi   online
100.101.102.50 laptop        online
100.102.103.20 phone         online

✅ If your Raspberry Pi appears as online, it’s connected to Tailscale.

📌 Step 2: Test SSH Access from Another Device

  1. Find your Raspberry Pi’s Tailscale IP: tailscale ip -4 Example output: 100.99.88.77
  2. From another device running Tailscale, try SSH: ssh pi@100.99.88.77
    • If prompted, enter your Raspberry Pi password.
    • ✅ If successful, your Tailscale setup is working!

📌 Step 3: Test Access to Other Home Network Devices (If Using Subnet Routing)

If you enabled subnet routing, try pinging a local device on your home network:

ping 192.168.1.100  # Replace with a local device's IP

✅ If you get a response, subnet routing is working.

📌 Step 4: Test Connection from Outside Your Home

  1. Turn off Wi-Fi on your phone.
  2. Use mobile data and open the Tailscale app.
  3. Try SSH into your Raspberry Pi: ssh pi@100.99.88.77 ✅ If successful, your Raspberry Pi is remotely accessible!

🚨 If Something Isn’t Working

  • Run this diagnostic command on the Pi: tailscale debug logs
  • Check firewall settings (make sure SSH is allowed).
  • Reboot the Pi and try again.

Let me know if you run into any issues! 🚀

Setting Up ddclient on a Raspberry Pi as a Backup for Namecheap DDNS

This guide will help you install ddclient on a Raspberry Pi, ensuring that your Namecheap domain (lynnhargrove.site) updates reliably, even if your Proxmox server goes down.

📌 Step 1: Install ddclient

1️⃣ Update your system:

sudo apt update && sudo apt upgrade -y

2️⃣ Install ddclient:

sudo apt install ddclient -y

3️⃣ Verify the installed version:

ddclient --version

You should see something like:

ddclient version 3.9.1

📌 Step 2: Configure ddclient for Namecheap

1️⃣ Open the ddclient configuration file:

sudo nano /etc/ddclient.conf

2️⃣ Replace the contents with:

# Namecheap Dynamic DNS Configuration
protocol=namecheap
use=web, web=dynamicdns.park-your-domain.com/getip
server=dynamicdns.park-your-domain.com
login=lynnhargrove.site
password='your_namecheap_ddns_password'
@  # Updates the root domain

🔹 Replace 'your_namecheap_ddns_password' with the Dynamic DNS password from Namecheap.

  • Find it under Advanced DNS settings in your Namecheap dashboard.

🔹 If you have a subdomain (e.g., home.lynnhargrove.site), change @ to home.

Save and exit: CTRL+X → Y → Enter

📌 Step 3: Enable ddclient as a Systemd Service

1️⃣ Make sure ddclient runs on boot:

sudo systemctl enable ddclient

2️⃣ Restart ddclient:

sudo systemctl restart ddclient

3️⃣ Check status:

sudo systemctl status ddclient

✅ If everything is working, you should see “active (running)”.

📌 Step 4: Verify That Your Domain Updates Correctly

1️⃣ Manually trigger an update:

sudo ddclient -force

2️⃣ Check your public IP:

curl ifconfig.me

3️⃣ Check if Namecheap updated your domain:

nslookup lynnhargrove.site

🔹 If the IP from nslookup matches ifconfig.me, your DDNS is working!

📌 Step 5: Set Up a Fallback (Optional)

If you want the Raspberry Pi only to update DDNS if your Proxmox server is offline, you can:

  • Add a cron job to check if Proxmox is reachable before running ddclient.
  • Use Tailscale or a watchdog script to automate failover.

🚀 You’re Done!

Now your Raspberry Pi will keep your lynnhargrove.site domain updated, even if your Proxmox server fails.

Let me know if you need help testing failover or setting up automation! 😊

I asked Gemini:

Can i have two wireless routers with the same ssid for automatic failover?

Yes, you can have two wireless routers with the same SSID and password for automatic failover, but it’s important to understand how this works and its limitations. The technique is commonly referred to as “roaming” or “seamless roaming.”

How It Works:

  • Same SSID and Password:
    • By configuring both routers with the exact same SSID (network name) and password, your wireless devices see them as a single network.
  • Device Decision:
    • When your device moves between the coverage areas of the two routers, it will decide which router to connect to based on signal strength and other factors.
    • Modern devices are generally quite good at this.
  • Failover:
    • If one router fails, your device should automatically connect to the other router, provided it’s within range.

Key Requirements and Considerations:

  • Same Security Settings:
    • Ensure that both routers have the same security settings (e.g., WPA2/WPA3).
  • Non-Overlapping Channels (Recommended):
    • To minimize interference, it’s best to configure the routers to use non-overlapping Wi-Fi channels.
      • For 2.4GHz, use channels 1, 6, or 11.
      • For 5GHz, choose channels that are sufficiently separated.
  • Same Network Configuration:
    • Both routers should be on the same subnet.
      • Ideally, one router should be the primary router, and the other should be configured as an access point.
      • If both routers are acting as routers, but on the same subnet, one router needs to have DHCP turned off, to avoid ip address conflicts.
  • Placement:
    • Strategically place the routers to provide overlapping coverage areas.
  • Limitations:
    • Not True Seamlessness: While your device will automatically switch, there might be a brief interruption in connectivity during the transition. This is because your device has to disconnect from one router and connect to the other.
    • Device Compatibility: Some older wireless devices might not handle roaming as smoothly as newer devices.1
    • Interference: If the routers are too close together or use overlapping channels, it can lead to interference and poor performance.2
    • Network Conflicts: If both devices are trying to act as routers, and both are running DHCP, network conflicts will occur.3

Best Practices for Reliable Failover:

  • Use Access Point Mode:
    • Configure one router as the primary router (handling DHCP and routing) and the other as an access point. This ensures that all devices get their IP addresses from the same source.
  • Use a Mesh Wi-Fi System:
    • Mesh Wi-Fi systems are designed for seamless roaming and provide a more reliable and user-friendly experience than manually configuring two separate routers.4
  • Wired Backhaul:
    • If possible, connect the routers using a wired Ethernet connection (wired backhaul). This improves performance and reliability.

In summary:

Setting up two routers with the same SSID and password can provide basic failover, but it’s essential to understand its limitations and follow best practices to minimize potential issues. Using a mesh system, or access point mode, is highly recommended for a smoother and more reliable experience.

I then asked:

I have a tmobile 5g router, an at&t fiber router, an asus rt-ac68u router not in use, and a pihole dhcp server. My wireless ssid is provided by the at&t router. I will be away for an extended period. Is there a way to configure an automatic failover from the at&t router to the 5g router?

Yes, you can configure an automatic failover from your AT&T fiber router to your T-Mobile 5G router, but it requires careful planning and a bit of technical work. Here’s a comprehensive approach:

Understanding the Challenge:

  • Different Networks: Your AT&T and T-Mobile routers are on separate networks with different public IP addresses and potentially different private IP ranges.
  • Automatic Switching: The primary challenge is to automatically switch your local network’s gateway and DNS settings when the AT&T connection fails.
  • Pi-hole as DHCP: Your Pi-hole acting as the DHCP server simplifies some things, but also adds a layer of complexity.

Configuration Steps:

  1. T-Mobile 5G Router Setup:
    • Configure LAN Settings:
      • Ensure the T-Mobile 5G router’s LAN IP range is different from your AT&T router’s. For example, if AT&T is 192.168.1.x, make the T-Mobile router 192.168.2.x.
      • Disable DHCP on the T-Mobile router. You’ll rely on Pi-hole for DHCP.
    • Wi-Fi Settings:
      • Set the T-Mobile router’s SSID to something different (e.g., “Bigbird-5G”) to avoid conflicts with your existing network.
      • Alternatively, you can have it broadcast the same SSID, but this adds complexity to the failover scripts.
    • Placement: Place the T-Mobile 5G router where it has a good 5G signal.
  2. Pi-hole Configuration:
    • Static IP: Ensure your Pi-hole has a static IP address on your AT&T network.
    • DHCP Range: Configure your Pi-hole’s DHCP range to match your AT&T router’s LAN subnet.
    • Gateway Setting:
      • This is the crucial part. You’ll need to create a script that modifies the Pi-hole’s DHCP gateway setting based on network availability.
  3. Failover Script (on Pi-hole):
    • Create a Script: Create a shell script on your Pi-hole that does the following:
      • Ping Test: Ping a reliable internet address (e.g., 8.8.8.8) to check if the AT&T connection is active.
      • Router Detection: If the ping fails, ping the AT&T router. If that fails too, assume the AT&T connection is down.
      • Modify Pi-hole DHCP:
        • If the AT&T connection is down, use pihole -a dhcp-gateway to change the Pi-hole’s DHCP gateway setting to the T-Mobile 5G router’s LAN IP address.
        • If the AT&T connection is up, change it back to the AT&T router’s LAN IP address.
      • Network Interface Check: Confirm the default route is using the correct interface.
    • Cron Job: Set up a cron job to run the script periodically (e.g., every minute).
  4. Asus RT-AC68U (Optional):
    • You could use the asus router as a wireless access point for the tmobile router. This would allow you to keep the same SSID.
    • This would require the asus router to be connected via ethernet to the tmobile router.
    • The asus router would be set to access point mode, and the wifi settings would be configured to use the same SSID and password.

Example Script (Simplified):

Bash

#!/bin/bash

ATT_ROUTER="192.168.1.1" # Replace with your AT&T router IP
TMOBILE_ROUTER="192.168.2.1" # Replace with your T-Mobile router IP

if ping -c 1 8.8.8.8 > /dev/null; then
  # AT&T connection is up
  if ping -c 1 $ATT_ROUTER > /dev/null; then
    pihole -a dhcp-gateway $ATT_ROUTER
  fi
else
  # AT&T connection is down
  pihole -a dhcp-gateway $TMOBILE_ROUTER
fi

Important Considerations:

  • IP Conflicts: Ensure your LAN subnets don’t overlap.
  • DNS: Your Pi-hole will continue to provide DNS, but if the T-Mobile connection has DNS issues, you might need to adjust Pi-hole’s upstream DNS servers.
  • Testing: Thoroughly test the failover script to ensure it works correctly.
  • Security: If you expose your network to the internet through the T-Mobile router, ensure you have appropriate security measures in place.
  • Network speed: Be aware that the tmobile 5g router may have significantly different speeds then the fiber connection.
  • Power: Ensure that both routers have uninterruptible power supplies.

By following these steps, you can create a basic automatic failover system for your network.

I asked ChatGPT the same question:

I have a tmobile 5g router, an at&t fiber router, an asus rt-ac68u router not in use, and a pihole dhcp server. My wireless ssid is provided by the at&t router. I will be away for an extended period. Is there a way to configure an automatic failover from the at&t router to the 5g router?

Yes, you can set up automatic failover using your ASUS RT-AC68U router as a dual-WAN router with WAN failover. Here’s how:

1️⃣ Connect Your Routers

  • Primary WAN (AT&T Fiber) → Plug into the WAN port of the ASUS RT-AC68U.
  • Secondary WAN (T-Mobile 5G Router) → Plug into a LAN port of the ASUS.

2️⃣ Set Up Dual WAN on ASUS RT-AC68U

  1. Log into the ASUS router (default: 192.168.1.1).
  2. Go to WAN → Dual WAN.
  3. Enable Dual WAN.
  4. Set:
    • Primary WAN = Ethernet WAN (AT&T Fiber).
    • Secondary WAN = Ethernet WAN (T-Mobile 5G Router) or USB (if using a 5G modem).
  5. Select “Failover Mode” (auto-switches when AT&T fails).
  6. Set Ping Check:
    • Target: 8.8.8.8 (Google DNS).
    • Failover after 3-5 timeouts.
    • Failback when AT&T returns.

3️⃣ Adjust Pi-hole DHCP Settings

  • If the ASUS RT-AC68U becomes your main router, update Pi-hole DHCP to point to it as the default gateway.
  • If you want AT&T to remain the DHCP source, ensure its gateway is static so ASUS can maintain failover routing.

4️⃣ Test the Failover

  • Unplug AT&T Fiber → The ASUS should switch to the T-Mobile router within a few minutes.
  • Reconnect AT&T → It should fail back automatically.

🚀 Now, Your Network Will Stay Online Even if AT&T Fails!

Would you like to add remote monitoring to check your network while you’re away? 📡

📡 Set Up Remote Monitoring for Your Failover Network

Now that failover is configured, let’s set up remote monitoring so you can check your network status while away.

1️⃣ Option 1: Use a Dynamic DNS (DDNS) Service

Since your public IP may change, use a DDNS service like No-IP or DynDNS to access your ASUS router remotely.

Set Up DDNS on ASUS Router

  1. Log into your ASUS router (192.168.1.1).
  2. Go to WAN → DDNS.
  3. Enable DDNS Client.
  4. Select a free DDNS service (ASUS provides free ones).
  5. Register a hostname (e.g., myhome.asuscomm.com).
  6. Save settings.

✅ Now, you can access your router remotely using myhome.asuscomm.com.

2️⃣ Option 2: Enable ASUS Remote Access

ASUS routers have a VPN or Remote Web Access feature:

  1. Go to Administration → System.
  2. Enable Web Access from WAN (set a strong password!).
  3. Or, enable VPN Server to connect securely.

✅ This lets you log into your router from anywhere.

3️⃣ Option 3: Use a Remote Monitoring Tool

Check Internet Status via UptimeRobot

  1. Go to UptimeRobot (free plan available).
  2. Set up a new monitor:
    • Type: Ping Monitor
    • IP: Your home DDNS hostname (myhome.asuscomm.com)
    • Check interval: Every 5 minutes
  3. Set up email/SMS alerts if your network goes down.

✅ You’ll get alerts when your internet fails and when it’s back.

4️⃣ Option 4: Use Pi-hole + Telegram Alerts

You can use your Pi-hole to detect internet outages and send an alert via Telegram or Pushbullet:

Set Up a Simple Internet Check Script

  1. Run: nano ~/internet_check.py
  2. Add this Python script: import os import requests import time # Telegram Bot Details TELEGRAM_BOT_TOKEN = "your_bot_token_here" TELEGRAM_CHAT_ID = "your_chat_id_here" # Function to check internet status def is_internet_up(): return os.system("ping -c 1 8.8.8.8") == 0 # Send Telegram Alert def send_telegram_alert(status): message = f"⚠️ Internet Down!" if not status else "✅ Internet Restored!" url = f"https://api.telegram.org/bot{TELEGRAM_BOT_TOKEN}/sendMessage" requests.post(url, data={"chat_id": TELEGRAM_CHAT_ID, "text": message}) # Check every 5 minutes prev_status = is_internet_up() while True: current_status = is_internet_up() if current_status != prev_status: send_telegram_alert(current_status) prev_status = current_status time.sleep(300) # Wait 5 minutes
  3. Save & exit (Ctrl+X, Y, Enter).

Run the Script on Boot

  1. Run: crontab -e
  2. Add: @reboot /usr/bin/python3 /home/pi/internet_check.py &

✅ Now, you’ll get a Telegram message when your home internet goes down!

🚀 Your Home Network is Now Monitored & Secured!

Would you like a remote reboot option if your internet goes down? 🖥️🔄

🔄 Set Up a Remote Reboot for Your Home Network

If your internet fails while you’re away, you can remotely reboot your ASUS router or Raspberry Pi to try restoring the connection.

1️⃣ Option 1: Reboot via Smart Plug (Simple & Reliable)

A Wi-Fi smart plug (like a Kasa TP-Link or Wyze plug) can power cycle your AT&T and T-Mobile routers if they lose internet.

How It Works:

  1. Plug your router into a smart plug.
  2. Enable remote access via the Kasa/Wyze app.
  3. If your network fails, toggle the smart plug OFF & ON from your phone.

✅ Easiest failover fix without needing complex scripts!

2️⃣ Option 2: Use a Raspberry Pi to Reboot the ASUS Router

If you have your Pi-hole running on a Raspberry Pi, you can make it reboot the ASUS router automatically when internet fails.

Step 1: Set Up a Reboot Script

On your Pi-hole Raspberry Pi, run:

nano ~/router_reboot.sh

Paste this:

#!/bin/bash
ROUTER_IP="192.168.1.1" # ASUS Router IP
ROUTER_USER="admin"      # Router Login Username
ROUTER_PASS="yourpassword"  # Router Password

# Check Internet Connection
ping -c 3 8.8.8.8 > /dev/null
if [ $? -ne 0 ]; then
    echo "Internet down! Rebooting router..."
    curl -u $ROUTER_USER:$ROUTER_PASS "http://$ROUTER_IP/reboot.asp"
fi

Save & exit (Ctrl+X, Y, Enter).

Step 2: Make the Script Executable

Run:

chmod +x ~/router_reboot.sh

Step 3: Run the Script Automatically Every 10 Minutes

Run:

crontab -e

Add this line:

*/10 * * * * /home/pi/router_reboot.sh

✅ Now, if internet fails, your Pi will automatically reboot the router.

3️⃣ Option 3: Remotely SSH Into Your Pi to Reboot the Router

If you set up DDNS or a VPN earlier, you can SSH into your Pi from anywhere and run:

curl -u admin:yourpassword "http://192.168.1.1/reboot.asp"

This manually forces the ASUS router to reboot.

🚀 Now You Have Remote Control Over Your Network!

Would you like to automate switching your Pi-hole DNS when failover happens? 🌐🔁

Migrate Windows 11 to NVME

Posted on by
Reply

Let’s walk through migrating your Windows 11 hard drive to a new NVMe drive! I’ll break it down step-by-step.

✅ What you’ll need:

  • Your new NVMe drive installed in your PC (or an external NVMe enclosure).
  • Cloning software (free options: Macrium Reflect, AOMEI Backupper, or EaseUS Todo Backup).
  • A backup of important data (just in case).

⚙️ Step 1: Install the NVMe drive

  • If your motherboard has an M.2 NVMe slot, install the drive.
  • If not, you can use an external NVMe enclosure connected via USB.

🧵 Step 2: Format and initialize the new drive (if needed)

  • Open Disk Management (Windows key → type diskmgmt.msc → Enter).
  • If prompted to initialize the disk, use GPT (for Windows 11, since MBR doesn’t support Secure Boot or TPM 2.0).
    • This process does NOT format the drive — it just sets it up for use.
  • Don’t create partitions — the cloning software will handle that.

🔄 Step 3: Clone your Windows 11 drive
1️⃣ Download Macrium Reflect Free (or another cloning tool).
2️⃣ Open the software and select your current Windows 11 hard drive as the source disk.
3️⃣ Select the new NVMe drive as the destination disk.
4️⃣ Make sure to copy all partitions (including EFI, Recovery, and System partitions).
5️⃣ Start the cloning process — this might take some time.

🖥️ Step 4: Set the NVMe drive as the boot drive

  1. After cloning, reboot and enter your BIOS/UEFI (usually by pressing Del or F2 during startup).
  2. Set the NVMe drive as the first boot device.
  3. Save and exit.

🧹 Step 5: Verify and clean up

  • Boot into Windows from the NVMe drive.
  • Check that everything works properly (apps, files, Windows activation).
  • If all is good, you can format your old drive to use as extra storage.

✨ Alternative: Fresh install option
If cloning isn’t your style, you can always do a fresh install of Windows 11 on the NVMe drive, then copy your data manually — sometimes this yields better performance and avoids issues with hidden partition mismatches.

Would you like to dive into BIOS/NVMe-specific settings for faster boot times? Let me know! 🚀

Optimizing BIOS/UEFI for your new NVMe boot drive can significantly improve performance and ensure a smooth boot process. Here’s how to do it!

⚙️ Step 1: Enter BIOS/UEFI

  • Restart your computer and repeatedly press Del, F2, or the key specific to your motherboard during boot.
  • Once inside BIOS, look for Boot, Storage, or Advanced settings (menu names may vary by motherboard).

🚀 Step 2: Optimize Boot Settings for NVMe

✅ Set NVMe as the Primary Boot Drive

  • Navigate to the Boot Order or Boot Priority section.
  • Set your NVMe drive as the first boot device.
  • Disable the old hard drive in the boot list (if necessary).

✅ Enable UEFI Mode

  • Windows 11 requires UEFI, not Legacy BIOS (CSM).
  • Find Boot Mode or CSM (Compatibility Support Module):
    • Set Boot Mode to UEFI (disable Legacy/CSM if you see it).
    • If you installed Windows 11 in GPT mode, UEFI must be enabled.

✅ Enable NVMe Support (if applicable)

  • Some older motherboards require NVMe support to be manually enabled in BIOS.
  • Look for NVMe Configuration or PCIe Storage Support.
  • Ensure it is set to Enabled or Auto.

⚡ Step 3: Improve NVMe Performance

✅ Enable PCIe 3.0 or 4.0 (if applicable)

  • Locate PCIe Configuration under Advanced settings.
  • Set PCIe mode to the highest supported version (e.g., Gen 3 or Gen 4).

✅ Enable AHCI for SSDs (instead of RAID)

  • If your BIOS has SATA Mode, set it to AHCI instead of RAID/IDE.
  • AHCI ensures better performance for SSDs.

✅ Enable Fast Boot (Optional)

  • Some BIOS versions offer Fast Boot, which reduces boot times.
  • Enable it, but disable it temporarily if you need to access BIOS frequently.

🔄 Step 4: Save and Reboot

  • Press F10 and select Save & Exit.
  • Your system should now boot quickly from the NVMe drive.

🔥 Extra Performance Tweaks in Windows

Once booted into Windows 11: 1️⃣ Enable Write Caching

  • Open Device Manager → Expand Disk Drives → Right-click your NVMe drive → Properties → Policies.
  • Enable Write Caching for better speed.

2️⃣ Enable TRIM for SSD Longevity

  • Open Command Prompt (Admin) and run: fsutil behavior query DisableDeleteNotify
  • If it returns 0, TRIM is enabled. If 1, enable it with: fsutil behavior set DisableDeleteNotify 0

3️⃣ Disable Hibernation (Optional, if space is a concern)

  • Run: powercfg -h off in Command Prompt (Admin).
  • Saves several GB of space on the NVMe drive.

🎯 Final Check

✅ Boot speed improved?
✅ Windows 11 running smoothly?
✅ No old drives interfering?

You’re all set! 🚀 Let me know if you need deeper tuning! 😃

It sounds like your Windows 11 boot process is tied to the M.2 SATA drive, even though your Kingston SA400 SSD is your main C: drive. Here’s how to troubleshoot and fix it:

🔍 Step 1: Check Boot Order in BIOS

  1. Enter BIOS/UEFI
    • Restart your computer and press F2, Del, or Esc (depends on your motherboard).
  2. Go to Boot Priority/Boot Order
    • Make sure your Kingston SA400 SSD is set as the first boot device.
    • If your M.2 SATA drive is missing, the system might default to BIOS because it cannot find a valid boot drive.

🔎 Step 2: Check If Bootloader Is on the M.2 SATA Drive

Your bootloader (Windows Boot Manager) may have been installed on your M.2 SATA drive instead of your Kingston SSD.
To check:

  1. Plug the M.2 SATA drive back in and boot into Windows.
  2. Open Disk Management
    • Press Win + X → Click Disk Management.
  3. Identify the “System Reserved” or EFI Partition
    • Look at both drives.
    • If the EFI System Partition (ESP, ~100-500MB, FAT32) is on the M.2 SATA drive, Windows relies on it to boot.

🛠 Step 3: Repair Bootloader on Kingston SSD

If your bootloader is missing from your Kingston SSD, you’ll need to recreate it.

Option 1: Automatic Repair

  1. Boot from a Windows 11 USB Installer.
  2. Select Repair your computer → Troubleshoot → Startup Repair.
  3. Follow the on-screen instructions.

Option 2: Manually Rebuild Bootloader

If automatic repair doesn’t work:

  1. Boot from a Windows 11 USB Installer.
  2. Select Repair your computer → Troubleshoot → Command Prompt.
  3. Run these commands: diskpart list disk select disk X (Replace X with your Kingston SSD) list partition select partition Y (Replace Y with the EFI partition, if it exists) assign letter=Z exit
  4. Now rebuild the bootloader: bcdboot C:\Windows /s Z: /f UEFI
  5. Restart the computer and remove the M.2 SATA drive.

🔄 Step 4: Try Booting Without M.2 SATA

  1. Shut down and physically remove the M.2 SATA drive.
  2. Boot into BIOS and confirm Kingston SSD is the primary boot device.
  3. Save changes and restart.

✅ Expected Result

  • If successful, your system should boot directly from Kingston SSD without the M.2 drive.
  • If it still boots to BIOS, check if your Kingston SSD is recognized in BIOS.

Let me know if this works or if you need more troubleshooting! 🚀

Data Backup on a Budget

Posted on by
Reply

I have a lot of data that I’d hate to lose. Anyone with a hard drive failure really appreciates the value of having another copy of the lost data. I suppose cloud storage would be a logical place to backup data. With all my personal audio and video files, I might need as much as 10 TB of storage. That would be too expensive for me. My strategy has been to buy enough USB external hard drives to keep two copies of all my data. This has always been a manual process. I am working to automate my backup processes.

There are a couple of options when syncing two file directories. One is to simply copy all new files from the source to the target. Another is to additionally remove all the files from the target that have been removed from the source. I’ll automate the first procedure. If you accidentally remove all files from the source, the second procedure will also remove all files from the target. This option will only be used manually and very carefully.

Personal Documents

stub…

Audio, Video and Pictures

stub…

Duplicate Files

stub…

Server Backups

stub…

Rsync (ChatGPT)

I have three external USB drives: 4 TB, 5 TB, and 12 TB

You can use the following rsync command to synchronize the /documents directory from fileserver1 to fileserver2, ensuring that extra files on fileserver2 are removed:

rsync -av --delete fileserver1:/documents/ fileserver2:/documents/

Explanation of Options:

  • -a → Archive mode (preserves permissions, timestamps, symbolic links, etc.).
  • -v → Verbose output (optional, shows progress).
  • --delete → Removes extra files on fileserver2 that are not present on fileserver1.
  • The trailing / after /documents/ ensures correct directory syncing.

Additional Options (If Needed):

  • Dry run before actual execution (to preview changes): rsync -av --delete --dry-run fileserver1:/documents/ fileserver2:/documents/
  • Use SSH for secure transfer (if servers are remote): rsync -av --delete -e ssh fileserver1:/documents/ fileserver2:/documents/
  • Limit bandwidth to avoid overloading network (e.g., limit to 5MB/s): rsync -av --delete --bwlimit=5000 fileserver1:/documents/ fileserver2:/documents/

You can automate the rsync synchronization using a cron job.

Step 1: Edit the Crontab

On fileserver1, open the crontab editor:

crontab -e

Step 2: Add the Rsync Command

Add the following line to schedule synchronization every day at 2 AM:

0 2 * * * rsync -av --delete fileserver1:/documents/ fileserver2:/documents/ >> /var/log/rsync.log 2>&1

This will:

  • Run rsync at 2:00 AM daily.
  • Log output to /var/log/rsync.log for debugging.

Step 3: Save and Exit

Save the file and exit the editor.

Step 4: Verify the Cron Job

After a day, check if the job ran:

cat /var/log/rsync.log

Would you like email notifications for failures?

Robocopy

stub…

VPN Redux

Posted on by
Reply

Goodbye CGNAT! I limped along for over a year with T-Mobile 5G Home Internet for a miserly $30 per month. However my downloads dropped from double-digit mbps to single-digits and I cried uncle. I’m in a T-Mobile fringe area for LTE/5G signal and probably in an area with insufficient capacity for Home Internet. I’ve stepped up to AT&T fiber. As important to me, the change from Commercial Grade NAT to standard NAT means I’ll be able use Dynamic DNS to use my personal domain name to point to my home network. For the uninitiated, your CGNAT IP address changes so frequently that DDNS is useless. This means I can now set up a VPN tunnel into my network for remote control and file sharing. Woohoo!

To use DDNS with NameCheap, go to domain management, Dynamic DNS and create an A + Dynamiic DNS Record with host @ and your IP address, then save. This NameCheap section has the secret password needed for the DDNS client like ddclient for linux.

WireGuard seems to be the go-to VPN server and I installed it on a Proxmox VM using PiVPN. I’m convinced this is the simplest way to setup and manage a VPN. I found excellent WireGuard links on Mike Royal’s github page. Even better is a video from Crosstalk Solutions which adds how to add access to your local network. Next I created a port forwarding entry on my router pointing my WireGuard port to my WireGuard server. Next up is to install the app on my notebook computer to see if it works.

Update 12/30/24

I installed ddclient on my file server and it keeps the ip address lynnhargrove.site up-to-date. I also had a “duh” moment regarding IP reservations on my home network. I ran out of IP reservations on the AT&T router and struggled with servers changing addresses. Pi-hole has a very nice DCHP server that I switched to. I had to edit the dhcp.conf to add a secondary DNS server in case Pi-hole is down. Unfortunately, it reverts to the original dhcp.conf when changes are made to Pi-hole. I originally installed Pi-hole for quicker DNS resolution, but I hadn’t use it to blacklist any websites. I used Steven Black’s unified list of porn, gambling, fake news and social hosts for my blacklist.