Protecting the Pi: Restricting SSH to the USB gadget

Why?

ssh is a remote login that is pretty much the only way to work on a headless Raspberry Pi without an attached keyboard/monitor or serial console cable. Remote login capability is a security risk even with something like SSH.  There are some good articles that describe how to lock down SSH via password change or other security measures.

I decided to limit SSH to specific network interfaces. Raspberry PIs can have several network interfaces. The only hardwire network interfaces Raspberry Pi Zero when it is in Network Gadget mode. I only truly trust the USB private LAN (usb0) since it requires a direct connection and cannot be directly seen by any other device.

Interfaces

Our Raspberry PI can have several different Network Interfaces.  All of them are candidates for allowing or denying SSH or other inbound access.

Linux lets us manage ssh by letting us configure port related ALLOW / DENY on all network interfaces. We can create inbound iptables firewall rules.  The script below blocks ssh wireless traffic and leave SSH enabled for wired connections. It is based on iptables / ip6tables. Some folks may wish to disable SSH over specific hardwire ethernet connections. These are some of the standard Raspberry Pi network devices.
  1. usb0: Raspberry Pi Zero running in gadget mode with a hard wire (USB) connection to a PC.
  2. eth0:  Raspberry Pi 3 and 3+ hard-wire Ethernet jacks.
  3. wlan0: Raspberry Pi Zero W, Raspberry Pi 3, 3+ with wi-fi enabled or Raspberry Pi (2, Zero...) with USB network adapter and no built-in wi-fi networking.
  4. wlan1: Raspberry Pi Zero W, Pi 3 with built-in wi-fi networking and USB wi-fi adapter

Finding the Right Interface

A Raspberry Pi running the network gadget has more than one active network interface.  They also have two network addresses per interface, one IPV4 and one IPV6. The Pi broadcasts its' DNS information for all interfaces so it can be picked up by mDNS / Bonjour.  A Pi with two network interfaces will have 4 addresses, two IP4 and two IP6. You normally don't care about this because the target service is enabled across all devices. 

Our ssh driven change removes ssh access from some of the devices. Now we have forced ssh to the interface/ip that is still enabled. You must use dns-sd to explicitly find the set of IPs for all active network devices on the Raspberry Pi.  
  • dns-sd -G v4v6 <hostname>.local
The following dns-sd command shows two network interfaces. I know I want to use the 169.x.x.x address because it represents the private connection between the PC and Raspberry PI over the USB gadget link.  In this case, we would ssh into the 169.254.107.129 IP address.

C:\Users\joe>dns-sd -G v4v6 pi-520863f1.local
Timestamp     A/R Flags if Hostname                  Address                                      TTL
21:57:00.586  Add     3 20 pi-520863f1.local.        FE80:0000:0000:0000:1B31:2156:F706:AFB8%ethernet_32785 120
21:57:00.590  Add     2 20 pi-520863f1.local.        169.254.107.129                              120
21:57:00.674  Add     3 13 pi-520863f1.local.        FE80:0000:0000:0000:1BE6:83EB:185C:D72F%ethernet_32777 120
21:57:00.676  Add     2 13 pi-520863f1.local.        192.168.1.3

Connection delays on the private LAN

I have found that ssh can take a while the first time. It feels like the machine first tries to use the main interface.  Then there is some kind of network timeout after which ssh quickly connects. Normally I just ping <hostname.local> until the ping is successful. 

Better living through scripting

The following script enables ssh and then blocks inbound ssh on wireless connections wlan0 and wlan1.  It can also block ssh on eth0 for those that only allow ssh over the gadget interface. The most recent source code can be found on GitHub . 

#!/bin/bash
#!/bin/bash
# is rerunable
#
# Portions styled from 
# ref: https://kerneltalks.com/virtualization/how-to-reset-iptables-to-default-settings/
# ref: https://makezine.com/2017/09/07/secure-your-raspberry-pi-against-attackers/
# ref: https://www.raspberrypi.org/documentation/remote-access/ssh/
# ref: https://serverfault.com/questions/475717/iptables-block-incoming-on-eth1-and-allow-all-from-eth0
# ref: https://askubuntu.com/a/30157/8698

# eth0 hardware, 
# wlan0 PZero, P3, P3+ builtin or P2 USB Wi-fi
# wlan1 P3, P3+ USB Wi-Fi
disable_eth0=false
disable_wlan0=true
disable_wlan1=true

if ! [ $(id -u) = 0 ]; then
   echo "The script need to be run as root." >&2
   exit 1
fi

if [ $SUDO_USER ]; then
    real_user=$SUDO_USER
else
    real_user=$(whoami)
fi

# Commands that you don't want run with root would be invoked
# with: sudo -u $real_user
# So they will be ran as the user who invoked the sudo command
# Keep in mind, if the user is using a root shell (they're logged in as root),
# then $real_user is actually root
# sudo -u $real_user non-root-command
# Commands that need to be ran with root would be invoked without sudo

# log the current rules
echo "----------------------------------------------"
echo inbound rules before more reset and ssh block
iptables --list --verbose
echo
ip6tables --list --verbose
echo "----------------------------------------------"
# reset the inbound rules
iptables -F INPUT
ip6tables -F INPUT
echo "Reset Input Rules"

# enable iptables inbound rule blocking SSH on wlan0/eth0
if [ "$disable_wlan0" = true ]; then
    if [ -e /sys/class/net/wlan0 ]; then
        echo "wlan0: Blocking port 22"
        iptables -A INPUT -p tcp --dport 22 -i wlan0 -j DROP
        ip6tables -A INPUT -p tcp --dport 22 -i wlan0 -j DROP
    fi
else
    echo "wlan0: No block requested"
fi
if [ "$disable_wlan1" = true ]; then
    if [ -e /sys/class/net/wlan1 ]; then
        echo "wlan1: Blocking port 22"
        iptables -A INPUT -p tcp --dport 22 -i wlan1 -j DROP
        ip6tables -A INPUT -p tcp --dport 22 -i wlan1 -j DROP
    fi
else
    echo "wlan1: No block requested"
fi
if [ "$disable_eth0" = true ]; then
    if [ -e /sys/class/net/eth0 ]; then
        echo "eth0: Blocking port 22"
        iptables -A INPUT -p tcp --dport 22 -i eth0 -j DROP
        ip6tables -A INPUT -p tcp --dport 22 -i eth0 -j DROP
    fi
else 
    echo "eth0: No block requested"
fi

# log the current rules
echo "----------------------------------------------"
echo inbound rules after blocking ssh on  wlan0/eth0
iptables --list --verbose
echo
ip6tables --list --verbose
echo "----------------------------------------------"

# install if not already there
PKG_OK=$(dpkg-query -W --showformat='${Status}\n' iptables-persistent | grep "install ok installed")
#echo Determined status for iptables-persistent: '$PKG_OK'
if [ -z "$PKG_OK" ]; then
  echo "No iptables-persistent. Installing iptables-persistent."
  apt-get --yes install iptables-persistent
else
  # persist the changes.
  echo "iptables-persistent present so just reconfigure."
  iptables-save >/etc/iptables/rules.v4
  ip6tables-save >/etc/iptables/rules.v6
fi

if [ "`systemctl is-active ssh`" != "active" ] 
then
    echo "starting ssh"
    systemctl enable ssh
    systemctl start ssh
else
    echo "ssh already started"
fi

Related

  • https://joe.blog.freemansoft.com/2018/03/setting-up-raspberry-pi-zero-without.html
  • https://joe.blog.freemansoft.com/2018/03/protecting-pi-restricting-ssh-to-usb.html
  • https://joe.blog.freemansoft.com/2022/11/installing-rndis-driver-on-windows-11.htm

Comments

Post a Comment

Popular posts from this blog

Installing the RNDIS driver on Windows 11 to use USB Raspberry Pi as network attached

Image
I do a lot of my development and configuration via ssh into my Raspberry Pi Zero over the RNDIS connection. Some models of the Raspberry PIs can be configured with  gadget drivers  that let the Raspberry pi emulate different devices when plugged into computers via USB. My favorite gadget  is the network profile that makes a Raspberry Pi look like an RNDIS-attached network device.  All types of network services travel over an RNDIS device without knowing it is a USB hardware connection. A Raspberry Pi shows up as a Remote NDIS (RNDIS) device when you plug the Pi into a PC or Mac via a USB cable. The gadget  in the Windows Device Manager picture shows this RNDIS Gadget connectivity between a Windows machine and a Raspberry Pi. The Problem Windows 11 and Windows 10 no longer auto-installs the RNDIS driver that makes magic happen. Windows recognizes that the Raspberry Pi is some type of generic USB COM device. Manually running W indows Update  or  Upd...
Read more

Understanding your WSL2 RAM and swap - Changing the default 50%-25%

Image
The Windows Subsystem for Linux operates as a virtual machine that can dynamically grow the amount of RAM to a maximum set at startup time.  Microsoft sets a default maximum RAM available to 50% of the physical memory and a swap-space that is 1/4 of the maximum WSL RAM.  You can scale those numbers up or down to allocate more or less RAM to the Linux instance. The first drawing shows the default WSL memory and swap space sizing. The images below show a developer machine that is running a dev environment in WSL2 and Docker Desktop.  Docker Desktop has two of its own WSL modules that need to be accounted for.  You can see that the memory would actually be oversubscribed, 3 x 50% if every VM used its maximum memory.  The actual amount of memory used is significantly smaller allowing every piece to fit. Click to Enlarge The second drawing shows the memory allocation on my 64GB laptop. WSL Linux defaul...
Read more

Get AI code assist VSCode with local LLMs using LM Studio and the Continue.dev extension - Windows

Image
This is about running VSCode AI code assist locally replacing Copilot or some other service.  You may run local models to guarantee none of your code ends up on external servers. Or, you may not want to maintain an ongoing AI subscription. We are going to use  LM  Studio and VS Code.  This was tested on Windows 11 with an RTX 3060 TI with 8GB of VRAM. 8GB really limits the number and size of the models we can use. LM Studio's simple hosting model of 1 LLM and an embedding works for us in this situation.  You want a big card. 8GB is a tiny card. Related blog articles and videos Several related blogs and videos that cover VSCode and local LLMs Blog  Get AI code assist VSCode with local LLMs using Ollama and the Continue.dev extension - Mac Get AI code assist VSCode with local LLMs using LM Studio and the Continue.dev extension - Windows Rocking an older Titan RTX 24GB as my local AI Code assist on Windows 11, Ollama and VS Code YouTube Video  Using loc...
Read more