How-to: Ultra-secure VNC to computer on home network via Synology NAS using SSH tunnel


Copyright Jösé
Copyright Jösé

As you may know from other articles here, I have a Synology DS214Play NAS, and I’m a big fan. It can do so much for so little money. Well, today I’m going to show you a new trick – it will work for most Synology models.

There are a few different ways of remotely connecting to and controlling computers on your home network. LogMeIn used to be a big favourite, until they discontinued the free version. TeamViewer is really the next best thing, but I find it pretty slow and erratic in operation. It’s also not free for commercial use, whereas the system I describe here is completely free.

Many people extol the virtues of VNC, but it does have a big drawback in terms of security, with various parts of the communication being transmitted unencrypted over the network. That’s obviously a bit of a no-no.

The solution is to set up a secure SSH tunnel first. Don’t worry if you don’t know what that means. Just think about this metaphor: imagine you had your own private tunnel, from your home to your office, with locked gates at either end. There are no other exits from this tunnel. So no one can peek into it to see what traffic (cars) is coming and going. An SSH tunnel is quite like that. You pass your VNC “traffic” (data) through this tunnel and it is then inaccessible to any prying eyes.


This guide assumes the following things:

  1. You have a Synology NAS, with users and file storage already configured.
  2. You have at home a Windows computer that is left switched on and connected to your home network while you’re off-site.
  3. Your home PC has a static IP address (or a DHCP reservation).
  4. You have some means of knowing your home’s IP address. In my case, my ISP has given me a static IP address, but you can use something like if you’re on a dynamic address. (Full instructions are available at that link.)
  5. You can redirect ports on your home router and ideally add firewall rules.
  6. You are able to use vi/vim. Sorry, but that knowledge is really beyond the scope of this tutorial, and you do need to use vi to edit config files on your NAS.
  7. You have a public/private key pair. If you’re not sure what that means, read this.

Install VNC

There are a few different implementations of VNC. I prefer TightVNC for various reasons – in particular it has a built-in file transfer module.

When installing TightVNC on your home PC, make sure you enable at least the “TightVNC Server” option.


Check all the boxes in the “Select Additional Tasks” window.


You will be prompted to create “Remote Access” and “Administrative” passwords. You should definitely set the remote access password, otherwise anyone with access to your home network (e.g. someone who might have cracked your wireless password) could easily gain access to your PC.


At work, you’ll just need to install the viewer component.

Configure Synology SSH server

Within Synology DiskStation Manager, enable the SSH server. In DSM 5, this option is found at Control Panel > System > Terminal & SNMP > Terminal.


I urge you not to enable Telnet, unless you really understand the risks.

Next, login to your NAS as root, using SSH. Normally I would use PuTTY for this purpose.

You’ll be creating an SSH tunnel using your normal day-to-day Synology user. (You don’t normally connect using admin do you? Do you?!) Use vi to edit /etc/passwd. Find the line with your user name and change /sbin/nologin to /bin/sh. E.g.:


Save the changes.

As part of this process, we are going to make it impossible for root to log in. This is a security advantage. Instead if you need root permissions, you’ll log in as an ordinary user and use “su” to escalate privileges. su doesn’t work by default. You need to add setuid to the busybox binary. If you don’t know what that means, don’t worry. If you do know what this means and it causes you concern, let me just say that according to my tests, busybox has been built in a way that does not allow users to bypass security via the setuid flag. So, run this command:

chmod 4755 /bin/busybox

Please don’t proceed until you’ve done this bit, otherwise you can lock root out of your NAS.

Next, we need to edit the configuration of SSH. We have to uncomment some lines (that normally being with #) and change the default values. So use vi to edit /etc/ssh/sshd_config. The values you need to change should end up looking like this, with no # at the start of the lines:

AllowTcpForwarding yes
LoginGraceTime 5m
MaxAuthTries 6
PubkeyAuthentication yes
AuthorizedKeysFile .ssh/authorized_keys

In brief these changes do the following, line by line:

  1. Allow using SSH to go from the SSH server (the NAS box) to another machine (e.g. your home PC)
  2. If you foul up your login password loads of times, restrict further attempts for 5 minutes.
  3. Give you 6 attempts before forcing you to wait to retry your logon.
  4. Allow authentication using a public/private key pair (which can enable password-less logons).
  5. Point the SSH daemon to the location of the list of authorized keys – this is relative to an individual user’s home directory.

Having saved these changes, you can force SSH to load the new configuration by uttering the following somewhat convoluted and slightly OCD incantation (OCD, because I hate leaving empty nohup.out files all over the place). We use nohup, because nine times out of ten this stops your SSH connection from dropping while the SSH daemon reloads:

nohup kill -HUP `cat /var/run/`; rm nohup.out &

SSH keys

You need to have a public/private SSH key pair. I’ve written about creating these elsewhere. Make sure you keep your private key safely protected. This is particularly important if, like me, you use an empty key phrase, enabling you to log on without a password.

In your home directory on the Synology server, create (if it doesn’t already exist) a directory, “.ssh”. You may need to enable the display of hidden files/folders, if you’re doing this from Windows.

Within the .ssh directory, create a new file “authorized_keys”, and paste in your public key. The file will then normally have contents similar to this:

ssh-rsa AAAAB3NzaC1yc2EAAAABJQAAAQEArLX5FlwhHJankhIoZcsIEgmHkOtsSR6eJINGgb4N3/4XQAHpmYPhlAy6Hg2hH8VqNLXgkVia+yMDaDOFQKXX6Ue+hOQt7Q5zB3W1NgVCsyIn9JBu3u6R8rDPBma248DhQ3yfac1iEZWa+3BrHaIM2dLVGu99C5z3Kh1NhDB83xetq08bHayzv39wuwZUZOohDzsCK29ZaEYU9ZctN/RZR4rW7A7odJbbgqG82IZXhUhiam2utpjszLJ+sMOw6z7tcpgnF5CLDys2xvE6ekLjEPA2b9KkrU6e+ILXM85s7/HP9aTkTwFyyBcPAvmO7i0xYyotu58DKf++nM2ZtpNBPQ== Rob's SSH key

This is all on a single line. For RSA keys, the line must begin ssh-rsa.

SSH is very fussy about file permissions. Log in to the NAS as root and then su to your normal user (e.g. su rob). Make sure permissions for these files are set correctly with the following commands:

chmod 755 ~
chmod 700 ~/.ssh
chmod 600 ~/.ssh/authorized_keys

If you encounter any errors at this point, make sure you fix them before proceeding. Now test your SSH login. If it works and you can also su to root, you can now safely set the final two settings in sshd_config:

PermitRootLogin no
PasswordAuthentication no

The effect of these:

  • Disallow direct logging in by root.
  • Disallow ordinary password-based logging in.

Reload SSH with nohup kill -HUP `cat /var/run/`; rm nohup.out & as before.

Setting up your router

There are so many routers out there that I can’t really help you out with this one. You will need to port forward a port number of your choosing to port 22 on your Synology NAS. If you’re not sure where to start, Port Forward is probably the most helpful resource on the Internet.

I used a high-numbered port on the outer edge of my router. I.e. I forwarded port 53268 to port 22 (SSH). This is only very mild protection, but it does reduce the number of script kiddie attacks. To put that in context, while I was testing this process I just forwarded the normal port 22 to port 22. Within 2 minutes, my NAS was emailing me about failed login attempts. Using a high-numbered port makes most of this noise go away.

To go one better however, I also used my router’s firewall to prevent unknown IP addresses from connecting to SSH. Since I’m only ever doing this from work, I can safely limit this to the IP range of my work’s leased line. This means it’s highly unlikely anyone will ever be able to brute force their way into my SSH connection, if I ever carelessly leave password-based logins enabled.

Create a PuTTY configuration

I recommend creating a PuTTY configuration using PuTTY’s interface. This is the easiest way of setting all the options that will later be used by plink in my batch script. plink is the stripped down command-line interface for PuTTY.

Within this configuration, you need to set:

  1. Connection type: SSH
  2. Hostname and port: your home external IP address (or DNS name) and the port you’ve forwarded through your router, preferably a non-standard port number.
  3. Connection > Data > Auto-login username: Put your Synology user name here.
  4. Connection > SSH > Auth > Private key file for identification: here put the path to the location of your private key on your work machine, from where you’ll be initiating the connection back to home.
  5. Connection > SSH > Tunnels: This bears some explanation. When you run VNC viewer on your work machine, you’ll connect to a port on your work machine. PuTTY forwards this then through the SSH tunnel. So here you need to choose a random “source port” (not the normal VNC port, if you’re also running VNC server on your work machine). This is the port that’s opened on your work machine. Then in the destination, put the LAN address of your home PC and add the normal VNC port. Like this:
    Make sure you click Add.
  6. Finally, go back to Session, type a name in the “Saved Session” field and click “Save”. You will then be able to use this configuration with plink for a fully-automatic login and creation of the SSH tunnel.

Now would be a good time to fire up this connection and check that you can login okay, without any password prompts or errors.

Using username "rob".
Authenticating with public key "Rob's SSH key"

BusyBox v1.16.1 (2014-05-29 11:29:12 CST) built-in shell (ash)
Enter ‘help’ for a list of built-in commands.


Making VNC connection

I would suggest keeping your PuTTY session open while you’re setting up and testing your VNC connection through the tunnel. This is really the easy bit, but there a couple of heffalump pits, which I’ll now warn you about. So, assuming your VNC server is running on your home PC and your SSH tunnel is up, let’s now configure the VNC viewer at the work end. Those heffalump pits:

  1. When you’re entering the “Remote Host”, you need to specify “localhost” or “”. You’re connecting to the port on your work machine. Don’t enter your work machine’s LAN ip address – PuTTY is not listening on the LAN interface, just on the local loopback interface.
  2. You need to specify the random port you chose when configuring tunnel forwarding (5990 in my case) and you need to separate that from “localhost” with double colons. A single colon is used for a different purpose, so don’t get tripped up by this subtle semantic difference.


If you have a working VNC session at this point, congratulations! That’s the hard work out of the way.

It would be nice to automate the whole connection process. While you have your VNC session established, it is worth saving a VNC configuration file, so you can use this in a batch script. Click the VNC logo in the top left of the VNC session, then “Save session to a .vnc file”. You have the option to store the VNC password in this file, which I’ve chosen to do.


Before saving the session, you might want to tweak some optimization settings. This will really vary depending on your preferences and the speed of your connection. On this subject, this page is worth a read. I found I had the best results when using Tight encoding, with best compression (9), JPEG quality 4 and Allow CopyRect.

One batch script to rule them all

To automate the entire process, bringing up the tunnel and connecting via VNC, you might like to amend the following batch script to fit your own environment:

@echo off
start /min "SSH tunnel home" "C:\Program Files (x86)\PuTTY\plink.exe" -N -batch -load "Home with VNC tunnel"
REM Use ping to pause for 2 seconds while connection establishes
ping -n 2 localhost > NUL
"C:\Batch scripts\HomePC.vnc"

I suggest creating a shortcut to this batch file in your Start menu and setting its properties such that it starts minimised. While your SSH tunnel is up, you will have a PuTTY icon on your task bar. Don’t forget to close this after you close VNC, to terminate the tunnel link. An alternative approach is to use the free tool MyEnTunnel to ensure your SSH tunnel is always running in the background if that’s what you want. I’ll leave that up to you.

DSM Upgrades

After a DSM upgrade, you may find that your SSH config resets and you can no longer use VNC remotely. In that eventuality, log into your NAS via your LAN (as admin) and change the config back as above. You’ll also probably need to chmod busybox again.

root locked out of SSH?

For the first time in my experience, during May 2015, a DSM upgrade reset the suid bit on Busybox (meaning no more su), but didn’t reset the PermitRootLogin setting. That meant that root could not log in via SSH. Nor could you change to root (using su). If you find yourself in this position, follow these remedial steps:

  1. Go to Control Panel > Terminal & SNMP
  2. Check the “Enable Telnet service” box.
  3. Click “Apply”.
  4. Log in as root, using Telnet. You can either use PuTTY (selecting Telnet/port 23 instead of SSH/port 22) or a built-in Telnet client.
  5. At the prompt, enter chmod 4755 /bin/busybox.
  6. Go to Control Panel > Terminal & SNMP
  7. Uncheck the “Enable Telnet service” box.
  8. Click “Apply”.

Do make sure you complete the whole process; leaving Telnet enabled is a security risk, partly because passwords are sent in plain text, which is a Very Bad Thing.


So, what do you think? Better than LogMeIn/TeamViewer? Personally I prefer it, because I’m no longer tied to a third party service. There are obvious drawbacks (it’s harder to set up for a start! and if you firewall your incoming SSH connection, you can’t use this from absolutely anywhere on the Internet) but I like it for its benefits, including what I consider to be superior security.

Anyway, I hope you find this useful. Until next time.

Subway Tunnel image copyright © Jösé, licensed under Creative Commons. Used with permission.

How-to: Laravel 4 tutorial; part 6 – virtualised development environment – Laravel Homestead

[easyreview title=”Complexity rating” icon=”geek” cat1title=”Level of experience required, to follow this how-to.” cat1detail=”We’re pulling together a few sophisticated components here, but keep your eye on the ball and you’ll be okay.” cat1rating=”4″ overall=”false”]

Laravel Tutorials


It has been a while since I have had time to work on Laravel development or indeed write a tutorial. And since then, I have decommissioned my main web development server in favour of a Synology NAS. Dummy and I use a third party hosting service for hosting our clients’ web sites and our own. This shared hosting service comes with limitations that make it impossible to install Laravel through conventional means

So instead, I’m setting up a virtual development environment, that will run on the same laptop that I use for code development. Once development is complete, I can upload the whole thing to the shared hosting service. Getting this set up is surprisingly complicated, but once you’ve worked through all these steps, you’ll have a flexible and easy-to-use environment for developing and testing your Laravel websites. This article assumes we’re running Windows.


  • virtualbox_iconVirtualBox enables you to run other operating systems on existing hardware, without wiping anything out. Your computer can run Windows and then through VirtualBox, it can run one or more other systems at the same time. A computer within a computer. Or in this case, a server within a computer. Download VirtualBox here and install it.
  • vagrant_iconVagrant enables the automation of much of the process of creating these “virtual computers” (usually called virtual machines). Download Vagrant here and install it.
  • git_iconGit for Windows. If you’re a developer, chances are you know a bit about Git, so I’ll not go into detail here. Suffice it to say that you’ll need Git for Windows for this project: here.
  • PuTTY_iconPuTTY comes with an SSH key pair generator, which you’ll need if you don’t already have a public/private key pair. Grab the installer here.
  • php_iconPHP for Windows. This is not used for powering your websites, but is used by Composer (next step). I suggest downloading the “VC11 x86 Thread Safe” zip file from here. Extract all files to C:\php. There’s no setup file for this. Rename the file php.ini-development to php.ini and remove the semicolon from the line ;extension=php_openssl.dll. Find a line containing “;extension_dir” and change it to extension_dir = "ext"
  • composer_iconComposer for Windows. Composer is a kind of software component manager for PHP and we use it to install and set up Laravel. Download the Windows installer here and install.

SSH key pair

You’ll need an SSH key pair later in this tutorial. If you don’t already have this, generate as follows:

  1. Start PuTTY Key Generator. (It may be called “PuTTYgen” in your Start Menu.)
  2. I would suggest accepting the defaults of a 2048-bit SSH-2 RSA key pair. Click “Generate” and move your mouse around as directed by the program.
    PuTTY key generation
  3. You can optionally give the key a passphrase. If you leave the passphrase blank, you can ultimately use your key for password-less logins. The drawback is that if your private key ever falls into the wrong hands, an intruder can use the key in the same way. Make your decision, then save the public and private keys. You should always protect your private key. If you left the passphrase blank, treat it like a plain text password. I tend to save my key pairs in a directory .ssh, under my user folder.
  4. Use the “Save private key” button to save your public key (I call it id_rsa).
  5. Don’t use the “Save public key” button – that produces a key that won’t work well in a Linux/Unix environment (which your virtual development box will be). Instead, copy the text from the “Key” box, under where it says “Public key for pasting into OpenSSH authorized_keys file:”. Save this into a new text file. I call my public key file “”.

Install the Laravel Installer (sounds clumsy, huh?!)

  1. Load Git Bash.
    Git Bash window
  2. Download the Laravel installer with this command: composer global require "laravel/installer=~1.1". This will take a few minutes, depending on the speed of your connection.
  3. Ideally, you want the Laravel executable in your system path. On Windows 7/8, from Windows/File Explorer, right-click “My Computer”/”This PC”, then click Properties. Click Advanced System Settings. Click the Environment Variables button. Clik Path in the System variables section (lower half of the dialogue) then click Edit. At the very end of the Variable value field, add “;%APPDATA%\Composer\vendor\bin”.
    Set PATH
    Click OK as needed to save changes. Git Bash won’t have access to that new PATH variable until you’ve exited and restarted.

Create Laravel Project

All your Laravel projects will be contained and edited within your Windows file system. I use NetBeans for development and tend to keep my development sites under (e.g.): C:\Users\Geek\Documents\NetBeansProjects\Project World Domination. Create this project as follows:

  1. Fire up Git Bash. This makes sure everything happens in the right place. The remaining commands shown below are from this shell.
  2. Change to the directory above wherever you want the new project to be created.
    cd ~/NetBeansProjects
  3. Install Laravel:
    laravel new "Project World Domination"
    Note: if the directory “Project World Domination” already exists, this command will fail with an obscure error.
  4. That’s it for this stage. Were you expecting something more complicated?

Laravel Homestead

Homestead is a pre-built development environment, consisting of Ubuntu, a web server, PHP, MySQL and a few other bits and bobs. It’s a place to host your Laravel websites while you’re testing them locally. Follow these steps to get it up and running:

  1. From a Git Bash prompt, change to your user folder. Make sure this location has sufficient space for storing virtual machines (800MB+ free).
    cd ~
  2. Make the Homestead Vagrant “box” available to your system.
    vagrant box add laravel/homestead
    This downloads 800MB or so and may take a while.
  3. Clone the Homestead repository into your user folder.
    git clone Homestead
    This should be pretty quick and results in a new Homestead folder containing various scripts and configuration items for the Homestead virtual machine.
  4. Edit the Homestead.yaml file inside the Homestead directory. In the section “authorize”, enter the path to your public SSH key (see above). Similarly, enter the path to your private key in the “keys” section.
  5. Vagrant can easily synchronise files between your PC and the virtual machine. Any changes in one place instantly appear in the other. So you could for example in the “folders” section, map C:\Users\Fred\Code (on your Windows machine) to /home/vagrant/code (on the virtual machine). In my case, I’ve got this:
    - map: ~/Documents/NetBeansProjects
    to: /home/vagrant/Projects
  6. We’re going to create a fake domain name for each project. Do something like this in the Homestead.yaml file:
    - map: pwd.localdev
    to: /home/vagrant/Projects/Project World Domination/public

    Of course, if you put “http://pwd.localdev” in your browser, it will have no idea where to go. See the next section “Acrylic DNS proxy” for the clever trick that will make this all possible.
  7. To fire up the Homestead virtual environment, issue the command vagrant up from the Homestead directory. This can take a while and may provoke a few security popups.

Here’s the complete Homestead.yaml file:

ip: ""
memory: 2048
cpus: 1

authorize: ~/.ssh/

- ~/.ssh/id_rsa

- map: ~/Documents/NetBeansProjects
to: /home/vagrant/Projects

- map: pwd.localdev
to: /home/vagrant/Projects/Project World Domination/public

- key: APP_ENV
value: local

At this point, you should be able to point your browser to If you have created a Laravel project as above, and everything has gone well, you’ll see the standard Laravel “you have arrived” message. The Homestead virtual machine runs the Nginx webserver and that webserver will by default give you the first-mentioned web site if you connect by IP address.

Laravel landing page

VirtualBox is configured to forward port 8000 on your computer to port 80 (the normal port for web browsing) on the virtual machine. In most cases, you can connect directly to your virtual machine instead of via port forwarding. You’ll see in the Homestead.yaml file that the virtual machine’s IP address is set to So generally (if there are no firewall rules in the way), you can browse to or (the port number 80 is assumed, if omitted). Both should work. Personally I prefer the latter.

Acrylic DNS proxy

Of course we want to be able to host multiple development websites on this virtual machine. To do this, you need to be able to connect to the web server by DNS name (, not just by IP address. Many tutorials on Homestead suggest editing the native Windows hosts file, but to be honest that can be a bit of a pain. Why? Because you can’t use wildcards. So your hosts file ends up looking something like this: pwd.local webapp1.local webapp2.local

(If you’re using, just replace in the above example.) So this can go on and on, if you’re developing a load of different sites/apps on the same Vagrant box. Wouldn’t it be nice if you could just put a single line, *.local? This simply doesn’t work in a Windows hosts file.

And this is where the Acrylic DNS proxy server comes in. It has many other great features, but the one I’m particularly interested in is the ability to deal with wildcard entries. All DNS requests go through Acrylic and any it can’t respond to, it sends out to whichever other DNS servers you configure. So it sits transparently between your computer and whatever other DNS servers you normally use.

The Acrylic website has instructions for Windows OSes – you have to configure your network to use Acrylic instead of any other DNS server. Having followed those instructions, we’re now most interested in is the Acrylic hosts file. You should have an entry in your Start menu saying “Edit Acrylic hosts file”. Click that link to open the file.

Into that file, I add a couple of lines (for both scenarios, port forwarding and direct connection, so that both work): *.localdev *.localdev

I prefer using *.localdev, rather than *.local for technical reasons (.local has some peculiarities).

This now means that I can now put in my Homestead.yaml file the following:

- map: site1.localdev
to: /home/vagrant/Projects/site1/public
- map: site2.localdev
to: /home/vagrant/Projects/site2/public
- map: site3.localdev
to: /home/vagrant/Projects/site3/public
- map: site4.localdev
to: /home/vagrant/Projects/site4/public
- map: site5.localdev
to: /home/vagrant/Projects/site5/public
- map: site6.localdev
to: /home/vagrant/Projects/site6/public
- map: site7.localdev
to: /home/vagrant/Projects/site7/public

and they will all work. No need to add a corresponding hosts file entry for each web site. Just create your Laravel project at each of those directories.

Managing MySQL databases

I would recommend managing your databases by running software on your laptop that communicates with the MySQL server on the virtual machine. Personally I would use MySQL Workbench, but some people find HeidiSQL easier to use. HeidiSQL can manage PostgreSQL and Microsoft SQL databases too. You can connect via a forwarded port. If you wish to connect directly to the virtual machine, you’ll need to reconfigure MySQL in the virtual machine, as follows:

  1. Start the Git Bash prompt
  2. Open a shell on the virtual machine by issuing the command vagrant ssh
  3. Assuming you know how to use vi/vim, type vim /etc/my.cnf. If you’re not familiar with vim, try nano, which displays help (keystrokes) at the bottom of the terminal: nano /etc/my.cnf
  4. Look for the line bind-address = and change it to bind-address = *
  5. Save my.cnf and exit the editor.
  6. Issue the command service mysql restart
  7. You can now connect to MySQL using the VM’s normal MySQL port. Exit the shell with Ctrl-D or exit.

Okay, okay, why go to all this trouble? I just prefer it. So sue me.

Forwarded port Direct to VM
Port: 33060
User name: homestead
Password: secret
Port: 3306
User name: homestead
Password: secret

Managing your environment

Each time you change the Homestead.yaml file, run the command vagrant provision from the Homestead directory, to push the changes through to the virtual machine. And once you’ve finished your development session, run vagrant suspend, to pause the virtual machine. (vagrant up starts it again.) If you want to tear the thing apart and start again, run the command vagrant destroy followed by vagrant up.