One of the most common issues for integrators is after-sales support. Geoff Meads looks at dynamic DNS and how it could be the solution.

By nature, the systems we install are both complex and customised. They are built from technologically advanced products to form intricate systems that mere mortals (like end users) would struggle to understand.

While designing and installing these systems is hard enough, supporting them when something goes wrong can be even more tricky. Especially when the system is located a long way from the installer’s company office. In fact, for many installers, projects might be in a different state or even country.

So how do we replicate, diagnose and fix problems with systems that are far away? The answer is, of course, in remote support via an internet connection.

The equipment manufacturer I tend to use for customer networks has an inbuilt system for that. It allows remote diagnosis of most issues via the ‘cloud’, as long as the main Internet connection is live at the remote site. However, with systems built from components from multiple manufacturers, this is often not the case. So, just how do we connect to a remote system via the Internet, regardless of the components used?

The problem

While the internet is a wonderful thing, it is not just a single network – it’s a network of networks. Traversing the internet from one site to another means traversing multiple networks to a specific destination IP address – the customer’s remote IP address on the internet or ‘WAN’.

If we have that address, and it always remains the same, then things are pretty easy. However, most clients will be connected via a ‘domestic’ internet supply from a local ISP, and those rarely come with a fixed or ‘static’ IP address.

Possible solutions

Sometimes it’s possible to pay a little more each month and get a static WAN IP address for the client’s home. Sadly though, with the global shortage of IPv4 addresses, many providers have stopped offering this option. In the UK, for example, it isn’t possible to get a static IP address from the larger ISPs unless it is for a registered business premises.

There is a way around this though. It has been around for a long time, and it’s supported by just about every home router that you’ll come across. It’s called Dynamic DNS or DDNS for short.

DNS recap

Before we learn about DDNS, let’s look at the system on which it is based – the Domain Name System (DNS). If you have a full back catalogue of this magazine then you may remember I wrote extensive articles on this subject back in 2020. If not, here’s the basics…

DNS is a system that assigns a ‘human-readable’ name to an IP address. This makes it possible for you to type ‘https://www.facebook.com’ into a web browser rather than typing in an IP address. This one piece of technology makes the web far more usable. DNS is, essentially, a huge telephone directory attaching readable names to IP addresses for all sorts of resources. Your own company domain will normally have multiple attached DNS records. For example, you have a record for your website, a record for your email server and probably others too.

There are many copies of the DNS record ‘library’ all over the internet, Your ISP probably has one. Some records are often cached locally to minimise the number of ‘lookups’ that take place in the libraries. In fact, your router keeps cached DNS records, and your devices do too. This is one reason why the websites you look at most loaded a little faster than new ones.

Intro to DDNS

Where ‘regular’ DNS is designed for records that don’t change much, DDNS (as the name suggests) was born to look after records that change more often. As we discussed earlier in this article, the IP address of your router on the Internet (or ‘WAN side’) can change at any time (unless you’re able to obtain a rare ‘static’ address). So, to connect to a remote client’s router, we need some way of keeping track of the router’s WAN IP address. DDNS does just that.

So, what do we need to make DDNS provide a stable route to the customer’s network? Across the internet? Firstly, you’ll need an account with a DDNS provider. Typical examples are DynDNS (dyn.com) or No-IP (noip.com) but there are, of course, others.

Once you have a DDNS account, you can set up a dynamic domain that will link to the remote router’s WAN IP. Actually, this will typically be a sub-domain like ‘mycompany.ddnsprovider.com’ where ‘mycompany’ is your chosen DDNS sub-domain and ‘ddnsprovider.com’ is the main domain for the provider.

At this point a word of caution. Not all routers support all DDNS provider domains. So, it’s worth looking at the DDNS setup area of your favourite router brand (or the ones most used by your clients) to see which they support and cross-check this with your chosen DDNS provider before signing up to a DDNS account.

Once you have your DDNS subdomain set up then it’s time to go into the menu of the remote router and add in your DDNS domain, your DDNS username and the inevitable password.

What happens now?

From now on, whenever the remote router gets a new IP address (typically when it reboots), it will send the new IP address to the DDNS provider which will, in turn, connect that IP to your DDNS subdomain. This is great news because, whenever you need to connect to the WAN side of the remote router, you can now use the DDNS subdomain (e.g. ‘mycompany.ddnsprovider.com’) rather than the IP address, and you will always be connected to the right place!

The next question might be: “Is this needed if I’m using a VPN to connect?”

The answer is yes! While a VPN server on the remote router will secure your connection to it and let you ‘appear’ to be on the LAN of the remote network once connected, a VPN client still needs to know the IP address of the remote VPN server (the router). So, just pop your new DDNS domain into the remote server address field of the VPN client setup and voila. You’re connected!