Curing Wireless Woes
As more organisations adapt their working practices to cope with an unpredictable office situation, so wireless networking becomes ever more important to allow access to networked services. Pervasive mobility is the name of the game with hot-desking and working from home being the requirements for more people than ever before. And so how do you maximise service availability for users when they are in the office?
How does wireless go wrong?
Forgive me for stating the obvious but wireless networking is not the same as wired. Due to the way that wifi works, your connectivity is much more likely to be affected by transient issues like localised electromagnetic interference, contention or even weather conditions! And so we need to consider a different set of problems when troubleshooting wireless connectivity.
The majority of wifi connectivity problems are typically down to client issues, either location, proximity or roaming. The simplest way to resolve these is – yes, you guessed it, to turn it off and back on again! The science behind that is simple enough – if you are in a good position for wifi signal, turning your wireless adapter off then on will allow it to renegotiate with the nearest wifi access point on the network you’re looking to connect to.
But if that doesn’t fix your issue, how can your network support team help?
First things first, you need to scope the issue to determine how widespread the issue might be. Is it a problem with only accessing a particular application, the internal network, or the Internet? Is your neighbour having the same problem, everybody on your floor, or all users in the building? This allows you to understand whether or not the issue might be localised to a particular Access Point (AP), group of APs or controller, or indeed nothing at all to do with the wireless infrastructure!
If the problem is localised in any way, you’ll need to look into – amongst other things:
- how the APs are registered with their controller;
- how the APs are broadcasting SSIDs and on which bands;
- which AP(s) the affected client(s) is/are connected to;
- signal strength and signal-to-noise ratio of the connection from client to AP;
- how the AP(s) and controller connect to the wired network;
- how the SSIDs are associated with VLANs in the wired network;
- whether the client is receiving IP addresses from the DHCP server;
- whether the client is allowed to get access to the service they are trying to access.
Collating all of that information is going to be difficult and long-winded: information needs to be gathered, collated and analysed from the wireless clients themselves, the APs, controllers, and the wired network.
And you’ll need to be able to determine whether the situation in any of these areas has changed over time as wireless connectivity issues are notoriously transient.
How can IP Fabric help?
When it captures a snapshot of the network, IP Fabric harvests thousands of data points from every network device it connects to, regardless of vendor or platform. It then builds a database of relationships and connections between those devices and the endpoints connected to them. In this way it gives deep visibility and insight into the operation of the network.
In the case of wireless networks, it captures as much information as is available about:
- clients and their connection status;
- APs and their radios;
- controllers and their connections to the wired network.
The platform also collects detailed information about forwarding and policy through the wired network and so can give a full picture of connectivity for wireless clients connecting to network services. And of course, with regular snapshots, we are able to compare state over time.
The most effective way of using IP Fabric to troubleshoot issues in the wireless network is to carry out and end-to-end path simulation from wireless client to service.
In this example, IP Fabric has identified that the wireless client is connected to the specific AP on the Testing SSID. The AP is connected on VLAN 109 to the switch stack, and builds a tunnel over the management interface to the Wireless LAN Controller. The SSID is then mapped to VLAN 101 for the client to reach its default gateway. Traffic from the client is then routed over the transit network to site 31 where the target endpoint is connected.
As IP Fabric is collecting state data from all of the devices in that path, we are able to evaluate where errors and issues may impact the path. By overlaying intent validation rules we can view the status directly in the map.
And by clicking through on devices which are shown in a colour other than grey, we are able to identify whether issues identified in the validation rules are impacting the operation of that device.
For example in this case, the wireless client has a slightly lower signal strength than expected.
So you can see that IP Fabric can point you towards issues very quickly and simply by using the end-to-end path check to highlight relevant config and state, wired or wireless.
If you have found this article helpful, please follow our company’s LinkedIn or Blog, where more content will be emerging. If you would like to test our solution to see for yourself how IP Fabric can help you manage your network more effectively, please contact us through www.ipfabric.io.