“It’s not on my end,” or something similar is not what you want to hear when troubleshooting issues related to user experience and the network. This is an occurrence that IT teams often face while looking for the root cause of issues. Unfortunately, using manual processes and legacy tools to rule out what is not the problem is a common tactic today. But, there’s hope.
For instance, during Facebook’s recent October 2021 outage, Aruba User Experience Insight (UXI) customers running Facebook performance tests saw that the issue wasn’t on their end within 5 minutes! As help desk calls started coming in, healthcare, hospitality, education, and other verticals where social media is used prominently were quickly able to see that their inability to get to Facebook was not due to their networks.
In fact, on that day, Aruba UXI sensors detected Incidents for 40+ customers directly related to Facebook. AI-powered Incident Detection identified that baseline performance metrics had drastically changed in real-time for sensors across these customer’s networks, which triggered notifications.
The “red” indicators are specific to Incidents and make it easy to see the sensors that are experiencing problems and attention is warranted. In this particular case, seeing that every sensor was exhibiting a problem was an indication of a bigger issue. In addition to quickly spotting where problems are, and where they’re not, Incident Detection really shines because an admin receives a consolidated notification that cuts down on alert fatigue.
A by-product of the Facebook outage was the uptick in DNS issues seen around the world. DNS servers (both public and private) saw massive increases in queries, putting pressure on them and degrading their performance. While writing this, someone shared a blog from Cloudflare with me ("Understanding how Facebook disappeared from the Internet"), where the following thought was pulled from a sentence—“the Facebook DNS outage caused a tsunami of additional DNS traffic across DNS servers.”
For the DNS issue, the AI models used by Incident Detection were able to help around 30 customers that were running DNS performance tests see that the problem was something outside of their control.
In closing, if you’re tired of chasing issues only to find that the problem is not on your end, Aruba User Experience Insight gets you to that Mean Time to Innocence (MTTI) faster so that you can focus on the things you can control. And, Incident Detection does help when there are problems on your end as well. Either way, the goal is to resolve problems more quickly and efficiently using AI.
For a brief overview, and an outline of how a large retailer is now using Aruba UXI and Incident Detection, click here.
