An Update on Today’s DNS Outage: A Message From Our CEO

by Ken Carnesi on Jan 25, 2022 12:00:00 AM

As a recursive DNS resolver, we take any incident or perceived outage very seriously as a company. We know that you rely on us to be highly available as your DNS provider and DNS security. We recognize DNS is a critical aspect of your infrastructure, and for our MSP partners, that of your customers as well.

On Thursday January 13th at 20:08pm UTC and again on Tuesday, January 25th at 14:33 UTC, we began to receive tickets from users in the Northeastern United States stating they were experiencing DNS timeouts. Based on our monitoring, it seems that just under 1% of our network was impacted. Our network monitoring also showed a BGP route peering change which was shifting a heavy volume of DNS traffic to our nodes in Eastern Europe. The initial form of notifications to our customers was posted on our status page.

Our engineers discovered the root cause was a provider that made an upstream peering change. In both incidents, we immediately stopped advertising to that node and restored traffic right away. Service was completely restored by 20:27 UTC and 14:47 UTC, respectively.

When we select service providers and begin the process of deploying new anycast nodes, we tune our BGP advertisements and blocks based on the network paths which give our customers the best experience. In this particular case, our provider added a new peering connection that introduced routing changes that negatively impacted some of our customers. In response, we removed that node from service until we could fully understand the change and tune our BGP advertisements. We are actively improving our observability to detect and respond to these types of changes more quickly, ahead of any customer impact.

We will further vet this new provider to ensure they’re truly able to meet our SLAs. We hold our providers to a very stringent process—any providers unable to meet our SLAs will be eliminated from the DNSFilter network.

I sincerely apologize to our customers for this inconvenience, especially all those impacted. We remain highly committed to operating a reliable and innovative platform that you can trust.

Sincerely,

Ken Carnesi
Founder & CEO

Topics: DNSFilter Updates

Why Scaling Your MSP Doesn’t Mean Hiring More Technicians

Growth should feel like progress. But for a lot of MSPs, there comes a point where growth starts to feel heavier instead. New clients are coming in, and revenue is rising, yet the day-to-day operation feels more stretched, not more efficient. The service desk is constantly busy. Senior techs keep getting pulled into escalations. The team is working harder just to maintain the same standard of delivery.The usual response is to hire more people. On...

The Hidden Cost of “Good Enough” Security in MSP Environments

“Good enough” security checks the boxes and keeps the dashboards green. It covers the basics and gets you through onboarding. But in MSP environments, “good enough” usually means nothing breaks badly enough to force action. And that’s exactly the problem.The tooling system doesn’t fail. It just becomes more expensive to run, gradually turning your service desk into a permanent cleanup crew.Over time, reactive security tools create a profitability...

SASE vs SSE: What's the Difference and Why It Matters for Your Security Stack

If you’ve spent any time researching modern network security, you’ve likely come across SASE and SSE used interchangeably, sometimes even in vendor messaging. The result is a lot of confusion around two concepts that are closely related but not identical.