All over the internet, there are ominous rumblings that “768K Day” is on its way—and it’s likely going to happen before May is over.
If you’re an MSP who hasn’t heard about 768K Day yet, brace yourself—you could be in for a serious connectivity headache.
What’s 768K Day?
The internet is made up of routers maintained by different ISPs. Those routers are responsible for sending packets all over the world at different speeds and across providers.
To know where to send specific packets, routers use the BGP routing protocol, which maintains a list of known networks and directions on how to reach them. But the global BGP routing table has a threshold for the number of routes it can handle, which is—you guessed it—768,000.
As more ISPs and users come online, we’re on the brink of passing that number. And when that happens, it will officially be 768K Day.
What happens when we exceed 768,000 route entries?
To understand what will happen when we pass the threshold, we just need to turn the clock back to 2014.
On Aug 12, 2014—known as 512K Day—we exceeded 512,000 routes when Verizon added 15,000 new routes to the global BGP routing table without warning. While ISPs knew 512K Day was on the horizon, most hadn’t taken the simple steps to increase their routers’ routing entry capacity.
As a result, we experienced the “mother of all internet outages” as older routers experienced memory overflows, causing their CPUs to crash. This created mass packet loss and traffic outages around the world—and cost ISPs millions of dollars while network administrators scrambled to up the route entry threshold to 768,000.
This time around, large ISPs are better prepared and have taken steps to update old routers. But network intelligence company Thousand Eyes anticipates 768K Day will be “an echo of 512K Day,” and we won’t fully escape connectivity issues and outages when it happens.
What does it all mean for MSPs?
We know internet outages of this scale are completely out of an MSP’s control, but that won’t stop your phone from ringing itself off the hook if clients lose access to business-critical applications and websites throughout the day.
So how can you manage the potential chaos?
First, before 768K Day happens, set expectations with your clients. Give them a heads up about what 768K Day is, how their internet access may be temporarily affected, and how you plan to respond if they run into problems.
Then, when the day arrives, make an effort to engage with ISPs. If a client experiences an outage issue you can’t replicate successfully, escalate it quickly to your client’s ISP, since they’re the ones who can remediate it.
And as a final word of advice: Don’t forget to breathe—it’ll all be over before you know it. Until we pass the next threshold, at least.