Hello, welcome to the Wednesday, October 6, 2021 edition of the Sansonet Storm Center's Stormcast. My name is Johannes Ulrich,

and today I'm recording from Jacksonville, Florida. One of the tools that I used yesterday to get to the bottom

of the Facebook outage was a looking class website.

So today I wrote a quick diary article about how to use these looking class websites.

Essentially, what they are is different network operators are maintaining these sites.

And you can essentially sort of run commands on various routers

using these websites to interrogate the routing table to do things like a trace route and such

essentially to look at what the routing table looks like from different viewpoints across the

internet so wrote up a quick introduction how to use these looks like from different viewpoints across the internet.

So wrote up a quick introduction how to use these sites.

Now with the focus on BGP in this incident, some people were jokingly calling October now the BGP awareness month.

I'm not really talking much about BGP here. There are plenty of other

great resources that can probably explain it better than I do. So this article really just focuses

on how to use these looking class websites. And I mean, got a little bit more detail from

Facebook's engineering team about yesterday's

outage. Apparently, what happened was that at the beginning, they were running some routine

maintenance jobs as they're calling it on their internal backbone. So the network that connects

Facebook's systems worldwide. And apparently that took down Facebook's systems worldwide.

And apparently that took down Facebook's internal backbone.

Now, at that point, some systems still had access to the Internet.

They just couldn't communicate with each other.

And as part of their sort of failover, which is actually not that unusual,

...