 time here from Lawrence Systems and what is BGP is a popular question being searched today on October 5th because on October 4th approximately 1130 Eastern Standard Time the world got more productive when Facebook went down. Now this was obviously filled with conspiracies and crazy things like well did they get attacked it's somehow Mark Zuckerberg have to be rebooted we're not really sure what's going on and of course everyone's speculating going on and on about it and we now have a D briefly understand 24 hours later why it broke and why it's back up now and well I don't really know why it's back up other than so people can post memes and share terrible ideas anyways Facebook's back and the reality of it is BGP is one of those things that I think people should take some more time to understand and also pull their hair out screaming like is this really how the internet works is it really this delicate and yeah it is I've done a few videos on this topic before now Cloudflare and Facebook both have postings that I'll link to down below kind of a debrief and write up but Cloudflare also has a really good write up on how BGP works so initially people thought it's DNS because it's always DNS is the joke in the IT community and the name servers of course were knocked offline but it was a lot more complicated than that to be very basic BGP is the routing protocol so all the different autonomous systems that each of these companies control can talk to each other or at least know how to talk to each other this is what they call a BGP route announcement so Facebook has their autonomous systems Cloudflare has their autonomous systems Comcast Verizon level 3 networks and many many companies you've never heard of that make up the backbone of the internet all have their own autonomous systems companies big companies small and lots of in between now in order for us to know how to get from point A to point B this is where the BGP route announcement system comes in and plays a role in this companies have to announce where they're peered with what other autonomous systems and this is how the resiliency of the internet is supposed to work so if you're peered with several different places you announce your peers and the peers accept the routes and everyone can talk and from the end user standpoint this is very invisible when you go to facebook.com it always finds the most efficient route wherever Facebook has these autonomous systems peered with the other autonomous systems that are related eventually back to the ISP you use that transported you across to get to these but with great power comes great risk and that great risk is if you push the route wrong and you announce them all wrong simultaneously because you built some type of automation tool that didn't validate these route changes you can suddenly point to nothing and no one knows how to get to you and you can't get to anything now this caused a cascading calamity of errors over at Facebook that essentially it sounds like they were unable to even access the tools that they needed to revert this this is why the outage was not just like hey undo and hit that reverse button somehow when this system got automated pushed these route changes that were not valid were accepted by all of Facebook's global data centers and all their different autonomous systems and uh yeah they went down and they had no way to do it other than to go into each of these places and kind of you know plug in and hit the undo button but then as i understood that it became a bit more challenging because people that were physically at those locations didn't have authorization or understanding of how to undo what got pushed from the main headquarters out to these systems so this whole back and forth uh took roughly five hours six hours of outage and you gotta remember it even after fixing it there's a series of queries that are all hitting these because everyone's refreshing their Facebook on their phone and looking for it so it eventually came back online for better or worse so that's what happened no big conspiracy no real the various things happen oops accounts for more outages in the tech industry than any type of threat actor if you work in it long enough you'll just shake your head a lot at how delicate these systems actually are and i actually had a fun time explaining to my dad who does not really use technology he heard that the internet broke and i thought it was funny that it reached someone like my father who's not very tech savvy and uh he said ah the internet breaks son because you know he thought i had something to do with it so it was amusing and it was fun i told him i would do a video explaining that it's really kind of uh more delicate that this system that was designed by darpa to survive nuclear wars and attacks um can be a couple button presses away from partial outages that a system that people find really important and if you follow back through my bgp video that i have when forizen accidentally accepted a bad route change and uh yeah you can see that we lost internet and part of the eastern seaboard so this incredibly robust system can really fall victim to really simple commands and people going oops and uh the famous words of well now we know that can happen that probably were said that people go well yeah this is uh it's a lot more delicate thing you think and uh it's not really anything you have to have a conspiracy for other than people sometimes press the wrong set of keys and that happens so i'll leave links to the cloudflare and facebook post uh and do take the time to read if you want to get a deeper understanding of how bgp works cloudflare has a really good write up on that you know it's an interesting protocol it's something i find kind of fascinating and uh yeah hopefully as we need more people in this industry that don't do oopsies and can look over the shoulders of the people uh before they push these oopsies that'd be great or maybe if there was some way to validate changes you know well there is but hey you know that's throw safety to the wind don't they have a policy over there like break things fast maybe policy might have changed now all right thanks and thank you for making it to the end of this video if you enjoyed this content please give it a thumbs up if you'd like to see more content from this channel hit the subscribe button and the bell icon to hire a sharp project head over to laurancesystems.com and click on the highest button right at the top to help this channel out in other ways there's a join button here for youtube and a patreon page where your support is greatly appreciated for deals discounts and offers check out our affiliate links and descriptions of all of our videos including a link to our shirt store where we have a wide variety of shirts and new designs come out well randomly so check back frequently and finally our forums forums.laurancesystems.com is where you can have a more in-depth discussion about this video and other tech topics covered on this channel thank you again and we look forward to hearing from you in the meantime check out some of our other videos