_Matt_ 26 Posted October 4, 2021 30 mins out right now - how people must be loosing their minds. At least we have the forum to continue our car content 😄   Quote Share this post Link to post Share on other sites
fendervg 29 Posted October 4, 2021 Nothing like a giant BGP cock-up to take your service completely down - from looking at what info is out there, it looks as if all the routes pointing to the Zuckerberg data centers were withdrawn at the same time, also taking out their name servers as these were co-located at the same sites. Nice. Quote Share this post Link to post Share on other sites
_Matt_ 26 Posted October 5, 2021 Yeah I didn't really understand the problem - was it just a case of someone plugging their charger in to the server port 😄  Quote Share this post Link to post Share on other sites
fendervg 29 Posted October 5, 2021 Looks more like someone applied a faulty configuration to their backbone routers that direct all traffic to, from and between the FB data centres. Can easily happen, have done it myself in the past on a much smaller scale! But I would have thought that they would have had some kind of out of band management access over dial-up or 4G for emergencies like this - that's what we put in afterwards. Apparently they had to actually send a team to the data centre to get access and fix it (read reboot). Quote Share this post Link to post Share on other sites
_Matt_ 26 Posted October 5, 2021 ha - I bet they will be having a disaster recovery review after this event. Just goes to show you can never be too big to learn. Quote Share this post Link to post Share on other sites
