NSONE – Data-Driven DNS on Software Gone Wild
DNS is a crucial component in modern scale-out application architectures, so when Alex Vayl and Kris Beevers from NSONE contacted me just as I was starting to work on my Active-Active Data Centers presentation, I was more than interested to hear what their solution can do.
The result: Episode 29 of Software Gone Wild in which we discussed a number of topics including:
- The history of DNS server evolution, from simple BIND-based servers to database-based solutions;
- How can you measure service availability without deploying agents on servers?
- How can you get around the well-known problems of DNS-based geolocation?
- How do you reconcile the geolocation data used by DNS with routing realities of Internet?
- How can you use end-user browsers as monitoring agents?
- Why are most managed DNS name servers still not using IPv6?
I listened to the podcast and found it interesting, thank you.
One thing that stuck in my mind was the mention that IPv6 anycast is "hard".
I know from experience that anycasting DNS (IPv4) does introduce complexity and multi-homing servers can sometimes be tricky.
I am contemplating IPv6 anycast for our company's new DNS deployment and I want to know, are there any significant challenges over and above those of IPv4 anycast?
Thank you!