Chapter 6
DNS
The internet's directory — a globally distributed, hierarchical, cached database that turns names into addresses without any central server.
Every connection starts with a name and ends at an address, and DNS is the machinery in between. It is not one server but a hierarchy: the root delegates to top-level domains, TLDs delegate to authoritative servers, and a recursive resolver chains those referrals into a single answer that it then caches. That structure is why one namespace can hold billions of names with no central bottleneck and no single point of failure.
This chapter treats DNS as the distributed protocol it is, not as any one operating system's resolver. You will walk the namespace and its delegation, follow a query from root to authoritative server, learn which record type each job needs, and see how TTLs — not "propagation" — govern how fast a change takes effect. The last two topics turn DNS into a security surface and a traffic-management tool: DNSSEC, DoT/DoH, split-horizon views, and GeoDNS steering.
Topics in This Chapter
NS records, and why your laptop never talks to a root server directly.A/AAAA for addresses, CNAME for aliases, MX and TXT for mail, SRV, NS, and PTR — and the apex-CNAME rule that trips everyone up.SOA serial, split-horizon views, GeoDNS steering, and why DNS-based failover is coarse, not sub-second.