In the third part of DNS series here I will be covering Glue Records in DNS, During normal DNS resolutions, when a resolver attempts to resolve a domain name, it first queries the root, which provides the top-level domain. Next, it queries the top-level domain servers, which provide the domain’s authoritative nameserver's. Finally, it queries the authoritative nameserver's for the domain resolution. If the nameserver's for a domain being resolved exist inside the domain itself, then a helper is needed to resolve the domain. What is a Glue Record ? One may wonder what Glue records are, these are DNS records created at the domain’s registrar. This record provides a complete answer when the TLD nameserver returns a reference for an authoritative nameserver for a domain. For example, the domain name “exampledomain.net” has nameserver's “ns1.exampledomain.net” "ns2.exampledomain.net”. To resolve the domain name, the DNS resolver would query in order i.e root, TLD name...
In the Second Part of DNS series we would be looking into reverse Lookup, Reverse DNS as the name suggests is the process of using the DNS to translate IP addresses to hostnames. Reverse DNS is the opposite of forward lookup, which is used to translate human readable hostnames to IP addresses. In the case of reverse DNS, the query is similar to forward lookup it travels from the root DNS to the DNS servers of the Regional Internet Registry (RIR), which contains the authoritative information for the RIR-delegated IP block, for example for any query to ip from IP block 202.0.0.0/8 will travel from root to the respective RIR's DNS server then based on the subnet delegation and the authoritative DNS server registered with the RIR for the queried subnet ip it will forward the query to the respective authoritative DNS server. How this is done? For all the IP addresses allocated by Internet Assigned Numbers Authority (IANA) to RIR, RIR also delegates corresponding rever...
MTU stands for Maximum transfer unit, it is the largest data size in bytes that can be transferred by the physical port, it is different for each physical medium that we use, it is per port basis and cannot be negotiated. Ethernet by default has the value of 1500 bytes, This means that a single frame passing the Ethernet medium can be of 1500 bytes of data, on top of this data payload (i.e data 1460 bytes + TCP header 20 bytes + IP header 20 bytes) further details like Ethernet headers of 14 bytes, Frame check sequence (FCS) 4 bytes gets added making it 1518 byes if this frame is carried over tagged connection then further 4 bytes for dot1Q gets added making the Frame 1522 bytes in size. If for any reason the size of payload increases which results in increase of the Ethernet frame size then the device starts fragmenting the packets i.e the source device will break the payload to fit the data into multiple Ethernet frames and the destination device will have to reassemble t...
Comments
Post a Comment