The Domain Name System (DNS)
What Is The Domain Name System (DNS)?
The Domain Name System (DNS) is the system and protocol that the Internet mainly uses for managing domain names to IP addresses to translate and vice versa.
A DNS server or Domain Name Server is based on this technology, making these translations, so that computers not only works through the (which is very impractical) IP address and can be approached, but also works through a host name (computer name).
The reverse of this process is also possible (reverse DNS). IP address is translated into the set hostname in the reverse process. Although it has the most possibilities, DNS is also used in other ways, for example, with respect to the mail servers for a domain. In addition, the Sender Policy Framework (SPF) to TXT records in a DNS server can also be added. Mail Servers can work on the basis of this record to check the origin of the e-mail is correct. SPF is one of the instruments that are used worldwide to combat e-mail spam, also called the Junkmail.
Making a connection between IP address and the name was originally done with a file system. As networks grew, this was not practical and it was ultimately the DNS that helped manage network connections. This file is still reflected in many operating systems as hosts (. Txt).
DNS in practical implementations consists of three parts:
* The stub resolver
* The caching / recursing resolver (sometimes referred recursor)
* The authoritative name server
The retrieval of data using DNS is normally called a lookup. Software such as a web browser sends a lookup request to do it to the stub resolver. This is relatively simple software, depending on the configuration, the question to a recursor or may first look in a file (such as including in derivatives known Unix / etc / hosts).
The DNS stub resolver suggests a package together and sends it to the recursor. Often, the ISP a recursor and is used, even though networks are also regular internal recursor be established. The recursor is more advanced than the stub resolver and will initially start with the question to a DNS root server.
This can refer to other servers, from which in turn can be redirected to other servers, etc., until finally a server is reached that know the answer or know that the lookup is not possible. From the latter may occur if the name does not exist or servers not responding. The process of the ‘long walk’ of several authoritative servers is called recursion.
In search of a domain starts at the highest level (called root) and then increasingly more specific search. In the search for a domain is immediately to the DNS root server asked, for example nl.wikipedia.org.
There is no intermediate step involving only org requested. It is theoretically possible that the root server itself knows the answer for nl.wikipedia.org. Thus, for example, root servers know the answer for a.root-servers.net. The rule by the DNS root server, however, refer to the name servers for org This would in the case of nl.wikipedia.org refer to the name servers for wikipedia.org then know the answer.
Authoritative Name Servers
They are also relatively stupid and give simple answers. These answers are often in files or in a database. An authoritative name server can answer, as both a reference to another server or a direct answer to the question can be.
Both the recursor as the authoritative name server are often DNS server. It is possible to combine these two functions in one program. This is done in BIND, for example, one of the best known and most widely used DNS servers. There are also programs that only one of two functions. NSD is an example of a purely authoritative nameserver. In programs that combine both functions, it is often possible to either off or only open for the internal network.
Caching
To prevent recursors regularly make unnecessary queries (DNS data changes relatively little) is a recursor caching implementation. This means that once received a reply some time saved. This time, the administrative record by adapting and Time to live (TTL) specified. The rule is that between a few minutes and a few days.
Redundancy
In general, there are several authoritative servers for the same dates. This is to avoid the possible consequences of failure of a server limit.
Resource Records
Data is stored in a DNS Resource Record. Such a resource record contains a type, a TTL, a name and data. The data may include an IP address or another name. This depends on the type of resource record.
Common types are:
* A for the definition of the IPv4 address to a name
* AAAA to determine the IPv6 address to a name
* PTR to determine a name to an IPv4 or IPv6 address (see below in reverse lookups)
* MX to determine the mail servers for a domain, each mail server a priority is assigned
* NS to identify the authoritative name servers are name servers (also used to refer to other name servers)
* TXT originally used by the user for any desired comments. Now it is also in use by the SPF anti-spam initiative.
* A relatively new SRV record is used for services in general.
Reverse Lookups
Reverse or “reverse” lookups can be used to find out what name is given to an IP address.
Name Server Tools
The effective management of a particular server are several diagnostic tools available. The so-called BIND tools are the best known.
Source: Wikipedia.
