Domain Name System, DNS | Part 3
Using DNS Names
A domain name, such as it.wikipedia.org, may be part of a URL, like http://it.wikipedia.org/wiki/Treno, or an e-mail, such as apache @ it . wikipedia.org.
These are the tools used to identify a resource on the Internet, which explains the pervasiveness of domain names.
Many domain names are used for web server in the left side the string of characters “www”, but it is not always necessary to have it. In many cases, but not always, the private name of the prefix “www.” still leads to the same page, such as “ns.nl” and “www.ns.nl.
Realization
The DNS implements a hierarchical namespace to allow portions of a namespace, known as “zones” can be delegated to a name server to another name server that is located lower down the hierarchy.
Domain names are subject to certain limitations: for example, every part of the name (the one that is limited by the points in its name) can not exceed 63 characters total, and the name can not exceed 255 characters.
Domain names are limited to a subset of ASCII characters, in this way prevents write names and words with characters that not everyone on your keyboard. To overcome this limitation, the system of IDNA and Punycode-based model, detects Unicode strings into a set of valid DNS character, was approved by ICANN and adopted by some registries.
Zone, Proxies And Replicas
A DNS zone is a part of the namespace, consisting of a domain and its subdomains, which are in turn delegates, which is under the same administration, and then is handled by one or more servers.
The management of a zone is delegated through the upper region of the NS record. For example, in the area: (dot) org, there will be a proxy for the area wikipedia.org DNS servers that run it. For redundancy, each zone is replicated on multiple servers, and so the delegation is formed by a NS record, indicating that each of the servers listed contains the information for that area (that is authoritative for the zone). Within a zone may be delegated to lower level areas, such as might exist wikipedia.org proxies for devel.wikipedia.org or accounting.admin.wikipedia.org.
The various servers that are delegated to an area should contain the same information, so that either of them can answer a query for a record of the area.
The pattern of replication typically requires that there be a master server (primary), which is one on which the information is updated and one or more slave servers (secondary), which copy the information from the master when needed. To keep track of different “versions” of an area that may be in circulation, and in particular to allow a minor to decide whether to transfer the zone from the primary, each zone has a serial number that must be incremented each time modifications are made to the primary. To get the serial number of a zone on a server, you make a query of type SOA. Secondary compare your serial number with that of the primary, and if that is higher than the primary transfer the zone.
The copying of all records of a zone from master to slave is called a zone transfer, and may be complete (all the content in the area is copied) or incremental (only changed records are copied from the version already there.
Some implementations of DNS zones allow for editing of any authoritative server, propagate the changes on other servers.
The root (root) DNS name of the tree and the area. (dot), which is managed by a set of servers called root servers precisely.
Recursion
In general, to achieve the resolution of a name is necessary to start from the root, root server in question one of the top-level domain, obtain the server that runs it, questioned in the second-level domain until it reaches the server authoritative for the name desired. This technique is called “recursion”.
Caching
Some servers are suitable to perform recursive queries on behalf of some clients. Once you have received a response, stored in a cache all the information they have learned until they expire. Some implementations of DNS servers allow the creation of the so-called caching only, ie without its own database, but useful to redirect queries to a server authoritative resolution. This feature is especially useful when the resolution should be done through slow links (with speeds below 500 kbps) or firewall.
Functions of the DNS server
A DNS server can be configured to perform one or more of the following functions:
* Server is authoritative for one or more areas, or the servers that are configured zone data and which is responsible for managing through NS records inserted in the upper zone. Usually there are multiple servers authoritative for a zone. Many implementations allow you to change the zone data only on a server:
or primary – authoritative server on which data changes in an area
or secondary – authoritative server that copies zone data from a primary
* Recursive server – the server that is configured in a population of clients, which deals with resolving queries that query the server receives the original and keeping a cache of responses
or query forwarder – a server that is configured in a population of clients, resolving their queries directly, but by querying a recursive server
Source of data
The data contained in a zone can be configured by one or more operators, or they can be powered by automatic mechanisms:
* In the simplest implementations, the data area are stored in one or more files on the primary server
* Implementations finest stores data in a database. In some cases, this is accessible not only to operators of the service but also directly to customers (this is the case of DNS services trade)
Continued…
