Menu
Community /Pin to ProfileBookmarkThey have multiple web servers so they called them www1, www2, etc because you can't set up multiple web servers on the same domain name and name them all www because then there would be no way to tell which server to access.
www, www1, and www2 are all subdomains. Most sites through legacy use www as the default website rather than or as well as just the domain itself. Also it is quite possible to have different sites onwww.domain.com and domain.com
The complete format for URLs is:
http://userid:[email protected]/directoryname/filename
It would be extremely rare to include all of them when typing one in thouigh - in most cases you allow some to default.[URL=http://www.faqs.org/rfc/rfc882.txt]According to RFC882[/URL] :
"[I]A domain is identified by a domain name, and consists of that part
of the domain name space that is at or below the domain name which
specifies the domain. A domain is a subdomain of another domain
if it is contained within that domain. This relationship can be
tested by seeing if the subdomain's name has the containing
domain's name as the right part of its name. For example, A.B.C.D
is a subdomain of B.C.D, C.D, D, and " ". Potentially, each node
or leaf on the tree can create new subdomains ad infinitum. [/I]"
@NogDogApr 07.2006 — #The subdomain points the request to a particular document root directory. Whether that directory is on the same server as the other subdomains of that domain or on a separate piece of hardware is completely at the option of the network administrators, and can be a combination of separate servers and/or separate (or same) directories on those servers. The network server(s) will have configuration settings that point each valid subdomain/domain request to the desired server and directory. A well-configured domain server will also have a configuration which points requests to a default subdomain if none is provided. (As noted above, it does not have to be called "www", it could be called "nogdog" if you wanted, but by convention "www" is usually the default subdomain.)
When a site uses more than one computer for hosting then each computer needs to be given a different machine name and often www1, www2 etc are what gets used.
Example: With my web site athttp://www.felgall.com/ the top level domain is .com, my domain name is felgall, the machine name (which can be omitted) is www and it uses HyperText Transfer Protocol to access it. While I do have several sub-domains within my domain I do not have one called www and if I did then you would be able to access it at www.www.felgall.com .[color=red]felgall[/color] .com[/I] [I]felgall[/I] is a subdomain of [I].com[/I] and in [I]http://[color=red]www.[/color] felgall.com[/I] [I]www[/I] is a subdomain of [I]felgall.com[/I]. [I].com[/I] is the toplevel domain and as such does not have it's own root directory.
protocol://userid[b]:[/b]password[b]@[/b]machinename.subdomain.domain/path/filename
with everything except protocol and domain being optional.
Strict programming logic means that the default value for something is assumed to be the one you want to use if you don't specify something different. In this instance the Server Alias provides the default or assumed value.
As to where I got the information - I have seen it in a number of different books that deal with how the URL and domain naming was designed to work - that is the specifications for defining the component parts of the URL. Our disagreement has probably come about due to the differences between the specification and the implementation of this part of the URL processing since the www that is supposed to represent the machine name is internally treated as an alias for the main domain.
Those are seperated from the domain name as a whole by a "[B]@[/B]". Just to clarify my statement: every delimited group of characters not itself containing any delimiting characters and superceeded by a dot is a subdomain.The specifications state that the www is the machine name. That it is implemented as something else doesn't change what it is supposed to represent.
The [I]www.[/I] is a moot point and this discussion about webservers is a distraction from its real reason for its existence. The domain name system (DNS) translates domain names into "[I]A[/I]" records (IP addresses), either directly or using CNAMES or MX records. The problem is the protocol under which "[I]A[/I]" records are served has no method to determine for which protocol the IP address will be used. That means for each protocol that needs to be served to a different unique IP we need to set up a subdomain. The reason this is necessary to know is because, for example the authoritative mailserver for a domain and its corresponding webserver may be located at different IP addresses. After splitting the requests into their various types with subdomains
Time to make this a tag-team fight then (no pun intended)
Felgall; you are partly correct; where a username is given, then the portion immediatly following the @ symbol may well refer to a particular machine, but there is no specific requirement to do so.
In this discussion, there was no mention of a login-style URL anyway, not least because hardly anyone uses that format. Therefore, as Bokeh said, everything up to the first slash, (or as far as the port number if specified) is the Fully-Qualified Domain Name. Reading from right to left, everything to the left of the first dot is a sub-domain of the letters to the right.
Therefore:www.edinburgh.gov.uk is a sub domain of edinburgh.gov.uk, which is a sub-domain of gov.uk, etc.
Good practice says that the final sub-domain should NOT match the name of the machine on which it runs - tying it down in that way makes infrastructure changes confusing at best, although it is perfectly possible. Clearly, since the 'chain' of sub-domains can be extended infinatly, it would also be illogical to make "edinburgh.gov.uk" the name of a single server, as suggested by someone in this thread, and then create sub-domains from that point.
Hope that helps the original poster!
Web sites, especially dynamic Web sites, that handle large amounts of traffic often need more than one server to accommodate the many requests they receive as one server often cannot handle the multitude of requests. The numbers that follow the “WWW’ refer to different Web servers, often as elements of a server farm, that all contain the exact same information. The servers are used in coordination with each other for load balancing. An example of this system iswww.google.com , which uses multiple servers to handle all its traffic. Sometimes the user’s physical location determines which server receives the routed requests, and sometimes the different servers are used when one or more of the servers need to be taken offline in order to be updated with current information.For a normal person to understand this:
They are "aliases" just so the browser will find the right website.
It's not necissary to call them www or www1, but most websites do so since it's sort of a web standard.
www. is usually the main website.
www1,2,3,4 etc.. can be internal websites, an alternate website or just a file server.
This is ofcourse not set in stone. You can configure it in your webserver.
And now I will get corrected by some more experienced developers, but this is how I put it. The original intention was that the www on the front identify the particular server that the site was being delivered from. Those writing the web servers realised that they could achieve the same end result simply by treating it as if it were a sub-domain and therefore not have to create any additional code for server identification. The concept of a server identifier proved to not be needed and so the www effectively has become redundant.
The way I see it to run a domain to full effect you need subdomains. I.e.
If you use the domain itself for one of those that one subdomain becomes redundant, but the rest are needed if each service is served from a different i.p. address.All I know is at work we have www1,www2,www3, and so on. All are seperate servers with the same content on it located around the world. and anyone that goes to www or 'sans' www gets whatever one of those servers are closest to them (unless it is offline, then the next closest and so on).
That is effectively using the sub domains as the server names they were intended to be in the original concept. The server identifiers are www1, www2, etc and the generic www server is whichever of those that the request happens to get directed to first.
In those instances while the naming mechanism is making use of the sub domain functionality there are no actual sub domains being referenced.
If sub domains are also involved then you would have servername.subdomain.domain eg. www1.sub.example.com
Most sites do not use multiple hosting in different locations and so the server name concept of www only gets used by the biggest of sites.
With each of those the "sub domain" is actually identifying a different server (although all might be running on the same computer).
www - web server
ftp = ftp server
mail = mail server
etc.
As I mentioned before those writing the core code for domain processing in the first place decided that the sub domain processing would work equally well for server name processing and so did not distinguish the two.
In the case of www since that is the expected default for a web server it is redundant except for sites that use multiple web servers. ftp is equally redundant in your ftp program and mail is redundant in your mail program. Again unless you have multiple ftp or mail servers and want to reference a specific one rather than just any one.
One example of the multiple www setup that I have seen was a web page designed to display the results of a Google search from 17 specific Google servers each in a separate iframe so that you can compare the different servers and see how the currently in progress update is affecting a particular listing (since it will often appear in two different spots depending on which servers have/haven't been updated). To reference those 17 specific servers requires the specific server identifiers in place of the generic www
It seems that technicalities can be so various.
Need a voice from Google, Facebook or Quora's IT engineer, I think.http://www.webopedia.com/TERM/W/WWW1.html
my search
https://www.google.com/#q=what+is+the+difference+between+%22www%22+and+%22www1%22+
Searching is something we need to learn to do.
This a very old thread. 04-02-2006, closing it
what is the difference between "www" and "www1"
Hello everybody,
what is the difference between “www” and “www1” or “www2” in Domain Names?
Examples:
[url]www.domainname.com
www1.domainname.com
Sign in
to post a comment26 Comments(s) ↴
0
@felgallApr 02.2006 — #reply?
0
@bokehApr 06.2006 — #you can't set up multiple web servers on the same domain name and name them all www[/QUOTE]Of course you can; it's called load balancing.
www, www1, and www2 are all subdomains. Most sites through legacy use www as the default website rather than or as well as just the domain itself. Also it is quite possible to have different sites on
reply?
0
@felgallApr 06.2006 — #www,www1, and www2 are not sub domains - they are machine names and identify which computer to access. If you don't specify a machine name then the default computer is accessed. Sub-domain names follow the machine names in URLs.The complete format for URLs is:
It would be extremely rare to include all of them when typing one in thouigh - in most cases you allow some to default.
reply?
0
@bokehApr 06.2006 — #www,www1, and www2 [/QUOTE]They are not domains so therefore they must be subdomains. Anything that prefixes the domain an that is separated from it by a dot is a subdomain. Also www. is not default, it just so happens that many people use the www. subdomain for legacy reasons.
reply?
0
@bokehApr 06.2006 — #Further! "[I]A domain is identified by a domain name, and consists of that part
of the domain name space that is at or below the domain name which
specifies the domain. A domain is a subdomain of another domain
if it is contained within that domain. This relationship can be
tested by seeing if the subdomain's name has the containing
domain's name as the right part of its name. For example, A.B.C.D
is a subdomain of B.C.D, C.D, D, and " ". Potentially, each node
or leaf on the tree can create new subdomains ad infinitum. [/I]"
reply?
0
@NogDogApr 07.2006 — #reply?
0
@felgallApr 07.2006 — #Once you get to the domain or sub-domain you can then have one or more computers hosting that domain or sub-domain. Each of these computers has a machine name and where there is only one the machine name of www is usually used. Where a machine name is left off of the front of the domain or sub-domain name the default machine name of www is usually assumed so that the person will still reach the right computer to access the site.When a site uses more than one computer for hosting then each computer needs to be given a different machine name and often www1, www2 etc are what gets used.
Example: With my web site at
reply?
0
@bokehApr 07.2006 — #machine name[/QUOTE]I don't know where you are getting this term from. Everything to the left of a dot in a domain name is a sub domain, hence in [I]http://
Once you get to the domain or sub-domain you can then have one or more computers hosting that domain or sub-domain. Each of these computers has a machine name and where there is only one the machine name of www is usually used. [/QUOTE]This is just not true. In a typical virtual host www. is set up as an alias of its parent domain. Example virtual host:
[CODE]<VirtualHost *:80>
ServerName felgall.com
[color=red]ServerAlias www.[/color]felgall.com
DocumentRoot /path/to/felgall.com/public_html
</VirtualHost>[/CODE]The root directory can also be set as a wildcard of the domain so it processes all requests for that domain that the DNS has entries for. Example virtual host:[CODE]<VirtualHost *:80>
ServerName felgall.com
[color=red]ServerAlias *.[/color]felgall.com
DocumentRoot /path/to/felgall.com/public_html
</VirtualHost>[/CODE]the default machine name of www is usuallyServers do not assume anything. They strictly follow programmed logic.[color=red]assumed[/color] so that the person will still reach the right computer to access the site.[/QUOTE]
When a site uses more than one computer for hosting then each computer needs to be given a different machine name and often www1, www2 etc are what gets used.[/QUOTE]That is not true. For example [I]www.google.com[/I] resolves to two IP addresses ([I]66.249.87.104[/I] and [I]66.249.87.99[/I]) and at each of those IPs there could be any number of individual servers all serving requests for the same domain name ([I]www.google.com[/I]). If each machine had its own name how would the clients ever know which one to access?
While I do have several sub-domains within my domain I do not have one called www and if I did then you would be able to access it atIf that sub domain did not exist the server would return a 404. What you really mean is you do not have a unique virtual host set up for www.felgall.com but instead it is an [I]alias[/I] of another virtual host.www.www.felgall.com .[/QUOTE]
if I did then you would be able to access it atThat is not true either. Each subdomain must either be explicitely listed in a virtual host or be the subject of a wildcard. If it is neither the server will either pass the request to the first virtual host (which is normally configured with an error message) or return a 404.www.www.felgall.com .[/QUOTE]
reply?
0
@felgallApr 07.2006 — #Not everything to the left of the dot is a subdomain name. The userid password and machine name are not subdomains and neither is the protocol.protocol://userid[b]:[/b]password[b]@[/b]machinename.subdomain.domain/path/filename
with everything except protocol and domain being optional.
Strict programming logic means that the default value for something is assumed to be the one you want to use if you don't specify something different. In this instance the Server Alias provides the default or assumed value.
As to where I got the information - I have seen it in a number of different books that deal with how the URL and domain naming was designed to work - that is the specifications for defining the component parts of the URL. Our disagreement has probably come about due to the differences between the specification and the implementation of this part of the URL processing since the www that is supposed to represent the machine name is internally treated as an alias for the main domain.
reply?
0
@bokehApr 07.2006 — #Not everything to the left of the dot is a subdomain name. The userid password and machine name are not subdomains and neither is the protocol.
protocol://userid:[email protected] /path/filename[/QUOTE]
Those are seperated from the domain name as a whole by a "[B]@[/B]". Just to clarify my statement: every delimited group of characters not itself containing any delimiting characters and superceeded by a dot is a subdomain.
reply?
0
@felgallApr 07.2006 — #reply?
0
@bokehApr 07.2006 — #The specifications state that the www is the machine name. [/QUOTE]I've done an extensive search but I can seem to find any reference to [I]machinename/machine name[/I] used in this context. Nevertheless, since I'm sure everyone understands what we are talking about I don't see any need to debate that further.
The [I]www.[/I] is a moot point and this discussion about webservers is a distraction from its real reason for its existence. The domain name system (DNS) translates domain names into "[I]A[/I]" records (IP addresses), either directly or using CNAMES or MX records. The problem is the protocol under which "[I]A[/I]" records are served has no method to determine for which protocol the IP address will be used. That means for each protocol that needs to be served to a different unique IP we need to set up a subdomain. The reason this is necessary to know is because, for example the authoritative mailserver for a domain and its corresponding webserver may be located at different IP addresses. After splitting the requests into their various types with subdomains
- [list]
[/list]
reply?
0
@_lt_Eddie_gt_Apr 17.2006 — #You two are making my head spin. I reckon a fight would solve this. ?reply?
0
@omnicityApr 18.2006 — #You two are making my head spin. I reckon a fight would solve this. ?[/QUOTE]
Time to make this a tag-team fight then (no pun intended)
Felgall; you are partly correct; where a username is given, then the portion immediatly following the @ symbol may well refer to a particular machine, but there is no specific requirement to do so.
In this discussion, there was no mention of a login-style URL anyway, not least because hardly anyone uses that format. Therefore, as Bokeh said, everything up to the first slash, (or as far as the port number if specified) is the Fully-Qualified Domain Name. Reading from right to left, everything to the left of the first dot is a sub-domain of the letters to the right.
Therefore:
Good practice says that the final sub-domain should NOT match the name of the machine on which it runs - tying it down in that way makes infrastructure changes confusing at best, although it is perfectly possible. Clearly, since the 'chain' of sub-domains can be extended infinatly, it would also be illogical to make "edinburgh.gov.uk" the name of a single server, as suggested by someone in this thread, and then create sub-domains from that point.
Hope that helps the original poster!
reply?
0
@saptarshibOct 13.2008 — #The WWW prefix (short for World Wide Web) that precedes URL addresses is an indication that the Web address exists on the vast network of the World Wide Web. Sometimes in a URL the “WWW” is followed by a number, such as “WWW1” or “WWW2.” The number that follows the “WWW” indicates that the data being retrieved by the Web browser is gathering the information from a different Web server than the one that serves the typical “WWW” address. Web sites, especially dynamic Web sites, that handle large amounts of traffic often need more than one server to accommodate the many requests they receive as one server often cannot handle the multitude of requests. The numbers that follow the “WWW’ refer to different Web servers, often as elements of a server farm, that all contain the exact same information. The servers are used in coordination with each other for load balancing. An example of this system is
reply?
0
@bokehOct 13.2008 — #Sometimes in a URL the “WWW” is followed by a number, such as “WWW1” or “WWW2.” The number that follows the “WWW” indicates that the data being retrieved by the Web browser is gathering the information from a different Web server than the one that serves the typical “WWW” address.[/QUOTE]Not necessarily. One server could be serving all those subdomain, while, on the other hand, 100 servers could be serving requests from a single www subdomain. By the way there is nothing special about the www subdomain; it works just the same as any other subdomain and has no special meaning.
reply?
0
@QuidamOct 18.2008 — #They are "aliases" just so the browser will find the right website.
It's not necissary to call them www or www1, but most websites do so since it's sort of a web standard.
www. is usually the main website.
www1,2,3,4 etc.. can be internal websites, an alternate website or just a file server.
This is ofcourse not set in stone. You can configure it in your webserver.
And now I will get corrected by some more experienced developers, but this is how I put it.
reply?
0
@felgallOct 18.2008 — #reply?
0
@bokehOct 18.2008 — #The original intention was that the www on the front identify the particular server that the site was being delivered from.[/QUOTE]Can you expand on that.
The way I see it to run a domain to full effect you need subdomains. I.e.
- [LIST]
[/LIST]
If you use the domain itself for one of those that one subdomain becomes redundant, but the rest are needed if each service is served from a different i.p. address.
reply?
0
@ShortsOct 18.2008 — #reply?
0
@felgallOct 19.2008 — #All I know is at work we have www1,www2,www3, and so on. All are seperate servers with the same content on it located around the world. and anyone that goes to www or 'sans' www gets whatever one of those servers are closest to them (unless it is offline, then the next closest and so on).[/QUOTE]
That is effectively using the sub domains as the server names they were intended to be in the original concept. The server identifiers are www1, www2, etc and the generic www server is whichever of those that the request happens to get directed to first.
In those instances while the naming mechanism is making use of the sub domain functionality there are no actual sub domains being referenced.
If sub domains are also involved then you would have servername.subdomain.domain eg. www1.sub.example.com
Most sites do not use multiple hosting in different locations and so the server name concept of www only gets used by the biggest of sites.
reply?
0
@felgallOct 19.2008 — #
[LIST]
[*] www.domain.com [*]ftp.domain.com [*]mail.domain.com [*]dns1.domain.com [*]etc.domain.com
[/LIST]
[/QUOTE]
With each of those the "sub domain" is actually identifying a different server (although all might be running on the same computer).
www - web server
ftp = ftp server
mail = mail server
etc.
As I mentioned before those writing the core code for domain processing in the first place decided that the sub domain processing would work equally well for server name processing and so did not distinguish the two.
In the case of www since that is the expected default for a web server it is redundant except for sites that use multiple web servers. ftp is equally redundant in your ftp program and mail is redundant in your mail program. Again unless you have multiple ftp or mail servers and want to reference a specific one rather than just any one.
One example of the multiple www setup that I have seen was a web page designed to display the results of a Google search from 17 specific Google servers each in a separate iframe so that you can compare the different servers and see how the currently in progress update is affecting a particular listing (since it will often appear in two different spots depending on which servers have/haven't been updated). To reference those 17 specific servers requires the specific server identifiers in place of the generic www
reply?
0
@DonHaromuntheJan 05.2016 — #You two are making my head spin. I reckon a fight would solve this. ?[/QUOTE]
It seems that technicalities can be so various.
Need a voice from Google, Facebook or Quora's IT engineer, I think.
reply?
0
@TrainJan 30.2016 — #my search
Searching is something we need to learn to do.
This a very old thread. 04-02-2006, closing it
reply?