WebCom precipitated consensus on use of “Host” header to enable virtual hosting…
Here’s Chris Schefler’s posting to the IETF HTTP Working Group elist, explaining why a header including the host name of the site being accessed needs to be passed by the http client.
“It is important to be able to determine which domain name was used in the URL in case a server answers to many domain names.”
“This is necessary to support ‘virtual hosting’.”
“[t]ypically, this is accomplished by assigning a different IP
address to each domain name, allowing the Web server to consult
a table or do a reverse DNS lookup to determine the domain name
and map to the appropriate home page.
However, this needlessly consumes IP addresses, requires an OS
which supports multiple IP addresses on the same network interface,
and has severe scalability problems (for instance, Solaris only
allows a maximum of 255 IP addresses per network interface).”
This posting precipitated quite a bit of discussion, as can be seen from the wg elist index page.
The result? Roy Fielding, the primary author of the protocol spec, posting a note to the HTTP working group, Status and Consensus on Host, five days later, stating that a rough consensus has been reached on the inclusion of the “Host” header in the HTTP 1.1 protocol as a mandatory item, and the 1.0 protocol as a strongly recommended option.
“…it seems that the WG has reached rough consensus
on defining the
Host: fully.qualified.domain.name
as the method for telling the server what hostname is being accessed
even when there are multiple hostnames per IP address. It will be
a required header on all HTTP/1.1 requests, and I encourage everyone
to go out and implement it on HTTP/1.0 clients. There are already
three implementations that I know of, and I expect it to be in all
the current browser releases by the end of the year.
Once a critical mass of clients have implemented Host, servers will
be able to add the multihome features as well.”
Yay, Chris! Yay, WebCom … the first “self-service”, “shared hosting”, “virtual host” in the world to have so many clients that this became absolutely essential to our continued success… the result of not having it to start out with is that years later, people are STILL linking to WebCom hosted sites as http://www.domainname.com/~username/ … or http://www.domainname.com/username/ (a later innovation we put in place to reduce the complexity of spelling out URLs)!!!
