At 10:22 AM 2/3/99 -0400, Kerry Miller wrote:
>
>
>Roeland,
>
> { Such round-robin DNS games is fine for load distribution among
> { different servers in a cluster and all content is the same company.
> { However, you'll
> { never find a marketing person who will agree to that, nor any other
> { business person. I want www.mhsc.com to come up with MY site
> { ALL the time, every time. You suggestion will break commercial use
> { of the Internet.
> {
>
>Would you be averse to being a portal for the others? To forming a co-op
>to share operational costs?
You obviously haven't really checked out MHSC. We're a security related
business. We run our own servers, and we are a virtual ISP. Many of our
customers are running on our private-label TLD, using SSH VPN technology.
Most of our machines are not visible on the Internet. Now, you want me to
share MHSC.COM with someone else? I don't think they can unless we host
them. Do I want some stranger's meat-hooks in the heart of our network?
Hah, our customers would fly to the nearest exit. It kind of defeats the
security aspect, you know?
There's a LOT more to a domain name than a mere web presence. There are
LOTs of OTHER services that expect a domain to be constant. If there's a
switch, round-robin style, it has to be done *real* careful and the other
host had BETTER be identical to the first one. We set our root-servers up
that way. They all give identical answers. Why is it done so? ... for
load-sharing, that's why. I don't know you, but I'm a little miffed at
gregbo for suggesting this, he knows better. I'm sure he's having some fun
out of it.
I'll suggest this, read up on computer systems network architectures then
review the DNS/BIND book <www.ora.com> and come back with this again.
>Greg,
> { The reason I made the suggestion was to demonstrate to some people
> { on the list that a technology might support a certain type of use, but
> {if significant numbers of people won't use that technology that way, it
> { won't be used.
>
>Whoo! Isnt this not just the problem in DNS, but right here, where the
>arguments are endless about 'only possible solutions,' and totally ignore
>the reality that the list itself can only *persuade 'significant (numbers
>of) people' to any solution at all?
>
>---------
>Earlier, y'all had written:
>{ >> The
>{ >>only solution possible is that trademark law can not apply to DNS
>{ >>entries a priori. They must be adjudicated ad hoc.
>{ >
>{ >Actually, one may have multiple references for the same name. People
>{ >do it all the time. My company has several IP addresses for its web
>{ >site, that are offered in a round-robin manner for load sharing.
>{ >
>
>
___________________________________________________
Roeland M.J. Meyer -
e-mail: mailto:[EMAIL PROTECTED]
Internet phone: hawk.lvrmr.mhsc.com
Personal web pages: http://staff.mhsc.com/~rmeyer
Company web-site: http://www.mhsc.com
___________________________________________________
KISS ... gotta love it!
