> > They are two virtual hosts on the same server (Apache
> > 2.0.46), same IP address (http/1.1 to distinguish by host
> > header), and under "Caching" in the CFMX Administrator we
> > have deselected "Cache web server paths".
>
> That's strange because, as far as I know, calling the CFC with
> "application.objects.crossover_test" isn't suppose to work, since CF
> Server
> has no notion of virtual server and will start looking for the CFC
from
> its
> root (I assume in your case CF root is D:\web\)
>
> Check a recent thread on this list:
>
> http://www.mail-archive.com/[EMAIL PROTECTED]/msg02677.html
I'm not sure about this one - I was told specifically by one the CF
Engineers that this is not the case.
It's not the CF root (the mapped slash) that's searched, it's the _WEB_
Root - and since every virtual server has a web root it can (or at least
should) find them that way.
>From the perspective of CFCs the mapped slash ("root") is just another
mapping. The web root is searched AFTER the server mappings (but
before Custom Tag mappings for some reason). So, considering our
hosting problem it SHOULD work such that if a host NEVER maps "com" then
anybody with a virtual server could use "com.mydomain.cfc".
The (perhaps assumed or pre-6.1 after these tests) problem was always
that if TWO people had the same directory structure/cfc name (as could
be the case with distributed/marketed applications) only one instance
would be found and that path cached. The wrong CFC would then be called
in the second instance.
Ben's tests seem to indicate that this is either no longer the case, or
that the behavior is more complex than we thought. I have a hope that
it's been fixed in 6.1, but I have a fear it's just more complex.
I truly think that CFC adoption would GREATLY benefit a qualified set of
recommendations to hosts. Although many people are on dedicated boxes
many, many more are on shared servers and many others are shunning CFCs
because they're building apps for possible/eventual resale/distribution.
A solid recommendation from Macromedia (or, lacking that, a solid
accepted community recommendation) on the matter would be greatly
appreciated. I get the feeling that hosts like CrystalTech are truly
trying to give their customers as much control as possible, but are just
totally confused by the issue.
> I am sorry for not being able to help you more, I can only share my
horror
> story and Jim did the same.
>
> Last winter I met Ben Forta on a CFUG event, I was discovering the
pain of
> using CFCs on shared hosts so, among the other things, I asked about
this
> too. He said there was the risk of CF server caching the location of
the
> CFC, just like it happens for the <cf_myTag> notation and custom tags.
>
> It would be nice to hear from a MM insider and see if they can confirm
it
> or
> not...
I'll see if I can get an answer on that (it not be for a while -
everybody I know that could find out is going to Max... and I'm not).
The real problem is that even the CF engineers seem more than a little
confused about how this REALLY works. ;^)
Jim Davis
----------------------------------------------------------
You are subscribed to cfcdev. To unsubscribe, send an email
to [EMAIL PROTECTED] with the word 'unsubscribe cfcdev'
in the message of the email.
CFCDev is run by CFCZone (www.cfczone.org) and supported
by Mindtool, Corporation (www.mindtool.com).
An archive of the CFCDev list is available at www.mail-archive.com/[EMAIL PROTECTED]
