Re: hop-count header

--MimeMultipartBoundary
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

Oskar Pearson wrote:
=20
> We have 3 caches - currently they all talk to one another with the 'pro=
xy-only'
> flag. If someone chose not to use this flag, though, couldn't you end
> up with a forwarding loop? (especially without this flag)

A hop-count is not a viable way to prevent forwarding loops. This is
much better done by "X-Forwarded-By" and to not cache objects that
already has passed by us one time before... (isn't this already there?
the detection is there at least...)

I am more worried about the situation that ICP choses a parent/neigbour,
but when we fetch the object our policy denies caching of the object..

---
Henrik Nordstr=F6m
--MimeMultipartBoundary--