https://bz.apache.org/bugzilla/show_bug.cgi?id=53099

--- Comment #4 from William A. Rowe Jr. <[email protected]> ---
This is a good suggestion, but I think we should take this one step further...

RemoteIPInternalHeader X-RemoteIP
RemoteIPHeader X-Forwarded-For, X-Forward

In the absence of an RemoteIPInternalHeader list, the RemoteIPHeader list would
be used instead to unwind the forwarded addresses for RemoteIPInternalProxy
members. When the RemoteIPInternalHeader list is exhausted, the
RemoteIPInternalProxy list would continue to be honored, but treated as trusted
external proxies, for purposes of dis-honoring private IP addresses. This
should ensure maximum compatibility with existing configurations and offer the
fewest unintended consequences of a misconfiguration.

It's unclear to me if https://tools.ietf.org/html/rfc7239 will see widespread
adoption and how it would interact with older conventions.

The header "Forwarded" could be treated as a special-case, supporting the new
RFC, but deciphered in the order it appears.  (I imagine it would typically be
first where the immediate proxies which are trusted have adopted the
convention).

Sadly, https://tools.ietf.org/html/rfc7239#section-7.4 is less than helpful on
this subject :)

-- 
You are receiving this mail because:
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to