[Bug libc/12377] getaddrinfo() should disregard link-local IPv6 addresses for AI_ADDRCONFIG purposes

["tore at fud dot no" <sourceware-bugzilla@sourceware.org>]

http://sourceware.org/bugzilla/show_bug.cgi?id=12377

--- Comment #8 from Tore Anderson <tore at fud dot no> 2011-07-09 11:28:55 UTC ---
(In reply to comment #7)
> (In reply to comment #6)
> 
> > The %ifname convention has nothing to do with getaddrinfo(), it's interpreted
> > by the application. When passing it to getaddrinfo() you'll get this result:
> 
> When did that change, or is that only for IP addresses passed-in to the call? 
> I was running netperf tests over IPv6 passing-in the likes of
> "fe80::207:43ff:fe05:5904%eth2" and I know that all netperf was doing was
> passing that to getaddrinfo() and getting scope that way (rhel5.mumble, sles10,
> sles11 sp1, lenny, and solaris 10)

I don't know. It's probably only if you're passing it IP literals - which
doesn't appear to generate any (reverse) DNS lookups in any case.

If I'm not mistaken this is not really relevant to the problem that I want to
see solved, i.e., making AI_ADDRCONFIG suppress unnecessary AAAA DNS lookups
from hosts that have no IPv6 connectivity. There are no shortage of DNS servers
out there that do not understand AAAA lookups, which in turn causes
getaddrinfo() to stall until a timeout has been reached, leading to a abysmal
end-user experience compared to the other major operating systems. The number
of comments and duplicates for bug reports such as
https://bugs.launchpad.net/ubuntu/+source/eglibc/+bug/417757 clearly shows that
this is a major problem.

Tore

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.