On Tue, 18 Jul 2000 [EMAIL PROTECTED] wrote:
> rfc-1123 6.1.3.4:
>
> When the host name-to-address function encounters a host
> with multiple addresses, it SHOULD rank or sort the
> addresses using knowledge of the immediately connected
> network number(s) and any other applicable performance or
> history information.
>
> Ok, it is only a SHOULD not a MUST but clearly this is the
> Right Thing. Of course this isn't RH's fault if glibc does
> the Wrong Thing (TM).
>
glibc had the code for sortlist a long time ago. It's strait out of
BIND. RedHat just didn't compile it in as it was considered experimental
at the time I reported this.
Apparently it's no longer experimental code.
--Mark
==========================================================================
Mark A Bentley -- Web Page: http://www.cs.umn.edu/~bentlema/
CSE/ITLabs Systems Staff -- Web Page: http://www.cs.umn.edu/systems-staff/
University of Minnesota -- EE/CSci Building -- Minneapolis (East Bank)
_______________________________________________
Redhat-devel-list mailing list
[EMAIL PROTECTED]
https://listman.redhat.com/mailman/listinfo/redhat-devel-list
