Already I can tell you that our identd server is safer to run. Really.
> Most people/single-user systems need ident running for IRC, typically to
> comply with the IRC server policy, or to remove the tilde sign from their
> username. Running a full-blown ident daemon, both bundled with the OS, or
> the IRC client, for this purpose seems overkill to me. A mate used Perl for
> this and out spawned some C code from the idea:
>
> $ cat /etc/inetd.conf | grep fakeident
> auth stream tcp nowait nobody /usr/local/bin/fakeident fakeident
>
> $ cat /usr/local/bin/fakeident
> #! /usr/local/bin/perl
> $in = <STDIN>;
> ($p1, $p2) = split(/\,/, $in);
> printf("%d , %d : USERID : UNIX : unix\n", $p1, $p2);
>
>
>
> $ #A simple way to do this in C:
> $ cat fakeident.c
> #include <stdio.h>
>
> main()
> {
> int u1, u2;
> char buff[512];
> fgets(buff,511,stdin);
> sscanf(buff,"%d,%d",&u1,&u2);
> printf("%d , %d : USERID : UNIX : unix\n",u1,u2);
> }
> /*EOF*/
>
>
> Since a lot of users just need to send a single auth/ident response to IRC
> servers, and only need like an alternative/fake ident daemon, invoked using
> inetd. This could be improved(?) by being a standalone program that binds to
> port 113.
> Most IRC clients come bundled with an ident daemon, but this is what I like,
> and use at my perimeter router on a NAT LAN.
> I'm no coder, and maybe this can be improved, and I'm open to constructive
> criticism, but is something like this a candidate for a new port?
>
>
>
> - S
>
