.\"      $OpenBSD: ipnat.4,v 1.17 2000/01/24 07:27:02 kjell Exp $
.Dd June 5, 1999
.Dt IPNAT 4
.Os
.Sh NAME
.Nm ipnat
.Nd Network Address Translation kernel interface
.Sh SYNOPSIS
.Fd #include <netinet/ip_fil_compat.h>
.Fd #include <netinet/ip_fil.h>
.Fd #include <netinets/ip_proxy.h>
.Fd #include <netinet/ip_nat.h>
.Sh DESCRIPTION
Unlike
.Xr ipf 4 ,
only a single list is supported by the kernel NAT
interface. An inactive list which can be swapped to is not currently
supported.
.Pp
.Pp
To add/delete rules to/from the NAT list, two
.Dq basic
ioctl's are provided:
.Bd -literal -offset indent
.Fn ioctl fd SIOCADNAT "struct ipnat *"
.Fn ioctl fd SIOCRMNAT "struct ipnat *"
.Ed
.Pp
To retrieve NAT statistics
.Bd -literal -offset indent
.Fn ioctl fd SIOCGNATS "struct natstat *"
.Ed
.Pp
is provided.
.Pp
These ioctl's are implemented as routing ioctl's, so the rules
for routing ioctl's and the file descriptor
.Ar fd
must be followed.
The most important rule is that the file descriptor
.Fa fd
must be for the device associated with the module (i.e.,
.Pa /dev/ipl ) .
.Pp
The structure
.Fa ipnat
and associated macros are defined as:
.Bd -literal -offset indent
typedef	struct	ipnat	{
	struct	ipnat	*in_next;
	void	*in_ifp;
	void	*in_apr;
	u_int	in_space;
	u_int	in_use;
	struct	in_addr	in_nextip;
	u_short	in_pnext;
	u_short	in_flags;
	u_short	in_port[2];
	struct	in_addr	in_in[2];
	struct	in_addr	in_out[2];
	int	in_redir;
	char	in_ifname[IFNAMSIZ];
	char	in_plabel[APR_LABELLEN];
	char	in_p;
	u_short	in_dport;
} ipnat_t;

#define	in_pmin		in_port[0]
#define	in_pmax		in_port[1]
#define	in_nip		in_nextip.s_addr
#define	in_inip		in_in[0].s_addr
#define	in_inmsk	in_in[1].s_addr
#define	in_outip	in_out[0].s_addr
#define	in_outmsk	in_out[1].s_addr
.Ed
.Pp
Where recognised values for
.Fa in_redir
are:
.Bd -literal -offset indent
#define	NAT_MAP		0x01
#define	NAT_REDIRECT	0x02
#define	NAT_BIMAP	(NAT_MAP|NAT_REDIRECT)
.Ed
.Pp
The structure
.Fa natstat
is defined as:
.Bd -literal -offset indent
typedef	struct	natstat	{
	u_long	ns_mapped[2];
	u_long	ns_rules;
	u_long	ns_added;
	u_long	ns_expire;
	u_long	ns_inuse;
	u_long	ns_logged;
	u_long	ns_logfail;
	nat_t	**ns_table[2];
	ipnat_t	*ns_list;
	void	*ns_apslist;
} natstat_t;
.Ed
.Pp
The NAT kernel tables are hash tables of size
.Dv NAT_SIZE
(default is 367).
.Sh FILES
.Bl -tag -width /dev/ipnat -compact
.It Pa /dev/ipnat
.El
.Sh SEE ALSO
.Xr ipf 8 ,
.Xr ipftest 1 ,
.Xr ipf 4 ,
.Xr ipl 4 ,
.Xr ipf 5 ,
.Xr ipnat 5 ,
.Xr ipnat 8 ,
.Xr ipfstat 8 ,
.Xr ipmon 8
.Pp
http://coombs.anu.edu.au/~avalon
.Sh BUGS
It would be nice if there were more flexibility when adding and deleting
filter rules.