Commit Graph

125 Commits

Author SHA1 Message Date
Bjoern A. Zeeb
0e81959d3a Conditionally add the interface name and address if available
so that a ppp running in `receiver' (server) mode can properly
update routes, for example to update the MTU.

Submitted by:	loos.br gmail.com (Luiz Otavio O Souza)
PR:		bin/130159
PR:		kern/125079, kern/122068, bin/126892
MFC after:	3 days
2009-04-20 14:38:48 +00:00
Bjoern A. Zeeb
0fb550caba While we currently still seem to have a gateway address, for future
stability, check for gw to be set before adding the flag and the address
to the routing message.

r186308, backed out in r191305, already tried to do that, and in addition
ignore AF_LINK types of gateway addresses to work around a problem that
r167797 had introduced on the kernel side always setting RTF_GATEWAY if a
gateway address was passed into the kernel.
The proper solution for this is still under discussion so I am hesitant to
re-add the special AF_LINK treatment for now.

MFC after:	3 days
2009-04-20 11:51:35 +00:00
Bjoern A. Zeeb
16f37cbd72 Back out r186308:
in case of AF_LINK, which the kernel still returns for an RTAX_GATEWAY
as an empty sockaddr_dl in the classic tun<n> case.
Copying the address into the message payload, but not the RTA_GATEWAY
flag results in rt_xaddrs() in the kernel tripping over that and parsing
the next attribute set with a flag, i.e. RTA_NETMASK, with the gateway
address, resulting in bogus route entry.

MFC after:	3 days
2009-04-20 11:22:51 +00:00
Qing Li
e657c679e1 The ppp application relies on the if_tun interface to properly
install a p2p host route between the end points. The ppp module
upates this router based on user configuration later on. The
rt_Update() seems to always set the RTF_GATEWAY flag, which is
broken.
2008-12-19 01:37:20 +00:00
Qing Li
6e6b3f7cbc This main goals of this project are:
1. separating L2 tables (ARP, NDP) from the L3 routing tables
2. removing as much locking dependencies among these layers as
   possible to allow for some parallelism in the search operations
3. simplify the logic in the routing code,

The most notable end result is the obsolescent of the route
cloning (RTF_CLONING) concept, which translated into code reduction
in both IPv4 ARP and IPv6 NDP related modules, and size reduction in
struct rtentry{}. The change in design obsoletes the semantics of
RTF_CLONING, RTF_WASCLONE and RTF_LLINFO routing flags. The userland
applications such as "arp" and "ndp" have been modified to reflect
those changes. The output from "netstat -r" shows only the routing
entries.

Quite a few developers have contributed to this project in the
past: Glebius Smirnoff, Luigi Rizzo, Alessandro Cerri, and
Andre Oppermann. And most recently:

- Kip Macy revised the locking code completely, thus completing
  the last piece of the puzzle, Kip has also been conducting
  active functional testing
- Sam Leffler has helped me improving/refactoring the code, and
  provided valuable reviews
- Julian Elischer setup the perforce tree for me and has helped
  me maintaining that branch before the svn conversion
2008-12-15 06:10:57 +00:00
Brian Somers
00d9db0314 Cast unsigned variables to int 2005-01-10 11:12:36 +00:00
Brian Somers
5d604c1161 Add a bunch of malloc() return checks
PR:		71592
Submitted by:	Dan Lukes <dan@obluda.cz> with further changes
2004-10-11 09:45:58 +00:00
Brian Somers
057f1760a8 Make ppp WARNS=5 clean 2004-09-05 01:46:52 +00:00
Hajimu UMEMOTO
93193fc76c We need filling scopeid to install routes for link-local
scope addresses.
2003-03-25 16:49:08 +00:00
Brian Somers
6eafd35305 Include the correct file (stdarg.h) and use va_list rather than _BSD_VA_LIST_
Suggested by: mike
2002-08-27 20:11:58 +00:00
Brian Somers
de59e178aa o Clean up some #includes
o Bump version number to 3.0.4
o When talking to a RADIUS server, provide a NAS-Port-Type.

  When the NAS-Port-Type is Ethernet, provide a NAS-Port value equal
  to the SESSIONID from the environment in direct mode or the
  NGM_PPPOE_SESSIONID message in other modes.  If no SESSIONID is found,
  default to the interface index in client mode or zero in server mode.

  When the NAS-Port-Type is ISDN, set the NAS-Port to the minor number
  of the physical device (ie, the N in /dev/i4brbchN).

  This makes it easier for the RADIUS server to identify the client
  WRT accounting data etc.

Prompted by:	lsz8425 <lsz8425@mail.cd.hn.cn>
2002-05-14 12:55:39 +00:00
Brian Somers
3089715715 Remove some extraneous whitespace 2002-03-31 01:51:15 +00:00
Brian Somers
afb28bf732 Don't use the RTF_GATEWAY flag and pass the RTA_GATEWAY address when updating
routes.
2002-03-30 12:50:49 +00:00
Brian Somers
fb11a9c23d Merge the NETGRAPH branch into HEAD. tty devices now use netgraph's line
discipline to do the async escaping, but no other benefits are available yet.

Change ``ifdef HAVE_DES'' to ``ifndef NODES'' for consistency.

Make the Makefile a little more sane WRT RELEASE_CRUNCH.
2002-03-30 12:30:09 +00:00
Brian Somers
63c6cac940 socket()s first argument should be a protocol family rather than an
address family.
2002-01-16 14:03:52 +00:00
Brian Somers
c87436f341 socket's first argument is an address family, not a protocol family. 2002-01-16 13:15:47 +00:00
Brian Somers
c4a913b6b8 Don't provide an RTA_GATEWAY sockaddr when we write RTM_CHANGE messages
to the routing socket.

The local address on a point-to-point interface is not actually a
gateway address - despite it appearing in the second column of
netstat -r's output.  Providing a gateway to an RTM_CHANGE will
currently change the route's interface so that it's using the
specified gateway - not what we want.

Patiently explained to me by:	ru
2001-11-30 14:01:20 +00:00
Brian Somers
0ed9882caa Remove the hack that ensures that rt_Update() works on FreeBSD. Now
that the ncpaddr code doesn't create default routes with non-zero
masks, everything works as it should.
2001-11-23 19:20:45 +00:00
Brian Somers
361a7b933f When writing messages to the routing socket, round sockaddr sizes
up in the same way that we expect them to be when we read them.

This is a no-op on i386 and probably on alphas, as we currently
only support AF_INET and AF_INET6.
2001-11-23 17:19:36 +00:00
Brian Somers
a3d71c3de1 Be paranoid about non-zero netmasks being associated with INET addresses
of 0.0.0.0.

The OpenBSD PF_ROUTE/NET_RT_DUMP sysctl is sending back routes with
RTAX_NETMASK set, but the corresponding sockaddr being 4 zero bytes
(with an address family of zero).  ppp was getting confused by this
and ending up interpreting it as a 0.0.0.0/32 routing table
destination and subsequently failing to do anything with the route.

Specifically, after this fix, ppp under OpenBSD can successfully
change and delete the default route again !
2001-11-23 17:19:27 +00:00
Brian Somers
846e7227be Remove an unused variable (oops) 2001-11-23 15:47:04 +00:00
Brian Somers
26dceef0a6 Don't adjust_linklocal() when pulling a sockaddr out of an ncpaddr or
ncprange structure.

Don't write() the netmask for IPv6 sockaddrs to the routing socket if
the prefixlen is 128.

It seems that messages written to the routing socket with the scopeid
set for link local addresses are not understood.  Instead, we have to
put the scopeid in the 5th and 6th bytes of the address (see
adjust_linklocal() in ncpaddr.c).  I think this may be a bug in the
KAME implementation - it should really understand both forms.
2001-11-23 12:39:20 +00:00
Brian Somers
156a04e466 Fix a diagnostic 2001-09-04 22:07:56 +00:00
Brian Somers
d42d9220c7 When attempting to change the default route, don't write the gateway
and mask to the routing socket, otherwise the update fails.

Warning provided by: markm

The code here was broken for FreeBSD when IPv6 support was added, but
was fixed for OpenBSD.  OpenBSD expects the gateway and mask to be
supplied and fails the update otherwise.
2001-08-20 00:46:33 +00:00
Brian Somers
1d3d76d47f Build properly with -DNOINET6 2001-08-18 12:31:54 +00:00
Brian Somers
5ae083292b Specify the gateway address when updating the MTU and send/recv pipe
sizes on a route.

IMHO this shouldn't be necessary (the destination & mask/prefixlen
should be enough), but without it, the default route update under
OpenBSD will fail.

Thanks to: Russell T Hunt <alaric@MIT.EDU>
2001-08-16 02:01:05 +00:00
Brian Somers
bcd13e2e67 Don't forget to attache the sticky route change to the ncp structure.
PR:	29726
2001-08-15 12:47:51 +00:00
Brian Somers
30949fd4b5 o Add ipv6 support, abstracting most NCP addresses into opaque
structures (well, they're treated as opaque).

  It's now possible to manage IPv6 interface addresses and routing
  table entries and to filter IPV6 traffic whether encapsulated or
  not.

  IPV6CP support is crude for now, and hasn't been tested against
  any other implementations.

  RADIUS and IPv6 are independent of eachother for now.

  ppp.linkup/ppp.linkdown aren't currently used by IPV6CP

o Understand all protocols(5) in filter rules rather than only a select
  few.

o Allow a mask specification for the ``delete'' command.  It's now
  possible to specifically delete one of two conflicting routes.

o When creating and deleting proxy arp entries, do it for all IPv4
  interface addresses rather than doing it just for the ``current''
  peer address.

o When iface-alias isn't in effect, don't blow away manually (via ``iface
  add'') added interface addresses.

o When listening on a tcp server (diagnostic) socket, bind so that a
  tcp46 socket is created -- allowing both IPv4 and IPv6 connections.

o When displaying ICMP traffic, don't display the icmp type twice.
  When display traffic, display at least some information about unrecognised
  traffic.

o Bump version

Inspired after filtering work by: Makoto MATSUSHITA <matusita@jp.FreeBSD.org>
2001-08-14 16:05:52 +00:00
Brian Somers
c8b9fb53a1 Handle hardware-imposed MTU/MRU limitations. PPPoE will no longer
allow MRU/MTU negotiations to exceed 1492.

Add an optional ``max'' specifier to ``set m[rt]u'', ie.

	set mtu max 1480

Bump the ppp version number.

Sponsored by:	Monzoon Networks AG and FreeBSD Services Limited
2001-06-18 14:59:36 +00:00
Brian Somers
65309e5cda Convert IIJ copyrights to BSD copyrights.
Approved by: Toshiharu OHNO <tohno@sirius.ocn.ne.jp>
2001-06-13 21:52:19 +00:00
Brian Somers
03a2501aa5 When we change the interface MTU, run through the routing table and tweak
all route MTUs too.
2001-04-05 02:23:48 +00:00
Brian Somers
c837ca39c6 Don't delete sticky routes when unconfiguring the interface.
Deleting all routes that match the interface is sufficient.
2000-08-14 08:43:40 +00:00
Brian Somers
759c81cbad Don't mis-interpret sockaddr_in6 structures as sockaddr_in structures
when purging routes.
2000-07-20 10:38:12 +00:00
Brian Somers
d568d6c405 Add the ``resolv'' command for telling ppp how to deal with resolv.conf.
You can now ``resolv restore'' in ppp.linkdown !
Add DNS0 and DNS1 macros.
2000-03-14 01:47:27 +00:00
Brian Somers
b9fa960d93 Don't include netinet6/in6.h directly. 2000-03-14 01:47:09 +00:00
Brian Somers
24731321e6 Don't ``break'' when we come accross a non-RTM_IFINFO type
entry in the block returned by the NET_RT_IFLIST mib,
``continue''

Broken a few minutes ago by: me
2000-03-14 01:46:56 +00:00
Brian Somers
d6d3eeab46 When ppp can't identify the relevant name, don't use "???", use
<nnn> or <0xxxx> instead.
2000-03-14 01:46:54 +00:00
Brian Somers
6b4579784c The interface list that comes back from the PF_ROUTE/NET_RT_IFLIST mib
is aligned.  Teach this to ``show route''.

Clean up some of the sockaddr parsing routines.
2000-03-14 01:46:44 +00:00
Brian Somers
6b38466448 Support IP6 addresses in ``show route''
Also, don't try to output routing entries if either the
RTA_DST or RTA_GATEWAY sockaddrs aren't present.
2000-03-14 01:46:38 +00:00
Brian Somers
b9391689ee Back out the bogus #ifdef __NetBSD__ #include <signal.h> lines.
The original report was due to a mis-installation of the NetBS
header files :-/

Submitted by:	 Kazuyoshi Kato <kazk@yyy.or.jp>
1999-09-21 19:37:00 +00:00
Brian Somers
7e795ebe38 NetBSD has moved ``extern int errno;'' to signal.h :-/
Submitted by:	Kazuyoshi Kato <kazk@yyy.or.jp>
1999-09-20 07:36:46 +00:00
Peter Wemm
97d92980a9 $Id$ -> $FreeBSD$ 1999-08-28 01:35:59 +00:00
Brian Somers
8473f372df Rebuild the list of interface numbers to names if we're trying
to evaluate an interface number that didn't previously exist.
1999-08-26 20:01:20 +00:00
Brian Somers
5d9e610366 o Redesign the layering mechanism and make the aliasing code part of
the layering.

  We now ``stack'' layers as soon as we open the device (when we figure
  out what we're dealing with).  A static set of `dispatch' routines are
  also declared for dealing with incoming packets after they've been
  `pulled' up through the stacked layers.

  Physical devices are now assigned handlers based on the device type
  when they're opened.  For the moment there are three device types;
  ttys, execs and tcps.

o Increment version number to 2.2
o Make an entry in [uw]tmp for non-tty -direct invocations (after
  pap/chap authentication).
o Make throughput counters quad_t's
o Account for the absolute number of mbuf malloc()s and free()s in
  ``show mem''.
o ``show modem'' becomes ``show physical''.
1999-05-08 11:07:56 +00:00
Brian Somers
972a1bcf5d Initial RADIUS support (using libradius). See the man page for
details.  Compiling with -DNORADIUS (the default for `release')
removes support.

TODO: The functionality in libradius::rad_send_request() needs
      to be supplied as a set of routines so that ppp doesn't
      have to wait indefinitely for the radius server(s).  Instead,
      we need to get a descriptor back, select() on the descriptor,
      and ask libradius to service it when necessary.
      For now, ppp blocks SIGALRM while in rad_send_request(), so
      it misses PAP/CHAP retries & timeouts if they occur.

      Only PAP is functional.  When CHAP is attempted, libradius
      complains that no User-Password has been specified... rfc2138
      says that it *mustn't* be used for CHAP :-(

Sponsored by: Internet Business Solutions Ltd., Switzerland
1999-01-28 01:56:34 +00:00
Brian Somers
8fa6ebe47d Solve the ``first connection'' problem that occurs on
demand-dial links with dynamic IP numbers where the program
that causes the dial bind()s to an interface address that is
subsequently changed after ppp negotiation.

The problem is defeated by adding negotiated addresses to the
tun interface as additional alias addresses and providing a set
of ``iface'' commands for managing the interface.  Libalias is
also required (and what a name clash!) - it happily IP-aliases
the address so that the source is that of the primary (negotiated)
interface and un-IP-aliases it on the way back.

An ``enable iface-alias'' is done implicitly by the -alias command
line switch.  If -alias isn't given, iface-aliasing is disabled by
default and can't be enabled 'till an ``alias enable yes'' is done.
``alias enable no'' silently disables iface-alias.

So, for dynamic-IP-type-connections, running ``ppp -alias -auto blah''
will work for the first connection, although existing bindings will
not survive a disconnect/connect as the TCP peer will be trying to
send to the old IP address - the packets won't route.

It's now a lot easier to add IPXCP to ppp with minor updates to
the new iface.[ch] (if anyone ever gets 'round to it).

It's also now possible to manually add interface aliases with
something like ``iface add 1.2.3.4/24 5.6.7.8''.  This allows
multi-homed ppp links :-)
1998-10-22 02:32:50 +00:00
Brian Somers
fc254be9bd Don't lose an allocated pointer if realloc() fails.
Free it instead.
Pointed out by: Theo de Raadt
1998-08-17 06:42:40 +00:00
Brian Somers
2062443fd7 Silence ``Network unreachable'' warnings when using
``add .... HISADDR''.  The network will never be
reachable at this point unless we're in -auto or reading
the command from ppp.linkup.

We can now run the following lines and get the expected
results:

  set ifaddr 1.2.3.4/0 5.6.7.8/0
  add default HISADDR

where a route is added immediately in auto mode and the
whole thing is delayed 'till the IP numbers have been
agreed in other modes.

Essentially, ppp.linkup is no longer required.
1998-07-28 21:54:54 +00:00
Brian Somers
3a2e4f621c o Fix remaining sizeof problems for 64 bit machines.
o Allow ``set ....'' when we have multiple links but aren't in
  multilink mode.
o Do a TLS when we receive a ``Open'' event in ``Closed'' state,
  despite the rfc state transition table.  This is clearly an
  error in the RFC as TLS cannot have yet been called (without
  TLF) in the ``Closed'' state.
  I've posted a message to comp.protocols.ppp for confirmation.
1998-06-27 23:48:54 +00:00
Brian Somers
3f06c5995f Don't assume ``sizeof(u_long) == 4''
Submitted by: Theo
1998-06-27 12:03:50 +00:00