mirror/freebsd
1
0
Fork 0
mirror of https://git.FreeBSD.org/src.git synced 2024-12-21 11:13:30 +00:00
Code Issues Releases Activity
b827692436
freebsd/usr.sbin/ppp/lcp.c

1292 lines
39 KiB
C
Raw Normal View History

Convert IIJ copyrights to BSD copyrights. Approved by: Toshiharu OHNO <tohno@sirius.ocn.ne.jp>
2001-06-13 21:52:19 +00:00
/*-
* Copyright (c) 1996 - 2001 Brian Somers <brian@Awfulhak.org>
* based on work by Toshiharu OHNO <tony-o@iij.ad.jp>
* Internet Initiative Japan, Inc (IIJ)
* All rights reserved.
1995-01-31 06:29:58 +00:00
*
Convert IIJ copyrights to BSD copyrights. Approved by: Toshiharu OHNO <tohno@sirius.ocn.ne.jp>
2001-06-13 21:52:19 +00:00
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
1995-01-31 06:29:58 +00:00
*
Convert IIJ copyrights to BSD copyrights. Approved by: Toshiharu OHNO <tohno@sirius.ocn.ne.jp>
2001-06-13 21:52:19 +00:00
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
Remove trailing whitespace.
1995-05-30 03:57:47 +00:00
*
$Id$ -> $FreeBSD$
1999-08-28 01:35:59 +00:00
* $FreeBSD$
1995-01-31 06:29:58 +00:00
*/
This commit was manufactured by cvs2svn to create branch 'MP'.
1998-01-29 00:44:16 +00:00
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
#include <sys/param.h>
Remove unused #includes. Make various bits static. Remove unused variables. Submitted by: eivind
1998-01-21 02:15:33 +00:00
#include <netinet/in.h>
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
#include <netinet/in_systm.h>
#include <netinet/ip.h>
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
#include <sys/socket.h>
o Add the link name to modem diagnostics. o Create struct mpserver as part of struct mp. mpserver creates a unix-domain socket based on the peers auth name and endpoint discriminator. If it already exists, ppp will ``pass the link'' over to the owner of the socket, joining it into the bundle of another ppp invocation, otherwise ppp waits for other invocations to pass it links through this socket. The final piece of code will be the code that flattens our datalink info and passes it down this channel (not yet implemented).
1998-04-28 01:25:46 +00:00
#include <sys/un.h>
Cosmetic (no functional changes): o Add missing $Id$s o Move extern decls from .c -> .h files o Staticize o Remove #includes from .h files o style(9)ify includes o bcopy -> memcpy bzero -> memset bcmp -> memcmp index -> strchr rindex -> strrchr o Move timeout.h -> timer.h (making it consistent w/ timer.c) o Add -Wmissing-prototypes
1997-10-26 01:04:02 +00:00
Don't pause for one second before starting LCP negotiation. Instead, incrementally pause after receiving LCPs with the same magic. We can now suffer a server that waits more than 1 second before responding. Pauses greater than a second get hopelessly confusing as when the server eventually starts, it sees a flood of Config Requests followed by config NAKs and changes of magic. This causes the server to change its magic over and over.....
1997-09-22 02:51:24 +00:00
#include <signal.h>
Cosmetic (no functional changes): o Add missing $Id$s o Move extern decls from .c -> .h files o Staticize o Remove #includes from .h files o style(9)ify includes o bcopy -> memcpy bzero -> memset bcmp -> memcmp index -> strchr rindex -> strrchr o Move timeout.h -> timer.h (making it consistent w/ timer.c) o Add -Wmissing-prototypes
1997-10-26 01:04:02 +00:00
#include <stdio.h>
#include <stdlib.h>
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
#include <string.h>
Cosmetic (no functional changes): o Add missing $Id$s o Move extern decls from .c -> .h files o Staticize o Remove #includes from .h files o style(9)ify includes o bcopy -> memcpy bzero -> memset bcmp -> memcmp index -> strchr rindex -> strrchr o Move timeout.h -> timer.h (making it consistent w/ timer.c) o Add -Wmissing-prototypes
1997-10-26 01:04:02 +00:00
#include <termios.h>
#include <unistd.h>
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
#include "layer.h"
Don't cast potentially unaligned addresses to pointers to non-char types on non-i386 architectures. On Alpha and Sparc we get a bus error if we do.
1998-09-04 18:26:00 +00:00
#include "ua.h"
Fix a rather nasty use of `static'. This caused a SEGV when running ``link * load label'' as we ended up recursing back into command_Interpret after nuking our command arg list.
1998-06-15 19:06:25 +00:00
#include "defs.h"
Fix prototypes. Remove extraneous decls. Add ``const'' to several places. Allow ``make NOALIAS=1'' to remove IP aliasing. Merge with OpenBSD - only the Makefiles vary. We can now survive a compile with -Wall -Wbad-function-cast -Wcast-align -Wcast-qual -Winline -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wpointer-arith -Wredundant-decls -Wshadow -Wstrict-prototypes -Wwrite-strings -Wchar-subscripts (although the Makefile just contains -Wall).
1997-11-22 03:37:54 +00:00
#include "command.h"
Cosmetic (no functional changes): o Add missing $Id$s o Move extern decls from .c -> .h files o Staticize o Remove #includes from .h files o style(9)ify includes o bcopy -> memcpy bzero -> memset bcmp -> memcmp index -> strchr rindex -> strrchr o Move timeout.h -> timer.h (making it consistent w/ timer.c) o Add -Wmissing-prototypes
1997-10-26 01:04:02 +00:00
#include "mbuf.h"
#include "log.h"
#include "timer.h"
1995-01-31 06:29:58 +00:00
#include "fsm.h"
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
#include "iplist.h"
o Move global/static IPCP data into IpcpInfo. o Use INADDR_ANY and INADDR_BROADCAST instead of 0 and 0xffffffff for IP addresses & masks. o Move struct compreq into ipcp.c (it's none of anyone elses business). o Rename MAX_STATES to MAX_VJ_STATES.
1998-01-29 23:11:44 +00:00
#include "throughput.h"
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
#include "proto.h"
Make struct bundle into a sort of `struct descriptor'. It does the fdsets/reads/writes for each of it's datalinks.
1998-03-20 19:48:28 +00:00
#include "descriptor.h"
De-staticise LQR information Increment OutPackets for any packet - not just LQRs MFC: o Fix a few comment typos. o Fix ``set timeout'' usage message and documentation. o Change ifOutPackets, ifOutOctets and ifOutLQRs to `u_int32_t's so that they wrap correctly. o Put the LQR in network byte order using the correct struct size (sizeof u_int32_t, not sizeof u_long). o Wrap LQR ECHO counters correctly. o Don't increment OutLQR count if the last LQR hasn't been replied to. o Initialise last received LQR in StartLqm. o Don't start the LQR timer if we're `disabled' and `accepted'. o Generate LQR responses when both sides are using a timer and we're not going to send our next LQR before the peers max timeout.
1998-03-13 00:44:26 +00:00
#include "lqr.h"
1995-01-31 06:29:58 +00:00
#include "hdlc.h"
Support link identification from rfc1570 Two new commands are available; ``ident'' and ``sendident''.
2000-07-19 02:10:35 +00:00
#include "lcp.h"
Final cleanup for now. -Wall is now silent. A couple of bogons found.
1996-01-11 17:48:59 +00:00
#include "ccp.h"
This commit was manufactured by cvs2svn to create branch 'MP'.
1998-01-29 00:44:16 +00:00
#include "async.h"
Create `struct link' - the logical link from which `struct physical' (and the future `struct logical') are derived.
1998-01-30 19:46:07 +00:00
#include "link.h"
Create a new MP branch for `multilink protocol'. Do lots of initial shuffling and grouping. Submitted by: Eivind Eklund <perhaps@yes.no>
1998-01-29 00:49:32 +00:00
#include "physical.h"
Move the terminal/diagnostic socket stuff out of main.c and into the new `prompt.c'. struct prompt is (of course) a `sort' of descriptor.
1998-02-10 03:23:50 +00:00
#include "prompt.h"
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
#include "slcompress.h"
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
#include "ncpaddr.h"
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
#include "ipcp.h"
#include "filter.h"
#include "mp.h"
o Defer setting up pap/chap based IP numbers & labels until after we've determined if we're going to join another ppp invocation. o Make ``show link'' show all link details, and ``show links'' just give a list of links and their current status. o Show our current label in ``show bundle''. o Allow link cloning and removal as soon as our MRRU is set. o Make ``show lcp'' require context as nothing will ever change in our MP LCP (it's auto-configured as per rfc1990). o Initialise our LQM owner in hdlc_Init(). o Store our endpoint discriminator and authentication name at both the datalink and multilink level and compare them when we've finished AUTHENTICATE phase and before entering NETWORK phase. If they don't match, close the link again. Display the information in the appropriate ``show'' command. o Initialise datalink::phone and datalink::fsmp.object properly when we're cloning the link. o Show which link we're passing LQRs on in our diagnostics. o Reject endpoint discriminator REQs at the logical multilink level. o Remove the rest of our CARRIER and LINK logging setup.
1998-04-24 19:16:15 +00:00
#include "chat.h"
New user Process PPP based on iij-ppp0.94beta2. o Supporting SYNC SIO device (But need a device driver) - add "set speed sync" o Fixing bug for Predictor-1 function. o Add new parameter that re-sent interval for set timeout commands. o Improving RTT (Round Trip Time) and reducing processor time. - Previous Timer service was using polling, and now using SIGALRM ;-) - A 0.94beta2 will not work correctly.... -- Follows are additinal feature not including 0.94beta2 o Support Proxy ARP - add "enable/disable proxy" commands o Marging common routine in CHAP/PAP. o Enhancing LCP/IPCP log information. o Support local Authfication connection on port 300x and tty. - You can set up pair of your "hostname -s" and password in ppp.secret. if either ppp.secret file nor your hostname line don't exist, It will notify a message and working as same as previous version.(Backword compatibility) - If you did set up them, It's allow connection but nothing to do except help and passwd command. - add "passwd yourpasswd" commands o Support afilter - keep Alive filter that a packet can send/receiving according to ifilter/ofilter but doesn't count it as preventing idle timer expires. - Same syntax of other filters. o Fixing bugs reported by current user for previous one. Thanks !! Reviewed by: Atsushi Murai (amurai@spec.co.jp)
1995-02-26 12:18:08 +00:00
#include "auth.h"
Cosmetic (no functional changes): o Add missing $Id$s o Move extern decls from .c -> .h files o Staticize o Remove #includes from .h files o style(9)ify includes o bcopy -> memcpy bzero -> memset bcmp -> memcmp index -> strchr rindex -> strrchr o Move timeout.h -> timer.h (making it consistent w/ timer.c) o Add -Wmissing-prototypes
1997-10-26 01:04:02 +00:00
#include "chap.h"
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
#include "cbcp.h"
o Defer setting up pap/chap based IP numbers & labels until after we've determined if we're going to join another ppp invocation. o Make ``show link'' show all link details, and ``show links'' just give a list of links and their current status. o Show our current label in ``show bundle''. o Allow link cloning and removal as soon as our MRRU is set. o Make ``show lcp'' require context as nothing will ever change in our MP LCP (it's auto-configured as per rfc1990). o Initialise our LQM owner in hdlc_Init(). o Store our endpoint discriminator and authentication name at both the datalink and multilink level and compare them when we've finished AUTHENTICATE phase and before entering NETWORK phase. If they don't match, close the link again. Display the information in the appropriate ``show'' command. o Initialise datalink::phone and datalink::fsmp.object properly when we're cloning the link. o Show which link we're passing LQRs on in our diagnostics. o Reject endpoint discriminator REQs at the logical multilink level. o Remove the rest of our CARRIER and LINK logging setup.
1998-04-24 19:16:15 +00:00
#include "datalink.h"
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
#ifndef NORADIUS
#include "radius.h"
#endif
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
#include "ipv6cp.h"
#include "ncp.h"
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
#include "bundle.h"
1995-01-31 06:29:58 +00:00
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
/* for received LQRs */
struct lqrreq {
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
struct fsm_opt_hdr hdr;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
u_short proto; /* Quality protocol */
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
u_int32_t period; /* Reporting interval */
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
};
o Create a new ``timer'' log level. This lets us ``set log debug'' without filling our filesystem/screen with junk that we don't really want to see. o change PHYS_STDIN to PHYS_DIRECT - we can handle incoming connections that aren't on STDIN_FILENO now. o Allow return values from our FSM LayerUp functions. If LayerUp() fails, the FSM does an immediate FsmDown() without calling the fsm_parent's Layer{Up,Down} functions. o Clear the close-on-exec flag of file descriptor 3 when executing chat programs so that our documented ability to communicate with /dev/tty via that descriptor works. Also document it as descriptor 3, not 4 :-O o Allow a ``rm'' command as an alias for ``remove''. o Fix the bind()/connect()/accept() calls made by the MP server. o Create bundle_SendDatalink() and bundle_ReceiveDatalink(). This allows `struct datalink's to flatten themselves, pass through a pipe (read: the eye of a needle !) and come alive at the other end. The donator then fork()s & exec()s pppmpipe, ``passing'' the connection to another ppp instance. *** PPP NOW TALKS MULTILINK :-))) *** Our link utilization is hideous, and lots of code needs tidying still. It's also probably riddled with bugs ! It's been tested against itself only, and has hung once, so confidence isn't high....
1998-04-30 23:53:56 +00:00
static int LcpLayerUp(struct fsm *);
Overhaul ppp: o Use syslog o Remove references to stdout/stderr (incl perror()) o Introduce VarTerm - the interactive terminal or zero o Allow "set timeout" to affect current session o Change "set debug" to "set log" o Allow "set log [+|-]flag" o Make MSEXT and PASSWDAUTH stuff the default o Move all #ifdef DEBUG stuff into the code - this shouldn't be too much overhead. It's now controlled with "set log +debug" o Add "set log command, debug, tun, warn, error, alert" o Remove cdefs.h, and assume an ansi compiler. o Improve all diagnostic output o Don't trap SIGSEGV o SIGHUP now terminates again (log files are controlled by syslog) o Call CloseModem() when changing devices o Fix parsing of third arg of "delete" I think this fixes the "magic is same" problems that some people have been experiencing. The man page is being rewritten. It'll follow soon.
1997-06-09 03:27:43 +00:00
static void LcpLayerDown(struct fsm *);
static void LcpLayerStart(struct fsm *);
static void LcpLayerFinish(struct fsm *);
Allow control over the number of ConfigREQ & TermREQ attempts that are made in each of the FSMs (LCP, CCP & IPCP) and the number of REQs/Challenges for PAP/CHAP by accepting more arguments in the ``set {c,ip,l}cpretry'' and ``set {ch,p}apretry'' commands. Change the non-convergence thresholds to 3 times the number of configured REQ tries (rather than the previous fixed ``10''). We now notice repeated NAKs and REJs rather than just REQs. Don't suggest that CHAP 0x05 isn't supported when it's not configured. Fix some bugs that expose themselves with smaller numbers of retries: o Handle instantaneous disconnects (set device /dev/null) correctly by stopping all fsm timers in fsm2initial. o Don't forget to uu_unlock() devices that are files but are not ttys (set device /dev/zero). Fix a *HORRENDOUS* bug in RFC1661 (already fixed for an Open event in state ``Closed''): According to the state transition table, a RCR+ or RCR- received in the ``Stopped'' state are supposed to InitRestartCounter, SendConfigReq and SendConfig{Ack,Nak}. However, in ``Stopped'', we haven't yet done a TLS (or the last thing we did is a TLF). We must therefore do the TLS at this point ! This was never noticed before because LCP and CCP used not use LayerStart() for anything interesting, and IPCP tends to go into Stopped then get a Down because of an LCP RTR rather than getting a RCR again.
1999-02-26 21:28:14 +00:00
static void LcpInitRestartCounter(struct fsm *, int);
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
static void LcpSendConfigReq(struct fsm *);
o Check the LCP Identifier field for correctness and drop dodgy packets by default. The old behaviour is still available with ``disable idcheck''. o Make all FSM diagnostics consistent and tidy up the way we build our LCP/CCP/IPCP requests. o Don't assume sizeof(u_long) == 4.
1998-03-20 19:47:10 +00:00
static void LcpSentTerminateReq(struct fsm *);
static void LcpSendTerminateAck(struct fsm *, u_char);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
static void LcpDecodeConfig(struct fsm *, u_char *, u_char *, int,
Remove more globals
1998-03-13 21:08:05 +00:00
struct fsm_decode *);
Isolate fsm callback functions Derive {lcp,ipcp,ccp}state structures from struct fsm where convenient. Create bundle_Down() and remove LcpDown()
1998-02-06 02:24:29 +00:00
static struct fsm_callbacks lcp_Callbacks = {
Create struct fsm_parent. FSMs are created with one of these, and the FSM passes subsequent events to them. The datalink now hides its CCP from the bundle layer.
1998-02-27 01:22:39 +00:00
LcpLayerUp,
LcpLayerDown,
LcpLayerStart,
LcpLayerFinish,
Isolate fsm callback functions Derive {lcp,ipcp,ccp}state structures from struct fsm where convenient. Create bundle_Down() and remove LcpDown()
1998-02-06 02:24:29 +00:00
LcpInitRestartCounter,
LcpSendConfigReq,
o Check the LCP Identifier field for correctness and drop dodgy packets by default. The old behaviour is still available with ``disable idcheck''. o Make all FSM diagnostics consistent and tidy up the way we build our LCP/CCP/IPCP requests. o Don't assume sizeof(u_long) == 4.
1998-03-20 19:47:10 +00:00
LcpSentTerminateReq,
Isolate fsm callback functions Derive {lcp,ipcp,ccp}state structures from struct fsm where convenient. Create bundle_Down() and remove LcpDown()
1998-02-06 02:24:29 +00:00
LcpSendTerminateAck,
Shuffle around our FSMs a bit. This'll make it easier to remove the CcpInfo, LcpInfo and IpcpInfo globals.
1998-02-21 01:45:26 +00:00
LcpDecodeConfig,
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
fsm_NullRecvResetReq,
fsm_NullRecvResetAck
Isolate fsm callback functions Derive {lcp,ipcp,ccp}state structures from struct fsm where convenient. Create bundle_Down() and remove LcpDown()
1998-02-06 02:24:29 +00:00
};
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
Add a bunch of `const's and fix a typo. Submitted by: Rich Neswold <rneswold@MCS.Net>
1999-12-27 11:54:57 +00:00
static const char * const lcp_TimerNames[] =
o Create a new ``timer'' log level. This lets us ``set log debug'' without filling our filesystem/screen with junk that we don't really want to see. o change PHYS_STDIN to PHYS_DIRECT - we can handle incoming connections that aren't on STDIN_FILENO now. o Allow return values from our FSM LayerUp functions. If LayerUp() fails, the FSM does an immediate FsmDown() without calling the fsm_parent's Layer{Up,Down} functions. o Clear the close-on-exec flag of file descriptor 3 when executing chat programs so that our documented ability to communicate with /dev/tty via that descriptor works. Also document it as descriptor 3, not 4 :-O o Allow a ``rm'' command as an alias for ``remove''. o Fix the bind()/connect()/accept() calls made by the MP server. o Create bundle_SendDatalink() and bundle_ReceiveDatalink(). This allows `struct datalink's to flatten themselves, pass through a pipe (read: the eye of a needle !) and come alive at the other end. The donator then fork()s & exec()s pppmpipe, ``passing'' the connection to another ppp instance. *** PPP NOW TALKS MULTILINK :-))) *** Our link utilization is hideous, and lots of code needs tidying still. It's also probably riddled with bugs ! It's been tested against itself only, and has hung once, so confidence isn't high....
1998-04-30 23:53:56 +00:00
{"LCP restart", "LCP openmode", "LCP stopped"};
1995-01-31 06:29:58 +00:00
When ppp can't identify the relevant name, don't use "???", use <nnn> or <0xxxx> instead.
2000-03-14 01:46:54 +00:00
static const char *
protoname(int proto)
{
static const char * const cftypes[] = {
/* Check out the latest ``Assigned numbers'' rfc (1700) */
NULL,
"MRU", /* 1: Maximum-Receive-Unit */
"ACCMAP", /* 2: Async-Control-Character-Map */
"AUTHPROTO", /* 3: Authentication-Protocol */
"QUALPROTO", /* 4: Quality-Protocol */
"MAGICNUM", /* 5: Magic-Number */
"RESERVED", /* 6: RESERVED */
"PROTOCOMP", /* 7: Protocol-Field-Compression */
"ACFCOMP", /* 8: Address-and-Control-Field-Compression */
"FCSALT", /* 9: FCS-Alternatives */
"SDP", /* 10: Self-Describing-Pad */
"NUMMODE", /* 11: Numbered-Mode */
"MULTIPROC", /* 12: Multi-Link-Procedure */
"CALLBACK", /* 13: Callback */
"CONTIME", /* 14: Connect-Time */
"COMPFRAME", /* 15: Compound-Frames */
"NDE", /* 16: Nominal-Data-Encapsulation */
"MRRU", /* 17: Multilink-MRRU */
"SHORTSEQ", /* 18: Multilink-Short-Sequence-Number-Header */
"ENDDISC", /* 19: Multilink-Endpoint-Discriminator */
"PROPRIETRY", /* 20: Proprietary */
"DCEID", /* 21: DCE-Identifier */
"MULTIPP", /* 22: Multi-Link-Plus-Procedure */
"LDBACP", /* 23: Link Discriminator for BACP */
};
if (proto < 0 || proto > sizeof cftypes / sizeof *cftypes ||
cftypes[proto] == NULL)
return HexStr(proto, NULL, 0);
return cftypes[proto];
}
Update the cftypes arrays according to rfc1700. Log IPCP reqests as IPCP.
1997-11-14 15:39:15 +00:00
1995-01-31 06:29:58 +00:00
int
o Don't immediately reply to the first LQR thinking it's a duplicate ('cos it compares with our initial values). o Move the LCP into struct datalink.
1998-03-13 21:07:14 +00:00
lcp_ReportStatus(struct cmdargs const *arg)
1995-01-31 06:29:58 +00:00
{
o Add the `set mode' command for change a links current mode. It is not possible to switch to or from dedicated or direct mode, but all other combinations are ok (eg. -auto -> -ddial). o Cope with the fact that commands with optional context may not be able to obtain a link with command_ChooseLink() (if all links have been deleted for example). o Allow `clone'ing in non-multilink mode. We may for example want to configure two links in unilink mode and dial them both, using the one that comes up first. It's also possible to rename ``deflink'' by cloning it, deleting the original, then setting the mode of the new link.
1998-05-15 23:58:30 +00:00
struct link *l;
struct lcp *lcp;
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
l = command_ChooseLink(arg);
o Add the `set mode' command for change a links current mode. It is not possible to switch to or from dedicated or direct mode, but all other combinations are ok (eg. -auto -> -ddial). o Cope with the fact that commands with optional context may not be able to obtain a link with command_ChooseLink() (if all links have been deleted for example). o Allow `clone'ing in non-multilink mode. We may for example want to configure two links in unilink mode and dial them both, using the one that comes up first. It's also possible to rename ``deflink'' by cloning it, deleting the original, then setting the mode of the new link.
1998-05-15 23:58:30 +00:00
lcp = &l->lcp;
o Move struct lcp and struct ccp into struct link. o Remove bundle2lcp(), bundle2ccp() and bundle2link(). They're too resource-hungry and we have `owner pointers' to do their job. o Make our FSM understand LCPs that are always ST_OPENED (with a minimum code that != 1). o Send FSM code rejects for invalid codes. o Make our bundle fsm_parent deal with multiple links. o Make timer diagnostics pretty and allow access via ~t in `term' mode (not just when logging debug) and `show timers'. Only show timers every second in debug mode, otherwise we get too many diagnostics to be useful (we probably still do). Also, don't restrict ~m in term mode to depend on debug logging. o Rationalise our bundles' phases. o Create struct mp (multilink protocol). This is both an NCP and a type of struct link. It feeds off other NCPs for output, passing fragmented packets into the queues of available datalinks. It also gets PROTO_MP input, reassembles the fragments into ppp frames, and passes them back to the HDLC layer that the fragments were passed from. ** It's not yet possible to enter multilink mode :-( ** o Add `set weight' (requires context) for deciding on a links weighting in multilink mode. Weighting is simplistic (and probably badly implemented) for now. o Remove the function pointers in struct link. They ended up only applying to physical links. o Configure our tun device with an MTU equal to the MRU from struct mp's LCP and a speed equal to the sum of our link speeds. o `show {lcp,ccp,proto}' and `set deflate' now have optional context and use ChooseLink() to decide on which `struct link' to use. This allows behaviour as before when in non-multilink mode, and allows access to the MP logical link in multilink mode. o Ignore reconnect and redial values when in -direct mode and when cleaning up. Always redial when in -ddial or -dedicated mode (unless cleaning up). o Tell our links to `staydown' when we close them due to a signal. o Remove remaining `#ifdef SIGALRM's (ppp doesn't function without alarms). o Don't bother strdup()ing our physical link name. o Various other cosmetic changes.
1998-04-03 19:21:56 +00:00
Deglobalise `struct prompt': o Our diagnostic socket has its password set in the `set socket' line only (not in ppp.secret). o Passwords are per server socket (*VarAuthKey are gone) o Authority is per prompt (VarLocalAuth is gone). o Local logging is per prompt. o Add a `show who' command to see who's connected. No identd routine - just a `where the connection came from' display. o SIGUSR1 is disabled for now - we have no way of choosing a password for the socket created :-( Prompts are attached as a list of `struct descriptor's in struct bundle, and serviced under the bundles descriptor service routines. Ultimately, everything should be done like this. Cosmetic: o alphabeticalise SRCS in Makefile. o Add a few comments in command.h TODO: Start checking that we don't overflow the descriptor sets in select() now that we can have any number of descriptors.
1998-04-03 19:26:02 +00:00
prompt_Printf(arg->prompt, "%s: %s [%s]\n", l->name, lcp->fsm.name,
o Move struct lcp and struct ccp into struct link. o Remove bundle2lcp(), bundle2ccp() and bundle2link(). They're too resource-hungry and we have `owner pointers' to do their job. o Make our FSM understand LCPs that are always ST_OPENED (with a minimum code that != 1). o Send FSM code rejects for invalid codes. o Make our bundle fsm_parent deal with multiple links. o Make timer diagnostics pretty and allow access via ~t in `term' mode (not just when logging debug) and `show timers'. Only show timers every second in debug mode, otherwise we get too many diagnostics to be useful (we probably still do). Also, don't restrict ~m in term mode to depend on debug logging. o Rationalise our bundles' phases. o Create struct mp (multilink protocol). This is both an NCP and a type of struct link. It feeds off other NCPs for output, passing fragmented packets into the queues of available datalinks. It also gets PROTO_MP input, reassembles the fragments into ppp frames, and passes them back to the HDLC layer that the fragments were passed from. ** It's not yet possible to enter multilink mode :-( ** o Add `set weight' (requires context) for deciding on a links weighting in multilink mode. Weighting is simplistic (and probably badly implemented) for now. o Remove the function pointers in struct link. They ended up only applying to physical links. o Configure our tun device with an MTU equal to the MRU from struct mp's LCP and a speed equal to the sum of our link speeds. o `show {lcp,ccp,proto}' and `set deflate' now have optional context and use ChooseLink() to decide on which `struct link' to use. This allows behaviour as before when in non-multilink mode, and allows access to the MP logical link in multilink mode. o Ignore reconnect and redial values when in -direct mode and when cleaning up. Always redial when in -ddial or -dedicated mode (unless cleaning up). o Tell our links to `staydown' when we close them due to a signal. o Remove remaining `#ifdef SIGALRM's (ppp doesn't function without alarms). o Don't bother strdup()ing our physical link name. o Various other cosmetic changes.
1998-04-03 19:21:56 +00:00
State2Nam(lcp->fsm.state));
Deglobalise `struct prompt': o Our diagnostic socket has its password set in the `set socket' line only (not in ppp.secret). o Passwords are per server socket (*VarAuthKey are gone) o Authority is per prompt (VarLocalAuth is gone). o Local logging is per prompt. o Add a `show who' command to see who's connected. No identd routine - just a `where the connection came from' display. o SIGUSR1 is disabled for now - we have no way of choosing a password for the socket created :-( Prompts are attached as a list of `struct descriptor's in struct bundle, and serviced under the bundles descriptor service routines. Ultimately, everything should be done like this. Cosmetic: o alphabeticalise SRCS in Makefile. o Add a few comments in command.h TODO: Start checking that we don't overflow the descriptor sets in select() now that we can have any number of descriptors.
1998-04-03 19:26:02 +00:00
prompt_Printf(arg->prompt,
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
" his side: MRU %d, ACCMAP %08lx, PROTOCOMP %s, ACFCOMP %s,\n"
" MAGIC %08lx, MRRU %u, SHORTSEQ %s, REJECT %04x\n",
lcp->his_mru, (u_long)lcp->his_accmap,
`int var : 1' -> `unsigned var : 1'. This made things look silly and made us mis-negotiate the number of VJ slots.
1998-04-05 19:02:49 +00:00
lcp->his_protocomp ? "on" : "off",
lcp->his_acfcomp ? "on" : "off",
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
(u_long)lcp->his_magic, lcp->his_mrru,
lcp->his_shortseq ? "on" : "off", lcp->his_reject);
Deglobalise `struct prompt': o Our diagnostic socket has its password set in the `set socket' line only (not in ppp.secret). o Passwords are per server socket (*VarAuthKey are gone) o Authority is per prompt (VarLocalAuth is gone). o Local logging is per prompt. o Add a `show who' command to see who's connected. No identd routine - just a `where the connection came from' display. o SIGUSR1 is disabled for now - we have no way of choosing a password for the socket created :-( Prompts are attached as a list of `struct descriptor's in struct bundle, and serviced under the bundles descriptor service routines. Ultimately, everything should be done like this. Cosmetic: o alphabeticalise SRCS in Makefile. o Add a few comments in command.h TODO: Start checking that we don't overflow the descriptor sets in select() now that we can have any number of descriptors.
1998-04-03 19:26:02 +00:00
prompt_Printf(arg->prompt,
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
" my side: MRU %d, ACCMAP %08lx, PROTOCOMP %s, ACFCOMP %s,\n"
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
" MAGIC %08lx, MRRU %u, SHORTSEQ %s, REJECT %04x\n",
o Move struct lcp and struct ccp into struct link. o Remove bundle2lcp(), bundle2ccp() and bundle2link(). They're too resource-hungry and we have `owner pointers' to do their job. o Make our FSM understand LCPs that are always ST_OPENED (with a minimum code that != 1). o Send FSM code rejects for invalid codes. o Make our bundle fsm_parent deal with multiple links. o Make timer diagnostics pretty and allow access via ~t in `term' mode (not just when logging debug) and `show timers'. Only show timers every second in debug mode, otherwise we get too many diagnostics to be useful (we probably still do). Also, don't restrict ~m in term mode to depend on debug logging. o Rationalise our bundles' phases. o Create struct mp (multilink protocol). This is both an NCP and a type of struct link. It feeds off other NCPs for output, passing fragmented packets into the queues of available datalinks. It also gets PROTO_MP input, reassembles the fragments into ppp frames, and passes them back to the HDLC layer that the fragments were passed from. ** It's not yet possible to enter multilink mode :-( ** o Add `set weight' (requires context) for deciding on a links weighting in multilink mode. Weighting is simplistic (and probably badly implemented) for now. o Remove the function pointers in struct link. They ended up only applying to physical links. o Configure our tun device with an MTU equal to the MRU from struct mp's LCP and a speed equal to the sum of our link speeds. o `show {lcp,ccp,proto}' and `set deflate' now have optional context and use ChooseLink() to decide on which `struct link' to use. This allows behaviour as before when in non-multilink mode, and allows access to the MP logical link in multilink mode. o Ignore reconnect and redial values when in -direct mode and when cleaning up. Always redial when in -ddial or -dedicated mode (unless cleaning up). o Tell our links to `staydown' when we close them due to a signal. o Remove remaining `#ifdef SIGALRM's (ppp doesn't function without alarms). o Don't bother strdup()ing our physical link name. o Various other cosmetic changes.
1998-04-03 19:21:56 +00:00
lcp->want_mru, (u_long)lcp->want_accmap,
`int var : 1' -> `unsigned var : 1'. This made things look silly and made us mis-negotiate the number of VJ slots.
1998-04-05 19:02:49 +00:00
lcp->want_protocomp ? "on" : "off",
lcp->want_acfcomp ? "on" : "off",
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
(u_long)lcp->want_magic, lcp->want_mrru,
lcp->want_shortseq ? "on" : "off", lcp->my_reject);
o Move default MRU, MTU, ACCMAP and OPENMODE config values into struct lcp and display them in `show lcp'. o Remove `show mru' and `show mtu' and make the data part of `show lcp'. Also merge `set m[tr]u' and `set openmode' implementations into the SetVariable function. o `set timeout' only accepts the idle timer value as an argument. o Move our lqr period into struct lcp, and create a `set lqrperiod' command. Display it in `show lcp'. o Remove VarRetryTimeout, and implement it at the LCP, PAP, CHAP, CCP and IPCP levels, creating individual `set XXXretry' commands for each. They must be separate because they have different context requirements in multilink mode. o Display default config values in `show ccp'. o Tart the man page up a bit (wrt PPP/TCP, compression and LQR) and explain the new commands.
1998-04-03 19:24:07 +00:00
Handle peer REQ/NAKs of >1500 byte MRUs when we have no preference. MFC after: 3 days
2001-07-26 11:33:53 +00:00
if (lcp->cfg.mru)
prompt_Printf(arg->prompt, "\n Defaults: MRU = %d (max %d), ",
lcp->cfg.mru, lcp->cfg.max_mru);
else
prompt_Printf(arg->prompt, "\n Defaults: MRU = any (max %d), ",
lcp->cfg.max_mru);
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
if (lcp->cfg.mtu)
prompt_Printf(arg->prompt, "MTU = %d (max %d), ",
lcp->cfg.mtu, lcp->cfg.max_mtu);
else
prompt_Printf(arg->prompt, "MTU = any (max %d), ", lcp->cfg.max_mtu);
Deglobalise `struct prompt': o Our diagnostic socket has its password set in the `set socket' line only (not in ppp.secret). o Passwords are per server socket (*VarAuthKey are gone) o Authority is per prompt (VarLocalAuth is gone). o Local logging is per prompt. o Add a `show who' command to see who's connected. No identd routine - just a `where the connection came from' display. o SIGUSR1 is disabled for now - we have no way of choosing a password for the socket created :-( Prompts are attached as a list of `struct descriptor's in struct bundle, and serviced under the bundles descriptor service routines. Ultimately, everything should be done like this. Cosmetic: o alphabeticalise SRCS in Makefile. o Add a few comments in command.h TODO: Start checking that we don't overflow the descriptor sets in select() now that we can have any number of descriptors.
1998-04-03 19:26:02 +00:00
prompt_Printf(arg->prompt, "ACCMAP = %08lx\n", (u_long)lcp->cfg.accmap);
prompt_Printf(arg->prompt, " LQR period = %us, ",
lcp->cfg.lqrperiod);
prompt_Printf(arg->prompt, "Open Mode = %s",
o Move default MRU, MTU, ACCMAP and OPENMODE config values into struct lcp and display them in `show lcp'. o Remove `show mru' and `show mtu' and make the data part of `show lcp'. Also merge `set m[tr]u' and `set openmode' implementations into the SetVariable function. o `set timeout' only accepts the idle timer value as an argument. o Move our lqr period into struct lcp, and create a `set lqrperiod' command. Display it in `show lcp'. o Remove VarRetryTimeout, and implement it at the LCP, PAP, CHAP, CCP and IPCP levels, creating individual `set XXXretry' commands for each. They must be separate because they have different context requirements in multilink mode. o Display default config values in `show ccp'. o Tart the man page up a bit (wrt PPP/TCP, compression and LQR) and explain the new commands.
1998-04-03 19:24:07 +00:00
lcp->cfg.openmode == OPEN_PASSIVE ? "passive" : "active");
if (lcp->cfg.openmode > 0)
Deglobalise `struct prompt': o Our diagnostic socket has its password set in the `set socket' line only (not in ppp.secret). o Passwords are per server socket (*VarAuthKey are gone) o Authority is per prompt (VarLocalAuth is gone). o Local logging is per prompt. o Add a `show who' command to see who's connected. No identd routine - just a `where the connection came from' display. o SIGUSR1 is disabled for now - we have no way of choosing a password for the socket created :-( Prompts are attached as a list of `struct descriptor's in struct bundle, and serviced under the bundles descriptor service routines. Ultimately, everything should be done like this. Cosmetic: o alphabeticalise SRCS in Makefile. o Add a few comments in command.h TODO: Start checking that we don't overflow the descriptor sets in select() now that we can have any number of descriptors.
1998-04-03 19:26:02 +00:00
prompt_Printf(arg->prompt, " (delay %ds)", lcp->cfg.openmode);
Allow control over the number of ConfigREQ & TermREQ attempts that are made in each of the FSMs (LCP, CCP & IPCP) and the number of REQs/Challenges for PAP/CHAP by accepting more arguments in the ``set {c,ip,l}cpretry'' and ``set {ch,p}apretry'' commands. Change the non-convergence thresholds to 3 times the number of configured REQ tries (rather than the previous fixed ``10''). We now notice repeated NAKs and REJs rather than just REQs. Don't suggest that CHAP 0x05 isn't supported when it's not configured. Fix some bugs that expose themselves with smaller numbers of retries: o Handle instantaneous disconnects (set device /dev/null) correctly by stopping all fsm timers in fsm2initial. o Don't forget to uu_unlock() devices that are files but are not ttys (set device /dev/zero). Fix a *HORRENDOUS* bug in RFC1661 (already fixed for an Open event in state ``Closed''): According to the state transition table, a RCR+ or RCR- received in the ``Stopped'' state are supposed to InitRestartCounter, SendConfigReq and SendConfig{Ack,Nak}. However, in ``Stopped'', we haven't yet done a TLS (or the last thing we did is a TLF). We must therefore do the TLS at this point ! This was never noticed before because LCP and CCP used not use LayerStart() for anything interesting, and IPCP tends to go into Stopped then get a Down because of an LCP RTR rather than getting a RCR again.
1999-02-26 21:28:14 +00:00
prompt_Printf(arg->prompt, "\n FSM retry = %us, max %u Config"
" REQ%s, %u Term REQ%s\n", lcp->cfg.fsm.timeout,
lcp->cfg.fsm.maxreq, lcp->cfg.fsm.maxreq == 1 ? "" : "s",
lcp->cfg.fsm.maxtrm, lcp->cfg.fsm.maxtrm == 1 ? "" : "s");
Support link identification from rfc1570 Two new commands are available; ``ident'' and ``sendident''.
2000-07-19 02:10:35 +00:00
prompt_Printf(arg->prompt, " Ident: %s\n", lcp->cfg.ident);
o Move the accept/deny/disable/enable globals either to the bundle (non-negotiated vars) or to their respective IPCP, LCP or CCP. o Enable rolling throughput statistics by default. o Remove the `display' command. These values now appear in `show bundle', `show ipcp', `show ccp' and `show lcp'. o Initialise auth name & key at bundle create time (oops). o Rename pppd-deflate (the id-24 hack) to deflate24. o Don't send both a REJ and a NAK to an IPCP or LCP REQ. Favour the REJ (already done at the CCP level). o Recurse in datalink_UpdateSet() when we change state, otherwise we end up setting no descriptors and getting jammed in the imminent select() instead of doing the dial/login/hangup. o Display our CHAP encryption method despite being built with DES. o Display VJ as not negotiated in ``show ipcp'' when necessary.
1998-04-16 00:26:21 +00:00
prompt_Printf(arg->prompt, "\n Negotiation:\n");
prompt_Printf(arg->prompt, " ACFCOMP = %s\n",
command_ShowNegval(lcp->cfg.acfcomp));
prompt_Printf(arg->prompt, " CHAP = %s\n",
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
command_ShowNegval(lcp->cfg.chap05));
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
#ifndef NODES
Call MSCHAP CHAP80 in ``show lcp'' so that it's consistent with CHAP81.
2000-11-29 15:47:35 +00:00
prompt_Printf(arg->prompt, " CHAP80 = %s\n",
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
command_ShowNegval(lcp->cfg.chap80nt));
prompt_Printf(arg->prompt, " LANMan = %s\n",
command_ShowNegval(lcp->cfg.chap80lm));
Add MPPE and MSChap v2 support (denied and disabled by default) Submitted by: Ustimenko Semen <semen@iclub.nsu.ru>
2000-10-30 00:15:04 +00:00
prompt_Printf(arg->prompt, " CHAP81 = %s\n",
command_ShowNegval(lcp->cfg.chap81));
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
#endif
o Move the accept/deny/disable/enable globals either to the bundle (non-negotiated vars) or to their respective IPCP, LCP or CCP. o Enable rolling throughput statistics by default. o Remove the `display' command. These values now appear in `show bundle', `show ipcp', `show ccp' and `show lcp'. o Initialise auth name & key at bundle create time (oops). o Rename pppd-deflate (the id-24 hack) to deflate24. o Don't send both a REJ and a NAK to an IPCP or LCP REQ. Favour the REJ (already done at the CCP level). o Recurse in datalink_UpdateSet() when we change state, otherwise we end up setting no descriptors and getting jammed in the imminent select() instead of doing the dial/login/hangup. o Display our CHAP encryption method despite being built with DES. o Display VJ as not negotiated in ``show ipcp'' when necessary.
1998-04-16 00:26:21 +00:00
prompt_Printf(arg->prompt, " LQR = %s\n",
command_ShowNegval(lcp->cfg.lqr));
prompt_Printf(arg->prompt, " PAP = %s\n",
command_ShowNegval(lcp->cfg.pap));
prompt_Printf(arg->prompt, " PROTOCOMP = %s\n",
command_ShowNegval(lcp->cfg.protocomp));
Overhaul ppp: o Use syslog o Remove references to stdout/stderr (incl perror()) o Introduce VarTerm - the interactive terminal or zero o Allow "set timeout" to affect current session o Change "set debug" to "set log" o Allow "set log [+|-]flag" o Make MSEXT and PASSWDAUTH stuff the default o Move all #ifdef DEBUG stuff into the code - this shouldn't be too much overhead. It's now controlled with "set log +debug" o Add "set log command, debug, tun, warn, error, alert" o Remove cdefs.h, and assume an ansi compiler. o Improve all diagnostic output o Don't trap SIGSEGV o SIGHUP now terminates again (log files are controlled by syslog) o Call CloseModem() when changing devices o Fix parsing of third arg of "delete" I think this fixes the "magic is same" problems that some people have been experiencing. The man page is being rewritten. It'll follow soon.
1997-06-09 03:27:43 +00:00
return 0;
1995-01-31 06:29:58 +00:00
}
Make things work when sizeof(long) != 32 (hopefully)
1998-01-11 17:50:49 +00:00
static u_int32_t
Fix prototypes. Remove extraneous decls. Add ``const'' to several places. Allow ``make NOALIAS=1'' to remove IP aliasing. Merge with OpenBSD - only the Makefiles vary. We can now survive a compile with -Wall -Wbad-function-cast -Wcast-align -Wcast-qual -Winline -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wpointer-arith -Wredundant-decls -Wshadow -Wstrict-prototypes -Wwrite-strings -Wchar-subscripts (although the Makefile just contains -Wall).
1997-11-22 03:37:54 +00:00
GenerateMagic(void)
1995-01-31 06:29:58 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* Generate random number which will be used as magic number */
Cosmetic (no functional changes): o Add missing $Id$s o Move extern decls from .c -> .h files o Staticize o Remove #includes from .h files o style(9)ify includes o bcopy -> memcpy bzero -> memset bcmp -> memcmp index -> strchr rindex -> strrchr o Move timeout.h -> timer.h (making it consistent w/ timer.c) o Add -Wmissing-prototypes
1997-10-26 01:04:02 +00:00
randinit();
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
return random();
1995-01-31 06:29:58 +00:00
}
void
o Create a new ``timer'' log level. This lets us ``set log debug'' without filling our filesystem/screen with junk that we don't really want to see. o change PHYS_STDIN to PHYS_DIRECT - we can handle incoming connections that aren't on STDIN_FILENO now. o Allow return values from our FSM LayerUp functions. If LayerUp() fails, the FSM does an immediate FsmDown() without calling the fsm_parent's Layer{Up,Down} functions. o Clear the close-on-exec flag of file descriptor 3 when executing chat programs so that our documented ability to communicate with /dev/tty via that descriptor works. Also document it as descriptor 3, not 4 :-O o Allow a ``rm'' command as an alias for ``remove''. o Fix the bind()/connect()/accept() calls made by the MP server. o Create bundle_SendDatalink() and bundle_ReceiveDatalink(). This allows `struct datalink's to flatten themselves, pass through a pipe (read: the eye of a needle !) and come alive at the other end. The donator then fork()s & exec()s pppmpipe, ``passing'' the connection to another ppp instance. *** PPP NOW TALKS MULTILINK :-))) *** Our link utilization is hideous, and lots of code needs tidying still. It's also probably riddled with bugs ! It's been tested against itself only, and has hung once, so confidence isn't high....
1998-04-30 23:53:56 +00:00
lcp_SetupCallbacks(struct lcp *lcp)
1995-01-31 06:29:58 +00:00
{
o Create a new ``timer'' log level. This lets us ``set log debug'' without filling our filesystem/screen with junk that we don't really want to see. o change PHYS_STDIN to PHYS_DIRECT - we can handle incoming connections that aren't on STDIN_FILENO now. o Allow return values from our FSM LayerUp functions. If LayerUp() fails, the FSM does an immediate FsmDown() without calling the fsm_parent's Layer{Up,Down} functions. o Clear the close-on-exec flag of file descriptor 3 when executing chat programs so that our documented ability to communicate with /dev/tty via that descriptor works. Also document it as descriptor 3, not 4 :-O o Allow a ``rm'' command as an alias for ``remove''. o Fix the bind()/connect()/accept() calls made by the MP server. o Create bundle_SendDatalink() and bundle_ReceiveDatalink(). This allows `struct datalink's to flatten themselves, pass through a pipe (read: the eye of a needle !) and come alive at the other end. The donator then fork()s & exec()s pppmpipe, ``passing'' the connection to another ppp instance. *** PPP NOW TALKS MULTILINK :-))) *** Our link utilization is hideous, and lots of code needs tidying still. It's also probably riddled with bugs ! It's been tested against itself only, and has hung once, so confidence isn't high....
1998-04-30 23:53:56 +00:00
lcp->fsm.fn = &lcp_Callbacks;
lcp->fsm.FsmTimer.name = lcp_TimerNames[0];
lcp->fsm.OpenTimer.name = lcp_TimerNames[1];
lcp->fsm.StoppedTimer.name = lcp_TimerNames[2];
}
1995-01-31 06:29:58 +00:00
This commit was manufactured by cvs2svn to create branch 'MP'.
1998-01-29 00:44:16 +00:00
void
o Move struct lcp and struct ccp into struct link. o Remove bundle2lcp(), bundle2ccp() and bundle2link(). They're too resource-hungry and we have `owner pointers' to do their job. o Make our FSM understand LCPs that are always ST_OPENED (with a minimum code that != 1). o Send FSM code rejects for invalid codes. o Make our bundle fsm_parent deal with multiple links. o Make timer diagnostics pretty and allow access via ~t in `term' mode (not just when logging debug) and `show timers'. Only show timers every second in debug mode, otherwise we get too many diagnostics to be useful (we probably still do). Also, don't restrict ~m in term mode to depend on debug logging. o Rationalise our bundles' phases. o Create struct mp (multilink protocol). This is both an NCP and a type of struct link. It feeds off other NCPs for output, passing fragmented packets into the queues of available datalinks. It also gets PROTO_MP input, reassembles the fragments into ppp frames, and passes them back to the HDLC layer that the fragments were passed from. ** It's not yet possible to enter multilink mode :-( ** o Add `set weight' (requires context) for deciding on a links weighting in multilink mode. Weighting is simplistic (and probably badly implemented) for now. o Remove the function pointers in struct link. They ended up only applying to physical links. o Configure our tun device with an MTU equal to the MRU from struct mp's LCP and a speed equal to the sum of our link speeds. o `show {lcp,ccp,proto}' and `set deflate' now have optional context and use ChooseLink() to decide on which `struct link' to use. This allows behaviour as before when in non-multilink mode, and allows access to the MP logical link in multilink mode. o Ignore reconnect and redial values when in -direct mode and when cleaning up. Always redial when in -ddial or -dedicated mode (unless cleaning up). o Tell our links to `staydown' when we close them due to a signal. o Remove remaining `#ifdef SIGALRM's (ppp doesn't function without alarms). o Don't bother strdup()ing our physical link name. o Various other cosmetic changes.
1998-04-03 19:21:56 +00:00
lcp_Init(struct lcp *lcp, struct bundle *bundle, struct link *l,
Create struct fsm_parent. FSMs are created with one of these, and the FSM passes subsequent events to them. The datalink now hides its CCP from the bundle layer.
1998-02-27 01:22:39 +00:00
const struct fsm_parent *parent)
This commit was manufactured by cvs2svn to create branch 'MP'.
1998-01-29 00:44:16 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* Initialise ourselves */
o Move struct lcp and struct ccp into struct link. o Remove bundle2lcp(), bundle2ccp() and bundle2link(). They're too resource-hungry and we have `owner pointers' to do their job. o Make our FSM understand LCPs that are always ST_OPENED (with a minimum code that != 1). o Send FSM code rejects for invalid codes. o Make our bundle fsm_parent deal with multiple links. o Make timer diagnostics pretty and allow access via ~t in `term' mode (not just when logging debug) and `show timers'. Only show timers every second in debug mode, otherwise we get too many diagnostics to be useful (we probably still do). Also, don't restrict ~m in term mode to depend on debug logging. o Rationalise our bundles' phases. o Create struct mp (multilink protocol). This is both an NCP and a type of struct link. It feeds off other NCPs for output, passing fragmented packets into the queues of available datalinks. It also gets PROTO_MP input, reassembles the fragments into ppp frames, and passes them back to the HDLC layer that the fragments were passed from. ** It's not yet possible to enter multilink mode :-( ** o Add `set weight' (requires context) for deciding on a links weighting in multilink mode. Weighting is simplistic (and probably badly implemented) for now. o Remove the function pointers in struct link. They ended up only applying to physical links. o Configure our tun device with an MTU equal to the MRU from struct mp's LCP and a speed equal to the sum of our link speeds. o `show {lcp,ccp,proto}' and `set deflate' now have optional context and use ChooseLink() to decide on which `struct link' to use. This allows behaviour as before when in non-multilink mode, and allows access to the MP logical link in multilink mode. o Ignore reconnect and redial values when in -direct mode and when cleaning up. Always redial when in -ddial or -dedicated mode (unless cleaning up). o Tell our links to `staydown' when we close them due to a signal. o Remove remaining `#ifdef SIGALRM's (ppp doesn't function without alarms). o Don't bother strdup()ing our physical link name. o Various other cosmetic changes.
1998-04-03 19:21:56 +00:00
int mincode = parent ? 1 : LCP_MINMPCODE;
1995-01-31 06:29:58 +00:00
Allow control over the number of ConfigREQ & TermREQ attempts that are made in each of the FSMs (LCP, CCP & IPCP) and the number of REQs/Challenges for PAP/CHAP by accepting more arguments in the ``set {c,ip,l}cpretry'' and ``set {ch,p}apretry'' commands. Change the non-convergence thresholds to 3 times the number of configured REQ tries (rather than the previous fixed ``10''). We now notice repeated NAKs and REJs rather than just REQs. Don't suggest that CHAP 0x05 isn't supported when it's not configured. Fix some bugs that expose themselves with smaller numbers of retries: o Handle instantaneous disconnects (set device /dev/null) correctly by stopping all fsm timers in fsm2initial. o Don't forget to uu_unlock() devices that are files but are not ttys (set device /dev/zero). Fix a *HORRENDOUS* bug in RFC1661 (already fixed for an Open event in state ``Closed''): According to the state transition table, a RCR+ or RCR- received in the ``Stopped'' state are supposed to InitRestartCounter, SendConfigReq and SendConfig{Ack,Nak}. However, in ``Stopped'', we haven't yet done a TLS (or the last thing we did is a TLF). We must therefore do the TLS at this point ! This was never noticed before because LCP and CCP used not use LayerStart() for anything interesting, and IPCP tends to go into Stopped then get a Down because of an LCP RTR rather than getting a RCR again.
1999-02-26 21:28:14 +00:00
fsm_Init(&lcp->fsm, "LCP", PROTO_LCP, mincode, LCP_MAXCODE, LogLCP,
o Create a new ``timer'' log level. This lets us ``set log debug'' without filling our filesystem/screen with junk that we don't really want to see. o change PHYS_STDIN to PHYS_DIRECT - we can handle incoming connections that aren't on STDIN_FILENO now. o Allow return values from our FSM LayerUp functions. If LayerUp() fails, the FSM does an immediate FsmDown() without calling the fsm_parent's Layer{Up,Down} functions. o Clear the close-on-exec flag of file descriptor 3 when executing chat programs so that our documented ability to communicate with /dev/tty via that descriptor works. Also document it as descriptor 3, not 4 :-O o Allow a ``rm'' command as an alias for ``remove''. o Fix the bind()/connect()/accept() calls made by the MP server. o Create bundle_SendDatalink() and bundle_ReceiveDatalink(). This allows `struct datalink's to flatten themselves, pass through a pipe (read: the eye of a needle !) and come alive at the other end. The donator then fork()s & exec()s pppmpipe, ``passing'' the connection to another ppp instance. *** PPP NOW TALKS MULTILINK :-))) *** Our link utilization is hideous, and lots of code needs tidying still. It's also probably riddled with bugs ! It's been tested against itself only, and has hung once, so confidence isn't high....
1998-04-30 23:53:56 +00:00
bundle, l, parent, &lcp_Callbacks, lcp_TimerNames);
o Move default MRU, MTU, ACCMAP and OPENMODE config values into struct lcp and display them in `show lcp'. o Remove `show mru' and `show mtu' and make the data part of `show lcp'. Also merge `set m[tr]u' and `set openmode' implementations into the SetVariable function. o `set timeout' only accepts the idle timer value as an argument. o Move our lqr period into struct lcp, and create a `set lqrperiod' command. Display it in `show lcp'. o Remove VarRetryTimeout, and implement it at the LCP, PAP, CHAP, CCP and IPCP levels, creating individual `set XXXretry' commands for each. They must be separate because they have different context requirements in multilink mode. o Display default config values in `show ccp'. o Tart the man page up a bit (wrt PPP/TCP, compression and LQR) and explain the new commands.
1998-04-03 19:24:07 +00:00
Handle peer REQ/NAKs of >1500 byte MRUs when we have no preference. MFC after: 3 days
2001-07-26 11:33:53 +00:00
lcp->cfg.mru = 0;
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
lcp->cfg.max_mru = MAX_MRU;
lcp->cfg.mtu = 0;
lcp->cfg.max_mtu = MAX_MTU;
o Move default MRU, MTU, ACCMAP and OPENMODE config values into struct lcp and display them in `show lcp'. o Remove `show mru' and `show mtu' and make the data part of `show lcp'. Also merge `set m[tr]u' and `set openmode' implementations into the SetVariable function. o `set timeout' only accepts the idle timer value as an argument. o Move our lqr period into struct lcp, and create a `set lqrperiod' command. Display it in `show lcp'. o Remove VarRetryTimeout, and implement it at the LCP, PAP, CHAP, CCP and IPCP levels, creating individual `set XXXretry' commands for each. They must be separate because they have different context requirements in multilink mode. o Display default config values in `show ccp'. o Tart the man page up a bit (wrt PPP/TCP, compression and LQR) and explain the new commands.
1998-04-03 19:24:07 +00:00
lcp->cfg.accmap = 0;
lcp->cfg.openmode = 1;
lcp->cfg.lqrperiod = DEF_LQRPERIOD;
Allow control over the number of ConfigREQ & TermREQ attempts that are made in each of the FSMs (LCP, CCP & IPCP) and the number of REQs/Challenges for PAP/CHAP by accepting more arguments in the ``set {c,ip,l}cpretry'' and ``set {ch,p}apretry'' commands. Change the non-convergence thresholds to 3 times the number of configured REQ tries (rather than the previous fixed ``10''). We now notice repeated NAKs and REJs rather than just REQs. Don't suggest that CHAP 0x05 isn't supported when it's not configured. Fix some bugs that expose themselves with smaller numbers of retries: o Handle instantaneous disconnects (set device /dev/null) correctly by stopping all fsm timers in fsm2initial. o Don't forget to uu_unlock() devices that are files but are not ttys (set device /dev/zero). Fix a *HORRENDOUS* bug in RFC1661 (already fixed for an Open event in state ``Closed''): According to the state transition table, a RCR+ or RCR- received in the ``Stopped'' state are supposed to InitRestartCounter, SendConfigReq and SendConfig{Ack,Nak}. However, in ``Stopped'', we haven't yet done a TLS (or the last thing we did is a TLF). We must therefore do the TLS at this point ! This was never noticed before because LCP and CCP used not use LayerStart() for anything interesting, and IPCP tends to go into Stopped then get a Down because of an LCP RTR rather than getting a RCR again.
1999-02-26 21:28:14 +00:00
lcp->cfg.fsm.timeout = DEF_FSMRETRY;
lcp->cfg.fsm.maxreq = DEF_FSMTRIES;
lcp->cfg.fsm.maxtrm = DEF_FSMTRIES;
o Move default MRU, MTU, ACCMAP and OPENMODE config values into struct lcp and display them in `show lcp'. o Remove `show mru' and `show mtu' and make the data part of `show lcp'. Also merge `set m[tr]u' and `set openmode' implementations into the SetVariable function. o `set timeout' only accepts the idle timer value as an argument. o Move our lqr period into struct lcp, and create a `set lqrperiod' command. Display it in `show lcp'. o Remove VarRetryTimeout, and implement it at the LCP, PAP, CHAP, CCP and IPCP levels, creating individual `set XXXretry' commands for each. They must be separate because they have different context requirements in multilink mode. o Display default config values in `show ccp'. o Tart the man page up a bit (wrt PPP/TCP, compression and LQR) and explain the new commands.
1998-04-03 19:24:07 +00:00
o Move the accept/deny/disable/enable globals either to the bundle (non-negotiated vars) or to their respective IPCP, LCP or CCP. o Enable rolling throughput statistics by default. o Remove the `display' command. These values now appear in `show bundle', `show ipcp', `show ccp' and `show lcp'. o Initialise auth name & key at bundle create time (oops). o Rename pppd-deflate (the id-24 hack) to deflate24. o Don't send both a REJ and a NAK to an IPCP or LCP REQ. Favour the REJ (already done at the CCP level). o Recurse in datalink_UpdateSet() when we change state, otherwise we end up setting no descriptors and getting jammed in the imminent select() instead of doing the dial/login/hangup. o Display our CHAP encryption method despite being built with DES. o Display VJ as not negotiated in ``show ipcp'' when necessary.
1998-04-16 00:26:21 +00:00
lcp->cfg.acfcomp = NEG_ENABLED|NEG_ACCEPTED;
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
lcp->cfg.chap05 = NEG_ACCEPTED;
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
#ifndef NODES
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
lcp->cfg.chap80nt = NEG_ACCEPTED;
Don't accept chap80lm by default - it doesn't work. MFC after: 1 week
2001-07-03 22:20:11 +00:00
lcp->cfg.chap80lm = 0;
Accept MSCHAPv2 by default. Enable and accept MPPE by default.
2000-11-19 16:53:50 +00:00
lcp->cfg.chap81 = NEG_ACCEPTED;
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
#endif
o Move the accept/deny/disable/enable globals either to the bundle (non-negotiated vars) or to their respective IPCP, LCP or CCP. o Enable rolling throughput statistics by default. o Remove the `display' command. These values now appear in `show bundle', `show ipcp', `show ccp' and `show lcp'. o Initialise auth name & key at bundle create time (oops). o Rename pppd-deflate (the id-24 hack) to deflate24. o Don't send both a REJ and a NAK to an IPCP or LCP REQ. Favour the REJ (already done at the CCP level). o Recurse in datalink_UpdateSet() when we change state, otherwise we end up setting no descriptors and getting jammed in the imminent select() instead of doing the dial/login/hangup. o Display our CHAP encryption method despite being built with DES. o Display VJ as not negotiated in ``show ipcp'' when necessary.
1998-04-16 00:26:21 +00:00
lcp->cfg.lqr = NEG_ACCEPTED;
lcp->cfg.pap = NEG_ACCEPTED;
lcp->cfg.protocomp = NEG_ENABLED|NEG_ACCEPTED;
Support link identification from rfc1570 Two new commands are available; ``ident'' and ``sendident''.
2000-07-19 02:10:35 +00:00
*lcp->cfg.ident = '\0';
o Move the accept/deny/disable/enable globals either to the bundle (non-negotiated vars) or to their respective IPCP, LCP or CCP. o Enable rolling throughput statistics by default. o Remove the `display' command. These values now appear in `show bundle', `show ipcp', `show ccp' and `show lcp'. o Initialise auth name & key at bundle create time (oops). o Rename pppd-deflate (the id-24 hack) to deflate24. o Don't send both a REJ and a NAK to an IPCP or LCP REQ. Favour the REJ (already done at the CCP level). o Recurse in datalink_UpdateSet() when we change state, otherwise we end up setting no descriptors and getting jammed in the imminent select() instead of doing the dial/login/hangup. o Display our CHAP encryption method despite being built with DES. o Display VJ as not negotiated in ``show ipcp'' when necessary.
1998-04-16 00:26:21 +00:00
o Move default MRU, MTU, ACCMAP and OPENMODE config values into struct lcp and display them in `show lcp'. o Remove `show mru' and `show mtu' and make the data part of `show lcp'. Also merge `set m[tr]u' and `set openmode' implementations into the SetVariable function. o `set timeout' only accepts the idle timer value as an argument. o Move our lqr period into struct lcp, and create a `set lqrperiod' command. Display it in `show lcp'. o Remove VarRetryTimeout, and implement it at the LCP, PAP, CHAP, CCP and IPCP levels, creating individual `set XXXretry' commands for each. They must be separate because they have different context requirements in multilink mode. o Display default config values in `show ccp'. o Tart the man page up a bit (wrt PPP/TCP, compression and LQR) and explain the new commands.
1998-04-03 19:24:07 +00:00
lcp_Setup(lcp, lcp->cfg.openmode);
1995-01-31 06:29:58 +00:00
}
Shuffle around our FSMs a bit. This'll make it easier to remove the CcpInfo, LcpInfo and IpcpInfo globals.
1998-02-21 01:45:26 +00:00
void
lcp_Setup(struct lcp *lcp, int openmode)
1995-01-31 06:29:58 +00:00
{
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
struct physical *p = link2physical(lcp->fsm.link);
Shuffle around our FSMs a bit. This'll make it easier to remove the CcpInfo, LcpInfo and IpcpInfo globals.
1998-02-21 01:45:26 +00:00
lcp->fsm.open_mode = openmode;
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
lcp->his_mru = DEF_MRU;
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
lcp->his_mrru = 0;
Shuffle around our FSMs a bit. This'll make it easier to remove the CcpInfo, LcpInfo and IpcpInfo globals.
1998-02-21 01:45:26 +00:00
lcp->his_magic = 0;
lcp->his_lqrperiod = 0;
lcp->his_acfcomp = 0;
lcp->his_auth = 0;
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
lcp->his_authtype = 0;
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
lcp->his_callback.opmask = 0;
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
lcp->his_shortseq = 0;
When the peer fails to specify an MRU and a 1500 byte MRU is not allowed either because of the transport or configuration, send a MRU NAK only once, then allow the negotiations to proceed. rfc1661 says that 1500 should always be allowed and rfc2516 says that 1492 is the maximum for PPPoE. This changes ppp so that it only weakly suggests 1492, then goes with the default (leaving the problem in the hands of the peer WRT how they set their MTU). MFC after: 1 week
2001-10-18 08:47:56 +00:00
lcp->mru_req = 0;
Shuffle around our FSMs a bit. This'll make it easier to remove the CcpInfo, LcpInfo and IpcpInfo globals.
1998-02-21 01:45:26 +00:00
Handle peer REQ/NAKs of >1500 byte MRUs when we have no preference. MFC after: 3 days
2001-07-26 11:33:53 +00:00
if ((lcp->want_mru = lcp->cfg.mru) == 0)
lcp->want_mru = DEF_MRU;
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
lcp->want_mrru = lcp->fsm.bundle->ncp.mp.cfg.mrru;
lcp->want_shortseq = IsEnabled(lcp->fsm.bundle->ncp.mp.cfg.shortseq) ? 1 : 0;
o Move the accept/deny/disable/enable globals either to the bundle (non-negotiated vars) or to their respective IPCP, LCP or CCP. o Enable rolling throughput statistics by default. o Remove the `display' command. These values now appear in `show bundle', `show ipcp', `show ccp' and `show lcp'. o Initialise auth name & key at bundle create time (oops). o Rename pppd-deflate (the id-24 hack) to deflate24. o Don't send both a REJ and a NAK to an IPCP or LCP REQ. Favour the REJ (already done at the CCP level). o Recurse in datalink_UpdateSet() when we change state, otherwise we end up setting no descriptors and getting jammed in the imminent select() instead of doing the dial/login/hangup. o Display our CHAP encryption method despite being built with DES. o Display VJ as not negotiated in ``show ipcp'' when necessary.
1998-04-16 00:26:21 +00:00
lcp->want_acfcomp = IsEnabled(lcp->cfg.acfcomp) ? 1 : 0;
Shuffle around our FSMs a bit. This'll make it easier to remove the CcpInfo, LcpInfo and IpcpInfo globals.
1998-02-21 01:45:26 +00:00
Initialise multilink LCPs according to rfc1990.
1998-04-03 19:26:12 +00:00
if (lcp->fsm.parent) {
lcp->his_accmap = 0xffffffff;
lcp->want_accmap = lcp->cfg.accmap;
lcp->his_protocomp = 0;
o Move the accept/deny/disable/enable globals either to the bundle (non-negotiated vars) or to their respective IPCP, LCP or CCP. o Enable rolling throughput statistics by default. o Remove the `display' command. These values now appear in `show bundle', `show ipcp', `show ccp' and `show lcp'. o Initialise auth name & key at bundle create time (oops). o Rename pppd-deflate (the id-24 hack) to deflate24. o Don't send both a REJ and a NAK to an IPCP or LCP REQ. Favour the REJ (already done at the CCP level). o Recurse in datalink_UpdateSet() when we change state, otherwise we end up setting no descriptors and getting jammed in the imminent select() instead of doing the dial/login/hangup. o Display our CHAP encryption method despite being built with DES. o Display VJ as not negotiated in ``show ipcp'' when necessary.
1998-04-16 00:26:21 +00:00
lcp->want_protocomp = IsEnabled(lcp->cfg.protocomp) ? 1 : 0;
Initialise multilink LCPs according to rfc1990.
1998-04-03 19:26:12 +00:00
lcp->want_magic = GenerateMagic();
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
if (IsEnabled(lcp->cfg.chap05)) {
lcp->want_auth = PROTO_CHAP;
lcp->want_authtype = 0x05;
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
#ifndef NODES
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
} else if (IsEnabled(lcp->cfg.chap80nt) ||
IsEnabled(lcp->cfg.chap80lm)) {
lcp->want_auth = PROTO_CHAP;
lcp->want_authtype = 0x80;
Add MPPE and MSChap v2 support (denied and disabled by default) Submitted by: Ustimenko Semen <semen@iclub.nsu.ru>
2000-10-30 00:15:04 +00:00
} else if (IsEnabled(lcp->cfg.chap81)) {
lcp->want_auth = PROTO_CHAP;
lcp->want_authtype = 0x81;
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
#endif
} else if (IsEnabled(lcp->cfg.pap)) {
lcp->want_auth = PROTO_PAP;
lcp->want_authtype = 0;
} else {
lcp->want_auth = 0;
lcp->want_authtype = 0;
}
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
if (p->type != PHYS_DIRECT)
Various whitespace changes. Make some functions static.
2000-10-30 00:15:29 +00:00
memcpy(&lcp->want_callback, &p->dl->cfg.callback,
sizeof(struct callback));
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
else
lcp->want_callback.opmask = 0;
o Move the accept/deny/disable/enable globals either to the bundle (non-negotiated vars) or to their respective IPCP, LCP or CCP. o Enable rolling throughput statistics by default. o Remove the `display' command. These values now appear in `show bundle', `show ipcp', `show ccp' and `show lcp'. o Initialise auth name & key at bundle create time (oops). o Rename pppd-deflate (the id-24 hack) to deflate24. o Don't send both a REJ and a NAK to an IPCP or LCP REQ. Favour the REJ (already done at the CCP level). o Recurse in datalink_UpdateSet() when we change state, otherwise we end up setting no descriptors and getting jammed in the imminent select() instead of doing the dial/login/hangup. o Display our CHAP encryption method despite being built with DES. o Display VJ as not negotiated in ``show ipcp'' when necessary.
1998-04-16 00:26:21 +00:00
lcp->want_lqrperiod = IsEnabled(lcp->cfg.lqr) ?
lcp->cfg.lqrperiod * 100 : 0;
Initialise multilink LCPs according to rfc1990.
1998-04-03 19:26:12 +00:00
} else {
lcp->his_accmap = lcp->want_accmap = 0;
lcp->his_protocomp = lcp->want_protocomp = 1;
lcp->want_magic = 0;
lcp->want_auth = 0;
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
lcp->want_authtype = 0;
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
lcp->want_callback.opmask = 0;
Initialise multilink LCPs according to rfc1990.
1998-04-03 19:26:12 +00:00
lcp->want_lqrperiod = 0;
}
Shuffle around our FSMs a bit. This'll make it easier to remove the CcpInfo, LcpInfo and IpcpInfo globals.
1998-02-21 01:45:26 +00:00
lcp->his_reject = lcp->my_reject = 0;
lcp->auth_iwait = lcp->auth_ineed = 0;
o Move struct lcp and struct ccp into struct link. o Remove bundle2lcp(), bundle2ccp() and bundle2link(). They're too resource-hungry and we have `owner pointers' to do their job. o Make our FSM understand LCPs that are always ST_OPENED (with a minimum code that != 1). o Send FSM code rejects for invalid codes. o Make our bundle fsm_parent deal with multiple links. o Make timer diagnostics pretty and allow access via ~t in `term' mode (not just when logging debug) and `show timers'. Only show timers every second in debug mode, otherwise we get too many diagnostics to be useful (we probably still do). Also, don't restrict ~m in term mode to depend on debug logging. o Rationalise our bundles' phases. o Create struct mp (multilink protocol). This is both an NCP and a type of struct link. It feeds off other NCPs for output, passing fragmented packets into the queues of available datalinks. It also gets PROTO_MP input, reassembles the fragments into ppp frames, and passes them back to the HDLC layer that the fragments were passed from. ** It's not yet possible to enter multilink mode :-( ** o Add `set weight' (requires context) for deciding on a links weighting in multilink mode. Weighting is simplistic (and probably badly implemented) for now. o Remove the function pointers in struct link. They ended up only applying to physical links. o Configure our tun device with an MTU equal to the MRU from struct mp's LCP and a speed equal to the sum of our link speeds. o `show {lcp,ccp,proto}' and `set deflate' now have optional context and use ChooseLink() to decide on which `struct link' to use. This allows behaviour as before when in non-multilink mode, and allows access to the MP logical link in multilink mode. o Ignore reconnect and redial values when in -direct mode and when cleaning up. Always redial when in -ddial or -dedicated mode (unless cleaning up). o Tell our links to `staydown' when we close them due to a signal. o Remove remaining `#ifdef SIGALRM's (ppp doesn't function without alarms). o Don't bother strdup()ing our physical link name. o Various other cosmetic changes.
1998-04-03 19:21:56 +00:00
lcp->LcpFailedMagic = 0;
1995-01-31 06:29:58 +00:00
}
This commit was manufactured by cvs2svn to create branch 'MP'.
1998-01-29 00:44:16 +00:00
static void
Allow control over the number of ConfigREQ & TermREQ attempts that are made in each of the FSMs (LCP, CCP & IPCP) and the number of REQs/Challenges for PAP/CHAP by accepting more arguments in the ``set {c,ip,l}cpretry'' and ``set {ch,p}apretry'' commands. Change the non-convergence thresholds to 3 times the number of configured REQ tries (rather than the previous fixed ``10''). We now notice repeated NAKs and REJs rather than just REQs. Don't suggest that CHAP 0x05 isn't supported when it's not configured. Fix some bugs that expose themselves with smaller numbers of retries: o Handle instantaneous disconnects (set device /dev/null) correctly by stopping all fsm timers in fsm2initial. o Don't forget to uu_unlock() devices that are files but are not ttys (set device /dev/zero). Fix a *HORRENDOUS* bug in RFC1661 (already fixed for an Open event in state ``Closed''): According to the state transition table, a RCR+ or RCR- received in the ``Stopped'' state are supposed to InitRestartCounter, SendConfigReq and SendConfig{Ack,Nak}. However, in ``Stopped'', we haven't yet done a TLS (or the last thing we did is a TLF). We must therefore do the TLS at this point ! This was never noticed before because LCP and CCP used not use LayerStart() for anything interesting, and IPCP tends to go into Stopped then get a Down because of an LCP RTR rather than getting a RCR again.
1999-02-26 21:28:14 +00:00
LcpInitRestartCounter(struct fsm *fp, int what)
1995-01-31 06:29:58 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* Set fsm timer load */
o Move default MRU, MTU, ACCMAP and OPENMODE config values into struct lcp and display them in `show lcp'. o Remove `show mru' and `show mtu' and make the data part of `show lcp'. Also merge `set m[tr]u' and `set openmode' implementations into the SetVariable function. o `set timeout' only accepts the idle timer value as an argument. o Move our lqr period into struct lcp, and create a `set lqrperiod' command. Display it in `show lcp'. o Remove VarRetryTimeout, and implement it at the LCP, PAP, CHAP, CCP and IPCP levels, creating individual `set XXXretry' commands for each. They must be separate because they have different context requirements in multilink mode. o Display default config values in `show ccp'. o Tart the man page up a bit (wrt PPP/TCP, compression and LQR) and explain the new commands.
1998-04-03 19:24:07 +00:00
struct lcp *lcp = fsm2lcp(fp);
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
Allow control over the number of ConfigREQ & TermREQ attempts that are made in each of the FSMs (LCP, CCP & IPCP) and the number of REQs/Challenges for PAP/CHAP by accepting more arguments in the ``set {c,ip,l}cpretry'' and ``set {ch,p}apretry'' commands. Change the non-convergence thresholds to 3 times the number of configured REQ tries (rather than the previous fixed ``10''). We now notice repeated NAKs and REJs rather than just REQs. Don't suggest that CHAP 0x05 isn't supported when it's not configured. Fix some bugs that expose themselves with smaller numbers of retries: o Handle instantaneous disconnects (set device /dev/null) correctly by stopping all fsm timers in fsm2initial. o Don't forget to uu_unlock() devices that are files but are not ttys (set device /dev/zero). Fix a *HORRENDOUS* bug in RFC1661 (already fixed for an Open event in state ``Closed''): According to the state transition table, a RCR+ or RCR- received in the ``Stopped'' state are supposed to InitRestartCounter, SendConfigReq and SendConfig{Ack,Nak}. However, in ``Stopped'', we haven't yet done a TLS (or the last thing we did is a TLF). We must therefore do the TLS at this point ! This was never noticed before because LCP and CCP used not use LayerStart() for anything interesting, and IPCP tends to go into Stopped then get a Down because of an LCP RTR rather than getting a RCR again.
1999-02-26 21:28:14 +00:00
fp->FsmTimer.load = lcp->cfg.fsm.timeout * SECTICKS;
switch (what) {
case FSM_REQ_TIMER:
fp->restart = lcp->cfg.fsm.maxreq;
break;
case FSM_TRM_TIMER:
fp->restart = lcp->cfg.fsm.maxtrm;
break;
default:
fp->restart = 1;
break;
}
1995-01-31 06:29:58 +00:00
}
static void
Create `struct link' - the logical link from which `struct physical' (and the future `struct logical') are derived.
1998-01-30 19:46:07 +00:00
LcpSendConfigReq(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* Send config REQ please */
Create `struct link' - the logical link from which `struct physical' (and the future `struct logical') are derived.
1998-01-30 19:46:07 +00:00
struct physical *p = link2physical(fp->link);
Rename ccpstate to ccp, ipcpstate to ipcp and lcpstate to lcp.
1998-02-08 11:05:01 +00:00
struct lcp *lcp = fsm2lcp(fp);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
u_char buff[200];
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
struct fsm_opt *o;
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
struct mp *mp;
Don't cast potentially unaligned addresses to pointers to non-char types on non-i386 architectures. On Alpha and Sparc we get a bus error if we do.
1998-09-04 18:26:00 +00:00
u_int16_t proto;
Ignore (with a warning message) mtu/mru configurations that are greater than the maximum physical values. MFC after: 1 week
2001-07-17 01:06:13 +00:00
u_short maxmru;
This commit was manufactured by cvs2svn to create branch 'MP'.
1998-01-29 00:44:16 +00:00
Create `struct link' - the logical link from which `struct physical' (and the future `struct logical') are derived.
1998-01-30 19:46:07 +00:00
if (!p) {
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogERROR, "%s: LcpSendConfigReq: Not a physical link !\n",
o Remove LINK level logging. o change the default link name to ``deflink'' rather than ``default''. o Prepend the link name to CCP and LCP FSM diagnostics. o Protect against 0 length options in CCP and IPCP REQ interpreters (already done for LCP). o Allow optional context for the `show' command. o Use MPs link when interpreting commands if the multilink mrru is configured rather than when multilink is active. This means that once we've ``set mrru xxx'', we then need to ``link deflink show ccp'' etc if we want to do link-level stuff (based on the command requiring optional or manditory context). o Use the ifconfig'd interface address in `set enddisc {ip,mac}' if it's there, otherwise the configuration file value.
1998-04-24 19:15:48 +00:00
fp->link->name);
Create `struct link' - the logical link from which `struct physical' (and the future `struct logical') are derived.
1998-01-30 19:46:07 +00:00
return;
}
1995-01-31 06:29:58 +00:00
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
o = (struct fsm_opt *)buff;
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
if (!physical_IsSync(p)) {
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
if (lcp->want_acfcomp && !REJECTED(lcp, TY_ACFCOMP))
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_ACFCOMP, 2, o);
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
if (lcp->want_protocomp && !REJECTED(lcp, TY_PROTOCOMP))
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_PROTOCOMP, 2, o);
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
o Check the LCP Identifier field for correctness and drop dodgy packets by default. The old behaviour is still available with ``disable idcheck''. o Make all FSM diagnostics consistent and tidy up the way we build our LCP/CCP/IPCP requests. o Don't assume sizeof(u_long) == 4.
1998-03-20 19:47:10 +00:00
if (!REJECTED(lcp, TY_ACCMAP)) {
Don't cast potentially unaligned addresses to pointers to non-char types on non-i386 architectures. On Alpha and Sparc we get a bus error if we do.
1998-09-04 18:26:00 +00:00
ua_htonl(&lcp->want_accmap, o->data);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_ACCMAP, 6, o);
o Check the LCP Identifier field for correctness and drop dodgy packets by default. The old behaviour is still available with ``disable idcheck''. o Make all FSM diagnostics consistent and tidy up the way we build our LCP/CCP/IPCP requests. o Don't assume sizeof(u_long) == 4.
1998-03-20 19:47:10 +00:00
}
1995-01-31 06:29:58 +00:00
}
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
Ignore (with a warning message) mtu/mru configurations that are greater than the maximum physical values. MFC after: 1 week
2001-07-17 01:06:13 +00:00
maxmru = p ? physical_DeviceMTU(p) : 0;
if (lcp->cfg.max_mru && (!maxmru || maxmru > lcp->cfg.max_mru))
maxmru = lcp->cfg.max_mru;
if (maxmru && lcp->want_mru > maxmru) {
log_Printf(LogWARN, "%s: Reducing configured MRU from %u to %u\n",
fp->link->name, lcp->want_mru, maxmru);
lcp->want_mru = maxmru;
}
If the peer REJects our MRU REQ, stop REQing it -- *EVEN* if we're doing PPPoE and the default MRU is therefore too big. When negotiating with win2k, we ask for MRU 1492 and the win2k box NAKs us saying ``MRU 1492''. This doesn't make sense to me. When we continue to request MRU 1492, the win2k box eventually REJs our MRU. This fix allows negotiations to continue at that point, bringing the link up and potentially allowing the win2k box to send us frames that are too large. AFAICT this is better than failing to bring the link up.... probably ! I have no idea how to do the equivalent of ``route get'' or ``ifconfig -a'' under win2k, so I can't tell what MTU it actually ends up using. I believe the bug is in win2k (it's certainly mis-negotiating). I'll MFC given the release engineers permission as code freeze begins on August 1. PR: 29277 MFC after: 3 days
2001-07-30 17:04:39 +00:00
if (!REJECTED(lcp, TY_MRU)) {
Don't cast potentially unaligned addresses to pointers to non-char types on non-i386 architectures. On Alpha and Sparc we get a bus error if we do.
1998-09-04 18:26:00 +00:00
ua_htons(&lcp->want_mru, o->data);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_MRU, 4, o);
o Check the LCP Identifier field for correctness and drop dodgy packets by default. The old behaviour is still available with ``disable idcheck''. o Make all FSM diagnostics consistent and tidy up the way we build our LCP/CCP/IPCP requests. o Don't assume sizeof(u_long) == 4.
1998-03-20 19:47:10 +00:00
}
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
o Check the LCP Identifier field for correctness and drop dodgy packets by default. The old behaviour is still available with ``disable idcheck''. o Make all FSM diagnostics consistent and tidy up the way we build our LCP/CCP/IPCP requests. o Don't assume sizeof(u_long) == 4.
1998-03-20 19:47:10 +00:00
if (lcp->want_magic && !REJECTED(lcp, TY_MAGICNUM)) {
Don't cast potentially unaligned addresses to pointers to non-char types on non-i386 architectures. On Alpha and Sparc we get a bus error if we do.
1998-09-04 18:26:00 +00:00
ua_htonl(&lcp->want_magic, o->data);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_MAGICNUM, 6, o);
o Check the LCP Identifier field for correctness and drop dodgy packets by default. The old behaviour is still available with ``disable idcheck''. o Make all FSM diagnostics consistent and tidy up the way we build our LCP/CCP/IPCP requests. o Don't assume sizeof(u_long) == 4.
1998-03-20 19:47:10 +00:00
}
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
o Check the LCP Identifier field for correctness and drop dodgy packets by default. The old behaviour is still available with ``disable idcheck''. o Make all FSM diagnostics consistent and tidy up the way we build our LCP/CCP/IPCP requests. o Don't assume sizeof(u_long) == 4.
1998-03-20 19:47:10 +00:00
if (lcp->want_lqrperiod && !REJECTED(lcp, TY_QUALPROTO)) {
Don't cast potentially unaligned addresses to pointers to non-char types on non-i386 architectures. On Alpha and Sparc we get a bus error if we do.
1998-09-04 18:26:00 +00:00
proto = PROTO_LQR;
ua_htons(&proto, o->data);
ua_htonl(&lcp->want_lqrperiod, o->data + 2);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_QUALPROTO, 8, o);
o Check the LCP Identifier field for correctness and drop dodgy packets by default. The old behaviour is still available with ``disable idcheck''. o Make all FSM diagnostics consistent and tidy up the way we build our LCP/CCP/IPCP requests. o Don't assume sizeof(u_long) == 4.
1998-03-20 19:47:10 +00:00
}
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
1995-01-31 06:29:58 +00:00
switch (lcp->want_auth) {
case PROTO_PAP:
Don't cast potentially unaligned addresses to pointers to non-char types on non-i386 architectures. On Alpha and Sparc we get a bus error if we do.
1998-09-04 18:26:00 +00:00
proto = PROTO_PAP;
ua_htons(&proto, o->data);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_AUTHPROTO, 4, o);
1995-01-31 06:29:58 +00:00
break;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
1995-01-31 06:29:58 +00:00
case PROTO_CHAP:
Don't cast potentially unaligned addresses to pointers to non-char types on non-i386 architectures. On Alpha and Sparc we get a bus error if we do.
1998-09-04 18:26:00 +00:00
proto = PROTO_CHAP;
ua_htons(&proto, o->data);
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
o->data[2] = lcp->want_authtype;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_AUTHPROTO, 5, o);
1995-01-31 06:29:58 +00:00
break;
}
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
if (!REJECTED(lcp, TY_CALLBACK)) {
if (lcp->want_callback.opmask & CALLBACK_BIT(CALLBACK_AUTH)) {
*o->data = CALLBACK_AUTH;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_CALLBACK, 3, o);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
} else if (lcp->want_callback.opmask & CALLBACK_BIT(CALLBACK_CBCP)) {
*o->data = CALLBACK_CBCP;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_CALLBACK, 3, o);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
} else if (lcp->want_callback.opmask & CALLBACK_BIT(CALLBACK_E164)) {
int sz = strlen(lcp->want_callback.msg);
if (sz > sizeof o->data - 1) {
sz = sizeof o->data - 1;
log_Printf(LogWARN, "Truncating E164 data to %d octets (oops!)\n", sz);
}
*o->data = CALLBACK_E164;
memcpy(o->data + 1, lcp->want_callback.msg, sz);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_CALLBACK, sz + 3, o);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
}
}
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
if (lcp->want_mrru && !REJECTED(lcp, TY_MRRU)) {
Don't cast potentially unaligned addresses to pointers to non-char types on non-i386 architectures. On Alpha and Sparc we get a bus error if we do.
1998-09-04 18:26:00 +00:00
ua_htons(&lcp->want_mrru, o->data);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_MRRU, 4, o);
1995-01-31 06:29:58 +00:00
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
if (lcp->want_shortseq && !REJECTED(lcp, TY_SHORTSEQ))
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_SHORTSEQ, 2, o);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
}
1995-01-31 06:29:58 +00:00
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
mp = &lcp->fsm.bundle->ncp.mp;
Allow our endpoint discriminator to be enabled, disabled, accepted and denied. This is necessary for some MP implementations that get confused if you accept their endpoint discriminator but reject their MRRU.
1999-06-09 16:54:04 +00:00
if (mp->cfg.enddisc.class != 0 && IsEnabled(mp->cfg.negenddisc) &&
!REJECTED(lcp, TY_ENDDISC)) {
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
*o->data = mp->cfg.enddisc.class;
memcpy(o->data+1, mp->cfg.enddisc.address, mp->cfg.enddisc.len);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
INC_FSM_OPT(TY_ENDDISC, mp->cfg.enddisc.len + 3, o);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
}
1995-01-31 06:29:58 +00:00
o Alter the mbuf type as it's processed by different layers. o Show more information about missing MP fragments in ``show mp''. o Do away with mbuf_Log(). It was showing mbuf stats twice on receipt of LCP/CCP/IPCP packets.... ???!!? o Pre-allocate a bit extra when creating LQR packets to avoid having to allocate another mbuf in mbuf_Prepend().
1999-06-02 15:59:09 +00:00
fsm_Output(fp, CODE_CONFIGREQ, fp->reqid, buff, (u_char *)o - buff,
MB_LCPOUT);
1995-01-31 06:29:58 +00:00
}
This commit was manufactured by cvs2svn to create branch 'MP'.
1998-01-29 00:44:16 +00:00
void
o Don't immediately reply to the first LQR thinking it's a duplicate ('cos it compares with our initial values). o Move the LCP into struct datalink.
1998-03-13 21:07:14 +00:00
lcp_SendProtoRej(struct lcp *lcp, u_char *option, int count)
1995-01-31 06:29:58 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* Don't understand `option' */
o Alter the mbuf type as it's processed by different layers. o Show more information about missing MP fragments in ``show mp''. o Do away with mbuf_Log(). It was showing mbuf stats twice on receipt of LCP/CCP/IPCP packets.... ???!!? o Pre-allocate a bit extra when creating LQR packets to avoid having to allocate another mbuf in mbuf_Prepend().
1999-06-02 15:59:09 +00:00
fsm_Output(&lcp->fsm, CODE_PROTOREJ, lcp->fsm.reqid, option, count,
MB_LCPOUT);
1995-01-31 06:29:58 +00:00
}
Support link identification from rfc1570 Two new commands are available; ``ident'' and ``sendident''.
2000-07-19 02:10:35 +00:00
int
lcp_SendIdentification(struct lcp *lcp)
{
static u_char id; /* Use a private id */
u_char msg[DEF_MRU - 3];
const char *argv[2];
char *exp[2];
if (*lcp->cfg.ident == '\0')
return 0;
argv[0] = lcp->cfg.ident;
argv[1] = NULL;
command_Expand(exp, 1, argv, lcp->fsm.bundle, 1, getpid());
ua_htonl(&lcp->want_magic, msg);
strncpy(msg + 4, exp[0], sizeof msg - 5);
msg[sizeof msg - 1] = '\0';
fsm_Output(&lcp->fsm, CODE_IDENT, id++, msg, 4 + strlen(msg + 4), MB_LCPOUT);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
log_Printf(LogLCP, " MAGICNUM %08x\n", lcp->want_magic);
log_Printf(LogLCP, " TEXT %s\n", msg + 4);
Support link identification from rfc1570 Two new commands are available; ``ident'' and ``sendident''.
2000-07-19 02:10:35 +00:00
Add a ``log'' command for logging specific information. Add an ``UPTIME'' variable to indicate the bundle uptime. It's now possible to put something like this in ppp.linkdown for a server setup: MYADDR: log Session closing: User USER, address HISADDR, up UPTIME Fixed some memory leakage with commands that expand words. Made some functions static. Fixed a diagnostic bug (iface add .... SIOCDIFADDR)
2001-11-03 21:45:32 +00:00
command_Free(1, exp);
Support link identification from rfc1570 Two new commands are available; ``ident'' and ``sendident''.
2000-07-19 02:10:35 +00:00
return 1;
}
void
lcp_RecvIdentification(struct lcp *lcp, char *data)
{
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
log_Printf(LogLCP, " MAGICNUM %08x\n", lcp->his_magic);
log_Printf(LogLCP, " TEXT %s\n", data);
Support link identification from rfc1570 Two new commands are available; ``ident'' and ``sendident''.
2000-07-19 02:10:35 +00:00
}
New user Process PPP based on iij-ppp0.94beta2. o Supporting SYNC SIO device (But need a device driver) - add "set speed sync" o Fixing bug for Predictor-1 function. o Add new parameter that re-sent interval for set timeout commands. o Improving RTT (Round Trip Time) and reducing processor time. - Previous Timer service was using polling, and now using SIGALRM ;-) - A 0.94beta2 will not work correctly.... -- Follows are additinal feature not including 0.94beta2 o Support Proxy ARP - add "enable/disable proxy" commands o Marging common routine in CHAP/PAP. o Enhancing LCP/IPCP log information. o Support local Authfication connection on port 300x and tty. - You can set up pair of your "hostname -s" and password in ppp.secret. if either ppp.secret file nor your hostname line don't exist, It will notify a message and working as same as previous version.(Backword compatibility) - If you did set up them, It's allow connection but nothing to do except help and passwd command. - add "passwd yourpasswd" commands o Support afilter - keep Alive filter that a packet can send/receiving according to ifilter/ofilter but doesn't count it as preventing idle timer expires. - Same syntax of other filters. o Fixing bugs reported by current user for previous one. Thanks !! Reviewed by: Atsushi Murai (amurai@spec.co.jp)
1995-02-26 12:18:08 +00:00
static void
Allow control over the number of ConfigREQ & TermREQ attempts that are made in each of the FSMs (LCP, CCP & IPCP) and the number of REQs/Challenges for PAP/CHAP by accepting more arguments in the ``set {c,ip,l}cpretry'' and ``set {ch,p}apretry'' commands. Change the non-convergence thresholds to 3 times the number of configured REQ tries (rather than the previous fixed ``10''). We now notice repeated NAKs and REJs rather than just REQs. Don't suggest that CHAP 0x05 isn't supported when it's not configured. Fix some bugs that expose themselves with smaller numbers of retries: o Handle instantaneous disconnects (set device /dev/null) correctly by stopping all fsm timers in fsm2initial. o Don't forget to uu_unlock() devices that are files but are not ttys (set device /dev/zero). Fix a *HORRENDOUS* bug in RFC1661 (already fixed for an Open event in state ``Closed''): According to the state transition table, a RCR+ or RCR- received in the ``Stopped'' state are supposed to InitRestartCounter, SendConfigReq and SendConfig{Ack,Nak}. However, in ``Stopped'', we haven't yet done a TLS (or the last thing we did is a TLF). We must therefore do the TLS at this point ! This was never noticed before because LCP and CCP used not use LayerStart() for anything interesting, and IPCP tends to go into Stopped then get a Down because of an LCP RTR rather than getting a RCR again.
1999-02-26 21:28:14 +00:00
LcpSentTerminateReq(struct fsm *fp)
New user Process PPP based on iij-ppp0.94beta2. o Supporting SYNC SIO device (But need a device driver) - add "set speed sync" o Fixing bug for Predictor-1 function. o Add new parameter that re-sent interval for set timeout commands. o Improving RTT (Round Trip Time) and reducing processor time. - Previous Timer service was using polling, and now using SIGALRM ;-) - A 0.94beta2 will not work correctly.... -- Follows are additinal feature not including 0.94beta2 o Support Proxy ARP - add "enable/disable proxy" commands o Marging common routine in CHAP/PAP. o Enhancing LCP/IPCP log information. o Support local Authfication connection on port 300x and tty. - You can set up pair of your "hostname -s" and password in ppp.secret. if either ppp.secret file nor your hostname line don't exist, It will notify a message and working as same as previous version.(Backword compatibility) - If you did set up them, It's allow connection but nothing to do except help and passwd command. - add "passwd yourpasswd" commands o Support afilter - keep Alive filter that a packet can send/receiving according to ifilter/ofilter but doesn't count it as preventing idle timer expires. - Same syntax of other filters. o Fixing bugs reported by current user for previous one. Thanks !! Reviewed by: Atsushi Murai (amurai@spec.co.jp)
1995-02-26 12:18:08 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* Term REQ just sent by FSM */
New user Process PPP based on iij-ppp0.94beta2. o Supporting SYNC SIO device (But need a device driver) - add "set speed sync" o Fixing bug for Predictor-1 function. o Add new parameter that re-sent interval for set timeout commands. o Improving RTT (Round Trip Time) and reducing processor time. - Previous Timer service was using polling, and now using SIGALRM ;-) - A 0.94beta2 will not work correctly.... -- Follows are additinal feature not including 0.94beta2 o Support Proxy ARP - add "enable/disable proxy" commands o Marging common routine in CHAP/PAP. o Enhancing LCP/IPCP log information. o Support local Authfication connection on port 300x and tty. - You can set up pair of your "hostname -s" and password in ppp.secret. if either ppp.secret file nor your hostname line don't exist, It will notify a message and working as same as previous version.(Backword compatibility) - If you did set up them, It's allow connection but nothing to do except help and passwd command. - add "passwd yourpasswd" commands o Support afilter - keep Alive filter that a packet can send/receiving according to ifilter/ofilter but doesn't count it as preventing idle timer expires. - Same syntax of other filters. o Fixing bugs reported by current user for previous one. Thanks !! Reviewed by: Atsushi Murai (amurai@spec.co.jp)
1995-02-26 12:18:08 +00:00
}
1995-01-31 06:29:58 +00:00
static void
o Check the LCP Identifier field for correctness and drop dodgy packets by default. The old behaviour is still available with ``disable idcheck''. o Make all FSM diagnostics consistent and tidy up the way we build our LCP/CCP/IPCP requests. o Don't assume sizeof(u_long) == 4.
1998-03-20 19:47:10 +00:00
LcpSendTerminateAck(struct fsm *fp, u_char id)
1995-01-31 06:29:58 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* Send Term ACK please */
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
struct physical *p = link2physical(fp->link);
if (p && p->dl->state == DATALINK_CBCP)
cbcp_ReceiveTerminateReq(p);
o Alter the mbuf type as it's processed by different layers. o Show more information about missing MP fragments in ``show mp''. o Do away with mbuf_Log(). It was showing mbuf stats twice on receipt of LCP/CCP/IPCP packets.... ???!!? o Pre-allocate a bit extra when creating LQR packets to avoid having to allocate another mbuf in mbuf_Prepend().
1999-06-02 15:59:09 +00:00
fsm_Output(fp, CODE_TERMACK, id, NULL, 0, MB_LCPOUT);
1995-01-31 06:29:58 +00:00
}
static void
Create `struct link' - the logical link from which `struct physical' (and the future `struct logical') are derived.
1998-01-30 19:46:07 +00:00
LcpLayerStart(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* We're about to start up ! */
o Don't immediately reply to the first LQR thinking it's a duplicate ('cos it compares with our initial values). o Move the LCP into struct datalink.
1998-03-13 21:07:14 +00:00
struct lcp *lcp = fsm2lcp(fp);
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
log_Printf(LogLCP, "%s: LayerStart\n", fp->link->name);
o Don't immediately reply to the first LQR thinking it's a duplicate ('cos it compares with our initial values). o Move the LCP into struct datalink.
1998-03-13 21:07:14 +00:00
lcp->LcpFailedMagic = 0;
Allow control over the number of ConfigREQ & TermREQ attempts that are made in each of the FSMs (LCP, CCP & IPCP) and the number of REQs/Challenges for PAP/CHAP by accepting more arguments in the ``set {c,ip,l}cpretry'' and ``set {ch,p}apretry'' commands. Change the non-convergence thresholds to 3 times the number of configured REQ tries (rather than the previous fixed ``10''). We now notice repeated NAKs and REJs rather than just REQs. Don't suggest that CHAP 0x05 isn't supported when it's not configured. Fix some bugs that expose themselves with smaller numbers of retries: o Handle instantaneous disconnects (set device /dev/null) correctly by stopping all fsm timers in fsm2initial. o Don't forget to uu_unlock() devices that are files but are not ttys (set device /dev/zero). Fix a *HORRENDOUS* bug in RFC1661 (already fixed for an Open event in state ``Closed''): According to the state transition table, a RCR+ or RCR- received in the ``Stopped'' state are supposed to InitRestartCounter, SendConfigReq and SendConfig{Ack,Nak}. However, in ``Stopped'', we haven't yet done a TLS (or the last thing we did is a TLF). We must therefore do the TLS at this point ! This was never noticed before because LCP and CCP used not use LayerStart() for anything interesting, and IPCP tends to go into Stopped then get a Down because of an LCP RTR rather than getting a RCR again.
1999-02-26 21:28:14 +00:00
fp->more.reqs = fp->more.naks = fp->more.rejs = lcp->cfg.fsm.maxreq * 3;
When the peer fails to specify an MRU and a 1500 byte MRU is not allowed either because of the transport or configuration, send a MRU NAK only once, then allow the negotiations to proceed. rfc1661 says that 1500 should always be allowed and rfc2516 says that 1492 is the maximum for PPPoE. This changes ppp so that it only weakly suggests 1492, then goes with the default (leaving the problem in the hands of the peer WRT how they set their MTU). MFC after: 1 week
2001-10-18 08:47:56 +00:00
lcp->mru_req = 0;
1995-01-31 06:29:58 +00:00
}
static void
Create `struct link' - the logical link from which `struct physical' (and the future `struct logical') are derived.
1998-01-30 19:46:07 +00:00
LcpLayerFinish(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* We're now down */
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
log_Printf(LogLCP, "%s: LayerFinish\n", fp->link->name);
1995-01-31 06:29:58 +00:00
}
o Create a new ``timer'' log level. This lets us ``set log debug'' without filling our filesystem/screen with junk that we don't really want to see. o change PHYS_STDIN to PHYS_DIRECT - we can handle incoming connections that aren't on STDIN_FILENO now. o Allow return values from our FSM LayerUp functions. If LayerUp() fails, the FSM does an immediate FsmDown() without calling the fsm_parent's Layer{Up,Down} functions. o Clear the close-on-exec flag of file descriptor 3 when executing chat programs so that our documented ability to communicate with /dev/tty via that descriptor works. Also document it as descriptor 3, not 4 :-O o Allow a ``rm'' command as an alias for ``remove''. o Fix the bind()/connect()/accept() calls made by the MP server. o Create bundle_SendDatalink() and bundle_ReceiveDatalink(). This allows `struct datalink's to flatten themselves, pass through a pipe (read: the eye of a needle !) and come alive at the other end. The donator then fork()s & exec()s pppmpipe, ``passing'' the connection to another ppp instance. *** PPP NOW TALKS MULTILINK :-))) *** Our link utilization is hideous, and lots of code needs tidying still. It's also probably riddled with bugs ! It's been tested against itself only, and has hung once, so confidence isn't high....
1998-04-30 23:53:56 +00:00
static int
Create `struct link' - the logical link from which `struct physical' (and the future `struct logical') are derived.
1998-01-30 19:46:07 +00:00
LcpLayerUp(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* We're now up */
Create `struct link' - the logical link from which `struct physical' (and the future `struct logical') are derived.
1998-01-30 19:46:07 +00:00
struct physical *p = link2physical(fp->link);
Rename ccpstate to ccp, ipcpstate to ipcp and lcpstate to lcp.
1998-02-08 11:05:01 +00:00
struct lcp *lcp = fsm2lcp(fp);
Create `struct link' - the logical link from which `struct physical' (and the future `struct logical') are derived.
1998-01-30 19:46:07 +00:00
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
log_Printf(LogLCP, "%s: LayerUp\n", fp->link->name);
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
physical_SetAsyncParams(p, lcp->want_accmap, lcp->his_accmap);
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
lqr_Start(lcp);
o Move struct lcp and struct ccp into struct link. o Remove bundle2lcp(), bundle2ccp() and bundle2link(). They're too resource-hungry and we have `owner pointers' to do their job. o Make our FSM understand LCPs that are always ST_OPENED (with a minimum code that != 1). o Send FSM code rejects for invalid codes. o Make our bundle fsm_parent deal with multiple links. o Make timer diagnostics pretty and allow access via ~t in `term' mode (not just when logging debug) and `show timers'. Only show timers every second in debug mode, otherwise we get too many diagnostics to be useful (we probably still do). Also, don't restrict ~m in term mode to depend on debug logging. o Rationalise our bundles' phases. o Create struct mp (multilink protocol). This is both an NCP and a type of struct link. It feeds off other NCPs for output, passing fragmented packets into the queues of available datalinks. It also gets PROTO_MP input, reassembles the fragments into ppp frames, and passes them back to the HDLC layer that the fragments were passed from. ** It's not yet possible to enter multilink mode :-( ** o Add `set weight' (requires context) for deciding on a links weighting in multilink mode. Weighting is simplistic (and probably badly implemented) for now. o Remove the function pointers in struct link. They ended up only applying to physical links. o Configure our tun device with an MTU equal to the MRU from struct mp's LCP and a speed equal to the sum of our link speeds. o `show {lcp,ccp,proto}' and `set deflate' now have optional context and use ChooseLink() to decide on which `struct link' to use. This allows behaviour as before when in non-multilink mode, and allows access to the MP logical link in multilink mode. o Ignore reconnect and redial values when in -direct mode and when cleaning up. Always redial when in -ddial or -dedicated mode (unless cleaning up). o Tell our links to `staydown' when we close them due to a signal. o Remove remaining `#ifdef SIGALRM's (ppp doesn't function without alarms). o Don't bother strdup()ing our physical link name. o Various other cosmetic changes.
1998-04-03 19:21:56 +00:00
hdlc_StartTimer(&p->hdlc);
Allow control over the number of ConfigREQ & TermREQ attempts that are made in each of the FSMs (LCP, CCP & IPCP) and the number of REQs/Challenges for PAP/CHAP by accepting more arguments in the ``set {c,ip,l}cpretry'' and ``set {ch,p}apretry'' commands. Change the non-convergence thresholds to 3 times the number of configured REQ tries (rather than the previous fixed ``10''). We now notice repeated NAKs and REJs rather than just REQs. Don't suggest that CHAP 0x05 isn't supported when it's not configured. Fix some bugs that expose themselves with smaller numbers of retries: o Handle instantaneous disconnects (set device /dev/null) correctly by stopping all fsm timers in fsm2initial. o Don't forget to uu_unlock() devices that are files but are not ttys (set device /dev/zero). Fix a *HORRENDOUS* bug in RFC1661 (already fixed for an Open event in state ``Closed''): According to the state transition table, a RCR+ or RCR- received in the ``Stopped'' state are supposed to InitRestartCounter, SendConfigReq and SendConfig{Ack,Nak}. However, in ``Stopped'', we haven't yet done a TLS (or the last thing we did is a TLF). We must therefore do the TLS at this point ! This was never noticed before because LCP and CCP used not use LayerStart() for anything interesting, and IPCP tends to go into Stopped then get a Down because of an LCP RTR rather than getting a RCR again.
1999-02-26 21:28:14 +00:00
fp->more.reqs = fp->more.naks = fp->more.rejs = lcp->cfg.fsm.maxreq * 3;
Support link identification from rfc1570 Two new commands are available; ``ident'' and ``sendident''.
2000-07-19 02:10:35 +00:00
lcp_SendIdentification(lcp);
o Create a new ``timer'' log level. This lets us ``set log debug'' without filling our filesystem/screen with junk that we don't really want to see. o change PHYS_STDIN to PHYS_DIRECT - we can handle incoming connections that aren't on STDIN_FILENO now. o Allow return values from our FSM LayerUp functions. If LayerUp() fails, the FSM does an immediate FsmDown() without calling the fsm_parent's Layer{Up,Down} functions. o Clear the close-on-exec flag of file descriptor 3 when executing chat programs so that our documented ability to communicate with /dev/tty via that descriptor works. Also document it as descriptor 3, not 4 :-O o Allow a ``rm'' command as an alias for ``remove''. o Fix the bind()/connect()/accept() calls made by the MP server. o Create bundle_SendDatalink() and bundle_ReceiveDatalink(). This allows `struct datalink's to flatten themselves, pass through a pipe (read: the eye of a needle !) and come alive at the other end. The donator then fork()s & exec()s pppmpipe, ``passing'' the connection to another ppp instance. *** PPP NOW TALKS MULTILINK :-))) *** Our link utilization is hideous, and lots of code needs tidying still. It's also probably riddled with bugs ! It's been tested against itself only, and has hung once, so confidence isn't high....
1998-04-30 23:53:56 +00:00
return 1;
1995-01-31 06:29:58 +00:00
}
This commit was manufactured by cvs2svn to create branch 'MP'.
1998-01-29 00:44:16 +00:00
static void
Create `struct bundle' - the top level control structure. This structure will eventually contain a list of NCPs (currently only IPCP is supported) and a list of physical `struct link's. It will also derive from a struct link itself. Make ModemTimeout() static - it's way to dangerous to be called from outside ! Bump version to 1.9. Our first MP release should be 2.0.
1998-02-02 19:32:16 +00:00
LcpLayerDown(struct fsm *fp)
1995-01-31 06:29:58 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* About to come down */
o Create `struct hdlc' - a part of struct physical. o Move our LCP report timer into struct hdlc - it's really a hdlc timer (fcs errors etc). o Make `show hdlc' require context and make the output more friendly. o Remove all non-const globals from hdlc.c o Output peer-rejected protocols by name - not just ones that we reject.
1998-02-18 19:35:59 +00:00
struct physical *p = link2physical(fp->link);
1995-01-31 06:29:58 +00:00
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
log_Printf(LogLCP, "%s: LayerDown\n", fp->link->name);
o Create `struct hdlc' - a part of struct physical. o Move our LCP report timer into struct hdlc - it's really a hdlc timer (fcs errors etc). o Make `show hdlc' require context and make the output more friendly. o Remove all non-const globals from hdlc.c o Output peer-rejected protocols by name - not just ones that we reject.
1998-02-18 19:35:59 +00:00
hdlc_StopTimer(&p->hdlc);
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
lqr_StopTimer(p);
Add ``ipcp'' as an optional argument to ``open'', and make open capable of re-negotiatiating the various layers. It is now possible to change various link options and then re-open the relevant layer, making the changes effective - for example, switching off VJ compression or starting ECHO LQRs on-the-fly.
1998-06-25 22:33:31 +00:00
lcp_Setup(fsm2lcp(fp), 0);
1995-01-31 06:29:58 +00:00
}
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
static int
E164ok(struct callback *cb, char *req, int sz)
{
char list[sizeof cb->msg], *next;
int len;
if (!strcmp(cb->msg, "*"))
return 1;
strncpy(list, cb->msg, sizeof list - 1);
list[sizeof list - 1] = '\0';
for (next = strtok(list, ","); next; next = strtok(NULL, ",")) {
len = strlen(next);
if (sz == len && !memcmp(list, req, sz))
return 1;
}
return 0;
}
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
static int
lcp_auth_nak(struct lcp *lcp, struct fsm_decode *dec)
{
struct fsm_opt nak;
nak.hdr.id = TY_AUTHPROTO;
if (IsAccepted(lcp->cfg.pap)) {
nak.hdr.len = 4;
nak.data[0] = (unsigned char)(PROTO_PAP >> 8);
nak.data[1] = (unsigned char)PROTO_PAP;
fsm_nak(dec, &nak);
return 1;
}
nak.hdr.len = 5;
nak.data[0] = (unsigned char)(PROTO_CHAP >> 8);
nak.data[1] = (unsigned char)PROTO_CHAP;
if (IsAccepted(lcp->cfg.chap05)) {
nak.data[2] = 0x05;
fsm_nak(dec, &nak);
#ifndef NODES
} else if (IsAccepted(lcp->cfg.chap80nt) ||
IsAccepted(lcp->cfg.chap80lm)) {
nak.data[2] = 0x80;
fsm_nak(dec, &nak);
} else if (IsAccepted(lcp->cfg.chap81)) {
nak.data[2] = 0x81;
fsm_nak(dec, &nak);
#endif
} else {
return 0;
}
return 1;
}
1995-01-31 06:29:58 +00:00
static void
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
LcpDecodeConfig(struct fsm *fp, u_char *cp, u_char *end, int mode_type,
Remove more globals
1998-03-13 21:08:05 +00:00
struct fsm_decode *dec)
1995-01-31 06:29:58 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* Deal with incoming PROTO_LCP */
Rename ccpstate to ccp, ipcpstate to ipcp and lcpstate to lcp.
1998-02-08 11:05:01 +00:00
struct lcp *lcp = fsm2lcp(fp);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
int sz, pos, op, callback_req, chap_type;
o Use __attribute__ (#ifdef __GNUC__) to type-check LogPrintf() and prompt_Printf(), and fix the bits identified as being wrong as a result.
1998-04-19 23:09:03 +00:00
u_int32_t magic, accmap;
Ignore (with a warning message) mtu/mru configurations that are greater than the maximum physical values. MFC after: 1 week
2001-07-17 01:06:13 +00:00
u_short mru, phmtu, maxmtu, maxmru, wantmtu, wantmru, proto;
1995-01-31 06:29:58 +00:00
struct lqrreq *req;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
char request[20], desc[22];
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
struct mp *mp;
o Defer setting up pap/chap based IP numbers & labels until after we've determined if we're going to join another ppp invocation. o Make ``show link'' show all link details, and ``show links'' just give a list of links and their current status. o Show our current label in ``show bundle''. o Allow link cloning and removal as soon as our MRRU is set. o Make ``show lcp'' require context as nothing will ever change in our MP LCP (it's auto-configured as per rfc1990). o Initialise our LQM owner in hdlc_Init(). o Store our endpoint discriminator and authentication name at both the datalink and multilink level and compare them when we've finished AUTHENTICATE phase and before entering NETWORK phase. If they don't match, close the link again. Display the information in the appropriate ``show'' command. o Initialise datalink::phone and datalink::fsmp.object properly when we're cloning the link. o Show which link we're passing LQRs on in our diagnostics. o Reject endpoint discriminator REQs at the logical multilink level. o Remove the rest of our CARRIER and LINK logging setup.
1998-04-24 19:16:15 +00:00
struct physical *p = link2physical(fp->link);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
struct fsm_opt *opt, nak;
1995-01-31 06:29:58 +00:00
When the peer fails to specify an MRU and a 1500 byte MRU is not allowed either because of the transport or configuration, send a MRU NAK only once, then allow the negotiations to proceed. rfc1661 says that 1500 should always be allowed and rfc2516 says that 1492 is the maximum for PPPoE. This changes ppp so that it only weakly suggests 1492, then goes with the default (leaving the problem in the hands of the peer WRT how they set their MTU). MFC after: 1 week
2001-10-18 08:47:56 +00:00
sz = op = callback_req = 0;
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
while (end - cp >= sizeof(opt->hdr)) {
if ((opt = fsm_readopt(&cp)) == NULL)
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
break;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
snprintf(request, sizeof request, " %s[%d]", protoname(opt->hdr.id),
opt->hdr.len);
switch (opt->hdr.id) {
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
case TY_MRRU:
mp = &lcp->fsm.bundle->ncp.mp;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
ua_ntohs(opt->data, &mru);
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, "%s %u\n", request, mru);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
switch (mode_type) {
case MODE_REQ:
if (mp->cfg.mrru) {
if (REJECTED(lcp, TY_MRRU))
/* Ignore his previous reject so that we REQ next time */
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject &= ~(1 << opt->hdr.id);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
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
if (mru > MAX_MRU) {
/* Push him down to MAX_MRU */
lcp->his_mrru = MAX_MRU;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.hdr.id = TY_MRRU;
nak.hdr.len = 4;
ua_htons(&lcp->his_mrru, nak.data);
fsm_nak(dec, &nak);
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
} else if (mru < MIN_MRU) {
Don't assign an MTU based on the peers first-link MRU in MP mode. Use the peers MRRU as we're supposed to.
2000-06-24 23:37:14 +00:00
/* Push him up to MIN_MRU */
lcp->his_mrru = MIN_MRU;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.hdr.id = TY_MRRU;
nak.hdr.len = 4;
ua_htons(&lcp->his_mrru, nak.data);
fsm_nak(dec, &nak);
} else {
Don't assign an MTU based on the peers first-link MRU in MP mode. Use the peers MRRU as we're supposed to.
2000-06-24 23:37:14 +00:00
lcp->his_mrru = mru;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_ack(dec, opt);
}
break;
} else {
fsm_rej(dec, opt);
lcp->my_reject |= (1 << opt->hdr.id);
}
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
break;
case MODE_NAK:
if (mp->cfg.mrru) {
if (REJECTED(lcp, TY_MRRU))
/* Must have changed his mind ! */
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject &= ~(1 << opt->hdr.id);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
if (mru > MAX_MRU)
lcp->want_mrru = MAX_MRU;
else if (mru < MIN_MRU)
lcp->want_mrru = MIN_MRU;
else
lcp->want_mrru = mru;
}
/* else we honour our config and don't send the suggested REQ */
break;
case MODE_REJ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject |= (1 << opt->hdr.id);
o Don't go multilink if the peer hasn't asked for it o Don't go multilink if the peer hasn't agreed o When negotiating multilink, don't try to push the peers MRU up according to our `set mtu' setting. Instead, accept whatever MRU they want and push their MRRU up instead. o When *not* negotiating multilink, REJ short sequence number REQs.
1998-04-23 18:55:50 +00:00
lcp->want_mrru = 0; /* Ah well, no multilink :-( */
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
break;
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
}
break;
1995-01-31 06:29:58 +00:00
case TY_MRU:
When the peer fails to specify an MRU and a 1500 byte MRU is not allowed either because of the transport or configuration, send a MRU NAK only once, then allow the negotiations to proceed. rfc1661 says that 1500 should always be allowed and rfc2516 says that 1492 is the maximum for PPPoE. This changes ppp so that it only weakly suggests 1492, then goes with the default (leaving the problem in the hands of the peer WRT how they set their MTU). MFC after: 1 week
2001-10-18 08:47:56 +00:00
lcp->mru_req = 1;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
ua_ntohs(opt->data, &mru);
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, "%s %d\n", request, mru);
1995-01-31 06:29:58 +00:00
Don't pass global vars as args. Remove local/global conflicts.
1997-11-08 00:28:11 +00:00
switch (mode_type) {
1995-01-31 06:29:58 +00:00
case MODE_REQ:
Ignore (with a warning message) mtu/mru configurations that are greater than the maximum physical values. MFC after: 1 week
2001-07-17 01:06:13 +00:00
maxmtu = p ? physical_DeviceMTU(p) : 0;
if (lcp->cfg.max_mtu && (!maxmtu || maxmtu > lcp->cfg.max_mtu))
maxmtu = lcp->cfg.max_mtu;
Handle peer REQ/NAKs of >1500 byte MRUs when we have no preference. MFC after: 3 days
2001-07-26 11:33:53 +00:00
wantmtu = lcp->cfg.mtu;
Ignore (with a warning message) mtu/mru configurations that are greater than the maximum physical values. MFC after: 1 week
2001-07-17 01:06:13 +00:00
if (maxmtu && wantmtu > maxmtu) {
log_Printf(LogWARN, "%s: Reducing configured MTU from %u to %u\n",
fp->link->name, wantmtu, maxmtu);
wantmtu = maxmtu;
}
if (maxmtu && mru > maxmtu) {
lcp->his_mru = maxmtu;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.hdr.id = TY_MRU;
nak.hdr.len = 4;
ua_htons(&lcp->his_mru, nak.data);
fsm_nak(dec, &nak);
Handle peer REQ/NAKs of >1500 byte MRUs when we have no preference. MFC after: 3 days
2001-07-26 11:33:53 +00:00
} else if (wantmtu && mru < wantmtu) {
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
/* Push him up to MTU or MIN_MRU */
Ignore (with a warning message) mtu/mru configurations that are greater than the maximum physical values. MFC after: 1 week
2001-07-17 01:06:13 +00:00
lcp->his_mru = wantmtu;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.hdr.id = TY_MRU;
nak.hdr.len = 4;
ua_htons(&lcp->his_mru, nak.data);
fsm_nak(dec, &nak);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
} else {
Handle peer REQ/NAKs of >1500 byte MRUs when we have no preference. MFC after: 3 days
2001-07-26 11:33:53 +00:00
lcp->his_mru = mru;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_ack(dec, opt);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
}
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
break;
1995-01-31 06:29:58 +00:00
case MODE_NAK:
Ignore (with a warning message) mtu/mru configurations that are greater than the maximum physical values. MFC after: 1 week
2001-07-17 01:06:13 +00:00
maxmru = p ? physical_DeviceMTU(p) : 0;
if (lcp->cfg.max_mru && (!maxmru || maxmru > lcp->cfg.max_mru))
maxmru = lcp->cfg.max_mru;
wantmru = lcp->cfg.mru > maxmru ? maxmru : lcp->cfg.mru;
Handle peer REQ/NAKs of >1500 byte MRUs when we have no preference. MFC after: 3 days
2001-07-26 11:33:53 +00:00
if (wantmru && mru > wantmru)
Ignore (with a warning message) mtu/mru configurations that are greater than the maximum physical values. MFC after: 1 week
2001-07-17 01:06:13 +00:00
lcp->want_mru = wantmru;
Handle peer REQ/NAKs of >1500 byte MRUs when we have no preference. MFC after: 3 days
2001-07-26 11:33:53 +00:00
else if (mru > maxmru)
lcp->want_mru = maxmru;
Ignore (with a warning message) mtu/mru configurations that are greater than the maximum physical values. MFC after: 1 week
2001-07-17 01:06:13 +00:00
else if (mru < MIN_MRU)
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
lcp->want_mru = MIN_MRU;
else
lcp->want_mru = mru;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
break;
1995-01-31 06:29:58 +00:00
case MODE_REJ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject |= (1 << opt->hdr.id);
break;
1995-01-31 06:29:58 +00:00
}
break;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
1995-01-31 06:29:58 +00:00
case TY_ACCMAP:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
ua_ntohl(opt->data, &accmap);
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, "%s 0x%08lx\n", request, (u_long)accmap);
1995-01-31 06:29:58 +00:00
Don't pass global vars as args. Remove local/global conflicts.
1997-11-08 00:28:11 +00:00
switch (mode_type) {
1995-01-31 06:29:58 +00:00
case MODE_REQ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_accmap = accmap;
fsm_ack(dec, opt);
break;
1995-01-31 06:29:58 +00:00
case MODE_NAK:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->want_accmap = accmap;
break;
1995-01-31 06:29:58 +00:00
case MODE_REJ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject |= (1 << opt->hdr.id);
break;
1995-01-31 06:29:58 +00:00
}
break;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
1995-01-31 06:29:58 +00:00
case TY_AUTHPROTO:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
ua_ntohs(opt->data, &proto);
chap_type = opt->hdr.len == 5 ? opt->data[2] : 0;
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
log_Printf(LogLCP, "%s 0x%04x (%s)\n", request, proto,
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
Auth2Nam(proto, chap_type));
1995-01-31 06:29:58 +00:00
Don't pass global vars as args. Remove local/global conflicts.
1997-11-08 00:28:11 +00:00
switch (mode_type) {
1995-01-31 06:29:58 +00:00
case MODE_REQ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
switch (proto) {
case PROTO_PAP:
if (opt->hdr.len == 4 && IsAccepted(lcp->cfg.pap)) {
lcp->his_auth = proto;
lcp->his_authtype = 0;
fsm_ack(dec, opt);
} else if (!lcp_auth_nak(lcp, dec)) {
lcp->my_reject |= (1 << opt->hdr.id);
fsm_rej(dec, opt);
}
break;
case PROTO_CHAP:
if ((chap_type == 0x05 && IsAccepted(lcp->cfg.chap05))
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
#ifndef NODES
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
|| (chap_type == 0x80 && (IsAccepted(lcp->cfg.chap80nt) ||
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
(IsAccepted(lcp->cfg.chap80lm))))
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
|| (chap_type == 0x81 && IsAccepted(lcp->cfg.chap81))
Allow Microsoft CHAP authentication. This is a combination of MD4 & DES. Submitted by: Gabor Kincses <gabor@acm.org>
1997-09-25 00:52:37 +00:00
#endif
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
) {
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_auth = proto;
lcp->his_authtype = chap_type;
fsm_ack(dec, opt);
} else {
#ifdef NODES
if (chap_type == 0x80) {
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
log_Printf(LogWARN, "CHAP 0x80 not available without DES\n");
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
} else if (chap_type == 0x81) {
Add MPPE and MSChap v2 support (denied and disabled by default) Submitted by: Ustimenko Semen <semen@iclub.nsu.ru>
2000-10-30 00:15:04 +00:00
log_Printf(LogWARN, "CHAP 0x81 not available without DES\n");
} else
Warn that the specific chap protocol isn't supported when rejecting CHAP because of byte 5. Suggested by: Daniel O'Callaghan <danny@hilink.com.au>
1998-10-17 12:28:02 +00:00
#endif
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
if (chap_type != 0x05)
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
log_Printf(LogWARN, "%s not supported\n",
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
Auth2Nam(PROTO_CHAP, chap_type));
if (!lcp_auth_nak(lcp, dec)) {
lcp->my_reject |= (1 << opt->hdr.id);
fsm_rej(dec, opt);
}
Warn that the specific chap protocol isn't supported when rejecting CHAP because of byte 5. Suggested by: Daniel O'Callaghan <danny@hilink.com.au>
1998-10-17 12:28:02 +00:00
}
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
break;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
default:
log_Printf(LogLCP, "%s 0x%04x - not recognised\n",
Fix PAP, CHAP & LQR req (I broke the byte ordering when I did the deflate re-org). Make PAP & CHAP negotiation prettier in the log file. If both PAP & CHAP are `enabled' and the peer NAKs CHAP suggesting PAP, be friendly and REQ PAP the next time. This is in line with the rfc.
1997-12-07 23:55:29 +00:00
request, proto);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
if (!lcp_auth_nak(lcp, dec)) {
lcp->my_reject |= (1 << opt->hdr.id);
fsm_rej(dec, opt);
}
break;
}
break;
1995-01-31 06:29:58 +00:00
case MODE_NAK:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
switch (proto) {
case PROTO_PAP:
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
if (IsEnabled(lcp->cfg.pap)) {
Isolate fsm callback functions Derive {lcp,ipcp,ccp}state structures from struct fsm where convenient. Create bundle_Down() and remove LcpDown()
1998-02-06 02:24:29 +00:00
lcp->want_auth = PROTO_PAP;
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
lcp->want_authtype = 0;
} else {
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, "Peer will only send PAP (not enabled)\n");
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject |= (1 << opt->hdr.id);
Fix PAP, CHAP & LQR req (I broke the byte ordering when I did the deflate re-org). Make PAP & CHAP negotiation prettier in the log file. If both PAP & CHAP are `enabled' and the peer NAKs CHAP suggesting PAP, be friendly and REQ PAP the next time. This is in line with the rfc.
1997-12-07 23:55:29 +00:00
}
break;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
case PROTO_CHAP:
if (chap_type == 0x05 && IsEnabled(lcp->cfg.chap05)) {
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
lcp->want_auth = PROTO_CHAP;
lcp->want_authtype = 0x05;
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
#ifndef NODES
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
} else if (chap_type == 0x80 && (IsEnabled(lcp->cfg.chap80nt) ||
IsEnabled(lcp->cfg.chap80lm))) {
Isolate fsm callback functions Derive {lcp,ipcp,ccp}state structures from struct fsm where convenient. Create bundle_Down() and remove LcpDown()
1998-02-06 02:24:29 +00:00
lcp->want_auth = PROTO_CHAP;
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
lcp->want_authtype = 0x80;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
} else if (chap_type == 0x81 && IsEnabled(lcp->cfg.chap81)) {
Add MPPE and MSChap v2 support (denied and disabled by default) Submitted by: Ustimenko Semen <semen@iclub.nsu.ru>
2000-10-30 00:15:04 +00:00
lcp->want_auth = PROTO_CHAP;
lcp->want_authtype = 0x81;
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
#endif
} else {
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
#ifdef NODES
if (chap_type == 0x80) {
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
log_Printf(LogLCP, "Peer will only send MSCHAP (not available"
" without DES)\n");
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
} else if (chap_type == 0x81) {
Add MPPE and MSChap v2 support (denied and disabled by default) Submitted by: Ustimenko Semen <semen@iclub.nsu.ru>
2000-10-30 00:15:04 +00:00
log_Printf(LogLCP, "Peer will only send MSCHAPV2 (not available"
" without DES)\n");
} else
Fully support both NT and LANMan CHAP type 0x80 as both authenticator and authenticatee.
1999-02-18 00:52:15 +00:00
#endif
Allow control over the number of ConfigREQ & TermREQ attempts that are made in each of the FSMs (LCP, CCP & IPCP) and the number of REQs/Challenges for PAP/CHAP by accepting more arguments in the ``set {c,ip,l}cpretry'' and ``set {ch,p}apretry'' commands. Change the non-convergence thresholds to 3 times the number of configured REQ tries (rather than the previous fixed ``10''). We now notice repeated NAKs and REJs rather than just REQs. Don't suggest that CHAP 0x05 isn't supported when it's not configured. Fix some bugs that expose themselves with smaller numbers of retries: o Handle instantaneous disconnects (set device /dev/null) correctly by stopping all fsm timers in fsm2initial. o Don't forget to uu_unlock() devices that are files but are not ttys (set device /dev/zero). Fix a *HORRENDOUS* bug in RFC1661 (already fixed for an Open event in state ``Closed''): According to the state transition table, a RCR+ or RCR- received in the ``Stopped'' state are supposed to InitRestartCounter, SendConfigReq and SendConfig{Ack,Nak}. However, in ``Stopped'', we haven't yet done a TLS (or the last thing we did is a TLF). We must therefore do the TLS at this point ! This was never noticed before because LCP and CCP used not use LayerStart() for anything interesting, and IPCP tends to go into Stopped then get a Down because of an LCP RTR rather than getting a RCR again.
1999-02-26 21:28:14 +00:00
log_Printf(LogLCP, "Peer will only send %s (not %s)\n",
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
Auth2Nam(PROTO_CHAP, chap_type),
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
#ifndef NODES
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
(chap_type == 0x80 || chap_type == 0x81) ? "configured" :
Allow control over the number of ConfigREQ & TermREQ attempts that are made in each of the FSMs (LCP, CCP & IPCP) and the number of REQs/Challenges for PAP/CHAP by accepting more arguments in the ``set {c,ip,l}cpretry'' and ``set {ch,p}apretry'' commands. Change the non-convergence thresholds to 3 times the number of configured REQ tries (rather than the previous fixed ``10''). We now notice repeated NAKs and REJs rather than just REQs. Don't suggest that CHAP 0x05 isn't supported when it's not configured. Fix some bugs that expose themselves with smaller numbers of retries: o Handle instantaneous disconnects (set device /dev/null) correctly by stopping all fsm timers in fsm2initial. o Don't forget to uu_unlock() devices that are files but are not ttys (set device /dev/zero). Fix a *HORRENDOUS* bug in RFC1661 (already fixed for an Open event in state ``Closed''): According to the state transition table, a RCR+ or RCR- received in the ``Stopped'' state are supposed to InitRestartCounter, SendConfigReq and SendConfig{Ack,Nak}. However, in ``Stopped'', we haven't yet done a TLS (or the last thing we did is a TLF). We must therefore do the TLS at this point ! This was never noticed before because LCP and CCP used not use LayerStart() for anything interesting, and IPCP tends to go into Stopped then get a Down because of an LCP RTR rather than getting a RCR again.
1999-02-26 21:28:14 +00:00
#endif
"supported");
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject |= (1 << opt->hdr.id);
Fix PAP, CHAP & LQR req (I broke the byte ordering when I did the deflate re-org). Make PAP & CHAP negotiation prettier in the log file. If both PAP & CHAP are `enabled' and the peer NAKs CHAP suggesting PAP, be friendly and REQ PAP the next time. This is in line with the rfc.
1997-12-07 23:55:29 +00:00
}
break;
default:
/* We've been NAK'd with something we don't understand :-( */
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject |= (1 << opt->hdr.id);
Fix PAP, CHAP & LQR req (I broke the byte ordering when I did the deflate re-org). Make PAP & CHAP negotiation prettier in the log file. If both PAP & CHAP are `enabled' and the peer NAKs CHAP suggesting PAP, be friendly and REQ PAP the next time. This is in line with the rfc.
1997-12-07 23:55:29 +00:00
break;
}
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
break;
1995-01-31 06:29:58 +00:00
case MODE_REJ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject |= (1 << opt->hdr.id);
break;
1995-01-31 06:29:58 +00:00
}
break;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
1995-01-31 06:29:58 +00:00
case TY_QUALPROTO:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
req = (struct lqrreq *)opt;
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
log_Printf(LogLCP, "%s proto %x, interval %lums\n",
request, ntohs(req->proto), (u_long)ntohl(req->period) * 10);
Don't pass global vars as args. Remove local/global conflicts.
1997-11-08 00:28:11 +00:00
switch (mode_type) {
1995-01-31 06:29:58 +00:00
case MODE_REQ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
if (ntohs(req->proto) != PROTO_LQR || !IsAccepted(lcp->cfg.lqr)) {
fsm_rej(dec, opt);
lcp->my_reject |= (1 << opt->hdr.id);
} else {
lcp->his_lqrperiod = ntohl(req->period);
if (lcp->his_lqrperiod < MIN_LQRPERIOD * 100)
lcp->his_lqrperiod = MIN_LQRPERIOD * 100;
req->period = htonl(lcp->his_lqrperiod);
fsm_ack(dec, opt);
}
break;
1995-01-31 06:29:58 +00:00
case MODE_NAK:
Allow the peer to modify the LQR interval
2002-05-01 07:49:07 +00:00
lcp->want_lqrperiod = ntohl(req->period);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
break;
1995-01-31 06:29:58 +00:00
case MODE_REJ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject |= (1 << opt->hdr.id);
break;
1995-01-31 06:29:58 +00:00
}
break;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
1995-01-31 06:29:58 +00:00
case TY_MAGICNUM:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
ua_ntohl(opt->data, &magic);
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, "%s 0x%08lx\n", request, (u_long)magic);
1995-01-31 06:29:58 +00:00
Don't pass global vars as args. Remove local/global conflicts.
1997-11-08 00:28:11 +00:00
switch (mode_type) {
1995-01-31 06:29:58 +00:00
case MODE_REQ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
if (lcp->want_magic) {
/* Validate magic number */
if (magic == lcp->want_magic) {
sigset_t emptyset;
sigpause() -> sigsuspend() sigblock() -> sigprocmask()
2001-09-13 10:03:30 +00:00
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
log_Printf(LogLCP, "Magic is same (%08lx) - %d times\n",
Isolate fsm callback functions Derive {lcp,ipcp,ccp}state structures from struct fsm where convenient. Create bundle_Down() and remove LcpDown()
1998-02-06 02:24:29 +00:00
(u_long)magic, ++lcp->LcpFailedMagic);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->want_magic = GenerateMagic();
fsm_nak(dec, opt);
Isolate fsm callback functions Derive {lcp,ipcp,ccp}state structures from struct fsm where convenient. Create bundle_Down() and remove LcpDown()
1998-02-06 02:24:29 +00:00
ualarm(TICKUNIT * (4 + 4 * lcp->LcpFailedMagic), 0);
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
sigemptyset(&emptyset);
sigsuspend(&emptyset);
} else {
lcp->his_magic = magic;
Isolate fsm callback functions Derive {lcp,ipcp,ccp}state structures from struct fsm where convenient. Create bundle_Down() and remove LcpDown()
1998-02-06 02:24:29 +00:00
lcp->LcpFailedMagic = 0;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_ack(dec, opt);
}
} else {
lcp->my_reject |= (1 << opt->hdr.id);
fsm_rej(dec, opt);
}
break;
1995-01-31 06:29:58 +00:00
case MODE_NAK:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
log_Printf(LogLCP, " Magic 0x%08lx is NAKed!\n", (u_long)magic);
lcp->want_magic = GenerateMagic();
break;
1995-01-31 06:29:58 +00:00
case MODE_REJ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
log_Printf(LogLCP, " Magic 0x%08x is REJected!\n", magic);
lcp->want_magic = 0;
lcp->his_reject |= (1 << opt->hdr.id);
break;
1995-01-31 06:29:58 +00:00
}
break;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
1995-01-31 06:29:58 +00:00
case TY_PROTOCOMP:
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, "%s\n", request);
1995-01-31 06:29:58 +00:00
Don't pass global vars as args. Remove local/global conflicts.
1997-11-08 00:28:11 +00:00
switch (mode_type) {
1995-01-31 06:29:58 +00:00
case MODE_REQ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
if (IsAccepted(lcp->cfg.protocomp)) {
lcp->his_protocomp = 1;
fsm_ack(dec, opt);
} else {
1995-01-31 06:29:58 +00:00
#ifdef OLDMST
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
/* MorningStar before v1.3 needs NAK */
fsm_nak(dec, opt);
1995-01-31 06:29:58 +00:00
#else
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_rej(dec, opt);
lcp->my_reject |= (1 << opt->hdr.id);
1995-01-31 06:29:58 +00:00
#endif
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
}
break;
1995-01-31 06:29:58 +00:00
case MODE_NAK:
case MODE_REJ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->want_protocomp = 0;
lcp->his_reject |= (1 << opt->hdr.id);
break;
1995-01-31 06:29:58 +00:00
}
break;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
1995-01-31 06:29:58 +00:00
case TY_ACFCOMP:
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, "%s\n", request);
Don't pass global vars as args. Remove local/global conflicts.
1997-11-08 00:28:11 +00:00
switch (mode_type) {
1995-01-31 06:29:58 +00:00
case MODE_REQ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
if (IsAccepted(lcp->cfg.acfcomp)) {
lcp->his_acfcomp = 1;
fsm_ack(dec, opt);
} else {
1995-01-31 06:29:58 +00:00
#ifdef OLDMST
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
/* MorningStar before v1.3 needs NAK */
fsm_nak(dec, opt);
1995-01-31 06:29:58 +00:00
#else
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_rej(dec, opt);
lcp->my_reject |= (1 << opt->hdr.id);
1995-01-31 06:29:58 +00:00
#endif
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
}
break;
1995-01-31 06:29:58 +00:00
case MODE_NAK:
case MODE_REJ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->want_acfcomp = 0;
lcp->his_reject |= (1 << opt->hdr.id);
break;
1995-01-31 06:29:58 +00:00
}
break;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
1995-01-31 06:29:58 +00:00
case TY_SDP:
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, "%s\n", request);
This commit was manufactured by cvs2svn to create branch 'MP'.
1998-01-29 00:44:16 +00:00
switch (mode_type) {
case MODE_REQ:
case MODE_NAK:
case MODE_REJ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
break;
This commit was manufactured by cvs2svn to create branch 'MP'.
1998-01-29 00:44:16 +00:00
}
break;
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
case TY_CALLBACK:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
if (opt->hdr.len == 2)
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
op = CALLBACK_NONE;
else
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
op = (int)opt->data[0];
sz = opt->hdr.len - 3;
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
switch (op) {
case CALLBACK_AUTH:
log_Printf(LogLCP, "%s Auth\n", request);
break;
case CALLBACK_DIALSTRING:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
log_Printf(LogLCP, "%s Dialstring %.*s\n", request, sz,
opt->data + 1);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
break;
case CALLBACK_LOCATION:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
log_Printf(LogLCP, "%s Location %.*s\n", request, sz, opt->data + 1);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
break;
case CALLBACK_E164:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
log_Printf(LogLCP, "%s E.164 (%.*s)\n", request, sz, opt->data + 1);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
break;
case CALLBACK_NAME:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
log_Printf(LogLCP, "%s Name %.*s\n", request, sz, opt->data + 1);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
break;
case CALLBACK_CBCP:
log_Printf(LogLCP, "%s CBCP\n", request);
break;
default:
log_Printf(LogLCP, "%s ???\n", request);
break;
}
switch (mode_type) {
case MODE_REQ:
callback_req = 1;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
if (p->type != PHYS_DIRECT) {
fsm_rej(dec, opt);
lcp->my_reject |= (1 << opt->hdr.id);
}
nak.hdr.id = opt->hdr.id;
nak.hdr.len = 3;
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
if ((p->dl->cfg.callback.opmask & CALLBACK_BIT(op)) &&
Fix AUTH callback negotiations. MFC after: 1 week
2001-10-18 08:48:04 +00:00
(op != CALLBACK_AUTH || p->link.lcp.want_auth) &&
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
(op != CALLBACK_E164 ||
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
E164ok(&p->dl->cfg.callback, opt->data + 1, sz))) {
lcp->his_callback.opmask = CALLBACK_BIT(op);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
if (sz > sizeof lcp->his_callback.msg - 1) {
sz = sizeof lcp->his_callback.msg - 1;
log_Printf(LogWARN, "Truncating option arg to %d octets\n", sz);
}
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
memcpy(lcp->his_callback.msg, opt->data + 1, sz);
lcp->his_callback.msg[sz] = '\0';
fsm_ack(dec, opt);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
} else if ((p->dl->cfg.callback.opmask & CALLBACK_BIT(CALLBACK_AUTH)) &&
p->link.lcp.auth_ineed) {
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.data[0] = CALLBACK_AUTH;
fsm_nak(dec, &nak);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
} else if (p->dl->cfg.callback.opmask & CALLBACK_BIT(CALLBACK_CBCP)) {
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.data[0] = CALLBACK_CBCP;
fsm_nak(dec, &nak);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
} else if (p->dl->cfg.callback.opmask & CALLBACK_BIT(CALLBACK_E164)) {
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.data[0] = CALLBACK_E164;
fsm_nak(dec, &nak);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
} else if (p->dl->cfg.callback.opmask & CALLBACK_BIT(CALLBACK_AUTH)) {
log_Printf(LogWARN, "Cannot insist on auth callback without"
" PAP or CHAP enabled !\n");
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.data[0] = 2;
fsm_nak(dec, &nak);
} else {
lcp->my_reject |= (1 << opt->hdr.id);
fsm_rej(dec, opt);
}
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
break;
case MODE_NAK:
If we've configured CBCP *and* another not-so-important CALLBACK protocol and end up agreeing CBCP, DTRT and go into CBCP phase rather than mistakenly terminating as if CBCP wasn't agreed. Problem reported by: Alexander Dubinin <alex@nstl.nnov.ru>
1999-01-12 21:50:20 +00:00
/* We don't do what he NAKs with, we do things in our preferred order */
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
if (lcp->want_callback.opmask & CALLBACK_BIT(CALLBACK_AUTH))
lcp->want_callback.opmask &= ~CALLBACK_BIT(CALLBACK_AUTH);
else if (lcp->want_callback.opmask & CALLBACK_BIT(CALLBACK_CBCP))
lcp->want_callback.opmask &= ~CALLBACK_BIT(CALLBACK_CBCP);
else if (lcp->want_callback.opmask & CALLBACK_BIT(CALLBACK_E164))
lcp->want_callback.opmask &= ~CALLBACK_BIT(CALLBACK_E164);
if (lcp->want_callback.opmask == CALLBACK_BIT(CALLBACK_NONE)) {
log_Printf(LogPHASE, "Peer NAKd all callbacks, trying none\n");
lcp->want_callback.opmask = 0;
} else if (!lcp->want_callback.opmask) {
log_Printf(LogPHASE, "Peer NAKd last configured callback\n");
fsm_Close(&lcp->fsm);
}
break;
case MODE_REJ:
if (lcp->want_callback.opmask & CALLBACK_BIT(CALLBACK_NONE)) {
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject |= (1 << opt->hdr.id);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
lcp->want_callback.opmask = 0;
} else {
log_Printf(LogPHASE, "Peer rejected *required* callback\n");
fsm_Close(&lcp->fsm);
}
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
break;
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
}
break;
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
case TY_SHORTSEQ:
mp = &lcp->fsm.bundle->ncp.mp;
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, "%s\n", request);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
Don't pass global vars as args. Remove local/global conflicts.
1997-11-08 00:28:11 +00:00
switch (mode_type) {
1995-01-31 06:29:58 +00:00
case MODE_REQ:
o Don't go multilink if the peer hasn't asked for it o Don't go multilink if the peer hasn't agreed o When negotiating multilink, don't try to push the peers MRU up according to our `set mtu' setting. Instead, accept whatever MRU they want and push their MRRU up instead. o When *not* negotiating multilink, REJ short sequence number REQs.
1998-04-23 18:55:50 +00:00
if (lcp->want_mrru && IsAccepted(mp->cfg.shortseq)) {
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
lcp->his_shortseq = 1;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_ack(dec, opt);
} else {
fsm_rej(dec, opt);
lcp->my_reject |= (1 << opt->hdr.id);
}
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
break;
1995-01-31 06:29:58 +00:00
case MODE_NAK:
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
/*
* He's trying to get us to ask for short sequence numbers.
* We ignore the NAK and honour our configuration file instead.
*/
break;
1995-01-31 06:29:58 +00:00
case MODE_REJ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject |= (1 << opt->hdr.id);
o Behave correctly when short sequence numbers are denied. o Determine if one seq is less than the other correctly, given the point at which they wrap (again).
1998-05-06 23:49:48 +00:00
lcp->want_shortseq = 0; /* For when we hit MP */
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
break;
1995-01-31 06:29:58 +00:00
}
break;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
case TY_ENDDISC:
Initialise `mp'.
1999-06-09 20:27:26 +00:00
mp = &lcp->fsm.bundle->ncp.mp;
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, "%s %s\n", request,
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
mp_Enddisc(opt->data[0], opt->data + 1, opt->hdr.len - 3));
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
switch (mode_type) {
case MODE_REQ:
o Defer setting up pap/chap based IP numbers & labels until after we've determined if we're going to join another ppp invocation. o Make ``show link'' show all link details, and ``show links'' just give a list of links and their current status. o Show our current label in ``show bundle''. o Allow link cloning and removal as soon as our MRRU is set. o Make ``show lcp'' require context as nothing will ever change in our MP LCP (it's auto-configured as per rfc1990). o Initialise our LQM owner in hdlc_Init(). o Store our endpoint discriminator and authentication name at both the datalink and multilink level and compare them when we've finished AUTHENTICATE phase and before entering NETWORK phase. If they don't match, close the link again. Display the information in the appropriate ``show'' command. o Initialise datalink::phone and datalink::fsmp.object properly when we're cloning the link. o Show which link we're passing LQRs on in our diagnostics. o Reject endpoint discriminator REQs at the logical multilink level. o Remove the rest of our CARRIER and LINK logging setup.
1998-04-24 19:16:15 +00:00
if (!p) {
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, " ENDDISC rejected - not a physical link\n");
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_rej(dec, opt);
lcp->my_reject |= (1 << opt->hdr.id);
} else if (!IsAccepted(mp->cfg.negenddisc)) {
lcp->my_reject |= (1 << opt->hdr.id);
fsm_rej(dec, opt);
} else if (opt->hdr.len - 3 < sizeof p->dl->peer.enddisc.address &&
opt->data[0] <= MAX_ENDDISC_CLASS) {
p->dl->peer.enddisc.class = opt->data[0];
p->dl->peer.enddisc.len = opt->hdr.len - 3;
memcpy(p->dl->peer.enddisc.address, opt->data + 1, opt->hdr.len - 3);
p->dl->peer.enddisc.address[opt->hdr.len - 3] = '\0';
o Defer setting up pap/chap based IP numbers & labels until after we've determined if we're going to join another ppp invocation. o Make ``show link'' show all link details, and ``show links'' just give a list of links and their current status. o Show our current label in ``show bundle''. o Allow link cloning and removal as soon as our MRRU is set. o Make ``show lcp'' require context as nothing will ever change in our MP LCP (it's auto-configured as per rfc1990). o Initialise our LQM owner in hdlc_Init(). o Store our endpoint discriminator and authentication name at both the datalink and multilink level and compare them when we've finished AUTHENTICATE phase and before entering NETWORK phase. If they don't match, close the link again. Display the information in the appropriate ``show'' command. o Initialise datalink::phone and datalink::fsmp.object properly when we're cloning the link. o Show which link we're passing LQRs on in our diagnostics. o Reject endpoint discriminator REQs at the logical multilink level. o Remove the rest of our CARRIER and LINK logging setup.
1998-04-24 19:16:15 +00:00
/* XXX: If mp->active, compare and NAK with mp->peer ? */
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_ack(dec, opt);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
} else {
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
if (opt->data[0] > MAX_ENDDISC_CLASS)
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, " ENDDISC rejected - unrecognised class %d\n",
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
opt->data[0]);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
else
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
log_Printf(LogLCP, " ENDDISC rejected - local max length is %ld\n",
(long)(sizeof p->dl->peer.enddisc.address - 1));
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_rej(dec, opt);
lcp->my_reject |= (1 << opt->hdr.id);
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
}
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
break;
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
Allow our endpoint discriminator to be enabled, disabled, accepted and denied. This is necessary for some MP implementations that get confused if you accept their endpoint discriminator but reject their MRRU.
1999-06-09 16:54:04 +00:00
case MODE_NAK: /* Treat this as a REJ, we don't vary our disc (yet) */
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
case MODE_REJ:
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
lcp->his_reject |= (1 << opt->hdr.id);
break;
Negotiate MRRU, SHORTSEQ and ENDDISC. ENDDISC doesn't imply multilink ('cos I've seen my ISP REQ it without multilink). Setting MRRU is ifdef'd out until it's debugged and we can merge -direct links with other running programs. Fix MTU setting.
1998-04-23 03:23:03 +00:00
}
break;
1995-01-31 06:29:58 +00:00
default:
Cosmetic (style): sizeof(var) -> sizeof var sizeof type -> sizeof(type) Suggested by: J Wunsch <j@uriah.heep.sax.de>
1997-12-24 09:29:17 +00:00
sz = (sizeof desc - 2) / 2;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
if (sz > opt->hdr.len - 2)
sz = opt->hdr.len - 2;
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
pos = 0;
desc[0] = sz ? ' ' : '\0';
for (pos = 0; sz--; pos++)
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
sprintf(desc+(pos<<1)+1, "%02x", opt->data[pos]);
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
Cosmetic: Make our external function names consistent.
1998-05-01 19:26:12 +00:00
log_Printf(LogLCP, "%s%s\n", request, desc);
Abstract the CCP layer a level. Add DEFLATE support.
1997-12-03 10:23:54 +00:00
Don't pass global vars as args. Remove local/global conflicts.
1997-11-08 00:28:11 +00:00
if (mode_type == MODE_REQ) {
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_rej(dec, opt);
lcp->my_reject |= (1 << opt->hdr.id);
New user Process PPP based on iij-ppp0.94beta2. o Supporting SYNC SIO device (But need a device driver) - add "set speed sync" o Fixing bug for Predictor-1 function. o Add new parameter that re-sent interval for set timeout commands. o Improving RTT (Round Trip Time) and reducing processor time. - Previous Timer service was using polling, and now using SIGALRM ;-) - A 0.94beta2 will not work correctly.... -- Follows are additinal feature not including 0.94beta2 o Support Proxy ARP - add "enable/disable proxy" commands o Marging common routine in CHAP/PAP. o Enhancing LCP/IPCP log information. o Support local Authfication connection on port 300x and tty. - You can set up pair of your "hostname -s" and password in ppp.secret. if either ppp.secret file nor your hostname line don't exist, It will notify a message and working as same as previous version.(Backword compatibility) - If you did set up them, It's allow connection but nothing to do except help and passwd command. - add "passwd yourpasswd" commands o Support afilter - keep Alive filter that a packet can send/receiving according to ifilter/ofilter but doesn't count it as preventing idle timer expires. - Same syntax of other filters. o Fixing bugs reported by current user for previous one. Thanks !! Reviewed by: Atsushi Murai (amurai@spec.co.jp)
1995-02-26 12:18:08 +00:00
}
1995-01-31 06:29:58 +00:00
break;
}
}
o Move the accept/deny/disable/enable globals either to the bundle (non-negotiated vars) or to their respective IPCP, LCP or CCP. o Enable rolling throughput statistics by default. o Remove the `display' command. These values now appear in `show bundle', `show ipcp', `show ccp' and `show lcp'. o Initialise auth name & key at bundle create time (oops). o Rename pppd-deflate (the id-24 hack) to deflate24. o Don't send both a REJ and a NAK to an IPCP or LCP REQ. Favour the REJ (already done at the CCP level). o Recurse in datalink_UpdateSet() when we change state, otherwise we end up setting no descriptors and getting jammed in the imminent select() instead of doing the dial/login/hangup. o Display our CHAP encryption method despite being built with DES. o Display VJ as not negotiated in ``show ipcp'' when necessary.
1998-04-16 00:26:21 +00:00
Make gcc-2.8.1 build ppp cleanly. Support OpenBSD again.
1998-04-25 10:49:52 +00:00
if (mode_type != MODE_NOP) {
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
if (mode_type == MODE_REQ && p && p->type == PHYS_DIRECT &&
p->dl->cfg.callback.opmask && !callback_req &&
!(p->dl->cfg.callback.opmask & CALLBACK_BIT(CALLBACK_NONE))) {
/* We *REQUIRE* that the peer requests callback */
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.hdr.id = TY_CALLBACK;
nak.hdr.len = 3;
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
if ((p->dl->cfg.callback.opmask & CALLBACK_BIT(CALLBACK_AUTH)) &&
Fix AUTH callback negotiations. MFC after: 1 week
2001-10-18 08:48:04 +00:00
p->link.lcp.want_auth)
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.data[0] = CALLBACK_AUTH;
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
else if (p->dl->cfg.callback.opmask & CALLBACK_BIT(CALLBACK_CBCP))
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.data[0] = CALLBACK_CBCP;
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
else if (p->dl->cfg.callback.opmask & CALLBACK_BIT(CALLBACK_E164))
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.data[0] = CALLBACK_E164;
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
else {
log_Printf(LogWARN, "Cannot insist on auth callback without"
" PAP or CHAP enabled !\n");
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.hdr.len = 2; /* XXX: Silly ! */
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
}
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_nak(dec, &nak);
o Support callback types NONE, E.164, AUTH and CBCP. (see the new ``set callback'' and ``set cbcp'' commands) o Add a ``cbcp'' log level and mbuf type. o Don't dump core when \T is given in ``set login'' or ``set hangup''. o Allow ``*'' and blanks as placeholders in ppp.secret and allow a fifth field for specifying auth/cbcp dialback parameters. o Remove a few extraneous #includes o Define the default number of REQs (restart counter) in defs.h rather than hardcoding ``5'' all over the place. o Fix a few man page inconsistencies.
1998-08-07 18:42:51 +00:00
}
When the peer fails to specify an MRU and a 1500 byte MRU is not allowed either because of the transport or configuration, send a MRU NAK only once, then allow the negotiations to proceed. rfc1661 says that 1500 should always be allowed and rfc2516 says that 1492 is the maximum for PPPoE. This changes ppp so that it only weakly suggests 1492, then goes with the default (leaving the problem in the hands of the peer WRT how they set their MTU). MFC after: 1 week
2001-10-18 08:47:56 +00:00
if (mode_type == MODE_REQ && !lcp->mru_req) {
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
mru = DEF_MRU;
phmtu = p ? physical_DeviceMTU(p) : 0;
if (phmtu && mru > phmtu)
mru = phmtu;
if (mru > lcp->cfg.max_mtu)
mru = lcp->cfg.max_mtu;
if (mru < DEF_MRU) {
/* Don't let the peer use the default MRU */
lcp->his_mru = lcp->cfg.mtu && lcp->cfg.mtu < mru ? lcp->cfg.mtu : mru;
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
nak.hdr.id = TY_MRU;
nak.hdr.len = 4;
ua_htons(&lcp->his_mru, nak.data);
fsm_nak(dec, &nak);
When the peer fails to specify an MRU and a 1500 byte MRU is not allowed either because of the transport or configuration, send a MRU NAK only once, then allow the negotiations to proceed. rfc1661 says that 1500 should always be allowed and rfc2516 says that 1492 is the maximum for PPPoE. This changes ppp so that it only weakly suggests 1492, then goes with the default (leaving the problem in the hands of the peer WRT how they set their MTU). MFC after: 1 week
2001-10-18 08:47:56 +00:00
lcp->mru_req = 1; /* Don't keep NAK'ing this */
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
}
}
Make the way FSM options are processed easier to read by using structures instead of u_char *. The changes are cosmetic except: RecvConfigAck() now displays the options that are being ACK'd Huge (bogus) options sent from the peer won't cause an infinite loop SendIdent and ReceiveIdent are displayed consistenlty with other FSM data LCP AUTHPROTO options that aren't understood are NAK'd, not REJ'd
2002-04-16 23:57:09 +00:00
fsm_opt_normalise(dec);
Make gcc-2.8.1 build ppp cleanly. Support OpenBSD again.
1998-04-25 10:49:52 +00:00
}
1995-01-31 06:29:58 +00:00
}
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
extern struct mbuf *
lcp_Input(struct bundle *bundle, struct link *l, struct mbuf *bp)
1995-01-31 06:29:58 +00:00
{
Move the global FSMs into their relevent structures. IPCP, CCP and LCP are now just derived FSMs. Comment each of the FSM implementations so that we can tell what's going on. Revise the state transitions so that CCP and IPCP actually send terminate REQs when appropriate. The OS & IPCP layers are still like spagetti (next job).
1998-01-31 02:48:30 +00:00
/* Got PROTO_LCP from link */
Cosmetic: Make struct mbuf more like kernel mbufs.
1999-12-20 20:29:47 +00:00
m_settype(bp, MB_LCPIN);
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
fsm_Input(&l->lcp.fsm, bp);
return NULL;
1995-01-31 06:29:58 +00:00
}
Reference in New Issue Copy Permalink