1PPPD(8) System Manager's Manual PPPD(8)
2
6 pppd - Point-to-Point Protocol Daemon
7
9 pppd [ options ]
10
12 PPP is the protocol used for establishing internet links over dial-up
13 modems, DSL connections, and many other types of point-to-point links.
14 The pppd daemon works together with the kernel PPP driver to establish
15 and maintain a PPP link with another system (called the peer) and to
16 negotiate Internet Protocol (IP) addresses for each end of the link.
17 Pppd can also authenticate the peer and/or supply authentication infor‐
18 mation to the peer. PPP can be used with other network protocols
19 besides IP, but such use is becoming increasingly rare.
20
22 ttyname
23 Use the serial port called ttyname to communicate with the peer.
24 If ttyname does not begin with a slash (/), the string "/dev/"
25 is prepended to ttyname to form the name of the device to open.
26 If no device name is given, or if the name of the terminal con‐
27 nected to the standard input is given, pppd will use that termi‐
28 nal, and will not fork to put itself in the background. A value
29 for this option from a privileged source cannot be overridden by
30 a non-privileged user.
31 speed An option that is a decimal number is taken as the desired baud
33 rate for the serial device. On systems such as 4.4BSD and Net‐
34 BSD, any speed can be specified. Other systems (e.g. Linux,
35 SunOS) only support the commonly-used baud rates.
36 asyncmap map
38 This option sets the Async-Control-Character-Map (ACCM) for this
39 end of the link. The ACCM is a set of 32 bits, one for each of
40 the ASCII control characters with values from 0 to 31, where a 1
41 bit indicates that the corresponding control character should
42 not be used in PPP packets sent to this system. The map is
43 encoded as a hexadecimal number (without a leading 0x) where the
44 least significant bit (00000001) represents character 0 and the
45 most significant bit (80000000) represents character 31. Pppd
46 will ask the peer to send these characters as a 2-byte escape
47 sequence. If multiple asyncmap options are given, the values
48 are ORed together. If no asyncmap option is given, the default
49 is zero, so pppd will ask the peer not to escape any control
50 characters. To escape transmitted characters, use the escape
51 option.
52 auth Require the peer to authenticate itself before allowing network
54 packets to be sent or received. This option is the default if
55 the system has a default route. If neither this option nor the
56 noauth option is specified, pppd will only allow the peer to use
57 IP addresses to which the system does not already have a route.
58 call name
60 Read additional options from the file /etc/ppp/peers/name. This
61 file may contain privileged options, such as noauth, even if
62 pppd is not being run by root. The name string may not begin
63 with / or include .. as a pathname component. The format of the
64 options file is described below.
65 connect script
67 Usually there is something which needs to be done to prepare the
68 link before the PPP protocol can be started; for instance, with
69 a dial-up modem, commands need to be sent to the modem to dial
70 the appropriate phone number. This option specifies an command
71 for pppd to execute (by passing it to a shell) before attempting
72 to start PPP negotiation. The chat (8) program is often useful
73 here, as it provides a way to send arbitrary strings to a modem
74 and respond to received characters. A value for this option
75 from a privileged source cannot be overridden by a non-privi‐
76 leged user.
77 crtscts
79 Specifies that pppd should set the serial port to use hardware
80 flow control using the RTS and CTS signals in the RS-232 inter‐
81 face. If neither the crtscts, the nocrtscts, the cdtrcts nor
82 the nocdtrcts option is given, the hardware flow control setting
83 for the serial port is left unchanged. Some serial ports (such
84 as Macintosh serial ports) lack a true RTS output. Such serial
85 ports use this mode to implement unidirectional flow control.
86 The serial port will suspend transmission when requested by the
87 modem (via CTS) but will be unable to request the modem to stop
88 sending to the computer. This mode retains the ability to use
89 DTR as a modem control line.
90 defaultroute
92 Add a default route to the system routing tables, using the peer
93 as the gateway, when IPCP negotiation is successfully completed.
94 This entry is removed when the PPP connection is broken. This
95 option is privileged if the nodefaultroute option has been spec‐
96 ified.
97 disconnect script
99 Execute the command specified by script, by passing it to a
100 shell, after pppd has terminated the link. This command could,
101 for example, issue commands to the modem to cause it to hang up
102 if hardware modem control signals were not available. The dis‐
103 connect script is not run if the modem has already hung up. A
104 value for this option from a privileged source cannot be over‐
105 ridden by a non-privileged user.
106 escape xx,yy,...
108 Specifies that certain characters should be escaped on transmis‐
109 sion (regardless of whether the peer requests them to be escaped
110 with its async control character map). The characters to be
111 escaped are specified as a list of hex numbers separated by com‐
112 mas. Note that almost any character can be specified for the
113 escape option, unlike the asyncmap option which only allows con‐
114 trol characters to be specified. The characters which may not
115 be escaped are those with hex values 0x20 - 0x3f or 0x5e.
116 file name
118 Read options from file name (the format is described below).
119 The file must be readable by the user who has invoked pppd.
120 init script
122 Execute the command specified by script, by passing it to a
123 shell, to initialize the serial line. This script would typi‐
124 cally use the chat(8) program to configure the modem to enable
125 auto answer. A value for this option from a privileged source
126 cannot be overridden by a non-privileged user.
127 lock Specifies that pppd should create a UUCP-style lock file for the
129 serial device to ensure exclusive access to the device. By
130 default, pppd will not create a lock file.
131 mru n Set the MRU [Maximum Receive Unit] value to n. Pppd will ask the
133 peer to send packets of no more than n bytes. The value of n
134 must be between 128 and 16384; the default is 1500. A value of
135 296 works well on very slow links (40 bytes for TCP/IP header +
136 256 bytes of data). Note that for the IPv6 protocol, the MRU
137 must be at least 1280.
138 mtu n Set the MTU [Maximum Transmit Unit] value to n. Unless the peer
140 requests a smaller value via MRU negotiation, pppd will request
141 that the kernel networking code send data packets of no more
142 than n bytes through the PPP network interface. Note that for
143 the IPv6 protocol, the MTU must be at least 1280.
144 passive
146 Enables the "passive" option in the LCP. With this option, pppd
147 will attempt to initiate a connection; if no reply is received
148 from the peer, pppd will then just wait passively for a valid
149 LCP packet from the peer, instead of exiting, as it would with‐
150 out this option.
151
153 <local_IP_address>:<remote_IP_address>
154 Set the local and/or remote interface IP addresses. Either one
155 may be omitted. The IP addresses can be specified with a host
156 name or in decimal dot notation (e.g. 150.234.56.78). The
157 default local address is the (first) IP address of the system
158 (unless the noipdefault option is given). The remote address
159 will be obtained from the peer if not specified in any option.
160 Thus, in simple cases, this option is not required. If a local
161 and/or remote IP address is specified with this option, pppd
162 will not accept a different value from the peer in the IPCP
163 negotiation, unless the ipcp-accept-local and/or
164 ipcp-accept-remote options are given, respectively.
165 ipv6 <local_interface_identifier>,<remote_interface_identifier>
167 Set the local and/or remote 64-bit interface identifier. Either
168 one may be omitted. The identifier must be specified in standard
169 ASCII notation of IPv6 addresses (e.g. ::dead:beef). If the
170 ipv6cp-use-ipaddr option is given, the local identifier is the
171 local IPv4 address (see above). On systems which supports a
172 unique persistent id, such as EUI-48 derived from the Ethernet
173 MAC address, ipv6cp-use-persistent option can be used to replace
174 the ipv6 <local>,<remote> option. Otherwise the identifier is
175 randomized.
176 active-filter filter-expression
178 Specifies a packet filter to be applied to data packets to
179 determine which packets are to be regarded as link activity, and
180 therefore reset the idle timer, or cause the link to be brought
181 up in demand-dialling mode. This option is useful in conjunc‐
182 tion with the idle option if there are packets being sent or
183 received regularly over the link (for example, routing informa‐
184 tion packets) which would otherwise prevent the link from ever
185 appearing to be idle. The filter-expression syntax is as
186 described for tcpdump(1), except that qualifiers which are inap‐
187 propriate for a PPP link, such as ether and arp, are not permit‐
188 ted. Generally the filter expression should be enclosed in sin‐
189 gle-quotes to prevent whitespace in the expression from being
190 interpreted by the shell. This option is currently only avail‐
191 able under Linux, and requires that the kernel was configured to
192 include PPP filtering support (CONFIG_PPP_FILTER). Note that it
193 is possible to apply different constraints to incoming and out‐
194 going packets using the inbound and outbound qualifiers.
195 allow-ip address(es)
197 Allow peers to use the given IP address or subnet without
198 authenticating themselves. The parameter is parsed as for each
199 element of the list of allowed IP addresses in the secrets files
200 (see the AUTHENTICATION section below).
201 allow-number number
203 Allow peers to connect from the given telephone number. A
204 trailing `*' character will match all numbers beginning with the
205 leading part.
206 bsdcomp nr,nt
208 Request that the peer compress packets that it sends, using the
209 BSD-Compress scheme, with a maximum code size of nr bits, and
210 agree to compress packets sent to the peer with a maximum code
211 size of nt bits. If nt is not specified, it defaults to the
212 value given for nr. Values in the range 9 to 15 may be used for
213 nr and nt; larger values give better compression but consume
214 more kernel memory for compression dictionaries. Alternatively,
215 a value of 0 for nr or nt disables compression in the corre‐
216 sponding direction. Use nobsdcomp or bsdcomp 0 to disable BSD-
217 Compress compression entirely.
218 cdtrcts
220 Use a non-standard hardware flow control (i.e. DTR/CTS) to con‐
221 trol the flow of data on the serial port. If neither the
222 crtscts, the nocrtscts, the cdtrcts nor the nocdtrcts option is
223 given, the hardware flow control setting for the serial port is
224 left unchanged. Some serial ports (such as Macintosh serial
225 ports) lack a true RTS output. Such serial ports use this mode
226 to implement true bi-directional flow control. The sacrifice is
227 that this flow control mode does not permit using DTR as a modem
228 control line.
229 chap-interval n
231 If this option is given, pppd will rechallenge the peer every n
232 seconds.
233 chap-max-challenge n
235 Set the maximum number of CHAP challenge transmissions to n
236 (default 10).
237 chap-restart n
239 Set the CHAP restart interval (retransmission timeout for chal‐
240 lenges) to n seconds (default 3).
241 child-timeout n
243 When exiting, wait for up to n seconds for any child processes
244 (such as the command specified with the pty command) to exit
245 before exiting. At the end of the timeout, pppd will send a
246 SIGTERM signal to any remaining child processes and exit. A
247 value of 0 means no timeout, that is, pppd will wait until all
248 child processes have exited.
249 connect-delay n
251 Wait for up to n milliseconds after the connect script finishes
252 for a valid PPP packet from the peer. At the end of this time,
253 or when a valid PPP packet is received from the peer, pppd will
254 commence negotiation by sending its first LCP packet. The
255 default value is 1000 (1 second). This wait period only applies
256 if the connect or pty option is used.
257 debug Enables connection debugging facilities. If this option is
259 given, pppd will log the contents of all control packets sent or
260 received in a readable form. The packets are logged through
261 syslog with facility daemon and level debug. This information
262 can be directed to a file by setting up /etc/syslog.conf appro‐
263 priately (see syslog.conf(5)).
264 default-asyncmap
266 Disable asyncmap negotiation, forcing all control characters to
267 be escaped for both the transmit and the receive direction.
268 default-mru
270 Disable MRU [Maximum Receive Unit] negotiation. With this
271 option, pppd will use the default MRU value of 1500 bytes for
272 both the transmit and receive direction.
273 deflate nr,nt
275 Request that the peer compress packets that it sends, using the
276 Deflate scheme, with a maximum window size of 2**nr bytes, and
277 agree to compress packets sent to the peer with a maximum window
278 size of 2**nt bytes. If nt is not specified, it defaults to the
279 value given for nr. Values in the range 9 to 15 may be used for
280 nr and nt; larger values give better compression but consume
281 more kernel memory for compression dictionaries. Alternatively,
282 a value of 0 for nr or nt disables compression in the corre‐
283 sponding direction. Use nodeflate or deflate 0 to disable
284 Deflate compression entirely. (Note: pppd requests Deflate com‐
285 pression in preference to BSD-Compress if the peer can do
286 either.)
287 demand Initiate the link only on demand, i.e. when data traffic is
289 present. With this option, the remote IP address must be speci‐
290 fied by the user on the command line or in an options file.
291 Pppd will initially configure the interface and enable it for IP
292 traffic without connecting to the peer. When traffic is avail‐
293 able, pppd will connect to the peer and perform negotiation,
294 authentication, etc. When this is completed, pppd will commence
295 passing data packets (i.e., IP packets) across the link.
296 The demand option implies the persist option. If this behaviour
298 is not desired, use the nopersist option after the demand
299 option. The idle and holdoff options are also useful in con‐
300 junction with the demand option.
301 domain d
303 Append the domain name d to the local host name for authentica‐
304 tion purposes. For example, if gethostname() returns the name
305 porsche, but the fully qualified domain name is
306 porsche.Quotron.COM, you could specify domain Quotron.COM. Pppd
307 would then use the name porsche.Quotron.COM for looking up
308 secrets in the secrets file, and as the default name to send to
309 the peer when authenticating itself to the peer. This option is
310 privileged.
311 dryrun With the dryrun option, pppd will print out all the option val‐
313 ues which have been set and then exit, after parsing the command
314 line and options files and checking the option values, but
315 before initiating the link. The option values are logged at
316 level info, and also printed to standard output unless the
317 device on standard output is the device that pppd would be using
318 to communicate with the peer.
319 dump With the dump option, pppd will print out all the option values
321 which have been set. This option is like the dryrun option
322 except that pppd proceeds as normal rather than exiting.
323 enable-session
325 Enables session accounting via PAM or wtwp/wtmpx, as appropri‐
326 ate. When PAM is enabled, the PAM "account" and "session" mod‐
327 ule stacks determine behavior, and are enabled for all PPP
328 authentication protocols. When PAM is disabled, wtmp/wtmpx
329 entries are recorded regardless of whether the peer name identi‐
330 fies a valid user on the local system, making peers visible in
331 the last(1) log. This feature is automatically enabled when the
332 pppd login option is used. Session accounting is disabled by
333 default.
334 endpoint <epdisc>
336 Sets the endpoint discriminator sent by the local machine to the
337 peer during multilink negotiation to <epdisc>. The default is
338 to use the MAC address of the first ethernet interface on the
339 system, if any, otherwise the IPv4 address corresponding to the
340 hostname, if any, provided it is not in the multicast or
341 locally-assigned IP address ranges, or the localhost address.
342 The endpoint discriminator can be the string null or of the form
343 type:value, where type is a decimal number or one of the strings
344 local, IP, MAC, magic, or phone. The value is an IP address in
345 dotted-decimal notation for the IP type, or a string of bytes in
346 hexadecimal, separated by periods or colons for the other types.
347 For the MAC type, the value may also be the name of an ethernet
348 or similar network interface. This option is currently only
349 available under Linux.
350 eap-interval n
352 If this option is given and pppd authenticates the peer with EAP
353 (i.e., is the server), pppd will restart EAP authentication
354 every n seconds. For EAP SRP-SHA1, see also the srp-interval
355 option, which enables lightweight rechallenge.
356 eap-max-rreq n
358 Set the maximum number of EAP Requests to which pppd will
359 respond (as a client) without hearing EAP Success or Failure.
360 (Default is 20.)
361 eap-max-sreq n
363 Set the maximum number of EAP Requests that pppd will issue (as
364 a server) while attempting authentication. (Default is 10.)
365 eap-restart n
367 Set the retransmit timeout for EAP Requests when acting as a
368 server (authenticator). (Default is 3 seconds.)
369 eap-timeout n
371 Set the maximum time to wait for the peer to send an EAP Request
372 when acting as a client (authenticatee). (Default is 20 sec‐
373 onds.)
374 hide-password
376 When logging the contents of PAP packets, this option causes
377 pppd to exclude the password string from the log. This is the
378 default.
379 holdoff n
381 Specifies how many seconds to wait before re-initiating the link
382 after it terminates. This option only has any effect if the
383 persist or demand option is used. The holdoff period is not
384 applied if the link was terminated because it was idle.
385 idle n Specifies that pppd should disconnect if the link is idle for n
387 seconds. The link is idle when no data packets (i.e. IP pack‐
388 ets) are being sent or received. Note: it is not advisable to
389 use this option with the persist option without the demand
390 option. If the active-filter option is given, data packets
391 which are rejected by the specified activity filter also count
392 as the link being idle.
393 ipcp-accept-local
395 With this option, pppd will accept the peer's idea of our local
396 IP address, even if the local IP address was specified in an
397 option.
398 ipcp-accept-remote
400 With this option, pppd will accept the peer's idea of its
401 (remote) IP address, even if the remote IP address was specified
402 in an option.
403 ipcp-max-configure n
405 Set the maximum number of IPCP configure-request transmissions
406 to n (default 10).
407 ipcp-max-failure n
409 Set the maximum number of IPCP configure-NAKs returned before
410 starting to send configure-Rejects instead to n (default 10).
411 ipcp-max-terminate n
413 Set the maximum number of IPCP terminate-request transmissions
414 to n (default 3).
415 ipcp-restart n
417 Set the IPCP restart interval (retransmission timeout) to n sec‐
418 onds (default 3).
419 ipparam string
421 Provides an extra parameter to the ip-up, ip-pre-up and ip-down
422 scripts. If this option is given, the string supplied is given
423 as the 6th parameter to those scripts.
424 ipv6cp-max-configure n
426 Set the maximum number of IPv6CP configure-request transmissions
427 to n (default 10).
428 ipv6cp-max-failure n
430 Set the maximum number of IPv6CP configure-NAKs returned before
431 starting to send configure-Rejects instead to n (default 10).
432 ipv6cp-max-terminate n
434 Set the maximum number of IPv6CP terminate-request transmissions
435 to n (default 3).
436 ipv6cp-restart n
438 Set the IPv6CP restart interval (retransmission timeout) to n
439 seconds (default 3).
440 ipx Enable the IPXCP and IPX protocols. This option is presently
442 only supported under Linux, and only if your kernel has been
443 configured to include IPX support.
444 ipx-network n
446 Set the IPX network number in the IPXCP configure request frame
447 to n, a hexadecimal number (without a leading 0x). There is no
448 valid default. If this option is not specified, the network
449 number is obtained from the peer. If the peer does not have the
450 network number, the IPX protocol will not be started.
451 ipx-node n:m
453 Set the IPX node numbers. The two node numbers are separated
454 from each other with a colon character. The first number n is
455 the local node number. The second number m is the peer's node
456 number. Each node number is a hexadecimal number, at most 10
457 digits long. The node numbers on the ipx-network must be unique.
458 There is no valid default. If this option is not specified then
459 the node numbers are obtained from the peer.
460 ipx-router-name <string>
462 Set the name of the router. This is a string and is sent to the
463 peer as information data.
464 ipx-routing n
466 Set the routing protocol to be received by this option. More
467 than one instance of ipx-routing may be specified. The 'none'
468 option (0) may be specified as the only instance of ipx-routing.
469 The values may be 0 for NONE, 2 for RIP/SAP, and 4 for NLSP.
470 ipxcp-accept-local
472 Accept the peer's NAK for the node number specified in the
473 ipx-node option. If a node number was specified, and non-zero,
474 the default is to insist that the value be used. If you include
475 this option then you will permit the peer to override the entry
476 of the node number.
477 ipxcp-accept-network
479 Accept the peer's NAK for the network number specified in the
480 ipx-network option. If a network number was specified, and non-
481 zero, the default is to insist that the value be used. If you
482 include this option then you will permit the peer to override
483 the entry of the node number.
484 ipxcp-accept-remote
486 Use the peer's network number specified in the configure request
487 frame. If a node number was specified for the peer and this
488 option was not specified, the peer will be forced to use the
489 value which you have specified.
490 ipxcp-max-configure n
492 Set the maximum number of IPXCP configure request frames which
493 the system will send to n. The default is 10.
494 ipxcp-max-failure n
496 Set the maximum number of IPXCP NAK frames which the local sys‐
497 tem will send before it rejects the options. The default value
498 is 3.
499 ipxcp-max-terminate n
501 Set the maximum number of IPXCP terminate request frames before
502 the local system considers that the peer is not listening to
503 them. The default value is 3.
504 kdebug n
506 Enable debugging code in the kernel-level PPP driver. The argu‐
507 ment values depend on the specific kernel driver, but in general
508 a value of 1 will enable general kernel debug messages. (Note
509 that these messages are usually only useful for debugging the
510 kernel driver itself.) For the Linux 2.2.x kernel driver, the
511 value is a sum of bits: 1 to enable general debug messages, 2 to
512 request that the contents of received packets be printed, and 4
513 to request that the contents of transmitted packets be printed.
514 On most systems, messages printed by the kernel are logged by
515 syslog(1) to a file as directed in the /etc/syslog.conf configu‐
516 ration file.
517 ktune Enables pppd to alter kernel settings as appropriate. Under
519 Linux, pppd will enable IP forwarding (i.e. set
520 /proc/sys/net/ipv4/ip_forward to 1) if the proxyarp option is
521 used, and will enable the dynamic IP address option (i.e. set
522 /proc/sys/net/ipv4/ip_dynaddr to 1) in demand mode if the local
523 address changes.
524 lcp-echo-failure n
526 If this option is given, pppd will presume the peer to be dead
527 if n LCP echo-requests are sent without receiving a valid LCP
528 echo-reply. If this happens, pppd will terminate the connec‐
529 tion. Use of this option requires a non-zero value for the
530 lcp-echo-interval parameter. This option can be used to enable
531 pppd to terminate after the physical connection has been broken
532 (e.g., the modem has hung up) in situations where no hardware
533 modem control lines are available.
534 lcp-echo-interval n
536 If this option is given, pppd will send an LCP echo-request
537 frame to the peer every n seconds. Normally the peer should
538 respond to the echo-request by sending an echo-reply. This
539 option can be used with the lcp-echo-failure option to detect
540 that the peer is no longer connected.
541 lcp-max-configure n
543 Set the maximum number of LCP configure-request transmissions to
544 n (default 10).
545 lcp-max-failure n
547 Set the maximum number of LCP configure-NAKs returned before
548 starting to send configure-Rejects instead to n (default 10).
549 lcp-max-terminate n
551 Set the maximum number of LCP terminate-request transmissions to
552 n (default 3).
553 lcp-restart n
555 Set the LCP restart interval (retransmission timeout) to n sec‐
556 onds (default 3).
557 linkname name
559 Sets the logical name of the link to name. Pppd will create a
560 file named ppp-name.pid in /var/run (or /etc/ppp on some sys‐
561 tems) containing its process ID. This can be useful in deter‐
562 mining which instance of pppd is responsible for the link to a
563 given peer system. This is a privileged option.
564 local Don't use the modem control lines. With this option, pppd will
566 ignore the state of the CD (Carrier Detect) signal from the
567 modem and will not change the state of the DTR (Data Terminal
568 Ready) signal. This is the opposite of the modem option.
569 logfd n
571 Send log messages to file descriptor n. Pppd will send log mes‐
572 sages to at most one file or file descriptor (as well as sending
573 the log messages to syslog), so this option and the logfile
574 option are mutually exclusive. The default is for pppd to send
575 log messages to stdout (file descriptor 1), unless the serial
576 port is already open on stdout.
577 logfile filename
579 Append log messages to the file filename (as well as sending the
580 log messages to syslog). The file is opened with the privileges
581 of the user who invoked pppd, in append mode.
582 login Use the system password database for authenticating the peer
584 using PAP, and record the user in the system wtmp file. Note
585 that the peer must have an entry in the /etc/ppp/pap-secrets
586 file as well as the system password database to be allowed
587 access. See also the enable-session option.
588 maxconnect n
590 Terminate the connection when it has been available for network
591 traffic for n seconds (i.e. n seconds after the first network
592 control protocol comes up).
593 maxfail n
595 Terminate after n consecutive failed connection attempts. A
596 value of 0 means no limit. The default value is 10.
597 modem Use the modem control lines. This option is the default. With
599 this option, pppd will wait for the CD (Carrier Detect) signal
600 from the modem to be asserted when opening the serial device
601 (unless a connect script is specified), and it will drop the DTR
602 (Data Terminal Ready) signal briefly when the connection is ter‐
603 minated and before executing the connect script. On Ultrix,
604 this option implies hardware flow control, as for the crtscts
605 option. This is the opposite of the local option.
606 mp Enables the use of PPP multilink; this is an alias for the `mul‐
608 tilink' option. This option is currently only available under
609 Linux.
610 mppe-stateful
612 Allow MPPE to use stateful mode. Stateless mode is still
613 attempted first. The default is to disallow stateful mode.
614 mpshortseq
616 Enables the use of short (12-bit) sequence numbers in multilink
617 headers, as opposed to 24-bit sequence numbers. This option is
618 only available under Linux, and only has any effect if multilink
619 is enabled (see the multilink option).
620 mrru n Sets the Maximum Reconstructed Receive Unit to n. The MRRU is
622 the maximum size for a received packet on a multilink bundle,
623 and is analogous to the MRU for the individual links. This
624 option is currently only available under Linux, and only has any
625 effect if multilink is enabled (see the multilink option).
626 ms-dns <addr>
628 If pppd is acting as a server for Microsoft Windows clients,
629 this option allows pppd to supply one or two DNS (Domain Name
630 Server) addresses to the clients. The first instance of this
631 option specifies the primary DNS address; the second instance
632 (if given) specifies the secondary DNS address. (This option
633 was present in some older versions of pppd under the name
634 dns-addr.)
635 ms-wins <addr>
637 If pppd is acting as a server for Microsoft Windows or "Samba"
638 clients, this option allows pppd to supply one or two WINS (Win‐
639 dows Internet Name Services) server addresses to the clients.
640 The first instance of this option specifies the primary WINS
641 address; the second instance (if given) specifies the secondary
642 WINS address.
643 multilink
645 Enables the use of the PPP multilink protocol. If the peer also
646 supports multilink, then this link can become part of a bundle
647 between the local system and the peer. If there is an existing
648 bundle to the peer, pppd will join this link to that bundle,
649 otherwise pppd will create a new bundle. See the MULTILINK sec‐
650 tion below. This option is currently only available under
651 Linux.
652 name name
654 Set the name of the local system for authentication purposes to
655 name. This is a privileged option. With this option, pppd will
656 use lines in the secrets files which have name as the second
657 field when looking for a secret to use in authenticating the
658 peer. In addition, unless overridden with the user option, name
659 will be used as the name to send to the peer when authenticating
660 the local system to the peer. (Note that pppd does not append
661 the domain name to name.)
662 noaccomp
664 Disable Address/Control compression in both directions (send and
665 receive).
666 noauth Do not require the peer to authenticate itself. This option is
668 privileged.
669 nobsdcomp
671 Disables BSD-Compress compression; pppd will not request or
672 agree to compress packets using the BSD-Compress scheme.
673 noccp Disable CCP (Compression Control Protocol) negotiation. This
675 option should only be required if the peer is buggy and gets
676 confused by requests from pppd for CCP negotiation.
677 nocrtscts
679 Disable hardware flow control (i.e. RTS/CTS) on the serial port.
680 If neither the crtscts nor the nocrtscts nor the cdtrcts nor the
681 nocdtrcts option is given, the hardware flow control setting for
682 the serial port is left unchanged.
683 nocdtrcts
685 This option is a synonym for nocrtscts. Either of these options
686 will disable both forms of hardware flow control.
687 nodefaultroute
689 Disable the defaultroute option. The system administrator who
690 wishes to prevent users from creating default routes with pppd
691 can do so by placing this option in the /etc/ppp/options file.
692 nodeflate
694 Disables Deflate compression; pppd will not request or agree to
695 compress packets using the Deflate scheme.
696 nodetach
698 Don't detach from the controlling terminal. Without this
699 option, if a serial device other than the terminal on the stan‐
700 dard input is specified, pppd will fork to become a background
701 process.
702 noendpoint
704 Disables pppd from sending an endpoint discriminator to the peer
705 or accepting one from the peer (see the MULTILINK section
706 below). This option should only be required if the peer is
707 buggy.
708 noip Disable IPCP negotiation and IP communication. This option
710 should only be required if the peer is buggy and gets confused
711 by requests from pppd for IPCP negotiation.
712 noipv6 Disable IPv6CP negotiation and IPv6 communication. This option
714 should only be required if the peer is buggy and gets confused
715 by requests from pppd for IPv6CP negotiation.
716 noipdefault
718 Disables the default behaviour when no local IP address is spec‐
719 ified, which is to determine (if possible) the local IP address
720 from the hostname. With this option, the peer will have to sup‐
721 ply the local IP address during IPCP negotiation (unless it
722 specified explicitly on the command line or in an options file).
723 noipx Disable the IPXCP and IPX protocols. This option should only be
725 required if the peer is buggy and gets confused by requests from
726 pppd for IPXCP negotiation.
727 noktune
729 Opposite of the ktune option; disables pppd from changing system
730 settings.
731 nolock Opposite of the lock option; specifies that pppd should not cre‐
733 ate a UUCP-style lock file for the serial device. This option
734 is privileged.
735 nolog Do not send log messages to a file or file descriptor. This
737 option cancels the logfd and logfile options.
738 nomagic
740 Disable magic number negotiation. With this option, pppd cannot
741 detect a looped-back line. This option should only be needed if
742 the peer is buggy.
743 nomp Disables the use of PPP multilink. This option is currently
745 only available under Linux.
746 nomppe Disables MPPE (Microsoft Point to Point Encryption). This is
748 the default.
749 nomppe-40
751 Disable 40-bit encryption with MPPE.
752 nomppe-128
754 Disable 128-bit encryption with MPPE.
755 nomppe-stateful
757 Disable MPPE stateful mode. This is the default.
758 nompshortseq
760 Disables the use of short (12-bit) sequence numbers in the PPP
761 multilink protocol, forcing the use of 24-bit sequence numbers.
762 This option is currently only available under Linux, and only
763 has any effect if multilink is enabled.
764 nomultilink
766 Disables the use of PPP multilink. This option is currently
767 only available under Linux.
768 nopcomp
770 Disable protocol field compression negotiation in both the
771 receive and the transmit direction.
772 nopersist
774 Exit once a connection has been made and terminated. This is
775 the default unless the persist or demand option has been speci‐
776 fied.
777 nopredictor1
779 Do not accept or agree to Predictor-1 compression.
780 noproxyarp
782 Disable the proxyarp option. The system administrator who
783 wishes to prevent users from creating proxy ARP entries with
784 pppd can do so by placing this option in the /etc/ppp/options
785 file.
786 noremoteip
788 Allow pppd to operate without having an IP address for the peer.
789 This option is only available under Linux. Normally, pppd will
790 request the peer's IP address, and if the peer does not supply
791 it, pppd will not bring up the link for IP traffic. With this
792 option, if the peer does not supply its IP address, pppd will
793 not ask the peer for it, and will not set the destination
794 address of the ppp interface. In this situation, the ppp inter‐
795 face can be used for routing by creating device routes, but the
796 peer itself cannot be addressed directly for IP traffic.
797 notty Normally, pppd requires a terminal device. With this option,
799 pppd will allocate itself a pseudo-tty master/slave pair and use
800 the slave as its terminal device. Pppd will create a child
801 process to act as a `character shunt' to transfer characters
802 between the pseudo-tty master and its standard input and output.
803 Thus pppd will transmit characters on its standard output and
804 receive characters on its standard input even if they are not
805 terminal devices. This option increases the latency and CPU
806 overhead of transferring data over the ppp interface as all of
807 the characters sent and received must flow through the character
808 shunt process. An explicit device name may not be given if this
809 option is used.
810 novj Disable Van Jacobson style TCP/IP header compression in both the
812 transmit and the receive direction.
813 novjccomp
815 Disable the connection-ID compression option in Van Jacobson
816 style TCP/IP header compression. With this option, pppd will
817 not omit the connection-ID byte from Van Jacobson compressed
818 TCP/IP headers, nor ask the peer to do so.
819 papcrypt
821 Indicates that all secrets in the /etc/ppp/pap-secrets file
822 which are used for checking the identity of the peer are
823 encrypted, and thus pppd should not accept a password which,
824 before encryption, is identical to the secret from the
825 /etc/ppp/pap-secrets file.
826 pap-max-authreq n
828 Set the maximum number of PAP authenticate-request transmissions
829 to n (default 10).
830 pap-restart n
832 Set the PAP restart interval (retransmission timeout) to n sec‐
833 onds (default 3).
834 pap-timeout n
836 Set the maximum time that pppd will wait for the peer to authen‐
837 ticate itself with PAP to n seconds (0 means no limit).
838 pass-filter filter-expression
840 Specifies a packet filter to applied to data packets being sent
841 or received to determine which packets should be allowed to
842 pass. Packets which are rejected by the filter are silently
843 discarded. This option can be used to prevent specific network
844 daemons (such as routed) using up link bandwidth, or to provide
845 a very basic firewall capability. The filter-expression syntax
846 is as described for tcpdump(1), except that qualifiers which are
847 inappropriate for a PPP link, such as ether and arp, are not
848 permitted. Generally the filter expression should be enclosed
849 in single-quotes to prevent whitespace in the expression from
850 being interpreted by the shell. Note that it is possible to
851 apply different constraints to incoming and outgoing packets
852 using the inbound and outbound qualifiers. This option is cur‐
853 rently only available under Linux, and requires that the kernel
854 was configured to include PPP filtering support (CONFIG_PPP_FIL‐
855 TER).
856 password password-string
858 Specifies the password to use for authenticating to the peer.
859 Use of this option is discouraged, as the password is likely to
860 be visible to other users on the system (for example, by using
861 ps(1)).
862 persist
864 Do not exit after a connection is terminated; instead try to
865 reopen the connection. The maxfail option still has an effect on
866 persistent connections.
867 plugin filename
869 Load the shared library object file filename as a plugin. This
870 is a privileged option. If filename does not contain a slash
871 (/), pppd will look in the /usr/lib/pppd/version directory for
872 the plugin, where version is the version number of pppd (for
873 example, 2.4.2).
874 predictor1
876 Request that the peer compress frames that it sends using Pre‐
877 dictor-1 compression, and agree to compress transmitted frames
878 with Predictor-1 if requested. This option has no effect unless
879 the kernel driver supports Predictor-1 compression.
880 privgroup group-name
882 Allows members of group group-name to use privileged options.
883 This is a privileged option. Use of this option requires care
884 as there is no guarantee that members of group-name cannot use
885 pppd to become root themselves. Consider it equivalent to
886 putting the members of group-name in the kmem or disk group.
887 proxyarp
889 Add an entry to this system's ARP [Address Resolution Protocol]
890 table with the IP address of the peer and the Ethernet address
891 of this system. This will have the effect of making the peer
892 appear to other systems to be on the local ethernet.
893 pty script
895 Specifies that the command script is to be used to communicate
896 rather than a specific terminal device. Pppd will allocate
897 itself a pseudo-tty master/slave pair and use the slave as its
898 terminal device. The script will be run in a child process with
899 the pseudo-tty master as its standard input and output. An
900 explicit device name may not be given if this option is used.
901 (Note: if the record option is used in conjunction with the pty
902 option, the child process will have pipes on its standard input
903 and output.)
904 receive-all
906 With this option, pppd will accept all control characters from
907 the peer, including those marked in the receive asyncmap. With‐
908 out this option, pppd will discard those characters as specified
909 in RFC1662. This option should only be needed if the peer is
910 buggy.
911 record filename
913 Specifies that pppd should record all characters sent and
914 received to a file named filename. This file is opened in
915 append mode, using the user's user-ID and permissions. This
916 option is implemented using a pseudo-tty and a process to trans‐
917 fer characters between the pseudo-tty and the real serial
918 device, so it will increase the latency and CPU overhead of
919 transferring data over the ppp interface. The characters are
920 stored in a tagged format with timestamps, which can be dis‐
921 played in readable form using the pppdump(8) program.
922 remotename name
924 Set the assumed name of the remote system for authentication
925 purposes to name.
926 remotenumber number
928 Set the assumed telephone number of the remote system for
929 authentication purposes to number.
930 refuse-chap
932 With this option, pppd will not agree to authenticate itself to
933 the peer using CHAP.
934 refuse-mschap
936 With this option, pppd will not agree to authenticate itself to
937 the peer using MS-CHAP.
938 refuse-mschap-v2
940 With this option, pppd will not agree to authenticate itself to
941 the peer using MS-CHAPv2.
942 refuse-eap
944 With this option, pppd will not agree to authenticate itself to
945 the peer using EAP.
946 refuse-pap
948 With this option, pppd will not agree to authenticate itself to
949 the peer using PAP.
950 require-chap
952 Require the peer to authenticate itself using CHAP [Challenge
953 Handshake Authentication Protocol] authentication.
954 require-mppe
956 Require the use of MPPE (Microsoft Point to Point Encryption).
957 This option disables all other compression types. This option
958 enables both 40-bit and 128-bit encryption. In order for MPPE
959 to successfully come up, you must have authenticated with either
960 MS-CHAP or MS-CHAPv2. This option is presently only supported
961 under Linux, and only if your kernel has been configured to
962 include MPPE support.
963 require-mppe-40
965 Require the use of MPPE, with 40-bit encryption.
966 require-mppe-128
968 Require the use of MPPE, with 128-bit encryption.
969 require-mschap
971 Require the peer to authenticate itself using MS-CHAP [Microsoft
972 Challenge Handshake Authentication Protocol] authentication.
973 require-mschap-v2
975 Require the peer to authenticate itself using MS-CHAPv2 [Micro‐
976 soft Challenge Handshake Authentication Protocol, Version 2]
977 authentication.
978 require-eap
980 Require the peer to authenticate itself using EAP [Extensible
981 Authentication Protocol] authentication.
982 require-pap
984 Require the peer to authenticate itself using PAP [Password
985 Authentication Protocol] authentication.
986 show-password
988 When logging the contents of PAP packets, this option causes
989 pppd to show the password string in the log message.
990 silent With this option, pppd will not transmit LCP packets to initiate
992 a connection until a valid LCP packet is received from the peer
993 (as for the `passive' option with ancient versions of pppd).
994 srp-interval n
996 If this parameter is given and pppd uses EAP SRP-SHA1 to authen‐
997 ticate the peer (i.e., is the server), then pppd will use the
998 optional lightweight SRP rechallenge mechanism at intervals of n
999 seconds. This option is faster than eap-interval reauthentica‐
1000 tion because it uses a hash-based mechanism and does not derive
1001 a new session key.
1002 srp-pn-secret string
1004 Set the long-term pseudonym-generating secret for the server.
1005 This value is optional and if set, needs to be known at the
1006 server (authenticator) side only, and should be different for
1007 each server (or poll of identical servers). It is used along
1008 with the current date to generate a key to encrypt and decrypt
1009 the client's identity contained in the pseudonym.
1010 srp-use-pseudonym
1012 When operating as an EAP SRP-SHA1 client, attempt to use the
1013 pseudonym stored in ~/.ppp_pseudonym first as the identity, and
1014 save in this file any pseudonym offered by the peer during
1015 authentication.
1016 sync Use synchronous HDLC serial encoding instead of asynchronous.
1018 The device used by pppd with this option must have sync support.
1019 Currently supports Microgate SyncLink adapters under Linux and
1020 FreeBSD 2.2.8 and later.
1021 unit num
1023 Sets the ppp unit number (for a ppp0 or ppp1 etc interface name)
1024 for outbound connections.
1025 updetach
1027 With this option, pppd will detach from its controlling terminal
1028 once it has successfully established the ppp connection (to the
1029 point where the first network control protocol, usually the IP
1030 control protocol, has come up).
1031 usehostname
1033 Enforce the use of the hostname (with domain name appended, if
1034 given) as the name of the local system for authentication pur‐
1035 poses (overrides the name option). This option is not normally
1036 needed since the name option is privileged.
1037 usepeerdns
1039 Ask the peer for up to 2 DNS server addresses. The addresses
1040 supplied by the peer (if any) are passed to the /etc/ppp/ip-up
1041 script in the environment variables DNS1 and DNS2, and the envi‐
1042 ronment variable USEPEERDNS will be set to 1. In addition, pppd
1043 will create an /var/run/ppp/resolv.conf file containing one or
1044 two nameserver lines with the address(es) supplied by the peer.
1045 user name
1047 Sets the name used for authenticating the local system to the
1048 peer to name.
1049 vj-max-slots n
1051 Sets the number of connection slots to be used by the Van Jacob‐
1052 son TCP/IP header compression and decompression code to n, which
1053 must be between 2 and 16 (inclusive).
1054 welcome script
1056 Run the executable or shell command specified by script before
1057 initiating PPP negotiation, after the connect script (if any)
1058 has completed. A value for this option from a privileged source
1059 cannot be overridden by a non-privileged user.
1060 xonxoff
1062 Use software flow control (i.e. XON/XOFF) to control the flow of
1063 data on the serial port.
1064
1066 Options can be taken from files as well as the command line. Pppd
1067 reads options from the files /etc/ppp/options, ~/.ppprc and
1068 /etc/ppp/options.ttyname (in that order) before processing the options
1069 on the command line. (In fact, the command-line options are scanned to
1070 find the terminal name before the options.ttyname file is read.) In
1071 forming the name of the options.ttyname file, the initial /dev/ is
1072 removed from the terminal name, and any remaining / characters are
1073 replaced with dots.
1074 An options file is parsed into a series of words, delimited by white‐
1076 space. Whitespace can be included in a word by enclosing the word in
1077 double-quotes ("). A backslash (\) quotes the following character. A
1078 hash (#) starts a comment, which continues until the end of the line.
1079 There is no restriction on using the file or call options within an
1080 options file.
1081
1083 pppd provides system administrators with sufficient access control that
1084 PPP access to a server machine can be provided to legitimate users
1085 without fear of compromising the security of the server or the network
1086 it's on. This control is provided through restrictions on which IP
1087 addresses the peer may use, based on its authenticated identity (if
1088 any), and through restrictions on which options a non-privileged user
1089 may use. Several of pppd's options are privileged, in particular those
1090 which permit potentially insecure configurations; these options are
1091 only accepted in files which are under the control of the system admin‐
1092 istrator, or if pppd is being run by root.
1093 The default behaviour of pppd is to allow an unauthenticated peer to
1095 use a given IP address only if the system does not already have a route
1096 to that IP address. For example, a system with a permanent connection
1097 to the wider internet will normally have a default route, and thus all
1098 peers will have to authenticate themselves in order to set up a connec‐
1099 tion. On such a system, the auth option is the default. On the other
1100 hand, a system where the PPP link is the only connection to the inter‐
1101 net will not normally have a default route, so the peer will be able to
1102 use almost any IP address without authenticating itself.
1103 As indicated above, some security-sensitive options are privileged,
1105 which means that they may not be used by an ordinary non-privileged
1106 user running a setuid-root pppd, either on the command line, in the
1107 user's ~/.ppprc file, or in an options file read using the file option.
1108 Privileged options may be used in /etc/ppp/options file or in an
1109 options file read using the call option. If pppd is being run by the
1110 root user, privileged options can be used without restriction.
1111 When opening the device, pppd uses either the invoking user's user ID
1113 or the root UID (that is, 0), depending on whether the device name was
1114 specified by the user or the system administrator. If the device name
1115 comes from a privileged source, that is, /etc/ppp/options or an options
1116 file read using the call option, pppd uses full root privileges when
1117 opening the device. Thus, by creating an appropriate file under
1118 /etc/ppp/peers, the system administrator can allow users to establish a
1119 ppp connection via a device which they would not normally have permis‐
1120 sion to access. Otherwise pppd uses the invoking user's real UID when
1121 opening the device.
1122
1124 Authentication is the process whereby one peer convinces the other of
1125 its identity. This involves the first peer sending its name to the
1126 other, together with some kind of secret information which could only
1127 come from the genuine authorized user of that name. In such an
1128 exchange, we will call the first peer the "client" and the other the
1129 "server". The client has a name by which it identifies itself to the
1130 server, and the server also has a name by which it identifies itself to
1131 the client. Generally the genuine client shares some secret (or pass‐
1132 word) with the server, and authenticates itself by proving that it
1133 knows that secret. Very often, the names used for authentication cor‐
1134 respond to the internet hostnames of the peers, but this is not essen‐
1135 tial.
1136 At present, pppd supports three authentication protocols: the Password
1138 Authentication Protocol (PAP), Challenge Handshake Authentication Pro‐
1139 tocol (CHAP), and Extensible Authentication Protocol (EAP). PAP
1140 involves the client sending its name and a cleartext password to the
1141 server to authenticate itself. In contrast, the server initiates the
1142 CHAP authentication exchange by sending a challenge to the client (the
1143 challenge packet includes the server's name). The client must respond
1144 with a response which includes its name plus a hash value derived from
1145 the shared secret and the challenge, in order to prove that it knows
1146 the secret. EAP supports CHAP-style authentication, and also includes
1147 the SRP-SHA1 mechanism, which is resistant to dictionary-based attacks
1148 and does not require a cleartext password on the server side.
1149 The PPP protocol, being symmetrical, allows both peers to require the
1151 other to authenticate itself. In that case, two separate and indepen‐
1152 dent authentication exchanges will occur. The two exchanges could use
1153 different authentication protocols, and in principle, different names
1154 could be used in the two exchanges.
1155 The default behaviour of pppd is to agree to authenticate if requested,
1157 and to not require authentication from the peer. However, pppd will
1158 not agree to authenticate itself with a particular protocol if it has
1159 no secrets which could be used to do so.
1160 Pppd stores secrets for use in authentication in secrets files
1162 (/etc/ppp/pap-secrets for PAP, /etc/ppp/chap-secrets for CHAP, MS-CHAP,
1163 MS-CHAPv2, and EAP MD5-Challenge, and /etc/ppp/srp-secrets for EAP
1164 SRP-SHA1). All secrets files have the same format. The secrets files
1165 can contain secrets for pppd to use in authenticating itself to other
1166 systems, as well as secrets for pppd to use when authenticating other
1167 systems to itself.
1168 Each line in a secrets file contains one secret. A given secret is
1170 specific to a particular combination of client and server - it can only
1171 be used by that client to authenticate itself to that server. Thus
1172 each line in a secrets file has at least 3 fields: the name of the
1173 client, the name of the server, and the secret. These fields may be
1174 followed by a list of the IP addresses that the specified client may
1175 use when connecting to the specified server.
1176 A secrets file is parsed into words as for a options file, so the
1178 client name, server name and secrets fields must each be one word, with
1179 any embedded spaces or other special characters quoted or escaped.
1180 Note that case is significant in the client and server names and in the
1181 secret.
1182 If the secret starts with an `@', what follows is assumed to be the
1184 name of a file from which to read the secret. A "*" as the client or
1185 server name matches any name. When selecting a secret, pppd takes the
1186 best match, i.e. the match with the fewest wildcards.
1187 Any following words on the same line are taken to be a list of accept‐
1189 able IP addresses for that client. If there are only 3 words on the
1190 line, or if the first word is "-", then all IP addresses are disal‐
1191 lowed. To allow any address, use "*". A word starting with "!" indi‐
1192 cates that the specified address is not acceptable. An address may be
1193 followed by "/" and a number n, to indicate a whole subnet, i.e. all
1194 addresses which have the same value in the most significant n bits. In
1195 this form, the address may be followed by a plus sign ("+") to indicate
1196 that one address from the subnet is authorized, based on the ppp net‐
1197 work interface unit number in use. In this case, the host part of the
1198 address will be set to the unit number plus one.
1199 Thus a secrets file contains both secrets for use in authenticating
1201 other hosts, plus secrets which we use for authenticating ourselves to
1202 others. When pppd is authenticating the peer (checking the peer's
1203 identity), it chooses a secret with the peer's name in the first field
1204 and the name of the local system in the second field. The name of the
1205 local system defaults to the hostname, with the domain name appended if
1206 the domain option is used. This default can be overridden with the
1207 name option, except when the usehostname option is used. (For EAP
1208 SRP-SHA1, see the srp-entry(8) utility for generating proper validator
1209 entries to be used in the "secret" field.)
1210 When pppd is choosing a secret to use in authenticating itself to the
1212 peer, it first determines what name it is going to use to identify
1213 itself to the peer. This name can be specified by the user with the
1214 user option. If this option is not used, the name defaults to the name
1215 of the local system, determined as described in the previous paragraph.
1216 Then pppd looks for a secret with this name in the first field and the
1217 peer's name in the second field. Pppd will know the name of the peer
1218 if CHAP or EAP authentication is being used, because the peer will have
1219 sent it in the challenge packet. However, if PAP is being used, pppd
1220 will have to determine the peer's name from the options specified by
1221 the user. The user can specify the peer's name directly with the
1222 remotename option. Otherwise, if the remote IP address was specified
1223 by a name (rather than in numeric form), that name will be used as the
1224 peer's name. Failing that, pppd will use the null string as the peer's
1225 name.
1226 When authenticating the peer with PAP, the supplied password is first
1228 compared with the secret from the secrets file. If the password
1229 doesn't match the secret, the password is encrypted using crypt() and
1230 checked against the secret again. Thus secrets for authenticating the
1231 peer can be stored in encrypted form if desired. If the papcrypt
1232 option is given, the first (unencrypted) comparison is omitted, for
1233 better security.
1234 Furthermore, if the login option was specified, the username and pass‐
1236 word are also checked against the system password database. Thus, the
1237 system administrator can set up the pap-secrets file to allow PPP
1238 access only to certain users, and to restrict the set of IP addresses
1239 that each user can use. Typically, when using the login option, the
1240 secret in /etc/ppp/pap-secrets would be "", which will match any pass‐
1241 word supplied by the peer. This avoids the need to have the same
1242 secret in two places.
1243 Authentication must be satisfactorily completed before IPCP (or any
1245 other Network Control Protocol) can be started. If the peer is
1246 required to authenticate itself, and fails to do so, pppd will termi‐
1247 nated the link (by closing LCP). If IPCP negotiates an unacceptable IP
1248 address for the remote host, IPCP will be closed. IP packets can only
1249 be sent or received when IPCP is open.
1250 In some cases it is desirable to allow some hosts which can't authenti‐
1252 cate themselves to connect and use one of a restricted set of IP
1253 addresses, even when the local host generally requires authentication.
1254 If the peer refuses to authenticate itself when requested, pppd takes
1255 that as equivalent to authenticating with PAP using the empty string
1256 for the username and password. Thus, by adding a line to the
1257 pap-secrets file which specifies the empty string for the client and
1258 password, it is possible to allow restricted access to hosts which
1259 refuse to authenticate themselves.
1260
1262 When IPCP negotiation is completed successfully, pppd will inform the
1263 kernel of the local and remote IP addresses for the ppp interface.
1264 This is sufficient to create a host route to the remote end of the
1265 link, which will enable the peers to exchange IP packets. Communica‐
1266 tion with other machines generally requires further modification to
1267 routing tables and/or ARP (Address Resolution Protocol) tables. In
1268 most cases the defaultroute and/or proxyarp options are sufficient for
1269 this, but in some cases further intervention is required. The
1270 /etc/ppp/ip-up script can be used for this.
1271 Sometimes it is desirable to add a default route through the remote
1273 host, as in the case of a machine whose only connection to the Internet
1274 is through the ppp interface. The defaultroute option causes pppd to
1275 create such a default route when IPCP comes up, and delete it when the
1276 link is terminated.
1277 In some cases it is desirable to use proxy ARP, for example on a server
1279 machine connected to a LAN, in order to allow other hosts to communi‐
1280 cate with the remote host. The proxyarp option causes pppd to look for
1281 a network interface on the same subnet as the remote host (an interface
1282 supporting broadcast and ARP, which is up and not a point-to-point or
1283 loopback interface). If found, pppd creates a permanent, published ARP
1284 entry with the IP address of the remote host and the hardware address
1285 of the network interface found.
1286 When the demand option is used, the interface IP addresses have already
1288 been set at the point when IPCP comes up. If pppd has not been able to
1289 negotiate the same addresses that it used to configure the interface
1290 (for example when the peer is an ISP that uses dynamic IP address
1291 assignment), pppd has to change the interface IP addresses to the nego‐
1292 tiated addresses. This may disrupt existing connections, and the use
1293 of demand dialling with peers that do dynamic IP address assignment is
1294 not recommended.
1295
1297 Multilink PPP provides the capability to combine two or more PPP links
1298 between a pair of machines into a single `bundle', which appears as a
1299 single virtual PPP link which has the combined bandwidth of the indi‐
1300 vidual links. Currently, multilink PPP is only supported under Linux.
1301 Pppd detects that the link it is controlling is connected to the same
1303 peer as another link using the peer's endpoint discriminator and the
1304 authenticated identity of the peer (if it authenticates itself). The
1305 endpoint discriminator is a block of data which is hopefully unique for
1306 each peer. Several types of data can be used, including locally-
1307 assigned strings of bytes, IP addresses, MAC addresses, randomly
1308 strings of bytes, or E-164 phone numbers. The endpoint discriminator
1309 sent to the peer by pppd can be set using the endpoint option.
1310 In some circumstances the peer may send no endpoint discriminator or a
1312 non-unique value. The bundle option adds an extra string which is
1313 added to the peer's endpoint discriminator and authenticated identity
1314 when matching up links to be joined together in a bundle. The bundle
1315 option can also be used to allow the establishment of multiple bundles
1316 between the local system and the peer. Pppd uses a TDB database in
1317 /var/run/pppd2.tdb to match up links.
1318 Assuming that multilink is enabled and the peer is willing to negotiate
1320 multilink, then when pppd is invoked to bring up the first link to the
1321 peer, it will detect that no other link is connected to the peer and
1322 create a new bundle, that is, another ppp network interface unit. When
1323 another pppd is invoked to bring up another link to the peer, it will
1324 detect the existing bundle and join its link to it.
1325 If the first link terminates (for example, because of a hangup or a
1327 received LCP terminate-request) the bundle is not destroyed unless
1328 there are no other links remaining in the bundle. Rather than exiting,
1329 the first pppd keeps running after its link terminates, until all the
1330 links in the bundle have terminated. If the first pppd receives a
1331 SIGTERM or SIGINT signal, it will destroy the bundle and send a SIGHUP
1332 to the pppd processes for each of the links in the bundle. If the
1333 first pppd receives a SIGHUP signal, it will terminate its link but not
1334 the bundle.
1335 Note: demand mode is not currently supported with multilink.
1337
1339 The following examples assume that the /etc/ppp/options file contains
1340 the auth option (as in the default /etc/ppp/options file in the ppp
1341 distribution).
1342 Probably the most common use of pppd is to dial out to an ISP. This
1344 can be done with a command such as
1345 pppd call isp
1347 where the /etc/ppp/peers/isp file is set up by the system administrator
1349 to contain something like this:
1350 ttyS0 19200 crtscts
1352 connect '/usr/sbin/chat -v -f /etc/ppp/chat-isp'
1353 noauth
1354 In this example, we are using chat to dial the ISP's modem and go
1356 through any logon sequence required. The /etc/ppp/chat-isp file con‐
1357 tains the script used by chat; it could for example contain something
1358 like this:
1359 ABORT "NO CARRIER"
1361 ABORT "NO DIALTONE"
1362 ABORT "ERROR"
1363 ABORT "NO ANSWER"
1364 ABORT "BUSY"
1365 ABORT "Username/Password Incorrect"
1366 "" "at"
1367 OK "at&d0&c1"
1368 OK "atdt2468135"
1369 "name:" "^Umyuserid"
1370 "word:" "\qmypassword"
1371 "ispts" "\q^Uppp"
1372 "~-^Uppp-~"
1373 See the chat(8) man page for details of chat scripts.
1375 Pppd can also be used to provide a dial-in ppp service for users. If
1377 the users already have login accounts, the simplest way to set up the
1378 ppp service is to let the users log in to their accounts and run pppd
1379 (installed setuid-root) with a command such as
1380 pppd proxyarp
1382 To allow a user to use the PPP facilities, you need to allocate an IP
1384 address for that user's machine and create an entry in
1385 /etc/ppp/pap-secrets, /etc/ppp/chap-secrets, or /etc/ppp/srp-secrets
1386 (depending on which authentication method the PPP implementation on the
1387 user's machine supports), so that the user's machine can authenticate
1388 itself. For example, if Joe has a machine called "joespc" that is to
1389 be allowed to dial in to the machine called "server" and use the IP
1390 address joespc.my.net, you would add an entry like this to
1391 /etc/ppp/pap-secrets or /etc/ppp/chap-secrets:
1392 joespc server "joe's secret" joespc.my.net
1394 (See srp-entry(8) for a means to generate the server's entry when
1396 SRP-SHA1 is in use.) Alternatively, you can create a username called
1397 (for example) "ppp", whose login shell is pppd and whose home directory
1398 is /etc/ppp. Options to be used when pppd is run this way can be put
1399 in /etc/ppp/.ppprc.
1400 If your serial connection is any more complicated than a piece of wire,
1402 you may need to arrange for some control characters to be escaped. In
1403 particular, it is often useful to escape XON (^Q) and XOFF (^S), using
1404 asyncmap a0000. If the path includes a telnet, you probably should
1405 escape ^] as well (asyncmap 200a0000). If the path includes an rlogin,
1406 you will need to use the escape ff option on the end which is running
1407 the rlogin client, since many rlogin implementations are not transpar‐
1408 ent; they will remove the sequence [0xff, 0xff, 0x73, 0x73, followed by
1409 any 8 bytes] from the stream.
1410
1412 Messages are sent to the syslog daemon using facility LOG_DAEMON.
1413 (This can be overridden by recompiling pppd with the macro LOG_PPP
1414 defined as the desired facility.) See the syslog(8) documentation for
1415 details of where the syslog daemon will write the messages. On most
1416 systems, the syslog daemon uses the /etc/syslog.conf file to specify
1417 the destination(s) for syslog messages. You may need to edit that file
1418 to suit.
1419 The debug option causes the contents of all control packets sent or
1421 received to be logged, that is, all LCP, PAP, CHAP, EAP, or IPCP pack‐
1422 ets. This can be useful if the PPP negotiation does not succeed or if
1423 authentication fails. If debugging is enabled at compile time, the
1424 debug option also causes other debugging messages to be logged.
1425 Debugging can also be enabled or disabled by sending a SIGUSR1 signal
1427 to the pppd process. This signal acts as a toggle.
1428
1430 The exit status of pppd is set to indicate whether any error was
1431 detected, or the reason for the link being terminated. The values used
1432 are:
1433 0 Pppd has detached, or otherwise the connection was successfully
1435 established and terminated at the peer's request.
1436 1 An immediately fatal error of some kind occurred, such as an
1438 essential system call failing, or running out of virtual memory.
1439 2 An error was detected in processing the options given, such as
1441 two mutually exclusive options being used.
1442 3 Pppd is not setuid-root and the invoking user is not root.
1444 4 The kernel does not support PPP, for example, the PPP kernel
1446 driver is not included or cannot be loaded.
1447 5 Pppd terminated because it was sent a SIGINT, SIGTERM or SIGHUP
1449 signal.
1450 6 The serial port could not be locked.
1452 7 The serial port could not be opened.
1454 8 The connect script failed (returned a non-zero exit status).
1456 9 The command specified as the argument to the pty option could
1458 not be run.
1459 10 The PPP negotiation failed, that is, it didn't reach the point
1461 where at least one network protocol (e.g. IP) was running.
1462 11 The peer system failed (or refused) to authenticate itself.
1464 12 The link was established successfully and terminated because it
1466 was idle.
1467 13 The link was established successfully and terminated because the
1469 connect time limit was reached.
1470 14 Callback was negotiated and an incoming call should arrive
1472 shortly.
1473 15 The link was terminated because the peer is not responding to
1475 echo requests.
1476 16 The link was terminated by the modem hanging up.
1478 17 The PPP negotiation failed because serial loopback was detected.
1480 18 The init script failed (returned a non-zero exit status).
1482 19 We failed to authenticate ourselves to the peer.
1484
1486 Pppd invokes scripts at various stages in its processing which can be
1487 used to perform site-specific ancillary processing. These scripts are
1488 usually shell scripts, but could be executable code files instead.
1489 Pppd does not wait for the scripts to finish (except for the ip-pre-up
1490 script). The scripts are executed as root (with the real and effective
1491 user-id set to 0), so that they can do things such as update routing
1492 tables or run privileged daemons. Be careful that the contents of
1493 these scripts do not compromise your system's security. Pppd runs the
1494 scripts with standard input, output and error redirected to /dev/null,
1495 and with an environment that is empty except for some environment vari‐
1496 ables that give information about the link. The environment variables
1497 that pppd sets are:
1498 DEVICE The name of the serial tty device being used.
1500 IFNAME The name of the network interface being used.
1502 IPLOCAL
1504 The IP address for the local end of the link. This is only set
1505 when IPCP has come up.
1506 IPREMOTE
1508 The IP address for the remote end of the link. This is only set
1509 when IPCP has come up.
1510 PEERNAME
1512 The authenticated name of the peer. This is only set if the
1513 peer authenticates itself.
1514 SPEED The baud rate of the tty device.
1516 ORIG_UID
1518 The real user-id of the user who invoked pppd.
1519 PPPLOGNAME
1521 The username of the real user-id that invoked pppd. This is
1522 always set.
1523 For the ip-down and auth-down scripts, pppd also sets the following
1525 variables giving statistics for the connection:
1526 CONNECT_TIME
1528 The number of seconds from when the PPP negotiation started
1529 until the connection was terminated.
1530 BYTES_SENT
1532 The number of bytes sent (at the level of the serial port) dur‐
1533 ing the connection.
1534 BYTES_RCVD
1536 The number of bytes received (at the level of the serial port)
1537 during the connection.
1538 LINKNAME
1540 The logical name of the link, set with the linkname option.
1541 DNS1 If the peer supplies DNS server addresses, this variable is set
1543 to the first DNS server address supplied.
1544 DNS2 If the peer supplies DNS server addresses, this variable is set
1546 to the second DNS server address supplied.
1547 Pppd invokes the following scripts, if they exist. It is not an error
1549 if they don't exist.
1550 /etc/ppp/auth-up
1552 A program or script which is executed after the remote system
1553 successfully authenticates itself. It is executed with the
1554 parameters
1555 interface-name peer-name user-name tty-device speed
1557 Note that this script is not executed if the peer doesn't
1559 authenticate itself, for example when the noauth option is used.
1560 /etc/ppp/auth-down
1562 A program or script which is executed when the link goes down,
1563 if /etc/ppp/auth-up was previously executed. It is executed in
1564 the same manner with the same parameters as /etc/ppp/auth-up.
1565 /etc/ppp/ip-pre-up
1567 A program or script which is executed just before the ppp net‐
1568 work interface is brought up. It is executed with the same
1569 parameters as the ip-up script (below). At this point the
1570 interface exists and has IP addresses assigned but is still
1571 down. This can be used to add firewall rules before any IP
1572 traffic can pass through the interface. Pppd will wait for this
1573 script to finish before bringing the interface up, so this
1574 script should run quickly.
1575 /etc/ppp/ip-up
1577 A program or script which is executed when the link is available
1578 for sending and receiving IP packets (that is, IPCP has come
1579 up). It is executed with the parameters
1580 interface-name tty-device speed local-IP-address
1582 remote-IP-address ipparam
1583 /etc/ppp/ip-down
1585 A program or script which is executed when the link is no longer
1586 available for sending and receiving IP packets. This script can
1587 be used for undoing the effects of the /etc/ppp/ip-up and
1588 /etc/ppp/ip-pre-up scripts. It is invoked in the same manner
1589 and with the same parameters as the ip-up script.
1590 /etc/ppp/ipv6-up
1592 Like /etc/ppp/ip-up, except that it is executed when the link is
1593 available for sending and receiving IPv6 packets. It is executed
1594 with the parameters
1595 interface-name tty-device speed local-link-local-address
1597 remote-link-local-address ipparam
1598 /etc/ppp/ipv6-down
1600 Similar to /etc/ppp/ip-down, but it is executed when IPv6 pack‐
1601 ets can no longer be transmitted on the link. It is executed
1602 with the same parameters as the ipv6-up script.
1603 /etc/ppp/ipx-up
1605 A program or script which is executed when the link is available
1606 for sending and receiving IPX packets (that is, IPXCP has come
1607 up). It is executed with the parameters
1608 interface-name tty-device speed network-number
1610 local-IPX-node-address remote-IPX-node-address local-IPX-rout‐
1611 ing-protocol remote-IPX-routing-protocol local-IPX-router-name
1612 remote-IPX-router-name ipparam pppd-pid
1613 The local-IPX-routing-protocol and remote-IPX-routing-protocol
1615 field may be one of the following:
1616 NONE to indicate that there is no routing protocol
1618 RIP to indicate that RIP/SAP should be used
1619 NLSP to indicate that Novell NLSP should be used
1620 RIP NLSP to indicate that both RIP/SAP and NLSP should be used
1621 /etc/ppp/ipx-down
1623 A program or script which is executed when the link is no longer
1624 available for sending and receiving IPX packets. This script
1625 can be used for undoing the effects of the /etc/ppp/ipx-up
1626 script. It is invoked in the same manner and with the same
1627 parameters as the ipx-up script.
1628
1630 /var/run/pppn.pid (BSD or Linux), /etc/ppp/pppn.pid (others)
1631 Process-ID for pppd process on ppp interface unit n.
1632 /var/run/ppp-name.pid (BSD or Linux),
1634 /etc/ppp/ppp-name.pid (others) Process-ID for pppd process for
1635 logical link name (see the linkname option).
1636 /var/run/pppd2.tdb
1638 Database containing information about pppd processes, interfaces
1639 and links, used for matching links to bundles in multilink oper‐
1640 ation. May be examined by external programs to obtain informa‐
1641 tion about running pppd instances, the interfaces and devices
1642 they are using, IP address assignments, etc.
1643 /etc/ppp/pap-secrets Usernames, passwords and IP addresses for
1644 PAP authentication. This file should be owned by root and not
1645 readable or writable by any other user. Pppd will log a warning
1646 if this is not the case.
1647 /etc/ppp/chap-secrets
1649 Names, secrets and IP addresses for CHAP/MS-CHAP/MS-CHAPv2
1650 authentication. As for /etc/ppp/pap-secrets, this file should
1651 be owned by root and not readable or writable by any other user.
1652 Pppd will log a warning if this is not the case.
1653 /etc/ppp/srp-secrets
1655 Names, secrets, and IP addresses for EAP authentication. As for
1656 /etc/ppp/pap-secrets, this file should be owned by root and not
1657 readable or writable by any other user. Pppd will log a warning
1658 if this is not the case.
1659 ~/.ppp_pseudonym
1661 Saved client-side SRP-SHA1 pseudonym. See the srp-use-pseudonym
1662 option for details.
1663 /etc/ppp/options
1665 System default options for pppd, read before user default
1666 options or command-line options.
1667 ~/.ppprc
1669 User default options, read before /etc/ppp/options.ttyname.
1670 /etc/ppp/options.ttyname
1672 System default options for the serial port being used, read
1673 after ~/.ppprc. In forming the ttyname part of this filename,
1674 an initial /dev/ is stripped from the port name (if present),
1675 and any slashes in the remaining part are converted to dots.
1676 /etc/ppp/peers
1678 A directory containing options files which may contain privi‐
1679 leged options, even if pppd was invoked by a user other than
1680 root. The system administrator can create options files in this
1681 directory to permit non-privileged users to dial out without
1682 requiring the peer to authenticate, but only to certain trusted
1683 peers.
1684
1686 chat(8), pppstats(8)
1687 RFC1144
1689 Jacobson, V. Compressing TCP/IP headers for low-speed serial
1690 links. February 1990.
1691 RFC1321
1693 Rivest, R. The MD5 Message-Digest Algorithm. April 1992.
1694 RFC1332
1696 McGregor, G. PPP Internet Protocol Control Protocol (IPCP).
1697 May 1992.
1698 RFC1334
1700 Lloyd, B.; Simpson, W.A. PPP authentication protocols. October
1701 1992.
1702 RFC1661
1704 Simpson, W.A. The Point-to-Point Protocol (PPP). July 1994.
1705 RFC1662
1707 Simpson, W.A. PPP in HDLC-like Framing. July 1994.
1708 RFC2284
1710 Blunk, L.; Vollbrecht, J., PPP Extensible Authentication Proto‐
1711 col (EAP). March 1998.
1712 RFC2472
1714 Haskin, D. IP Version 6 over PPP December 1998.
1715 RFC2945
1717 Wu, T., The SRP Authentication and Key Exchange System September
1718 2000.
1719 draft-ietf-pppext-eap-srp-03.txt
1721 Carlson, J.; et al., EAP SRP-SHA1 Authentication Protocol. July
1722 2001.
1723
1725 Some limited degree of control can be exercised over a running pppd
1726 process by sending it a signal from the list below.
1727 SIGINT, SIGTERM
1729 These signals cause pppd to terminate the link (by closing LCP),
1730 restore the serial device settings, and exit. If a connector or
1731 disconnector process is currently running, pppd will send the
1732 same signal to its process group, so as to terminate the connec‐
1733 tor or disconnector process.
1734 SIGHUP This signal causes pppd to terminate the link, restore the
1736 serial device settings, and close the serial device. If the
1737 persist or demand option has been specified, pppd will try to
1738 reopen the serial device and start another connection (after the
1739 holdoff period). Otherwise pppd will exit. If this signal is
1740 received during the holdoff period, it causes pppd to end the
1741 holdoff period immediately. If a connector or disconnector
1742 process is running, pppd will send the same signal to its
1743 process group.
1744 SIGUSR1
1746 This signal toggles the state of the debug option.
1747 SIGUSR2
1749 This signal causes pppd to renegotiate compression. This can be
1750 useful to re-enable compression after it has been disabled as a
1751 result of a fatal decompression error. (Fatal decompression
1752 errors generally indicate a bug in one or other implementation.)
1753
1756 Paul Mackerras (paulus@samba.org), based on earlier work by Drew
1757 Perkins, Brad Clements, Karl Fox, Greg Christy, and Brad Parker.
1758
1761 Pppd is copyrighted and made available under conditions which provide
1762 that it may be copied and used in source or binary forms provided that
1763 the conditions listed below are met. Portions of pppd are covered by
1764 the following copyright notices:
1765 Copyright (c) 1984-2000 Carnegie Mellon University. All rights
1767 reserved.
1768 Copyright (c) 1993-2004 Paul Mackerras. All rights reserved.
1769 Copyright (c) 1995 Pedro Roque Marques. All rights reserved.
1770 Copyright (c) 1995 Eric Rosenquist. All rights reserved.
1771 Copyright (c) 1999 Tommi Komulainen. All rights reserved.
1772 Copyright (C) Andrew Tridgell 1999
1773 Copyright (c) 2000 by Sun Microsystems, Inc. All rights reserved.
1774 Copyright (c) 2001 by Sun Microsystems, Inc. All rights reserved.
1775 Copyright (c) 2002 Google, Inc. All rights reserved.
1776 The copyright notices contain the following statements.
1778 Redistribution and use in source and binary forms, with or without mod‐
1780 ification, are permitted provided that the following conditions are
1781 met:
1782 1. Redistributions of source code must retain the above copyright
1784 notice, this list of conditions and the following disclaimer.
1785 2. Redistributions in binary form must reproduce the above copyright
1787 notice, this list of conditions and the following disclaimer in
1788 the documentation and/or other materials provided with the
1789 distribution.
1790 3. The name "Carnegie Mellon University" must not be used to
1792 endorse or promote products derived from this software without
1793 prior written permission. For permission or any legal
1794 details, please contact
1795 Office of Technology Transfer
1796 Carnegie Mellon University
1797 5000 Forbes Avenue
1798 Pittsburgh, PA 15213-3890
1799 (412) 268-4387, fax: (412) 268-7395
1800 tech-transfer@andrew.cmu.edu
1801 3b. The name(s) of the authors of this software must not be used to
1803 endorse or promote products derived from this software without
1804 prior written permission.
1805 4. Redistributions of any form whatsoever must retain the following
1807 acknowledgements:
1808 "This product includes software developed by Computing Services
1809 at Carnegie Mellon University (http://www.cmu.edu/computing/)."
1810 "This product includes software developed by Paul Mackerras
1811 <paulus@samba.org>".
1812 "This product includes software developed by Pedro Roque Marques
1813 <pedro_m@yahoo.com>".
1814 "This product includes software developed by Tommi Komulainen
1815 <Tommi.Komulainen@iki.fi>".
1816 CARNEGIE MELLON UNIVERSITY DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS
1818 SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FIT‐
1819 NESS, IN NO EVENT SHALL CARNEGIE MELLON UNIVERSITY BE LIABLE FOR ANY
1820 SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER
1821 RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF
1822 CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
1823 CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
1824 THE AUTHORS OF THIS SOFTWARE DISCLAIM ALL WARRANTIES WITH REGARD TO
1826 THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND
1827 FITNESS, IN NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY SPECIAL, INDI‐
1828 RECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM
1829 LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLI‐
1830 GENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH
1831 THE USE OR PERFORMANCE OF THIS SOFTWARE.
1832 PPPD(8)