Debian Codegraph @

docs/html/manual.html

raw
1 <h1 id="overview">Overview</h1>
2 <p><code>l2tpns</code> is half of a complete L2TP implementation. It
3 supports only the LNS side of the connection.</p>
4 <p>L2TP (Layer 2 Tunneling Protocol) is designed to allow any layer 2
5 protocol (e.g. Ethernet, PPP) to be tunneled over an IP connection.
6 <code>l2tpns</code> implements PPP over L2TP only.</p>
7 <p>There are a couple of other L2TP implementations, of which <a
8 href="http://sourceforge.net/projects/l2tpd">l2tpd</a> is probably the
9 most popular. l2tpd also will handle being either end of a tunnel, and
10 is a lot more configurable than <code>l2tpns</code>. However, due to the
11 way it works, it is nowhere near as scalable.</p>
12 <p><code>l2tpns</code> uses the TUN/TAP interface provided by the Linux
13 kernel to receive and send packets. Using some packet manipulation it
14 doesn't require a single interface per connection, as l2tpd does.</p>
15 <p>This allows it to scale extremely well to very high loads and very
16 high numbers of connections.</p>
17 <p>It also has a plugin architecture which allows custom code to be run
18 during processing. An example of this is in the walled garden module
19 included.</p>
20 <h1 id="installation">Installation</h1>
21 <h2 id="installation-requirements">Requirements</h2>
22 <ul>
23 <li><p>Linux kernel version 2.4 or above, with the Tun/Tap interface
24 either compiled in, or as a module.</p></li>
25 <li><p>libcli 1.8.5 or greater. You can get this from <a
26 href="http://sourceforge.net/projects/libcli">SourceForge</a></p></li>
27 </ul>
28 <h2 id="installation-compile">Compiling</h2>
29 <p>You can generally get away with just running <code>make</code> from
30 the source directory. This will compile the daemon, associated tools and
31 any modules shipped with the distribution.</p>
32 <h2 id="installation-install">Installing</h2>
33 <p>After you have successfully compiled everything, run
34 <code>make install</code> to install it. By default, the binaries are
35 installed into <code>/usr/sbin</code>, the configuration into
36 <code>/etc/l2tpns</code>, and the modules into
37 <code>/usr/lib/l2tpns</code>.</p>
38 <p>You will definately need to edit the configuration files before you
39 start. See <a href="#configuration">Configuration</a> for more
40 information.</p>
41 <h2 id="installation-run">Running</h2>
42 <p>You only need to run <code>/usr/sbin/l2tpns</code> as root to start
43 it. It does not normally detach to a daemon process (see the
44 <code>-d</code> option), so you should perhaps run it from
45 <code>init</code>.</p>
46 <p>By default there is no log destination set, so all log messages will
47 go to stdout.</p>
48 <h1 id="configuration">Configuration</h1>
49 <p>All configuration of the software is done from the files installed
50 into <code>/etc/l2tpns</code>.</p>
51 <h2 id="config-startup"><code>startup-config</code></h2>
52 <p>This is the main configuration file for <code>l2tpns</code>. The
53 format of the file is a list of commands that can be run through the
54 command-line interface. This file can also be written directly by the
55 <code>l2tpns</code> process if a user runs the <code>write memory</code>
56 command, so any comments will be lost. However if your policy is not to
57 write the config by the program, then feel free to comment the file with
58 a <code>#</code> or <code>!</code> at the beginning of the line.</p>
59 <p>A list of the possible configuration directives follows. Each of
60 these should be set by a line like: set configstring "value" set
61 ipaddress 192.168.1.1 set boolean true</p>
62 <dl>
63 <dt><code>debug</code> (int)</dt>
64 <dd>
65 <p>Sets the level of messages that will be written to the log file. The
66 value should be between 0 and 5, with 0 being no debugging, and 5 being
67 the highest. A rough description of the levels is:</p>
68 <dl>
69 <dt><code>0</code>: Critical Errors</dt>
70 <dd>
71 <p>Things are probably broken</p>
72 </dd>
73 <dt><code>1</code>: Errors</dt>
74 <dd>
75 <p>Things might have gone wrong, but probably will recover</p>
76 </dd>
77 <dt><code>2</code>: Warnings</dt>
78 <dd>
79 <p>Just in case you care what is not quite perfect</p>
80 </dd>
81 <dt><code>3</code>: Information</dt>
82 <dd>
83 <p>Parameters of control packets</p>
84 </dd>
85 <dt><code>4</code>: Calls</dt>
86 <dd>
87 <p>For tracing the execution of the code</p>
88 </dd>
89 <dt><code>5</code>: Packets</dt>
90 <dd>
91 <p>Everything, including a hex dump of all packets processed... probably
92 twice</p>
93 </dd>
94 </dl>
95 <p>Note that the higher you set the debugging level, the slower the
96 program will run. Also, at level 5 a <em>lot</em> of information will be
97 logged. This should only ever be used for working out why it doesn't
98 work at all.</p>
99 </dd>
100 <dt><code>log_file</code> (string)</dt>
101 <dd>
102 <p>This will be where all logging and debugging information is written
103 to. This may be either a filename, such as <code>/var/log/l2tpns</code>,
104 or the special magic string <code>syslog:</code>facility, where facility
105 is any one of the syslog logging facilities, such as
106 <code>local5</code>.</p>
107 </dd>
108 <dt><code>pid_file</code> (string)</dt>
109 <dd>
110 <p>If set, the process id will be written to the specified file. The
111 value must be an absolute path.</p>
112 </dd>
113 <dt><code>random_device</code> (string)</dt>
114 <dd>
115 <p>Path to random data source (default <code>/dev/urandom</code>). Use
116 "" to use the rand() library function.</p>
117 </dd>
118 <dt><code>l2tp_secret</code> (string)</dt>
119 <dd>
120 <p>The secret used by <code>l2tpns</code> for authenticating tunnel
121 request. Must be the same as the LAC, or authentication will fail. Only
122 actually be used if the LAC requests authentication.</p>
123 </dd>
124 <dt><code>l2tp_mtu</code> (int)</dt>
125 <dd>
126 <p>MTU of interface for L2TP traffic (default: <code>1500</code>). Used
127 to set link MRU and adjust TCP MSS.</p>
128 </dd>
129 <dt><code>mp_mrru</code> (int)</dt>
130 <dd>
131 <p>MRRU for MP traffic (default: <code>1614</code>). Can be set to 0 to
132 disable MP negociation.</p>
133 </dd>
134 <dt><code>ppp_restart_time</code> (int); <code>ppp_max_configure</code>
135 (int); <code>ppp_max_failure</code> (int)</dt>
136 <dd>
137 <p>PPP counter and timer values, as described in §4.1 of <a
138 href="ftp://ftp.rfc-editor.org/in-notes/rfc1661.txt">RFC1661</a>.</p>
139 </dd>
140 <dt><code>lcp_renegotiation</code> (string)</dt>
141 <dd>
142 <p>By default (<code>always</code>), we renegotiate LCP even if the LAC
143 already did with the client.</p>
144 <p>We can avoid the LCP renegotiation (proxy LCP negotiation), unless
145 what was already negotiated is not fine for us
146 (<code>on-mismatch</code>). To be noted: we will accept not using our
147 preferred RADIUS authentication method. This is notably useful when the
148 LAC cannot pass LCP configuration through, and thus we have to try to
149 accept what was already negotiated by the LAC with the client.</p>
150 </dd>
151 <dt><code>primary_dns</code> (ip address); <code>econdary_dns</code> (ip
152 address)</dt>
153 <dd>
154 <p>Whenever a PPP connection is established, DNS servers will be sent to
155 the user, both a primary and a secondary. If either is set to 0.0.0.0,
156 then that one will not be sent.</p>
157 </dd>
158 <dt><code>primary_radius</code> (ip address);
159 <code>secondary_radius</code> (ip address)</dt>
160 <dd>
161 <p>Sets the RADIUS servers used for both authentication and accounting.
162 If the primary server does not respond, then the secondary RADIUS server
163 will be tried.</p>
164 <div class="note">
165 <p>In addition to the source IP address and identifier, the RADIUS
166 server <em>must</em> include the source port when detecting duplicates
167 to suppress (in order to cope with a large number of sessions coming
168 on-line simultaneously <code>l2tpns</code> uses a set of udp sockets,
169 each with a separate identifier).</p>
170 </div>
171 </dd>
172 <dt><code>primary_radius_port</code> (short);
173 <code>secondary_radius_port</code> (short)</dt>
174 <dd>
175 <p>Sets the authentication ports for the primary and secondary RADIUS
176 servers. The accounting port is one more than the authentication port.
177 If no RADIUS ports are given, the authentication port defaults to 1812,
178 and the accounting port to 1813.</p>
179 </dd>
180 <dt><code>radius_accounting</code> (boolean)</dt>
181 <dd>
182 <p>If set to true, then RADIUS accounting packets will be sent. This
183 means that a Start record will be sent when the session is successfully
184 authenticated, and a Stop record will be sent when the session is
185 closed.</p>
186 </dd>
187 <dt><code>radius_interim</code> (int)</dt>
188 <dd>
189 <p>If <code>radius_accounting</code> is on, defines the interval between
190 sending of RADIUS interim accounting records (in seconds).</p>
191 </dd>
192 <dt><code>radius_secret</code> (string)</dt>
193 <dd>
194 <p>This secret will be used in all RADIUS queries. If this is not set
195 then RADIUS queries will fail.</p>
196 </dd>
197 <dt><code>radius_require_message_authenticator</code> (string)</dt>
198 <dd>
199 <p>If set to true, RADIUS answers to AccessRequests will have to contain
200 a valid MessageAuthenticator. If set to auto (default), if the first
201 RADIUS answer to AccessRequests contains a valid MessageAuthenticator,
202 subsequent answers will have to contain one. If set to no (not
203 recommended), RADIUS answers to AccessRequests do not have to contain a
204 valid MessageAuthenticator. It is advised to set this to true after
205 checking that your RADIUS server does send MessageAuthenticator.</p>
206 </dd>
207 <dt><code>radius_authtypes</code> (string)</dt>
208 <dd>
209 <p>A comma separated list of supported RADIUS authentication methods
210 (<code>pap</code> or <code>chap</code>), in order of preference (default
211 <code>pap</code>).</p>
212 </dd>
213 <dt><code>radius_bind_min</code> (short); <code>radius_bind_max</code>
214 (short)</dt>
215 <dd>
216 <p>Define a port range in which to bind sockets used to send and receive
217 RADIUS packets. Must be at least RADIUS_FDS (64) wide. Simplifies
218 firewalling of RADIUS ports (default: dynamically assigned).</p>
219 </dd>
220 <dt><code>radius_dae_port</code> (short)</dt>
221 <dd>
222 <p>Port for DAE RADIUS (Packet of Death/Disconnect, Change of
223 Authorization) requests (default: <code>3799</code>).</p>
224 </dd>
225 <dt><code>allow_duplicate_users</code> (boolean)</dt>
226 <dd>
227 <p>Allow multiple logins with the same username. If false (the default),
228 any prior session with the same username will be dropped when a new
229 session is established.</p>
230 </dd>
231 <dt><code>guest_account</code> (string)</dt>
232 <dd>
233 <p>Allow multiple logins matching this specific username.</p>
234 </dd>
235 <dt><code>bind_address</code> (ip address)</dt>
236 <dd>
237 <p>When the tun interface is created, it is assigned the address
238 specified here. If no address is given, 1.1.1.1 is used. Packets
239 containing user traffic should be routed via this address if given,
240 otherwise the primary address of the machine.</p>
241 </dd>
242 <dt><code>peer_address</code> (ip address)</dt>
243 <dd>
244 <p>Address to send to clients as the default gateway.</p>
245 </dd>
246 <dt><code>route_protocol</code> (short)</dt>
247 <dd>
248 <p>Protocol number to record when adding a route (see
249 <code>/usr/share/iproute2/rt_protos</code>). This allows to separate
250 routes added by several l2tpns instances. The default is 42.</p>
251 </dd>
252 <dt><code>route_metric</code> (int)</dt>
253 <dd>
254 <p>Metric of the route. Routes with lower values are preferred. This
255 should not be changed during execution, since l2tpns would then not
256 recognize the routes it added itself. The default is 1.</p>
257 </dd>
258 <dt><code>ipv6_prefix</code> (ipv6 address)</dt>
259 <dd>
260 <p>Enable negotiation of IPv6. This forms the the first 64 bits of the
261 client allocated address. The remaining 64 come from the allocated IPv4
262 address and 4 bytes of 0s.</p>
263 </dd>
264 <dt><code>send_garp</code> (boolean)</dt>
265 <dd>
266 <p>Determines whether or not to send a gratuitous ARP for the
267 bind_address when the server is ready to handle traffic (default:
268 <code>true</code>). This value is ignored if BGP is configured.</p>
269 </dd>
270 <dt><code>throttle_speed</code> (int)</dt>
271 <dd>
272 <p>Sets the default speed (in kbits/s) which sessions will be limited
273 to. If this is set to 0, then throttling will not be used at all. Note:
274 You can set this by the CLI, but changes will not affect currently
275 connected users.</p>
276 </dd>
277 <dt><code>throttle_buckets</code> (int)</dt>
278 <dd>
279 <p>Number of token buckets to allocate for throttling. Each throttled
280 session requires two buckets (in and out).</p>
281 </dd>
282 <dt><code>accounting_dir</code> (string)</dt>
283 <dd>
284 <p>If set to a directory, then every 5 minutes the current usage for
285 every connected use will be dumped to a file in this directory. Each
286 file dumped begins with a header, where each line is prefixed by
287 <code>#</code>. Following the header is a single line for every
288 connected user, fields separated by a space.</p>
289 <p>The fields are username, ip, qos, uptxoctets, downrxoctets. The qos
290 field is 1 if a standard user, and 2 if the user is throttled.</p>
291 </dd>
292 <dt><code>dump_speed</code> (boolean)</dt>
293 <dd>
294 <p>If set to true, then the current bandwidth utilization will be logged
295 every second. Even if this is disabled, you can see this information by
296 running the <code>uptime</code> command on the CLI.</p>
297 </dd>
298 <dt><code>multi_read_count</code> (int)</dt>
299 <dd>
300 <p>Number of packets to read off each of the UDP and TUN fds when
301 returned as readable by select (default: 10). Avoids incurring the
302 unnecessary system call overhead of select on busy servers.</p>
303 </dd>
304 <dt><code>scheduler_fifo</code> (boolean)</dt>
305 <dd>
306 <p>Sets the scheduling policy for the <code>l2tpns</code> process to
307 <code>SCHED_FIFO</code>. This causes the kernel to immediately preempt
308 any currently running <code>SCHED_OTHER</code> (normal) process in
309 favour of <code>l2tpns</code> when it becomes runnable. Ignored on
310 uniprocessor systems.</p>
311 </dd>
312 <dt><code>lock_pages</code> (boolean)</dt>
313 <dd>
314 <p>Keep all pages mapped by the <code>l2tpns</code> process in
315 memory.</p>
316 </dd>
317 <dt><code>icmp_rate</code> (int)</dt>
318 <dd>
319 <p>Maximum number of host unreachable ICMP packets to send per
320 second.</p>
321 </dd>
322 <dt><code>packet_limit</code> (int)</dt>
323 <dd>
324 <p>Maximum number of packets of downstream traffic to be handled each
325 tenth of a second per session. If zero, no limit is applied (default:
326 0). Intended as a DoS prevention mechanism and not a general throttling
327 control (packets are dropped, not queued).</p>
328 </dd>
329 <dt><code>cluster_address</code> (ip address)</dt>
330 <dd>
331 <p>Multicast cluster address (default: 239.192.13.13). See <a
332 href="#clustering">Clustering</a> for more information.</p>
333 </dd>
334 <dt><code>cluster_port</code> (udp port)</dt>
335 <dd>
336 <p>UDP cluster port (default: 32792). See <a
337 href="#clustering">Clustering</a> for more information.</p>
338 </dd>
339 <dt><code>cluster_interface</code> (string)</dt>
340 <dd>
341 <p>Interface for cluster packets (default: none). An IPv4 address can
342 also be specified, e.g. to run several l2tpns on the same host (but
343 beware of enabling kernel acceleration only on one instance, to set a
344 different <code>bind_portremotelns</code>, and to set a different
345 <code>route_metric</code>. It is also useful to set a different
346 <code>route_protocol</code> number)</p>
347 </dd>
348 <dt><code>cluster_mcast_ttl</code> (int)</dt>
349 <dd>
350 <p>TTL for multicast packets (default: 1).</p>
351 </dd>
352 <dt><code>cluster_hb_interval</code> (int)</dt>
353 <dd>
354 <p>Interval in tenths of a second between cluster heartbeat/pings.</p>
355 </dd>
356 <dt><code>cluster_hb_timeout</code> (int)</dt>
357 <dd>
358 <p>Cluster heartbeat timeout in tenths of a second. A new master will be
359 elected when this interval has been passed without seeing a heartbeat
360 from the master.</p>
361 </dd>
362 <dt><code>cluster_master_min_adv</code> (int)</dt>
363 <dd>
364 <p>Determines the minimum number of up to date slaves required before
365 the master will drop routes (default: 1).</p>
366 </dd>
367 </dl>
368 <h3 id="config-startup-bgp">BGP</h3>
369 <p>BGP routing configuration is entered by the command: router bgp as
370 where as specifies the local AS number.</p>
371 <p>Subsequent lines prefixed with neighbour peer define the attributes
372 of BGP neighhbours. Valid commands are: neighbour peer remote-as as
373 neighbour peer timers keepalive hold</p>
374 <p>Where peer specifies the BGP neighbour as either a hostname or IP
375 address, as is the remote AS number and keepalive, hold are the timer
376 values in seconds.</p>
377 <h3 id="config-startup-acl">Access Lists</h3>
378 <p>Named access-lists are configured using one of the commands: ip
379 access-list standard name ip access-list extended name</p>
380 <p>Subsequent lines prefixed with <code>permit</code> or
381 <code>deny</code> define the body of the access-list. Standard
382 access-list syntax:</p>
383 <p>{<code>permit</code>|<code>deny</code>} {host|source
384 source-wildcard|<code>any</code>} [{host|destination
385 destination-wildcard|<code>any</code>}]</p>
386 <p>Extended access-lists:</p>
387 <p>{<code>permit</code>|<code>deny</code>} <code>ip</code> {host|source
388 source-wildcard|<code>any</code>} {host|destination
389 destination-wildcard|<code>any</code>} [<code>fragments</code>]</p>
390 <p>{<code>permit</code>|<code>deny</code>} <code>udp</code> {host|source
391 source-wildcard|<code>any</code>}
392 [{<code>eq</code>|<code>neq</code>|<code>gt</code>|<code>lt</code>}
393 port|<code>range</code> from to] {host|destination
394 destination-wildcard|<code>any</code>}
395 [{<code>eq</code>|<code>neq</code>|<code>gt</code>|<code>lt</code>}
396 port|<code>range</code> from to] [<code>fragments</code>]</p>
397 <p>{<code>permit</code>|<code>deny</code>} <code>tcp</code> {host|source
398 source-wildcard|<code>any</code>}
399 [{<code>eq</code>|<code>neq</code>|<code>gt</code>|<code>lt</code>}
400 port|<code>range</code> from to] {host|destination
401 destination-wildcard|<code>any</code>}
402 [{<code>eq</code>|<code>neq</code>|<code>gt</code>|<code>lt</code>}
403 port|<code>range</code> from to]
404 [{<code>established</code>|{<code>match-any</code>|<code>match-all</code>}
405 {<code>+</code>|<code>-</code>}{<code>fin</code>|<code>syn</code>|<code>rst</code>|<code>psh</code>|<code>ack</code>|<code>urg</code>}
406 ...|<code>fragments</code>]</p>
407 <h2 id="config-users"><code>users</code></h2>
408 <p>Usernames and passwords for the command-line interface are stored in
409 this file. The format is username:password where password may either by
410 plain text, an MD5 digest (prefixed by
411 <code>$1</code>salt<code>$</code>) or a DES password, distinguished from
412 plain text by the prefix <code>{crypt}</code>.</p>
413 <p>The username <code>enable</code> has a special meaning and is used to
414 set the enable password.</p>
415 <div class="important">
416 <p>If this file doesn't exist, then anyone who can get to port 23 will
417 be allowed access without a username or password.</p>
418 </div>
419 <h2 id="config-ip-pool"><code>ip_pool</code></h2>
420 <p>This file is used to configure the IP address pool which user
421 addresses are assigned from. This file should contain either an IP
422 address or a CIDR network per line. e.g.:</p>
423 <pre><code>192.168.1.1
424 192.168.1.2
425 192.168.1.3
426 192.168.4.0/24
427 172.16.0.0/16
428 10.0.0.0/8</code></pre>
429 <p>Keep in mind that <code>l2tpns</code> can only handle 65535
430 connections per process, so don't put more than 65535 IP addresses in
431 the configuration file. They will be wasted.</p>
432 <h2 id="config-build-garden"><code>build-garden</code></h2>
433 <p>The garden plugin on startup creates a NAT table called "garden" then
434 sources the <code>build-garden</code> script to populate that table. All
435 packets from gardened users will be sent through this table.
436 Example:</p>
437 <pre><code>iptables -t nat -A garden -p tcp -m tcp --dport 25 -j DNAT --to 192.168.1.1
438 iptables -t nat -A garden -p udp -m udp --dport 53 -j DNAT --to 192.168.1.1
439 iptables -t nat -A garden -p tcp -m tcp --dport 53 -j DNAT --to 192.168.1.1
440 iptables -t nat -A garden -p tcp -m tcp --dport 80 -j DNAT --to 192.168.1.1
441 iptables -t nat -A garden -p tcp -m tcp --dport 110 -j DNAT --to 192.168.1.1
442 iptables -t nat -A garden -p tcp -m tcp --dport 443 -j DNAT --to 192.168.1.1
443 iptables -t nat -A garden -p icmp -m icmp --icmp-type echo-request -j DNAT --to 192.168.1.1
444 iptables -t nat -A garden -p icmp -j ACCEPT
445 iptables -t nat -A garden -j DROP</code></pre>
446 <h1 id="operation">Operation</h1>
447 <p>A running l2tpns process can be controlled in a number of ways. The
448 primary method of control is by the Command-Line Interface (CLI).</p>
449 <p>You can also remotely send commands to modules via the
450 <code>nsctl</code> client provided.</p>
451 <p>There are also a number of signals that l2tpns understands and takes
452 action when it receives them.</p>
453 <h2 id="operation-cli">Command-Line Interface</h2>
454 <p>You can access the command line interface by telneting to port 23.
455 There is no IP address restriction, so it's a good idea to firewall this
456 port off from anyone who doesn't need access to it. See <a
457 href="#config-users"></a> for information on restricting access based on
458 a username and password.</p>
459 <p>The CLI gives you real-time control over almost everything in the
460 process. The interface is designed to look like a Cisco device, and
461 supports things like command history, line editing and context sensitive
462 help. This is provided by linking with the <a
463 href="http://sourceforge.net/projects/libcli">libcli</a> library. Some
464 general documentation of the interface is <a
465 href="http://sourceforge.net/docman/display_doc.php?docid=20501&amp;group_id=79019">here</a>.</p>
466 <p>After you have connected to the telnet port (and perhaps logged in),
467 you will be presented with a <code>hostname&gt;</code> prompt.</p>
468 <p>Enter <code>help</code> to get a list of possible commands, or press
469 <code>?</code> for context-specific help.</p>
470 <p>A brief overview of the more important commands follows:</p>
471 <p><code>show session [ID]</code></p>
472 <p>: Detailed information for a specific session is presented if you
473 specify a session ID argument.</p>
474 <pre><code>If no ID is given, a summary of all connected sessions is produced.
475 Note that this summary list can be around 185 columns wide, so you
476 should probably use a wide terminal.
477
478 The columns listed in the summary are:
479
480 -------------- -------------------------------------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------------------------------------
481 `SID` Session ID
482 `TID` Tunnel ID See also the [show tunnel](#operation-cli-show-tunnel) CLI command.
483 `Username` The username given in the PPP authentication. If this is \*, then LCP authentication has not completed.
484 `IP` The IP address given to the session. If this is 0.0.0.0, IPCP negotiation has not completed
485 `I` Intercept Y or N: indicates whether the session is being snooped. See also the [snoop](#operation-cli-snoop) CLI command.
486 `T` Throttled Y or N: indicates whether the session is currently throttled. See also the [throttle](#operation-cli-throttle) CLI command.
487 `G` Walled Garden Y or N: indicates whether the user is trapped in the walled garden. This field is present even if the garden module is not loaded.
488 `6` IPv6 Y or N: indicates whether the session has IPv6 active (IPV6CP open)
489 `opened` The number of seconds since the session started
490 `downloaded` Number of bytes downloaded by the user
491 `uploaded` Number of bytes uploaded by the user
492 `idle` The number of seconds since traffic was detected on the session
493 `LAC` The IP address of the LAC the session is connected to.
494 `CLI` The Calling-Line-Identification field provided during the session setup. This field is generated by the LAC.
495 -------------- -------------------------------------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------------------------------------</code></pre>
496 <p><code>show users</code>; <code>show user username</code></p>
497 <p>: With no arguments, display a list of currently connected users. If
498 an argument is given, the session details for the given username are
499 displayed.</p>
500 <dl>
501 <dt><code>show tunnel [ID]</code></dt>
502 <dd>
503 <p>Produce a summary list of all open tunnels, or detail on a specific
504 tunnel ID.</p>
505 <p>The columns listed in the summary are:</p>
506 <table>
507 <tbody>
508 <tr class="odd">
509 <td style="text-align: left;">TID</td>
510 <td style="text-align: left;">Tunnel ID</td>
511 </tr>
512 <tr class="even">
513 <td style="text-align: left;">Hostname</td>
514 <td style="text-align: left;">The hostname for the tunnel as provided by
515 the LAC. This has no relation to DNS, it is just a text field.</td>
516 </tr>
517 <tr class="odd">
518 <td style="text-align: left;">IP</td>
519 <td style="text-align: left;">The IP address of the LAC</td>
520 </tr>
521 <tr class="even">
522 <td style="text-align: left;">State</td>
523 <td style="text-align: left;">Tunnel state: Free, Open, Dieing,
524 Opening</td>
525 </tr>
526 <tr class="odd">
527 <td style="text-align: left;">Sessions</td>
528 <td style="text-align: left;">The number of open sessions on the
529 tunnel</td>
530 </tr>
531 </tbody>
532 </table>
533 </dd>
534 <dt><code>show pool</code></dt>
535 <dd>
536 <p>Displays the current IP address pool allocation. This will only
537 display addresses that are in use, or are reserved for re-allocation to
538 a disconnected user.</p>
539 <p>If an address is not currently in use, but has been used, then in the
540 User column the username will be shown in square brackets, followed by
541 the time since the address was used:</p>
542 <pre><code>IP Address Used Session User
543 192.168.100.6 N [joe.user] 1548s</code></pre>
544 </dd>
545 <dt><code>show radius</code></dt>
546 <dd>
547 <p>Show a summary of the in-use RADIUS sessions. This list should not be
548 very long, as RADIUS sessions should be cleaned up as soon as they are
549 used. The columns listed are:</p>
550 <table>
551 <tbody>
552 <tr class="odd">
553 <td style="text-align: left;">Radius</td>
554 <td style="text-align: left;">The ID of the RADIUS request. This is sent
555 in the packet to the RADIUS server for identification</td>
556 </tr>
557 <tr class="even">
558 <td style="text-align: left;">State</td>
559 <td style="text-align: left;">The state of the request: WAIT, CHAP,
560 AUTH, IPCP, START, STOP or NULL</td>
561 </tr>
562 <tr class="odd">
563 <td style="text-align: left;">Session</td>
564 <td style="text-align: left;">The session ID that this RADIUS request is
565 associated with</td>
566 </tr>
567 <tr class="even">
568 <td style="text-align: left;">Retry</td>
569 <td style="text-align: left;">If a response does not appear to the
570 request, it will retry at this time. This is a Unix timestamp</td>
571 </tr>
572 <tr class="odd">
573 <td style="text-align: left;">Try</td>
574 <td style="text-align: left;">Retry count. The RADIUS request is
575 discarded after 3 retries</td>
576 </tr>
577 </tbody>
578 </table>
579 </dd>
580 <dt><code>show running-config</code></dt>
581 <dd>
582 <p>This will list the current running configuration. This is in a format
583 that can either be pasted into the configuration file, or run directly
584 at the command line.</p>
585 </dd>
586 <dt><code>show counters</code></dt>
587 <dd>
588 <p>Internally, counters are kept of key values, such as bytes and
589 packets transferred, as well as function call counters. This function
590 displays all these counters, and is probably only useful for
591 debugging.</p>
592 <p>You can reset these counters by running
593 <code>clear counters</code>.</p>
594 </dd>
595 <dt><code>show cluster</code></dt>
596 <dd>
597 <p>Show cluster status. Shows the cluster state for this server
598 (Master/Slave), information about known peers and (for slaves) the
599 master IP address, last packet seen and up-to-date status. See <a
600 href="#clustering">Clustering</a> for more information.</p>
601 </dd>
602 <dt><code>write memory</code></dt>
603 <dd>
604 <p>This will write the current running configuration to the config file
605 <code>startup-config</code>, which will be run on a restart.</p>
606 </dd>
607 </dl>
608 <p><code>snoop user IP port</code></p>
609 <p>: You must specify a username, IP address and port. All packets for
610 the current session for that username will be forwarded to the given
611 host/port. Specify <code>no snoop username</code> to disable
612 interception for the session.</p>
613 <pre><code>If you want interception to be permanent, you will have to modify
614 the RADIUS response for the user. See [Interception](#interception).</code></pre>
615 <p><code>throttle user [in|out] rate</code></p>
616 <p>: You must specify a username, which will be throttled for the
617 current session to rate Kbps. Prefix rate with <code>in</code> or
618 <code>out</code> to set different upstream and downstream rates.</p>
619 <pre><code>Specify `no throttle
620 username` to disable throttling for the current session.
621
622 If you want throttling to be permanent, you will have to modify the
623 RADIUS response for the user. See [Throttling](#throttling).</code></pre>
624 <dl>
625 <dt><code>drop session</code></dt>
626 <dd>
627 <p>This will cleanly disconnect the session specified by session ID.</p>
628 </dd>
629 <dt><code>drop tunnel</code></dt>
630 <dd>
631 <p>This will cleanly disconnect the tunnel specified by tunnel ID, as
632 well as all sessions on that tunnel.</p>
633 </dd>
634 <dt><code>uptime</code></dt>
635 <dd>
636 <p>This will show how long the <code>l2tpns</code> process has been
637 running, and the current bandwidth utilization:</p>
638 <pre><code>17:10:35 up 8 days, 2212 users, load average: 0.21, 0.17, 0.16
639 Bandwidth: UDP-ETH:6/6 ETH-UDP:13/13 TOTAL:37.6 IN:3033 OUT:2569</code></pre>
640 <p>The bandwidth line contains 4 sets of values:</p>
641 <table>
642 <tbody>
643 <tr class="odd">
644 <td style="text-align: left;">UDP-ETH</td>
645 <td style="text-align: left;">The current bandwidth going from the LAC
646 to the ethernet (user uploads), in mbits/sec.</td>
647 </tr>
648 <tr class="even">
649 <td style="text-align: left;">ETH-UDP</td>
650 <td style="text-align: left;">The current bandwidth going from ethernet
651 to the LAC (user downloads).</td>
652 </tr>
653 <tr class="odd">
654 <td style="text-align: left;">TOTAL</td>
655 <td style="text-align: left;">The total aggregate bandwidth in
656 mbits/s.</td>
657 </tr>
658 <tr class="even">
659 <td style="text-align: left;">IN and OUT</td>
660 <td style="text-align: left;">Packets/per-second going between UDP-ETH
661 and ETH-UDP.</td>
662 </tr>
663 </tbody>
664 </table>
665 <p>These counters are updated every second.</p>
666 </dd>
667 <dt><code>configure terminal</code></dt>
668 <dd>
669 <p>Enter configuration mode. Use <code>exit</code> or <code>^Z</code> to
670 exit this mode.</p>
671 <p>The following commands are valid in this mode:</p>
672 <p><code>load plugin name</code></p>
673 <p>: Load a plugin. You must specify the plugin name, and it will search
674 in <code>/usr/lib/l2tpns</code> for <code>name.so</code>. You can unload
675 a loaded plugin with
676 <code>remove plugin name</code>.</p>
677 <dl>
678 <dt><code>set</code> ...</dt>
679 <dd>
680 <p>Set a configuration variable. You must specify the variable name, and
681 the value. If the value contains any spaces, you should quote the value
682 with double (") or single (') quotes.</p>
683 <p>You can set any configuration value in this way, although some may
684 require a restart to take effect. See <a
685 href="#config-startup"></a>.</p>
686 </dd>
687 <dt><code>router bgp</code> ...</dt>
688 <dd>
689 <p>Configure BGP. See <a href="#config-startup-bgp">BGP</a>.</p>
690 </dd>
691 <dt><code>ip access-list</code> ...</dt>
692 <dd>
693 <p>Configure a named access list. See <a
694 href="#config-startup-acl">Access Lists</a>.</p>
695 </dd>
696 </dl>
697 </dd>
698 </dl>
699 <h2 id="operation-nsctl">nsctl</h2>
700 <p><code>nsctl</code> sends messages to a running <code>l2tpns</code>
701 instance to be control plugins.</p>
702 <p>Arguments are <code>command</code> and optional args. See
703 <code>nsctl(8)</code>.</p>
704 <p>Built-in command are <code>load_plugin</code>,
705 <code>unload_plugin</code> and <code>help</code>. Any other commands are
706 passed to plugins for processing by the <code>plugin_control</code>
707 function.</p>
708 <h2 id="operation-signals">Signals</h2>
709 <p>While the process is running, you can send it a few different
710 signals, using the <code>kill</code> command.</p>
711 <pre><code>killall -HUP l2tpns</code></pre>
712 <p>The signals understood are:</p>
713 <dl>
714 <dt>SIGHUP</dt>
715 <dd>
716 <p>Reload the config from disk and re-open log file.</p>
717 </dd>
718 <dt>SIGTERM; SIGINT</dt>
719 <dd>
720 <p>Stop process. Tunnels and sessions are not terminated. This signal
721 should be used to stop <code>l2tpns</code> on a cluster node where there
722 are other machines to continue handling traffic. See <a
723 href="#clustering">Clustering</a></p>
724 </dd>
725 <dt>SIGQUIT</dt>
726 <dd>
727 <p>Shut down tunnels and sessions, exit process when complete.</p>
728 </dd>
729 </dl>
730 <h1 id="throttling">Throttling</h1>
731 <p><code>l2tpns</code> contains support for slowing down user sessions
732 to whatever speed you desire. The global setting
733 <code>throttle_speed</code> defines the default throttle rate.</p>
734 <p>To throttle a sesion permanently, add a <code>Cisco-AVPair</code>
735 RADIUS attribute. The <code>autothrotle</code> module interprets the
736 following attributes:</p>
737 <table>
738 <tbody>
739 <tr class="odd">
740 <td style="text-align: left;"><code>throttle=yes</code></td>
741 <td style="text-align: left;">Throttle upstream/downstream traffic to
742 the configured <code>throttle_speed</code>.</td>
743 </tr>
744 <tr class="even">
745 <td style="text-align: left;"><code>throttle=rate</code></td>
746 <td style="text-align: left;">Throttle upstream/downstream traffic to
747 the specified rate Kbps.</td>
748 </tr>
749 <tr class="odd">
750 <td style="text-align: left;">`lcp:interface-config#1=service-policy
751 input</td>
752 <td style="text-align: left;">Alternate (Cisco) format: throttle
753 upstream/downstream to specified rate Kbps.</td>
754 </tr>
755 <tr class="even">
756 <td style="text-align: left;">rate`</td>
757 <td style="text-align: left;"></td>
758 </tr>
759 <tr class="odd">
760 <td style="text-align: left;">`lcp:interface-config#2=service-policy
761 output</td>
762 <td style="text-align: left;"></td>
763 </tr>
764 <tr class="even">
765 <td style="text-align: left;">rate`</td>
766 <td style="text-align: left;"></td>
767 </tr>
768 </tbody>
769 </table>
770 <p>You can also enable and disable throttling an active session using
771 the <a href="#operation-cli-throttle">throttle</a> CLI command.</p>
772 <h1 id="interception">Interception</h1>
773 <p>You may have to deal with legal requirements to be able to intercept
774 a user's traffic at any time. <code>l2tpns</code> allows you to begin
775 and end interception on the fly, as well as at authentication time.</p>
776 <p>When a user is being intercepted, a copy of every packet they send
777 and receive will be sent wrapped in a UDP packet to a specified
778 host.</p>
779 <p>The UDP packet contains just the raw IP frame, with no extra headers.
780 The script <code>scripts/l2tpns-capture</code> may be used as the
781 end-point for such intercepts, writing the data in PCAP format (suitable
782 for inspection with <code>tcpdump</code>).</p>
783 <p>To enable or disable interception of a connected user, use the <a
784 href="#operation-cli-snoop">snoop</a> and <code>no snoop</code>
785 CLI commands. These will enable interception immediately.</p>
786 <p>If you wish the user to be intercepted whenever they reconnect, you
787 will need to modify the RADIUS response to include the Vendor-Specific
788 value <code>Cisco-AVPair="intercept=ip:port"</code>. For this feature to
789 be enabled, you need to have the <code>autosnoop</code> module
790 loaded.</p>
791 <h1 id="plugins">Plugins</h1>
792 <p>So as to make <code>l2tpns</code> as flexible as possible, a plugin
793 API is include which you can use to hook into certain events.</p>
794 <p>There are a some standard modules included which may be used as
795 examples: <code>autosnoop</code>, <code>autothrottle</code>,
796 <code>garden</code>, <code>sessionctl</code>, <code>setrxspeed</code>,
797 <code>snoopctl</code>, <code>stripdomain</code> and
798 <code>throttlectl</code>.</p>
799 <p>When an event occurs that has a hook, <code>l2tpns</code> looks for a
800 predefined function name in every loaded module, and runs them in the
801 order the modules were loaded.</p>
802 <p>The function should return <code>PLUGIN_RET_OK</code> if it is all
803 OK. If it returns <code>PLUGIN_RET_STOP</code>, then it is assumed to
804 have worked, but that no further modules should be run for this
805 event.</p>
806 <p>A return of <code>PLUGIN_RET_ERROR</code> means that this module
807 failed, and no further processing should be done for this event.</p>
808 <div class="note">
809 <p>Use this with care.</p>
810 </div>
811 <p>Most event functions take a specific structure named
812 <code>param_event</code>, which varies in content with each event. The
813 function name for each event will be <code>plugin_event</code>, so for
814 the event timer, the function declaration should look like:</p>
815 <pre><code>int plugin_timer(struct param_timer *data);</code></pre>
816 <p>A list of the available events follows, with a list of all the fields
817 in the supplied structure:</p>
818 <table style="width:97%;">
819 <colgroup>
820 <col style="width: 29%" />
821 <col style="width: 29%" />
822 <col style="width: 29%" />
823 <col style="width: 8%" />
824 </colgroup>
825 <tbody>
826 <tr class="odd">
827 <td rowspan="2"><h1 id="event">Event</h1>
828 <p><code>plugin_init</code></p></td>
829 <td rowspan="2"><h1 id="description">Description</h1>
830 <p>Called when the plugin is loaded. A pointer to a struct containing
831 function pointers is passed as the only argument, allowing the plugin to
832 call back into the main code.</p>
833 <p>Prior to loading the plugin, <code>l2tpns</code> checks the API
834 version the plugin was compiled against. All plugins should contain:</p>
835 <pre><code> int</code></pre>
836 <p>plugin_api_version = PLUGIN_API_VERSION;</p></td>
837 <td rowspan="2"><h1 id="arguments">Arguments</h1>
838 <p><code>s truct pluginfuncs *</code></p></td>
839 <td></td>
840 </tr>
841 <tr class="even">
842 <td></td>
843 </tr>
844 <tr class="odd">
845 <td>See <code>pluginfuncs</code> structure in <code>plugin.h</code> for
846 available functions.</td>
847 <td></td>
848 <td></td>
849 <td></td>
850 </tr>
851 <tr class="even">
852 <td><code>plugin_done</code></td>
853 <td>Called when the plugin is unloaded or <code>l2tpns</code> is
854 shutdown.</td>
855 <td><code>void</code></td>
856 <td></td>
857 </tr>
858 <tr class="odd">
859 <td>No arguments.</td>
860 <td></td>
861 <td></td>
862 <td></td>
863 </tr>
864 <tr class="even">
865 <td><code>plugin_pre_auth</code></td>
866 <td>Called after a RADIUS response has been received, but before it has
867 been processed by the code. This will allow you to modify the response
868 in some way.</td>
869 <td><code>struct plug in param_pre_auth *</code></td>
870 <td></td>
871 </tr>
872 <tr class="odd">
873 <td><code>tunnelt *t</code></td>
874 <td>Tunnel.</td>
875 <td></td>
876 <td></td>
877 </tr>
878 <tr class="even">
879 <td><code>sessiont *s</code></td>
880 <td>Session.</td>
881 <td></td>
882 <td></td>
883 </tr>
884 <tr class="odd">
885 <td><code>char *username</code></td>
886 <td>User name.</td>
887 <td></td>
888 <td></td>
889 </tr>
890 <tr class="even">
891 <td><code>char *password</code></td>
892 <td>Password.</td>
893 <td></td>
894 <td></td>
895 </tr>
896 <tr class="odd">
897 <td><code>int protocol</code></td>
898 <td>Authentication protocol: <code>0xC023</code> for PAP,
899 <code>0xC223</code> for CHAP.</td>
900 <td></td>
901 <td></td>
902 </tr>
903 <tr class="even">
904 <td><code>int continue_auth</code></td>
905 <td>Set to 0 to stop processing authentication modules.</td>
906 <td></td>
907 <td></td>
908 </tr>
909 <tr class="odd">
910 <td><code>plugin_post_auth</code></td>
911 <td>Called after a RADIUS response has been received, and the basic
912 checks have been performed. This is what the <code>garden</code> module
913 uses to force authentication to be accepted.</td>
914 <td><code>struct plugi n param_post_auth *</code></td>
915 <td></td>
916 </tr>
917 <tr class="even">
918 <td><code>tunnelt *t</code></td>
919 <td>Tunnel.</td>
920 <td></td>
921 <td></td>
922 </tr>
923 <tr class="odd">
924 <td><code>sessiont *s</code></td>
925 <td>Session.</td>
926 <td></td>
927 <td></td>
928 </tr>
929 <tr class="even">
930 <td><code>char *username</code></td>
931 <td>User name.</td>
932 <td></td>
933 <td></td>
934 </tr>
935 <tr class="odd">
936 <td><code>short auth_allowed</code></td>
937 <td>Initially true or false depending on whether authentication has been
938 allowed so far. You can set this to 1 or 0 to force authentication to be
939 accepted or rejected.</td>
940 <td></td>
941 <td></td>
942 </tr>
943 <tr class="even">
944 <td><code>int protocol</code></td>
945 <td>Authentication protocol: <code>0xC023</code> for PAP,
946 <code>0xC223</code> for CHAP.</td>
947 <td></td>
948 <td></td>
949 </tr>
950 <tr class="odd">
951 <td><code>plugin_timer</code></td>
952 <td>Run once per second.</td>
953 <td><code>struct p lugin param_timer *</code></td>
954 <td></td>
955 </tr>
956 <tr class="even">
957 <td><code>time_t time_now</code></td>
958 <td>The current unix timestamp.</td>
959 <td></td>
960 <td></td>
961 </tr>
962 <tr class="odd">
963 <td><code>plugin_new_session</code></td>
964 <td>Called after a session is fully set up. The session is now ready to
965 handle traffic.</td>
966 <td><code>struct plugin param_new_session *</code></td>
967 <td></td>
968 </tr>
969 <tr class="even">
970 <td><code>tunnelt *t</code></td>
971 <td>Tunnel.</td>
972 <td></td>
973 <td></td>
974 </tr>
975 <tr class="odd">
976 <td><code>sessiont *s</code></td>
977 <td>Session.</td>
978 <td></td>
979 <td></td>
980 </tr>
981 <tr class="even">
982 <td><code>plugin_kill_session</code></td>
983 <td>Called when a session is about to be shut down. This may be called
984 multiple times for the same session.</td>
985 <td><code>struct plugin p aram_kill_session *</code></td>
986 <td></td>
987 </tr>
988 <tr class="odd">
989 <td><code>tunnelt *t</code></td>
990 <td>Tunnel.</td>
991 <td></td>
992 <td></td>
993 </tr>
994 <tr class="even">
995 <td><code>sessiont *s</code></td>
996 <td>Session.</td>
997 <td></td>
998 <td></td>
999 </tr>
1000 <tr class="odd">
1001 <td><code>plugin_control</code></td>
1002 <td><p>Called in whenever a <code>nsctl</code> packet is received. This
1003 should handle the packet and form a response if required.</p>
1004 <p>Plugin-specific help strings may be included in the output of
1005 <code>nsctl help</code> by defining a <code>NULL</code> terminated list
1006 of strings as follows:</p>
1007 <pre><code>char</code></pre>
1008 <p>*plugin_control_hel p[] = { …, NULL };</p></td>
1009 <td><code>struct plu gin param_control *</code></td>
1010 <td></td>
1011 </tr>
1012 <tr class="even">
1013 <td><code>int iam_master</code></td>
1014 <td>If true, this node is the cluster master.</td>
1015 <td></td>
1016 <td></td>
1017 </tr>
1018 <tr class="odd">
1019 <td><code>int argc</code></td>
1020 <td><code>nsctl</code> arguments.</td>
1021 <td></td>
1022 <td></td>
1023 </tr>
1024 <tr class="even">
1025 <td><code>char **argc</code></td>
1026 <td></td>
1027 <td></td>
1028 <td></td>
1029 </tr>
1030 <tr class="odd">
1031 <td><code>int response</code></td>
1032 <td>Response from control message (if handled): should be either
1033 <code>NSCTL_RES_OK</code> or <code>NSCTL_RES_ERR</code>.</td>
1034 <td></td>
1035 <td></td>
1036 </tr>
1037 <tr class="even">
1038 <td><code>char *additional</code></td>
1039 <td>Additional information, output by <code>nsctl</code> on receiving
1040 the response.</td>
1041 <td></td>
1042 <td></td>
1043 </tr>
1044 <tr class="odd">
1045 <td colspan="4"><code>plugin_radius_response</code> | Called whenever a
1046 |
1047 <code>struct plugin para | | RADIUS response | m_radius_response *</code>
1048 | | includes a | | | <code>Cisco-AVPair</code> | | | value. The value is
1049 | | | split into | | | key<code>=</code>value pairs. | | | Will be
1050 called once | | | for each pair in the | | | response. | |</td>
1051 </tr>
1052 <tr class="even">
1053 <td><code>tunnelt *t</code></td>
1054 <td>Tunnel.</td>
1055 <td></td>
1056 <td></td>
1057 </tr>
1058 <tr class="odd">
1059 <td><code>sessiont *s</code></td>
1060 <td>Session.</td>
1061 <td></td>
1062 <td></td>
1063 </tr>
1064 <tr class="even">
1065 <td><code>char *key</code></td>
1066 <td>Key and value.</td>
1067 <td></td>
1068 <td></td>
1069 </tr>
1070 <tr class="odd">
1071 <td><code>char *value</code></td>
1072 <td></td>
1073 <td></td>
1074 <td></td>
1075 </tr>
1076 <tr class="even">
1077 <td colspan="4"><code>plugin_radius_reset</code> | Called whenever a |
1078 <code>struct p | | RADIUS CoA request | aram_radius_reset *</code>
1079 | | is received to reset | | | any options to | | | default values | | |
1080 before the new | | | values are applied. | |</td>
1081 </tr>
1082 <tr class="odd">
1083 <td><code>tunnelt *t</code></td>
1084 <td>Tunnel.</td>
1085 <td></td>
1086 <td></td>
1087 </tr>
1088 <tr class="even">
1089 <td><code>sessiont *s</code></td>
1090 <td>Session.</td>
1091 <td></td>
1092 <td></td>
1093 </tr>
1094 <tr class="odd">
1095 <td colspan="4"><code>plugin_radius_account</code> | Called when |
1096 <code>struct par | | preparing a RADIUS | am_radius_account *</code>
1097 | | accounting record to | | | allow additional | | | data to be added
1098 to | | | the packet. | |</td>
1099 </tr>
1100 <tr class="even">
1101 <td><code>tunnelt *t</code></td>
1102 <td>Tunnel.</td>
1103 <td></td>
1104 <td></td>
1105 </tr>
1106 <tr class="odd">
1107 <td><code>sessiont *s</code></td>
1108 <td>Session.</td>
1109 <td></td>
1110 <td></td>
1111 </tr>
1112 <tr class="even">
1113 <td><code>uint8_t **packet</code></td>
1114 <td>Pointer to the end of the currently assembled packet buffer. The
1115 value should be incremented by the length of any data added.</td>
1116 <td></td>
1117 <td></td>
1118 </tr>
1119 <tr class="odd">
1120 <td colspan="4"><code>plugin_become_master</code> | Called when a node |
1121 <code>void</code> | | elects itself | | | cluster master. | |</td>
1122 </tr>
1123 <tr class="even">
1124 <td>No arguments.</td>
1125 <td></td>
1126 <td></td>
1127 <td></td>
1128 </tr>
1129 <tr class="odd">
1130 <td colspan="4"><code>plugin_new_session_master</code> | Called once for
1131 each | <code>sessiont *</code> | open session on | | | becoming cluster
1132 | | | master. | |</td>
1133 </tr>
1134 <tr class="even">
1135 <td>Session.</td>
1136 <td></td>
1137 <td></td>
1138 <td></td>
1139 </tr>
1140 </tbody>
1141 </table>
1142 <h1 id="walled-garden">Walled Garden</h1>
1143 <p>A "Walled Garden" is implemented so that you can provide perhaps
1144 limited service to sessions that incorrectly authenticate.</p>
1145 <p>Whenever a session provides incorrect authentication, and the RADIUS
1146 server responds with Auth-Reject, the walled garden module (if loaded)
1147 will force authentication to succeed, but set the
1148 <code>walled_garden</code> flag in the session structure, and adds an
1149 <code>iptables</code> rule to the <code>garden_users</code> chain to
1150 cause all packets for the session to traverse the <code>garden</code>
1151 chain.</p>
1152 <p>This doesn't <em>just work</em>. To set this all up, you will to
1153 setup the <code>garden</code> nat table with the <a
1154 href="#config-build-garden">build-garden</a> script with rules to limit
1155 user's traffic.</p>
1156 <p>For example, to force all traffic except DNS to be forwarded to
1157 192.168.1.1, add these entries to your <code>build-garden</code>
1158 script:</p>
1159 <pre><code>iptables -t nat -A garden -p tcp --dport ! 53 -j DNAT --to 192.168.1.1
1160 iptables -t nat -A garden -p udp --dport ! 53 -j DNAT --to 192.168.1.1</code></pre>
1161 <p><code>l2tpns</code> will add entries to the <code>garden_users</code>
1162 chain as appropriate.</p>
1163 <p>You can check the amount of traffic being captured using the
1164 following command:</p>
1165 <pre><code>iptables -t nat -L garden -nvx</code></pre>
1166 <h1 id="filtering">Filtering</h1>
1167 <p>Sessions may be filtered by specifying <code>Filter-Id</code>
1168 attributes in the RADIUS reply. filter.<code>in</code> specifies that
1169 the named access-list filter should be applied to traffic from the
1170 customer, filter.<code>out</code> specifies a list for traffic to the
1171 customer.</p>
1172 <h1 id="clustering">Clustering</h1>
1173 <p>An <code>l2tpns</code> cluster consists of one* or more servers
1174 configured with the same configuration, notably the multicast
1175 <code>cluster_address</code> and the <code>cluster_port</code></p>
1176 <p>*A stand-alone server is simply a degraded cluster.</p>
1177 <p>Initially servers come up as cluster slaves, and periodically (every
1178 <code>cluster_hb_interval</code>/10 seconds) send out ping packets
1179 containing the start time of the process to the multicast
1180 <code>cluster_address</code> on <code>cluster_port</code>.</p>
1181 <p>A cluster master sends heartbeat rather than ping packets, which
1182 contain those session and tunnel changes since the last heartbeat.</p>
1183 <p>When a slave has not seen a heartbeat within
1184 <code>cluster_hb_timeout</code>/10 seconds it "elects" a new master by
1185 examining the list of peers it has seen pings from and determines which
1186 of these and itself is the "best" candidate to be master. "Best" in this
1187 context means the server with the highest uptime (the highest IP address
1188 is used as a tie-breaker in the case of equal uptimes).</p>
1189 <p>After discovering a master, and determining that it is up-to-date
1190 (has seen an update for all in-use sessions and tunnels from heartbeat
1191 packets) will raise a route (see <a href="#routing">Routing</a>) for the
1192 <code>bind_address</code> and for all addresses/networks in
1193 <code>ip_pool</code>.</p>
1194 <p>Any packets received by the slave which would alter the session
1195 state, as well as packets for throttled or gardened sessions are
1196 forwarded to the master for handling. In addition, byte counters for
1197 session traffic are periodically forwarded.</p>
1198 <p>The heartbeat exchange use timing, so the master and slaves need to
1199 have their time synchronized, e.g. with ntp.</p>
1200 <p>Since all the masters and slaves should be able to receive all
1201 packets, <code>bind_address</code> needs to be used to make all of them
1202 agree on the IP to be used. If <code>l2tpns</code> is to behave as LAC,
1203 <code>bind_address_remotelns</code> also needs to be set.</p>
1204 <p>To balance the load, <code>bind_multi_address</code> can be used to
1205 make them support several IPs, and one can then prioritize one or the
1206 other on each server, and make e.g. radius pick up one of them randomly
1207 for <code>Tunnel-Server-Endpoint</code>. There is currently no support
1208 for such load balancing for LAC support.</p>
1209 <p>*Configurable with <code>cluster_master_min_adv</code></p>
1210 <p>Multiple clusters can be run on the same network by just using
1211 different multicast <code>cluster_address</code>. However, for a given
1212 host to be part of multiple clusters without mixing the clusters,
1213 <code>cluster_port</code> must be different for each cluster.</p>
1214 <h1 id="routing">Routing</h1>
1215 <p>If you are running a single instance, you may simply statically route
1216 the IP pools to the <code>bind_address</code> (<code>l2tpns</code> will
1217 send a gratuitous arp).</p>
1218 <p>For a cluster, configure the members as BGP neighbours on your router
1219 and configure multi-path load-balancing. Cisco uses
1220 <code>maximum-paths ibgp</code> for IBGP. If this is not supported by
1221 your IOS revision, you can use <code>maximum-paths</code> (which works
1222 for EBGP) and set <code>as_number</code> to a private value such as
1223 64512.</p>