summaryrefslogtreecommitdiff
path: root/src/pluto/foodgroups.c
blob: c92bdb3d481afa8c084062d2390d7d4010c95366 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
/* Implement policy groups-style control files (aka "foodgroups")
 * Copyright (C) 2002  D. Hugh Redelmeier.
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License as published by the
 * Free Software Foundation; either version 2 of the License, or (at your
 * option) any later version.  See <http://www.fsf.org/copyleft/gpl.txt>.
 *
 * This program is distributed in the hope that it will be useful, but
 * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
 * or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
 * for more details.
 *
 * RCSID $Id: foodgroups.c,v 1.2 2004/04/01 18:28:32 as Exp $
 */

#include <string.h>
#include <stdio.h>
#include <stddef.h>
#include <stdlib.h>
#include <sys/queue.h>

#include <freeswan.h>

#include "constants.h"
#include "defs.h"
#include "connections.h"
#include "foodgroups.h"
#include "kernel.h"
#include "lex.h"
#include "log.h"
#include "whack.h"


/* Food group config files are found in directory fg_path */

#ifndef POLICYGROUPSDIR
#define POLICYGROUPSDIR IPSEC_CONFDIR "/ipsec.d/policies"
#endif

const char *policygroups_dir = POLICYGROUPSDIR;

static char *fg_path = NULL;
static size_t fg_path_space = 0;


/* Groups is a list of connections that are policy groups.
 * The list is updated as group connections are added and deleted.
 */

struct fg_groups {
    struct fg_groups *next;
    struct connection *connection;
};

static struct fg_groups *groups = NULL;


/* Targets is a list of pairs: subnet and its policy group.
 * This list is bulk-updated on whack --listen and
 * incrementally updated when group connections are deleted.
 *
 * It is ordered by source subnet, and if those are equal, then target subnet.
 * A subnet is compared by comparing the network, and if those are equal,
 * comparing the mask.
 */

struct fg_targets {
    struct fg_targets *next;
    struct fg_groups *group;
    ip_subnet subnet;
    char *name;	/* name of instance of group conn */
};

static struct fg_targets *targets = NULL;

struct fg_targets *new_targets;

/* ipcmp compares the two ip_address values a and b.
 * It returns -1, 0, or +1 if a is, respectively,
 * less than, equal to, or greater than b.
 */
static int
ipcmp(ip_address *a, ip_address *b)
{
    if (addrtypeof(a) != addrtypeof(b))
    {
	return addrtypeof(a) < addrtypeof(b)? -1 : 1;
    }
    else if (sameaddr(a, b))
    {
	return 0;
    }
    else
    {
	const struct sockaddr *sa = sockaddrof(a)
	    , *sb = sockaddrof(b);

	passert(addrtypeof(a) == AF_INET);	/* not yet implemented IPv6 version :-( */
	return (ntohl(((const struct sockaddr_in *)sa)->sin_addr.s_addr)
	    < ntohl(((const struct sockaddr_in *)sb)->sin_addr.s_addr))
	    ? -1 : 1;
    }
}

/* subnetcmp compares the two ip_subnet values a and b.
 * It returns -1, 0, or +1 if a is, respectively,
 * less than, equal to, or greater than b.
 */
static int
subnetcmp(const ip_subnet *a, const ip_subnet *b)
{
    ip_address neta, maska, netb, maskb;
    int r;

    networkof(a, &neta);
    maskof(a, &maska);
    networkof(b, &netb);
    maskof(b, &maskb);
    r = ipcmp(&neta, &netb);
    if (r == 0)
	r = ipcmp(&maska, &maskb);
    return r;
}

static void
read_foodgroup(struct fg_groups *g)
{
    const char *fgn = g->connection->name;
    const ip_subnet *lsn = &g->connection->spd.this.client;
    size_t plen = strlen(policygroups_dir) + 1 + strlen(fgn) + 1;
    struct file_lex_position flp_space;

    if (plen > fg_path_space)
    {
	pfreeany(fg_path);
	fg_path_space = plen + 10;
	fg_path = alloc_bytes(fg_path_space, "policy group path");
    }
    snprintf(fg_path, fg_path_space, "%s/%s", policygroups_dir, fgn);
    if (!lexopen(&flp_space, fg_path, TRUE))
    {
	DBG(DBG_CONTROL, DBG_log("no group file \"%s\"", fg_path));
    }
    else
    {
	plog("loading group \"%s\"", fg_path);
	for (;;)
	{
	    switch (flp->bdry)
	    {
	    case B_none:
		{
		    /* !!! this test is not sufficient for distinguishing address families.
		     * We need a notation to specify that a FQDN is to be resolved to IPv6.
		     */
		    const struct af_info *afi = strchr(tok, ':') == NULL
			? &af_inet4_info: &af_inet6_info;
		    ip_subnet sn;
		    err_t ugh;

		    if (strchr(tok, '/') == NULL)
		    {
			/* no /, so treat as /32 or V6 equivalent */
			ip_address t;

			ugh = ttoaddr(tok, 0, afi->af, &t);
			if (ugh == NULL)
			    ugh = addrtosubnet(&t, &sn);
		    }
		    else
		    {
			ugh = ttosubnet(tok, 0, afi->af, &sn);
		    }

		    if (ugh != NULL)
		    {
			loglog(RC_LOG_SERIOUS, "\"%s\" line %d: %s \"%s\""
			    , flp->filename, flp->lino, ugh, tok);
		    }
		    else if (afi->af != AF_INET)
		    {
			loglog(RC_LOG_SERIOUS
			    , "\"%s\" line %d: unsupported Address Family \"%s\""
			    , flp->filename, flp->lino, tok);
		    }
		    else
		    {
			/* Find where new entry ought to go in new_targets. */
			struct fg_targets **pp;
			int r;

			for (pp = &new_targets; ; pp = &(*pp)->next)
			{
			    if (*pp == NULL)
			    {
				r = -1;	/* end of list is infinite */
				break;
			    }
			    r = subnetcmp(lsn, &(*pp)->group->connection->spd.this.client);
			    if (r == 0)
				r = subnetcmp(&sn, &(*pp)->subnet);
			    if (r <= 0)
				break;
			}

			if (r == 0)
			{
			    char source[SUBNETTOT_BUF];

			    subnettot(lsn, 0, source, sizeof(source));
			    loglog(RC_LOG_SERIOUS
				, "\"%s\" line %d: subnet \"%s\", source %s, already \"%s\""
				, flp->filename
				, flp->lino
				, tok
				, source
				, (*pp)->group->connection->name);
			}
			else
			{
			    struct fg_targets *f = alloc_thing(struct fg_targets, "fg_target");

			    f->next = *pp;
			    f->group = g;
			    f->subnet = sn;
			    f->name = NULL;
			    *pp = f;
			}
		    }
		}
		(void)shift();	/* next */
		continue;

	    case B_record:
		flp->bdry = B_none;	/* eat the Record Boundary */
		(void)shift();	/* get real first token */
		continue;

	    case B_file:
		break;	/* done */
	    }
	    break;	/* if we reach here, out of loop */
	}
	lexclose();
    }
}

static void
free_targets(void)
{
    while (targets != NULL)
    {
	struct fg_targets *t = targets;

	targets = t->next;
	pfreeany(t->name);
	pfree(t);
    }
}

void
load_groups(void)
{
    passert(new_targets == NULL);

    /* for each group, add config file targets into new_targets */
    {
	struct fg_groups *g;

	for (g = groups; g != NULL; g = g->next)
	    if (oriented(*g->connection))
		read_foodgroup(g);
    }

    /* dump new_targets */
    DBG(DBG_CONTROL,
	{
	    struct fg_targets *t;

	    for (t = new_targets; t != NULL; t = t->next)
	    {
		char asource[SUBNETTOT_BUF];
		char atarget[SUBNETTOT_BUF];

		subnettot(&t->group->connection->spd.this.client
		    , 0, asource, sizeof(asource));
		subnettot(&t->subnet, 0, atarget, sizeof(atarget));
		DBG_log("%s->%s %s"
		    , asource, atarget
		    , t->group->connection->name);
	    }
	});

    /* determine and deal with differences between targets and new_targets.
     * structured like a merge.
     */
    {
	struct fg_targets *op = targets
	    , *np = new_targets;

	while (op != NULL && np != NULL)
	{
	    int r = subnetcmp(&op->group->connection->spd.this.client
		, &np->group->connection->spd.this.client);

	    if (r == 0)
		r = subnetcmp(&op->subnet, &np->subnet);

	    if (r == 0 && op->group == np->group)
	    {
		/* unchanged -- steal name & skip over */
		np->name = op->name;
		op->name = NULL;
		op = op->next;
		np = np->next;
	    }
	    else
	    {
		/* note: following cases overlap! */
		if (r <= 0)
		{
		    remove_group_instance(op->group->connection, op->name);
		    op = op->next;
		}
		if (r >= 0)
		{
		    np->name = add_group_instance(np->group->connection, &np->subnet);
		    np = np->next;
		}
	    }
	}
	for (; op != NULL; op = op->next)
	    remove_group_instance(op->group->connection, op->name);
	for (; np != NULL; np = np->next)
	    np->name = add_group_instance(np->group->connection, &np->subnet);

	/* update: new_targets replaces targets */
	free_targets();
	targets = new_targets;
	new_targets = NULL;
    }
}


void
add_group(struct connection *c)
{
    struct fg_groups *g = alloc_thing(struct fg_groups, "policy group");

    g->next = groups;
    groups = g;

    g->connection = c;
}

static struct fg_groups *
find_group(const struct connection *c)
{
    struct fg_groups *g;

    for (g = groups; g != NULL && g->connection != c; g = g->next)
	;
    return g;
}

void
route_group(struct connection *c)
{
    /* it makes no sense to route a connection that is ISAKMP-only */
    if (!NEVER_NEGOTIATE(c->policy) && !HAS_IPSEC_POLICY(c->policy))
    {
	loglog(RC_ROUTE, "cannot route an ISAKMP-only group connection");
    }
    else
    {
	struct fg_groups *g = find_group(c);
	struct fg_targets *t;

	passert(g != NULL);
	g->connection->policy |= POLICY_GROUTED;
	for (t = targets; t != NULL; t = t->next)
	{
	    if (t->group == g)
	    {
		struct connection *ci = con_by_name(t->name, FALSE);

		if (ci != NULL)
		{
		    set_cur_connection(ci);
		    if (!trap_connection(ci))
			whack_log(RC_ROUTE, "could not route");
		    set_cur_connection(c);
		}
	    }
	}
    }
}

void
unroute_group(struct connection *c)
{
    struct fg_groups *g = find_group(c);
    struct fg_targets *t;

    passert(g != NULL);
    g->connection->policy &= ~POLICY_GROUTED;
    for (t = targets; t != NULL; t = t->next)
    {
	if (t->group == g)
	{
	    struct connection *ci = con_by_name(t->name, FALSE);

	    if (ci != NULL)
	    {
		set_cur_connection(ci);
		unroute_connection(ci);
		set_cur_connection(c);
	    }
	}
    }
}

void
delete_group(const struct connection *c)
{
    struct fg_groups *g;

    /* find and remove from groups */
    {
	struct fg_groups **pp;

	for (pp = &groups; (g = *pp)->connection != c; pp = &(*pp)->next)
	    ;

	*pp = g->next;
    }

    /* find and remove from targets */
    {
	struct fg_targets **pp;

	for (pp = &targets; *pp != NULL; )
	{
	    struct fg_targets *t = *pp;

	    if (t->group == g)
	    {
		*pp = t->next;
		remove_group_instance(t->group->connection, t->name);
		pfree(t);
		/* pp is ready for next iteration */
	    }
	    else
	    {
		pp = &t->next;
	    }
	}
    }

    pfree(g);
}