summaryrefslogtreecommitdiff
path: root/src/charon/network/socket.c
blob: dd231ebed7118ffff12bba82f0b46d3841a41bc6 (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
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
/**
 * @file socket.c
 *
 * @brief Implementation of socket_t.
 *
 */

/*
 * Copyright (C) 2006 Tobias Brunner, Daniel Roethlisberger
 * Copyright (C) 2005-2006 Martin Willi
 * Copyright (C) 2005 Jan Hutter
 * Hochschule fuer Technik Rapperswil
 *
 * 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.
 */

#include <pthread.h>
#include <sys/types.h>
#include <sys/socket.h>
#include <string.h>
#include <errno.h>
#include <unistd.h>
#include <stdlib.h>
#include <fcntl.h>
#include <sys/ioctl.h>
#include <netinet/in.h>
#include <netinet/ip.h>
#include <netinet/ip6.h>
#include <netinet/udp.h>
#include <linux/ipsec.h>
#include <linux/filter.h>
#include <net/if.h>

#include "socket.h"

#include <daemon.h>

/* constants for packet handling */
#define IP_LEN sizeof(struct iphdr)
#define IP6_LEN sizeof(struct ip6_hdr)
#define UDP_LEN sizeof(struct udphdr)
#define MARKER_LEN sizeof(u_int32_t)

/* offsets for packet handling */
#define IP_PROTO_OFFSET 9
#define IP6_PROTO_OFFSET 6
#define IKE_VERSION_OFFSET 17
#define IKE_LENGTH_OFFSET 24

/* from linux/in.h */
#ifndef IP_IPSEC_POLICY
#define IP_IPSEC_POLICY 16
#endif /*IP_IPSEC_POLICY*/

/* from linux/udp.h */
#ifndef UDP_ENCAP
#define UDP_ENCAP 100
#endif /*UDP_ENCAP*/

#ifndef UDP_ENCAP_ESPINUDP
#define UDP_ENCAP_ESPINUDP 2
#endif /*UDP_ENCAP_ESPINUDP*/

/* needed for older kernel headers */
#ifndef IPV6_2292PKTINFO
#define IPV6_2292PKTINFO 2
#endif /*IPV6_2292PKTINFO*/

/* missing on uclibc */
#ifndef IPV6_IPSEC_POLICY
#define IPV6_IPSEC_POLICY 34
#endif /*IPV6_IPSEC_POLICY*/

typedef struct private_socket_t private_socket_t;

/**
 * Private data of an socket_t object
 */
struct private_socket_t{
	/**
	 * public functions
	 */
	 socket_t public;

	 /**
	  * regular port
	  */
	 int port;

	 /**
	  * port used for nat-t
	  */
	 int natt_port;
	 
	 /**
	  * raw receiver socket for IPv4
	  */
	 int recv4;
	 
	 /**
	  * raw receiver socket for IPv6
	  */
	 int recv6;

	 /**
	  * send socket on regular port for IPv4
	  */
	 int send4;

	 /**
	  * send socket on regular port for IPv6
	  */
	 int send6;

	 /**
	  * send socket on nat-t port for IPv4
	  */
	 int send4_natt;

	 /**
	  * send socket on nat-t port for IPv6
	  */
	 int send6_natt;
};

/**
 * implementation of socket_t.receive
 */
static status_t receiver(private_socket_t *this, packet_t **packet)
{
	char buffer[MAX_PACKET];
	chunk_t data;
	packet_t *pkt;
	struct udphdr *udp;
	host_t *source = NULL, *dest = NULL;
	int bytes_read = 0;
	int data_offset, oldstate;
	fd_set rfds;

	FD_ZERO(&rfds);
	
	if (this->recv4)
	{
		FD_SET(this->recv4, &rfds);
	}
	if (this->recv6)
	{
		FD_SET(this->recv6, &rfds);
	}
	
	DBG2(DBG_NET, "waiting for data on raw sockets");
	
	pthread_setcancelstate(PTHREAD_CANCEL_ENABLE, &oldstate);
	if (select(max(this->recv4, this->recv6) + 1, &rfds, NULL, NULL, NULL) <= 0)
	{
		pthread_setcancelstate(oldstate, NULL);
		return FAILED;
	}
	pthread_setcancelstate(oldstate, NULL);
	
	if (this->recv4 && FD_ISSET(this->recv4, &rfds))
	{
		/* IPv4 raw sockets return the IP header. We read src/dest
		 * information directly from the raw header */
		struct iphdr *ip;
		struct sockaddr_in src, dst;
		
		bytes_read = recv(this->recv4, buffer, MAX_PACKET, 0);
		if (bytes_read < 0)
		{
			DBG1(DBG_NET, "error reading from IPv4 socket: %s", strerror(errno));
			return FAILED;
		}
		DBG3(DBG_NET, "received IPv4 packet %b", buffer, bytes_read);
		
		/* read source/dest from raw IP/UDP header */
		if (bytes_read < IP_LEN + UDP_LEN + MARKER_LEN)
		{
			DBG1(DBG_NET, "received IPv4 packet too short (%d bytes)",
				 bytes_read);
			return FAILED;
		}
		ip = (struct iphdr*) buffer;
		udp = (struct udphdr*) (buffer + IP_LEN);
		src.sin_family = AF_INET;
		src.sin_addr.s_addr = ip->saddr;
		src.sin_port = udp->source;
		dst.sin_family = AF_INET;
		dst.sin_addr.s_addr = ip->daddr;
		dst.sin_port = udp->dest;
		source = host_create_from_sockaddr((sockaddr_t*)&src);
		dest = host_create_from_sockaddr((sockaddr_t*)&dst);
		
		pkt = packet_create();
		pkt->set_source(pkt, source);
		pkt->set_destination(pkt, dest);
		DBG2(DBG_NET, "received packet: from %#H to %#H", source, dest);
		data_offset = IP_LEN + UDP_LEN;
		/* remove non esp marker */	
		if (dest->get_port(dest) == this->natt_port)
		{
			data_offset += MARKER_LEN;
		}
		/* fill in packet */
		data.len = bytes_read - data_offset;
		data.ptr = malloc(data.len);
		memcpy(data.ptr, buffer + data_offset, data.len);
		pkt->set_data(pkt, data);
	}
	else if (this->recv6 && FD_ISSET(this->recv6, &rfds))
	{
		/* IPv6 raw sockets return no IP header. We must query
		 * src/dest via socket options/ancillary data */
		struct msghdr msg;
		struct cmsghdr *cmsgptr;
		struct sockaddr_in6 src, dst;
		struct iovec iov;
		char ancillary[64];
		
		msg.msg_name = &src;
		msg.msg_namelen = sizeof(src);
		iov.iov_base = buffer;
		iov.iov_len = sizeof(buffer);
		msg.msg_iov = &iov;
		msg.msg_iovlen = 1;
		msg.msg_control = ancillary;
		msg.msg_controllen = sizeof(ancillary);
		msg.msg_flags = 0;
		
		bytes_read = recvmsg(this->recv6, &msg, 0);
		if (bytes_read < 0)
		{
			DBG1(DBG_NET, "error reading from IPv6 socket: %s", strerror(errno));
			return FAILED;
		}
		DBG3(DBG_NET, "received IPv6 packet %b", buffer, bytes_read);
		
		if (bytes_read < IP_LEN + UDP_LEN + MARKER_LEN)
		{
			DBG3(DBG_NET, "received IPv6 packet too short (%d bytes)",
				 bytes_read);
			return FAILED;
		}
		
		/* read ancillary data to get destination address */
		for (cmsgptr = CMSG_FIRSTHDR(&msg); cmsgptr != NULL;
			 cmsgptr = CMSG_NXTHDR(&msg, cmsgptr))
		{
			if (cmsgptr->cmsg_len == 0)
			{
				DBG1(DBG_NET, "error reading IPv6 ancillary data");
				return FAILED;
			}	
			if (cmsgptr->cmsg_level == SOL_IPV6 &&
				cmsgptr->cmsg_type == IPV6_2292PKTINFO)
			{
				struct in6_pktinfo *pktinfo;
				pktinfo = (struct in6_pktinfo*)CMSG_DATA(cmsgptr);
				
				memset(&dst, 0, sizeof(dst));
				memcpy(&dst.sin6_addr, &pktinfo->ipi6_addr, sizeof(dst.sin6_addr));
				dst.sin6_family = AF_INET6;
				udp = (struct udphdr*) (buffer);
				dst.sin6_port = udp->dest;
				src.sin6_port = udp->source;
				dest = host_create_from_sockaddr((sockaddr_t*)&dst);
			}
		}
		/* ancillary data missing? */
		if (dest == NULL)
		{
			DBG1(DBG_NET, "error reading IPv6 packet header");
			return FAILED;
		}
		
		source = host_create_from_sockaddr((sockaddr_t*)&src);
		
		pkt = packet_create();
		pkt->set_source(pkt, source);
		pkt->set_destination(pkt, dest);
		DBG2(DBG_NET, "received packet: from %#H to %#H", source, dest);
		data_offset = UDP_LEN;
		/* remove non esp marker */	
		if (dest->get_port(dest) == this->natt_port)
		{
			data_offset += MARKER_LEN;
		}
		/* fill in packet */
		data.len = bytes_read - data_offset;
		data.ptr = malloc(data.len);
		memcpy(data.ptr, buffer + data_offset, data.len);
		pkt->set_data(pkt, data);
	}
	else
	{
		/* oops, shouldn't happen */
		return FAILED;
	}
	
	/* return packet */
	*packet = pkt;
	return SUCCESS;
}

/**
 * implementation of socket_t.send
 */
status_t sender(private_socket_t *this, packet_t *packet)
{
	int sport, skt, family;
	ssize_t bytes_sent;
	chunk_t data, marked;
	host_t *src, *dst;
	struct msghdr msg;
	struct cmsghdr *cmsg;
	struct iovec iov;
	
	src = packet->get_source(packet);
	dst = packet->get_destination(packet);
	data = packet->get_data(packet);

	DBG2(DBG_NET, "sending packet: from %#H to %#H", src, dst);
	
	/* send data */
	sport = src->get_port(src);
	family = dst->get_family(dst);
	if (sport == this->port)
	{
		if (family == AF_INET)
		{
			skt = this->send4;
		}
		else
		{
			skt = this->send6;
		}
	}
	else if (sport == this->natt_port)
	{
		if (family == AF_INET)
		{
			skt = this->send4_natt;
		}
		else
		{
			skt = this->send6_natt;
		}
		/* NAT keepalives without marker */
		if (data.len != 1 || data.ptr[0] != 0xFF)
		{
			/* add non esp marker to packet */
			if (data.len > MAX_PACKET - MARKER_LEN)
			{
				DBG1(DBG_NET, "unable to send packet: it's too big (%d bytes)",
					 data.len);
				return FAILED;
			}
			marked = chunk_alloc(data.len + MARKER_LEN);
			memset(marked.ptr, 0, MARKER_LEN);
			memcpy(marked.ptr + MARKER_LEN, data.ptr, data.len);
			/* let the packet do the clean up for us */
			packet->set_data(packet, marked);
			data = marked;
		}
	}
	else
	{
		DBG1(DBG_NET, "unable to locate a send socket for port %d", sport);
		return FAILED;
	}
	
	memset(&msg, 0, sizeof(struct msghdr));
	msg.msg_name = dst->get_sockaddr(dst);;
	msg.msg_namelen = *dst->get_sockaddr_len(dst);
	iov.iov_base = data.ptr;
	iov.iov_len = data.len;
	msg.msg_iov = &iov;
	msg.msg_iovlen = 1;
	msg.msg_flags = 0;
	
	if (!dst->is_anyaddr(dst))
	{
		if (family == AF_INET)
		{
			char buf[CMSG_SPACE(sizeof(struct in_pktinfo))];
			struct in_pktinfo *pktinfo;
			struct sockaddr_in *sin;
			
			msg.msg_control = buf;
			msg.msg_controllen = sizeof(buf);
			cmsg = CMSG_FIRSTHDR(&msg);
			cmsg->cmsg_level = SOL_IP;
			cmsg->cmsg_type = IP_PKTINFO;
			cmsg->cmsg_len = CMSG_LEN(sizeof(struct in_pktinfo));
			pktinfo = (struct in_pktinfo*)CMSG_DATA(cmsg);
			memset(pktinfo, 0, sizeof(struct in_pktinfo));
			sin = (struct sockaddr_in*)src->get_sockaddr(src);
			memcpy(&pktinfo->ipi_spec_dst, &sin->sin_addr, sizeof(struct in_addr));
		}
		else
		{
			char buf[CMSG_SPACE(sizeof(struct in6_pktinfo))];
			struct in6_pktinfo *pktinfo;
			struct sockaddr_in6 *sin;
			
			msg.msg_control = buf;
			msg.msg_controllen = sizeof(buf);
			cmsg = CMSG_FIRSTHDR(&msg);
			cmsg->cmsg_level = SOL_IPV6;
			cmsg->cmsg_type = IPV6_2292PKTINFO;
			cmsg->cmsg_len = CMSG_LEN(sizeof(struct in6_pktinfo));
			pktinfo = (struct in6_pktinfo*)CMSG_DATA(cmsg);
			memset(pktinfo, 0, sizeof(struct in6_pktinfo));
			sin = (struct sockaddr_in6*)src->get_sockaddr(src);
			memcpy(&pktinfo->ipi6_addr, &sin->sin6_addr, sizeof(struct in6_addr));
		}
	}
	
	bytes_sent = sendmsg(skt, &msg, 0);

	if (bytes_sent != data.len)
	{
		DBG1(DBG_NET, "error writing to socket: %s", strerror(errno));
		return FAILED;
	}
	return SUCCESS;
}

/**
 * open a socket to send packets
 */
static int open_send_socket(private_socket_t *this, int family, u_int16_t port)
{
	int on = TRUE;
	int type = UDP_ENCAP_ESPINUDP;
	struct sockaddr_storage addr;
	u_int sol, ipsec_policy;
	struct sadb_x_policy policy;
	int skt;
	
	memset(&addr, 0, sizeof(addr));
	/* precalculate constants depending on address family */
	switch (family)
	{
		case AF_INET:
		{
			struct sockaddr_in *sin = (struct sockaddr_in *)&addr;
			sin->sin_family = AF_INET;
			sin->sin_addr.s_addr = INADDR_ANY;
			sin->sin_port = htons(port);
			sol = SOL_IP;
			ipsec_policy = IP_IPSEC_POLICY;
			break;
		}
		case AF_INET6:
		{
			struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)&addr;
			sin6->sin6_family = AF_INET6;
			memcpy(&sin6->sin6_addr, &in6addr_any, sizeof(in6addr_any));
			sin6->sin6_port = htons(port);
			sol = SOL_IPV6;
			ipsec_policy = IPV6_IPSEC_POLICY;
			break;
		}
		default:
			return 0;
	}
	
	skt = socket(family, SOCK_DGRAM, IPPROTO_UDP);
	if (skt < 0)
	{
		DBG1(DBG_NET, "could not open send socket: %s", strerror(errno));
		return 0;
	}
	
	if (setsockopt(skt, SOL_SOCKET, SO_REUSEADDR, (void*)&on, sizeof(on)) < 0)
	{
		DBG1(DBG_NET, "unable to set SO_REUSEADDR on send socket: %s",
			 strerror(errno));
		close(skt);
		return 0;
	}
	
	/* bypass outgoung IKE traffic on send socket */
	memset(&policy, 0, sizeof(policy));
	policy.sadb_x_policy_len = sizeof(policy) / sizeof(u_int64_t);
	policy.sadb_x_policy_exttype = SADB_X_EXT_POLICY;
	policy.sadb_x_policy_type = IPSEC_POLICY_BYPASS;
	policy.sadb_x_policy_dir = IPSEC_DIR_OUTBOUND;
	
	if (setsockopt(skt, sol, ipsec_policy, &policy, sizeof(policy)) < 0)
	{
		DBG1(DBG_NET, "unable to set IPSEC_POLICY on send socket: %s",
			 strerror(errno));
		close(skt);
		return 0;
	}
	
	/* We don't receive packets on the send socket, but we need a INBOUND policy.
	 * Otherwise, UDP decapsulation does not work!!! */
	policy.sadb_x_policy_dir = IPSEC_DIR_INBOUND;
	if (setsockopt(skt, sol, ipsec_policy, &policy, sizeof(policy)) < 0)
	{
		DBG1(DBG_NET, "unable to set IPSEC_POLICY on send socket: %s", 
			 strerror(errno));
		close(skt);
		return 0;
	}
	
	/* bind the send socket */
	if (bind(skt, (struct sockaddr *)&addr, sizeof(addr)) < 0)
	{
		DBG1(DBG_NET, "unable to bind send socket: %s",
			 strerror(errno));
		close(skt);
		return 0;
	}
	
	if (family == AF_INET)
	{
		/* enable UDP decapsulation globally, only for one socket needed */
		if (setsockopt(skt, SOL_UDP, UDP_ENCAP, &type, sizeof(type)) < 0)
		{
			DBG1(DBG_NET, "unable to set UDP_ENCAP: %s; NAT-T may fail",
				 strerror(errno));
		}
	}
	
	return skt;
}

/**
 * open a socket to receive packets
 */
static int open_recv_socket(private_socket_t *this, int family)
{
	int skt;
	int on = TRUE;
	u_int proto_offset, ip_len, sol, ipsec_policy, udp_header, ike_header;
	struct sadb_x_policy policy;
	
	/* precalculate constants depending on address family */
	switch (family)
	{
		case AF_INET:
			proto_offset = IP_PROTO_OFFSET;
			ip_len = IP_LEN;
			sol = SOL_IP;
			ipsec_policy = IP_IPSEC_POLICY;
			break;
		case AF_INET6:
			proto_offset = IP6_PROTO_OFFSET;
			ip_len = 0; /* IPv6 raw sockets contain no IP header */
			sol = SOL_IPV6;
			ipsec_policy = IPV6_IPSEC_POLICY;
			break;
		default:
			return 0;
	}
	udp_header = ip_len;
	ike_header = ip_len + UDP_LEN;
	
	/* This filter code filters out all non-IKEv2 traffic on
	 * a SOCK_RAW IP_PROTP_UDP socket. Handling of other
	 * IKE versions is done in pluto.
	 */
	struct sock_filter ikev2_filter_code[] =
	{
		/* Destination Port must be either port or natt_port */
		BPF_STMT(BPF_LD+BPF_H+BPF_ABS, udp_header + 2),
		BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, this->port, 1, 0),
		BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, this->natt_port, 5, 12),
		/* port */
			/* IKE version must be 2.0 */
			BPF_STMT(BPF_LD+BPF_B+BPF_ABS, ike_header + IKE_VERSION_OFFSET),
			BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x20, 0, 10),
			/* packet length is length in IKEv2 header + ip header + udp header */
			BPF_STMT(BPF_LD+BPF_W+BPF_ABS, ike_header + IKE_LENGTH_OFFSET),
			BPF_STMT(BPF_ALU+BPF_ADD+BPF_K, ip_len + UDP_LEN),
			BPF_STMT(BPF_RET+BPF_A, 0),
		/* natt_port */
			/* nat-t: check for marker */
			BPF_STMT(BPF_LD+BPF_W+BPF_ABS, ike_header),
			BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0, 0, 5),
			/* nat-t: IKE version must be 2.0 */
			BPF_STMT(BPF_LD+BPF_B+BPF_ABS, ike_header + MARKER_LEN + IKE_VERSION_OFFSET),
			BPF_JUMP(BPF_JMP+BPF_JEQ+BPF_K, 0x20, 0, 3),
			/* nat-t: packet length is length in IKEv2 header + ip header + udp header + non esp marker */
			BPF_STMT(BPF_LD+BPF_W+BPF_ABS, ike_header + MARKER_LEN + IKE_LENGTH_OFFSET),
			BPF_STMT(BPF_ALU+BPF_ADD+BPF_K, ip_len + UDP_LEN + MARKER_LEN),
			BPF_STMT(BPF_RET+BPF_A, 0),
		/* packet doesn't match, ignore */
		BPF_STMT(BPF_RET+BPF_K, 0),
	};

	/* Filter struct to use with setsockopt */
	struct sock_fprog ikev2_filter = {
		sizeof(ikev2_filter_code) / sizeof(struct sock_filter),
		ikev2_filter_code
	};
	
	/* set up a raw socket */
	skt = socket(family, SOCK_RAW, IPPROTO_UDP);
	if (skt < 0)
	{
		DBG1(DBG_NET, "unable to create raw socket: %s", strerror(errno));
		return 0;
	}
	
	if (setsockopt(skt, SOL_SOCKET, SO_ATTACH_FILTER,
				   &ikev2_filter, sizeof(ikev2_filter)) < 0)
	{
		DBG1(DBG_NET, "unable to attach IKEv2 filter to raw socket: %s",
			 strerror(errno));
		close(skt);
		return 0;
	}
	
	if (family == AF_INET6 &&
		/* we use IPV6_2292PKTINFO, as IPV6_PKTINFO is defined as
		 * 2 or 50 depending on kernel header version */
		setsockopt(skt, sol, IPV6_2292PKTINFO, &on, sizeof(on)) < 0)
	{
		DBG1(DBG_NET, "unable to set IPV6_PKTINFO on raw socket: %s",
			 strerror(errno));
		close(skt);
		return 0;
	}
	
	/* bypass incomining IKE traffic on this socket */
	memset(&policy, 0, sizeof(policy));
	policy.sadb_x_policy_len = sizeof(policy) / sizeof(u_int64_t);
	policy.sadb_x_policy_exttype = SADB_X_EXT_POLICY;
	policy.sadb_x_policy_type = IPSEC_POLICY_BYPASS;
	policy.sadb_x_policy_dir = IPSEC_DIR_INBOUND;
	
	if (setsockopt(skt, sol, ipsec_policy, &policy, sizeof(policy)) < 0)
	{
		DBG1(DBG_NET, "unable to set IPSEC_POLICY on raw socket: %s",
			 strerror(errno));
		close(skt);
		return 0;
	}
	
	return skt;
}

/**
 * implementation of socket_t.destroy
 */
static void destroy(private_socket_t *this)
{
	if (this->recv4)
	{
		close(this->recv4);
	}
	if (this->recv6)
	{
		close(this->recv6);
	}
	if (this->send4)
	{
		close(this->send4);
	}
	if (this->send6)
	{
		close(this->send6);
	}
	if (this->send4_natt)
	{
		close(this->send4_natt);
	}
	if (this->send6_natt)
	{
		close(this->send6_natt);
	}
	free(this);
}

/*
 * See header for description
 */
socket_t *socket_create(u_int16_t port, u_int16_t natt_port)
{
	int key;
	private_socket_t *this = malloc_thing(private_socket_t);

	/* public functions */
	this->public.send = (status_t(*)(socket_t*, packet_t*))sender;
	this->public.receive = (status_t(*)(socket_t*, packet_t**))receiver;
	this->public.destroy = (void(*)(socket_t*)) destroy;
	
	this->port = port;
	this->natt_port = natt_port;
	this->recv4 = 0;
	this->recv6 = 0;
	this->send4 = 0;
	this->send6 = 0;
	this->send4_natt = 0;
	this->send6_natt = 0;
	
	/* we open a AF_KEY socket to autoload the af_key module. Otherwise
	 * setsockopt(IPSEC_POLICY) won't work. */
	key = socket(AF_KEY, SOCK_RAW, PF_KEY_V2);
	if (key == 0)
	{
		charon->kill(charon, "could not open AF_KEY socket");
	}
	close(key);
	
	this->recv4 = open_recv_socket(this, AF_INET);
	if (this->recv4 == 0)
	{
		DBG1(DBG_NET, "could not open IPv4 receive socket, IPv4 disabled");
	}
	else
	{
		this->send4 = open_send_socket(this, AF_INET, this->port);
		if (this->send4 == 0)
		{
			DBG1(DBG_NET, "could not open IPv4 send socket, IPv4 disabled");
			close(this->recv4);
		}
		else
		{
			this->send4_natt = open_send_socket(this, AF_INET, this->natt_port);
			if (this->send4_natt == 0)
			{
				DBG1(DBG_NET, "could not open IPv4 NAT-T send socket");
			}
		}
	}
	
	this->recv6 = open_recv_socket(this, AF_INET6);
	if (this->recv6 == 0)
	{
		DBG1(DBG_NET, "could not open IPv6 receive socket, IPv6 disabled");
	}
	else
	{
		this->send6 = open_send_socket(this, AF_INET6, this->port);
		if (this->send6 == 0)
		{
			DBG1(DBG_NET, "could not open IPv6 send socket, IPv6 disabled");
			close(this->recv6);
		}
		else
		{
			this->send6_natt = open_send_socket(this, AF_INET6, this->natt_port);
			if (this->send6_natt == 0)
			{
				DBG1(DBG_NET, "could not open IPv6 NAT-T send socket");
			}
		}
	}
	
	if (!(this->send4 || this->send6) || !(this->recv4 || this->recv6))
	{
		DBG1(DBG_NET, "could not create any sockets");
		destroy(this);
		charon->kill(charon, "socket initialization failed");
	}
	
	return (socket_t*)this;
}