forked from traviscross/mtr
-
Notifications
You must be signed in to change notification settings - Fork 3
/
net.c
1285 lines (1160 loc) · 37.7 KB
/
net.c
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
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
mtr -- a network diagnostic tool
Copyright (C) 1997,1998 Matt Kimball
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License version 2 as
published by the Free Software Foundation.
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.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
#include <stddef.h>
#include <stdint.h>
#include <stdio.h>
#include <unistd.h>
#include <string.h>
#include <errno.h>
#include <limits.h>
#include <math.h>
#include <fcntl.h>
#include <time.h>
#include <sys/types.h>
#include <sys/socket.h>
#include <netinet/in_systm.h>
#include <netinet/in.h>
#include <netinet/ip.h>
#include <netinet/ip_icmp.h>
#include <netinet/tcp.h>
#include <netinet/udp.h>
#include <arpa/inet.h>
#include <assert.h>
#include "config.h" // IWYU pragma: keep
#if defined(LOG_NET) && !defined(LOGMOD)
#define LOGMOD
#endif
#if !defined(LOG_NET) && defined(LOGMOD)
#undef LOGMOD
#endif
#ifdef LIBCAP
#include <sys/capability.h>
#endif
#include "common.h"
#include "net.h"
#include "aux.h"
#include "mtr-poll.h"
#include "display.h"
#ifdef ENABLE_DNS
#include "dns.h"
#endif
#ifdef OUTPUT_FORMAT_RAW
#include "report.h"
#endif
#ifdef HAVE_ARC4RANDOM_UNIFORM
#ifdef HAVE_BSD_STDLIB_H
#include <bsd/stdlib.h>
#endif
#define RANDUNIFORM(base) arc4random_uniform(base)
#else // original version
#define RANDUNIFORM(base) ((base - 1) * (rand() / (RAND_MAX + 0.1)))
#endif
// no common field names
struct PACKIT _iphdr {
#if BYTE_ORDER == LITTLE_ENDIAN
uint8_t ihl:4;
uint8_t ver:4;
#elif BYTE_ORDER == BIG_ENDIAN
uint8_t ver:4;
uint8_t ihl:4;
#else
#error "Undefined byte order"
#endif
uint8_t tos;
uint16_t len, id, frag;
uint8_t ttl, proto;
uint16_t sum;
uint32_t saddr, daddr;
}; /* must be 20 bytes */
// no common field names
struct PACKIT _icmphdr {
uint8_t type, code;
uint16_t sum, id, seq;
}; /* must be 8 bytes */
#ifndef ICMP_TIME_EXCEEDED // not defined on old systems
#define ICMP_TIME_EXCEEDED 11
#endif
// struct tcphdr /* common because RFC793 */
// struct udphdr /* common because RFC768 */
// udp4 pseudo header
struct PACKIT _udpph {
uint32_t saddr, daddr;
uint8_t zero, proto;
uint16_t len;
}; /* must be 12 bytes */
#ifdef WITH_MPLS
struct PACKIT icmpext_struct { // RFC4884
#if BYTE_ORDER == LITTLE_ENDIAN
uint8_t res:4;
uint8_t ver:4;
#elif BYTE_ORDER == BIG_ENDIAN
uint8_t ver:4;
uint8_t res:4;
#else
#error "Undefined byte order"
#endif
uint8_t rest;
uint16_t sum;
}; /* must be 4 bytes */
struct PACKIT icmpext_object { // RFC4884
uint16_t len;
uint8_t class;
uint8_t type;
}; /* must be 4 bytes */
#define ICMP_EXT_VER 2
#define ICMP_EXT_CLASS_MPLS 1
#define ICMP_EXT_TYPE_MPLS 1
#define IES_SZ sizeof(struct icmpext_struct)
#define IEO_SZ sizeof(struct icmpext_object)
#define LAB_SZ sizeof(mpls_label_t)
#define MPLSMIN 120 // min after: [ip] icmp ip
#define MPLSFNTAIL(arg) , arg
#else
#define MPLSFNTAIL(arg)
#endif /*MPLS*/
#define LO_UDPPORT 33433 // start from LO_UDPPORT+1
#define UDPPORTS 90 // go thru udp:33434-33523 acl
#define TCP_DEFAULT_PORT 80
#define SET_UDP_UH_PORTS(uh, s, d) { (uh)->uh_sport = htons(s); (uh)->uh_dport = htons(d); }
// NOTE: don't forget to include sys/param.h
#if defined(__FreeBSD_version)
#if __FreeBSD_version >= 1100000
#define IPLEN_RAW(sz) htons(sz)
#endif
#elif defined(__OpenBSD__)
#define IPLEN_RAW(sz) htons(sz)
#elif defined(__HAIKU__)
#define IPLEN_RAW(sz) htons(sz)
#ifndef IPV6_CHECKSUM
#define IPV6_CHECKSUM 7
#endif
#else
#define IPLEN_RAW(sz) sz
#endif
#define EXIT(s) { int e = errno; display_close(true); ERRX(e, "%s: %s", s, strerror(e)); }
#define CLOSE(fd) if ((fd) >= 0) { close(fd); fd = -1; /*summ*/ sum_sock[1]++; }
#define FAIL_WITH_WARNX(rcode, fd, fmt, ...) { last_neterr = rcode; \
display_clear(); CLOSE(fd); \
WARNX(fmt ": %s", __VA_ARGS__, strerror(last_neterr)); \
snprintf(neterr_txt, ERRBYFN_SZ, fmt ": %s", __VA_ARGS__, strerror(last_neterr)); \
LOG_RE(false, fmt ": %s", __VA_ARGS__, strerror(last_neterr)); \
}
#define FAIL_WITH_WARN(fd, fmt, ...) FAIL_WITH_WARNX(errno, fd, fmt, __VA_ARGS__)
struct sequence {
int at;
bool transit;
struct timespec time;
#ifdef CURSESMODE
int saved_seq;
#endif
};
// global vars
t_ipaddr unspec_addr; // 0
bool (*addr_exist)(const void *a); // true if not 0
bool (*addr_equal)(const void *a, const void *b);
void* (*addr_copy)(void *dst, const void *src);
int af = AF_INET; // default address family
struct nethost host[MAXHOST];
char localaddr[MAX_ADDRSTRLEN];
unsigned long net_queries[4]; // number of queries (sum, icmp, udp, tcp)
unsigned long net_replies[4]; // number of replies (sum, icmp, udp, tcp)
enum { QR_SUM = 0 /*sure*/, QR_ICMP, QR_UDP, QR_TCP };
//
/* How many queries to unknown hosts do we send?
* It limits the amount of traffic generated if a host is not reachable
*/
#define MAX_UNKNOWN_HOSTS 10
static int bitpattern;
static int packetsize;
static struct sequence seqlist[MAXSEQ];
static int sendsock4 = -1;
static int recvsock4 = -1;
#ifdef ENABLE_IPV6
static int sendsock6 = -1;
static int sendsock6_icmp = -1;
static int sendsock6_udp = -1;
static int recvsock6 = -1;
#endif
static int sendsock = -1;
static int recvsock = -1;
static t_sockaddr lsa, rsa; // losal and remote sockaddr
static t_ipaddr *remote_ipaddr;
static int echo_reply, time_exceed, dst_unreach;
static size_t ipicmphdr_sz = sizeof(struct _icmphdr);
static size_t minfailsz = sizeof(struct _icmphdr);
static size_t hdr_minsz;
static size_t iphdr_sz;
static size_t sa_addr_offset;
static socklen_t sa_len;
static int batch_at;
static int numhosts = 10;
static int portpid;
static int stopper = MAXHOST;
enum { RE_PONG, RE_EXCEED, RE_UNREACH }; // reason of a pong response
bool addr4exist(const void *a) { return memcmp(a, &unspec_addr, sizeof(struct in_addr)) ? true : false; }
bool addr4equal(const void *a, const void *b) { return memcmp(a, b, sizeof(struct in_addr)) ? false : true; }
void* addr4copy(void *dst, const void *src) { return memcpy(dst, src, sizeof(struct in_addr)); }
#ifdef ENABLE_IPV6
bool addr6exist(const void *a) { return memcmp(a, &unspec_addr, sizeof(struct in6_addr)) ? true : false; }
bool addr6equal(const void *a, const void *b) { return memcmp(a, b, sizeof(struct in6_addr)) ? false : true; }
void* addr6copy(void *dst, const void *src) { return memcpy(dst, src, sizeof(struct in6_addr)); }
#endif
// return in 'tv' waittime before sending the next ping
void waitspec(struct timespec *tv) {
double wait = wait_time;
int n = numhosts;
int f = fstTTL - 1;
if ((f > 0) && (n > f))
n -= f;
wait /= n;
tv->tv_sec = trunc(wait);
tv->tv_nsec = (wait - tv->tv_sec) * NANO;
}
static uint16_t sum1616(const uint16_t *data, unsigned len, unsigned sum) {
for (; len; len--)
sum += *data++;
while (sum >> 16)
sum = (sum >> 16) + (sum & 0xffff);
return ~sum;
}
// Prepend pseudoheader to the udp datagram and calculate checksum
static uint16_t udpsum16(struct _iphdr *ip, void *udata, int udata_len, int dsize) {
unsigned tsize = sizeof(struct _udpph) + dsize;
char csumpacket[tsize];
memset(csumpacket, bitpattern, sizeof(csumpacket));
struct _udpph *prepend = (struct _udpph *)csumpacket;
prepend->saddr = ip->saddr;
prepend->daddr = ip->daddr;
prepend->zero = 0;
prepend->proto = ip->proto;
prepend->len = udata_len;
struct udphdr *content = (struct udphdr *)(csumpacket + sizeof(struct _udpph));
struct udphdr *data = (struct udphdr *)udata;
content->uh_sport = data->uh_sport;
content->uh_dport = data->uh_dport;
content->uh_ulen = data->uh_ulen;
content->uh_sum = data->uh_sum;
return sum1616((uint16_t*)csumpacket, tsize / 2, (tsize % 2) ? (bitpattern & 0xff) : 0);
}
static inline bool save_send_ts(int seq) {
int rc = clock_gettime(CLOCK_MONOTONIC, &seqlist[seq].time);
if (rc < 0) FAIL_CLOCK_GETTIME;
return true;
}
static void save_sequence(int seq, int at) {
LOGMSG("seq=%d at=%d", seq, at);
seqlist[seq].at = at;
seqlist[seq].transit = true;
if (host[at].transit)
host[at].up = false; // if previous packet is in transit too, then assume it's down
host[at].transit = true;
host[at].sent++;
#ifdef CURSESMODE
seqlist[seq].saved_seq = host[at].sent;
if (host[at].saved[SAVED_PINGS - 1] != CT_UNSENT) {
for (int at = 0; at < MAXHOST; at++) {
memmove(host[at].saved, host[at].saved + 1, (SAVED_PINGS - 1) * sizeof(int));
host[at].saved[SAVED_PINGS - 1] = CT_UNSENT;
host[at].saved_seq_offset += 1;
}
}
host[at].saved[SAVED_PINGS - 1] = CT_UNKN;
#endif
}
static int new_sequence(int at) {
static int max_seqs[] = { [IPPROTO_ICMP] = MAXSEQ, [IPPROTO_UDP] = UDPPORTS };
static int next_seq;
int seq = next_seq++;
if (next_seq >= max_seqs[mtrtype])
next_seq = 0;
save_sequence(seq, at);
return seq;
}
#define NET_SETTOS(PROTO_VERSION, TOS_TYPE) if (tos) \
if (setsockopt(sock, PROTO_VERSION, TOS_TYPE, &tos, sizeof(tos)) < 0) \
FAIL_WITH_WARN(sock, "%s(sock=%d, tos=%d)", __func__, sock, tos);
#define NET_SETTTL(PROTO_VERSION, TTL_TYPE) \
if (setsockopt(sock, PROTO_VERSION, TTL_TYPE, &ttl, sizeof(ttl)) < 0) \
FAIL_WITH_WARN(sock, "%s(sock=%d, ttl=%d)", __func__, sock, ttl);
static bool settosttl(int sock, int ttl) {
NET_SETTTL(IPPROTO_IP, IP_TTL);
#ifdef IP_TOS
NET_SETTOS(IPPROTO_IP, IP_TOS);
#endif
return true; }
#ifdef ENABLE_IPV6
static bool settosttl6(int sock, int ttl) {
NET_SETTTL(IPPROTO_IPV6, IPV6_UNICAST_HOPS);
#ifdef IPV6_TCLASS
NET_SETTOS(IPPROTO_IPV6, IPV6_TCLASS);
#endif
return true; }
#endif
#undef NET_SETTOS
#undef NET_SETTTL
// Create TCP socket for hop 'at', and try to connect (poll results later)
static bool net_send_tcp(int at) {
#define SET_TCP_ATTR(src_addr, ssa_addr, dst_addr, dst_port) { \
addr_copy(&src_addr, &ssa_addr); \
addr_copy(&dst_addr, remote_ipaddr); \
dst_port = htons((remoteport > 0) ? remoteport : TCP_DEFAULT_PORT); \
}
int sock = socket(af, SOCK_STREAM, 0);
if (sock < 0)
FAIL_WITH_WARN(sock, "socket[at=%d]", at);
/*summ*/ sum_sock[0]++;
t_sockaddr local = {0}, remote = {0};
local.SA_AF = remote.SA_AF = af;
socklen_t addrlen = sizeof(local);
switch (af) {
case AF_INET:
SET_TCP_ATTR(local.S_ADDR, lsa.S_ADDR, remote.S_ADDR, remote.S_PORT);
addrlen = sizeof(lsa.sin);
break;
#ifdef ENABLE_IPV6
case AF_INET6:
SET_TCP_ATTR(local.S6ADDR, lsa.S6ADDR, remote.S6ADDR, remote.S6PORT)
addrlen = sizeof(lsa.sin6);
break;
#endif
default:
FAIL_POSTPONE(ENOPROTOOPT, "address family %d", af);
}
if (bind(sock, &local.sa, addrlen))
FAIL_WITH_WARN(sock, "bind[at=%d]", at);
if (getsockname(sock, &local.sa, &addrlen))
FAIL_WITH_WARN(sock, "getsockname[at=%d]", at);
int flags = fcntl(sock, F_GETFL, 0);
if (fcntl(sock, F_SETFL, flags | O_NONBLOCK) < 0)
FAIL_WITH_WARN(sock, "fcntl(O_NONBLOCK) at=%d", at);
int ttl = at + 1, port;
switch (af) {
case AF_INET:
if (!settosttl(sock, ttl)) return false;
port = ntohs(local.S_PORT);
break;
#ifdef ENABLE_IPV6
case AF_INET6:
if (!settosttl6(sock, ttl)) return false;
port = ntohs(local.S6PORT);
break;
#endif
default:
FAIL_POSTPONE(ENOPROTOOPT, "address family %d", af);
}
int seq = port % MAXSEQ;
if (poll_reg_fd(sock, seq) < 0)
FAIL_WITH_WARNX(1, sock, "no place in pool for sockets (at=%d)", at);
save_sequence(seq, at);
if (!save_send_ts(seq)) return false;
connect(sock, &remote.sa, addrlen); // NOLINT(bugprone-unused-return-value)
#ifdef LOGMOD
{ struct timespec now;
int rc = clock_gettime(CLOCK_MONOTONIC, &now); // LOGMOD for debug only
LOGMSG("at=%d seq=%d sock=%d: ttl=%d (ts=%lld.%09ld)", at, seq, sock, ttl, rc ? 0 : (long long)now.tv_sec, rc ? 0 : now.tv_nsec); }
#endif
/*summ*/ net_queries[QR_SUM]++; net_queries[QR_TCP]++;
return true;
#undef SET_TCP_ATTR
}
static inline void net_fill_icmp_hdr(uint16_t seq, uint8_t type, uint8_t *data, uint16_t size) {
struct _icmphdr *icmp = (struct _icmphdr *)data;
icmp->type = type;
icmp->code = 0;
icmp->sum = 0;
icmp->id = mypid;
icmp->seq = seq;
icmp->sum = sum1616((uint16_t*)data, size / 2, (size % 2) ? (bitpattern & 0xff) : 0);
LOGMSG("icmp: seq=%d id=%u", icmp->seq, icmp->id);
}
static inline bool net_fill_udp_hdr(uint16_t seq, uint8_t *data, uint16_t size
#ifdef IP_HDRINCL
, struct _iphdr *ip
#endif
) {
struct udphdr *udp = (struct udphdr *)data;
udp->uh_sum = 0;
udp->uh_ulen = htons(size);
if (remoteport < 0)
SET_UDP_UH_PORTS(udp, portpid, LO_UDPPORT + seq)
else
SET_UDP_UH_PORTS(udp, LO_UDPPORT + seq, remoteport);
LOGMSG("udp: seq=%d port=%u", seq, ntohs(udp->uh_dport));
switch (af) {
case AF_INET:
#ifdef IP_HDRINCL
if (ip->saddr) { // checksum is not mandatory, calculate if source address is known
uint16_t sum = udpsum16(ip, udp, udp->uh_ulen, size);
udp->uh_sum = sum ? sum : 0xffff;
}
#endif
return true;
#ifdef ENABLE_IPV6
case AF_INET6: { // checksumming by kernel
int opt = 6;
if (setsockopt(sendsock, IPPROTO_IPV6, IPV6_CHECKSUM, &opt, sizeof(opt)))
FAIL_WITH_WARN(sendsock, "setsockopt6(sock=%d, IPV6_CHECKSUM)", sendsock);
} return true;
#endif
} return false;
}
// Send packet for hop 'at'
static bool net_send_icmp_udp(int at) {
if (packetsize < MINPACKET) packetsize = MINPACKET;
else if (packetsize > MAXPACKET) packetsize = MAXPACKET;
static uint8_t packet[MAXPACKET];
memset(packet, bitpattern, sizeof(packet));
int iphsize = 0, echotype = 0, salen = 0;
#ifdef IP_HDRINCL
struct _iphdr *ip = (struct _iphdr *)packet;
#endif
int ttl = at + 1;
switch (af) {
case AF_INET:
#ifdef IP_HDRINCL
iphsize = sizeof(struct _iphdr);
ip->ver = 4;
ip->ihl = 5;
ip->tos = tos;
ip->len = IPLEN_RAW(packetsize);
ip->id = 0;
ip->frag = 0;
ip->ttl = ttl;
ip->proto = mtrtype;
ip->sum = 0;
// BSD needs the source IPv4 address here
addr_copy(&ip->saddr, &lsa.S_ADDR);
addr_copy(&ip->daddr, &rsa.S_ADDR);
#else
if (!settosttl(sendsock, ttl)) return false;
iphsize = 0;
#endif
echotype = ICMP_ECHO;
salen = sizeof(struct sockaddr_in);
break;
#ifdef ENABLE_IPV6
case AF_INET6:
if (!settosttl6(sendsock, ttl)) return false;
iphsize = 0;
echotype = ICMP6_ECHO_REQUEST;
salen = sizeof(struct sockaddr_in6);
break;
#endif
default:
FAIL_POSTPONE(ENOPROTOOPT, "address family %d", af);
}
int seq = new_sequence(at);
uint8_t *ipdata = packet + iphsize;
uint16_t ipdatasize = packetsize - iphsize;
switch (mtrtype) {
case IPPROTO_ICMP:
net_fill_icmp_hdr(seq, echotype, ipdata, ipdatasize);
break;
case IPPROTO_UDP:
if (!net_fill_udp_hdr(seq, ipdata, ipdatasize
#ifdef IP_HDRINCL
, ip
#endif
)) return false;
break;
default:
FAIL_POSTPONE(EPROTONOSUPPORT, "protocol type %d", mtrtype);
}
if (!save_send_ts(seq)) return false;
if (sendto(sendsock, packet, packetsize, 0, &rsa.sa, salen) < 0) {
int rc = errno; const char *dst = strlongip(remote_ipaddr); errno = rc;
FAIL_WITH_WARN(sendsock, "sendto(%s)", dst ? dst : "");
}
/*summ*/ net_queries[QR_SUM]++; (mtrtype == IPPROTO_ICMP) ? net_queries[QR_ICMP]++ : net_queries[QR_UDP];
return true;
}
static void stats(int at, timemsec_t curr) {
double curr_f = msec2float(curr);
LOGMSG("prev=%lld.%09ld curr=%lld.%09ld", (long long)host[at].last.ms, host[at].last.frac, (long long)curr.ms, curr.frac);
if (host[at].recv < 1) {
host[at].best = host[at].worst = curr;
host[at].mean = curr_f;
host[at].var = host[at].jitter = host[at].jworst = host[at].jinta = host[at].avg = 0;
} else {
double jitter = float_sub_msec(curr, host[at].last);
host[at].jitter = (jitter < 0) ? -jitter : jitter; // abs()
}
host[at].last = curr;
if (mseccmp(curr, host[at].best, <))
host[at].best = curr;
if (mseccmp(curr, host[at].worst, >))
host[at].worst = curr;
if (host[at].jitter > host[at].jworst)
host[at].jworst = host[at].jitter;
host[at].recv++;
double davg = curr_f - host[at].avg;
host[at].avg += davg / host[at].recv;
host[at].var += davg * (curr_f - host[at].avg);
host[at].javg += (host[at].jitter - host[at].javg) / host[at].recv;
host[at].jinta += (host[at].jitter - host[at].jinta) / 16; /* RFC1889 A.8 */
if (host[at].recv > 1) {
double inv_recv = 1 / (double)host[at].recv;
host[at].mean = pow(host[at].mean, 1 - inv_recv) * pow(curr_f, inv_recv);
}
host[at].up = true;
host[at].transit = false;
if (cache_mode)
host[at].seen = time(NULL);
}
static int addr2ndx(int hop, const t_ipaddr *addr) { // return index of 'addr' at 'hop', otherwise -1
for (int i = 0; i < MAXPATH; i++)
if (addr_equal(&IP_AT_NDX(hop, i), addr))
return i;
return -1;
}
int at2next(int hop) { // return first free slot at 'hop', otherwise -1
for (int i = 0; i < MAXPATH; i++)
if (!addr_exist(&IP_AT_NDX(hop, i)))
return i;
return -1;
}
// set new ip-addr and clear associated data
static void set_new_addr(int at, int ndx, const t_ipaddr *ipaddr MPLSFNTAIL(const mpls_data_t *mpls)) {
addr_copy(&IP_AT_NDX(at, ndx), ipaddr);
#ifdef WITH_MPLS
mpls ? memcpy(&MPLS_AT_NDX(at, ndx), mpls, sizeof(mpls_data_t))
: memset(&MPLS_AT_NDX(at, ndx), 0, sizeof(mpls_data_t));
#endif
if (QPTR_AT_NDX(at, ndx)) {
free(QPTR_AT_NDX(at, ndx));
QPTR_AT_NDX(at, ndx) = NULL;
}
if (RPTR_AT_NDX(at, ndx)) {
free(RPTR_AT_NDX(at, ndx));
RPTR_AT_NDX(at, ndx) = NULL;
}
#ifdef WITH_IPINFO
if (QTXT_AT_NDX(at, ndx)) {
free(QTXT_AT_NDX(at, ndx));
QTXT_AT_NDX(at, ndx) = NULL;
}
for (int i = 0; i < MAX_TXT_ITEMS; i++) {
if (RTXT_AT_NDX(at, ndx, i)) {
free(RTXT_AT_NDX(at, ndx, i));
RTXT_AT_NDX(at, ndx, i) = NULL;
}
}
#endif
}
// Got a return
static int net_stat(unsigned port, const void *addr, struct timespec *recv_at, int reason MPLSFNTAIL(const mpls_data_t *mpls)) {
unsigned seq = port % MAXSEQ;
if (!seqlist[seq].transit)
return true;
#ifdef WITH_MPLS
LOGMSG("at=%d seq=%d (labels=%d)", seqlist[seq].at, seq, mpls ? mpls->n : 0);
#else
LOGMSG("at=%d seq=%d", seqlist[seq].at, seq);
#endif
seqlist[seq].transit = false;
int at = seqlist[seq].at;
if (reason == RE_UNREACH) {
if (at < stopper)
stopper = at; // set stopper
} else if (stopper == at)
stopper = MAXHOST; // clear stopper
if (at > stopper) // return unless reachable
return true;
t_ipaddr copy;
addr_copy(©, addr); // can it be overwritten?
int ndx = addr2ndx(at, ©);
if (ndx < 0) { // new one
ndx = at2next(at);
if (ndx < 0) { // no free slots? warn about it, and change the last one
WARNX("MAXPATH=%d is exceeded at hop=%d", MAXPATH, at);
ndx = MAXPATH - 1;
}
set_new_addr(at, ndx, © MPLSFNTAIL(mpls));
#ifdef OUTPUT_FORMAT_RAW
if (enable_raw)
raw_rawhost(at, &IP_AT_NDX(at, ndx));
#endif
}
#ifdef WITH_MPLS
else if (mpls && memcmp(&MPLS_AT_NDX(at, ndx), mpls, sizeof(mpls_data_t))) {
LOGMSG("update mpls at=%d ndx=%d (labels=%d)", at, ndx, mpls->n);
memcpy(&MPLS_AT_NDX(at, ndx), mpls, sizeof(mpls_data_t));
}
#endif
struct timespec tv, stated_now;
if (!recv_at) { // try again
recv_at = &stated_now;
if (clock_gettime(CLOCK_MONOTONIC, recv_at) < 0)
FAIL_CLOCK_GETTIME;
}
timespecsub(recv_at, &seqlist[seq].time, &tv);
timemsec_t curr = { .ms = time2msec(tv), .frac = time2mfrac(tv) };
stats(at, curr);
#ifdef CURSESMODE
int n = seqlist[seq].saved_seq - host[at].saved_seq_offset;
if ((n >= 0) && (n <= SAVED_PINGS))
host[at].saved[n] = time2usec(tv);
#endif
#ifdef OUTPUT_FORMAT_RAW
if (enable_raw)
raw_rawping(at, time2usec(tv));
#endif
return true;
}
#ifdef WITH_MPLS
static inline bool mplslike(int psize, int hsize) { return enable_mpls & ((psize - hsize) >= MPLSMIN); }
static mpls_data_t *decodempls(const uint8_t *data, int sz) {
static mpls_data_t mplsdata;
// given: icmpext_struct(4) icmpext_object(4) label(4) [label(4) ...]
static const size_t mplsoff = MPLSMIN - (IES_SZ + IEO_SZ + LAB_SZ);
static const size_t ieomin = IEO_SZ + LAB_SZ;
if (sz < MPLSMIN) {
LOGMSG("got %d bytes of data, whereas mpls min is %d", sz, MPLSMIN);
return NULL;
}
int off = mplsoff; // at least 12bytes ahead: icmp_ext_struct(4) icmp_ext_object(4) label(4) [label(4) ...]
// icmp extension structure
struct icmpext_struct *ies = (struct icmpext_struct *)&data[off];
if ((ies->ver != ICMP_EXT_VER) || ies->res || !ies->sum) {
LOGMSG("got ver=%d res=%d sum=%d, expected ver=%d res=0 sum!=0", ies->ver, ies->res, ies->sum, ICMP_EXT_VER);
return NULL;
}
off += IES_SZ;
// icmp extension object
struct icmpext_object *ieo = (struct icmpext_object *)&data[off];
ieo->len = ntohs(ieo->len);
if ((ieo->len < ieomin) || (ieo->class != ICMP_EXT_CLASS_MPLS) || (ieo->type != ICMP_EXT_TYPE_MPLS)) {
LOGMSG("got len=%d class=%d type=%d, expected len>=%zd class=%d type=%d",
ieo->len, ieo->class, ieo->type, ieomin, ICMP_EXT_CLASS_MPLS, ICMP_EXT_TYPE_MPLS);
return NULL;
}
uint8_t n = (ieo->len - IEO_SZ) / LAB_SZ;
off += IEO_SZ;
// limit number of MPLS labels
if (n > MAXLABELS) {
LOGMSG("got %d MPLS labels, limit=%d", n, MAXLABELS);
n = MAXLABELS;
}
memset(&mplsdata, 0, sizeof(mplsdata));
// mpls labels
while ((mplsdata.n < n) && ((off + LAB_SZ) <= sz)) {
mplsdata.label[mplsdata.n++].u32 = ntohl(*(uint32_t*)&data[off]);
off += LAB_SZ;
}
return &mplsdata;
}
#endif
void net_icmp_parse(struct timespec *recv_at) {
#define ICMPSEQID { seq = icmp->seq; if (icmp->id != (uint16_t)mypid) \
LOGRET("icmp(myid=%u): got unknown id=%u (type=%u seq=%u)", mypid, icmp->id, icmp->type, seq); }
uint8_t packet[MAXPACKET];
struct sockaddr_storage sa_in;
ssize_t sz = recvfrom(recvsock, packet, MAXPACKET, 0, (struct sockaddr *)&sa_in, &sa_len);
LOGMSG("got %zd bytes", sz);
if (sz < hdr_minsz)
LOGRET("incorrect packet size %zd [af=%d proto=%d minsize=%zd]", sz, af, mtrtype, hdr_minsz);
struct _icmphdr *icmp = (struct _icmphdr *)(packet + iphdr_sz);
uint8_t *data = ((uint8_t*)icmp) + ipicmphdr_sz;
#ifdef WITH_MPLS
bool mplson = false;
#endif
int seq = -1, reason = -1;
switch (mtrtype) {
case IPPROTO_ICMP: {
if (icmp->type == echo_reply) {
reason = RE_PONG;
ICMPSEQID;
} else if ((icmp->type == time_exceed) || (icmp->type == dst_unreach)) {
if (sz < minfailsz)
LOGRET("incorrect packet size %zd [af=%d proto=%d expect>=%zd]", sz, af, mtrtype, minfailsz);
reason = (icmp->type == time_exceed) ? RE_EXCEED : RE_UNREACH;
icmp = (struct _icmphdr *)data;
ICMPSEQID;
#ifdef WITH_MPLS
mplson = mplslike(sz, data - packet);
#endif
}
#ifdef WITH_MPLS
LOGMSG("icmp seq=%d type=%d mpls=%d", seq, icmp->type, mplson);
#else
LOGMSG("icmp seq=%d type=%d", seq, icmp->type);
#endif
/*summ*/ net_replies[QR_ICMP]++;
} break;
case IPPROTO_UDP: {
struct udphdr *uh = (struct udphdr *)data;
if (remoteport < 0) {
if (ntohs(uh->uh_sport) != portpid)
return;
seq = ntohs(uh->uh_dport);
} else {
if (ntohs(uh->uh_dport) != remoteport)
return;
seq = ntohs(uh->uh_sport);
}
seq -= LO_UDPPORT;
#ifdef WITH_MPLS
mplson = mplslike(sz, data - packet);
LOGMSG("udp seq=%d id=%d mpls=%d", seq, portpid, mplson);
#else
LOGMSG("udp seq=%d id=%d", seq, portpid);
#endif
/*summ*/ net_replies[QR_UDP]++;
} break;
case IPPROTO_TCP: {
struct tcphdr *th = (struct tcphdr *)data;
seq = ntohs(th->th_sport);
#ifdef WITH_MPLS
mplson = mplslike(sz, data - packet);
LOGMSG("tcp seq=%d mpls=%d", seq, mplson);
#else
LOGMSG("tcp seq=%d", seq);
#endif
/*summ*/ net_replies[QR_TCP]++;
} break;
default: LOGRET("Unsupported proto %d", mtrtype);
} /*end of switch(mtrtype)*/
/*summ*/ net_replies[QR_SUM]++;
if (seq >= 0) net_stat(seq, ((uint8_t*)&sa_in) + sa_addr_offset, recv_at,
reason MPLSFNTAIL(mplson ? decodempls(data, sz - (data - packet)) : NULL));
#undef ICMPSEQID
}
const char *net_elem(int at, char c) {
static char elemstr[16];
int ival = -1;
switch (c) {
case 'D': // Dropped Packets
ival = host[at].sent - host[at].recv - (int)host[at].transit; break;
case 'R': // Received Packets
ival = host[at].recv; break;
case 'S': // Sent Packets
ival = host[at].sent; break;
}
if (ival >= 0) {
snprintf(elemstr, sizeof(elemstr), "%d", ival);
return elemstr;
}
double val;
char *suffix = NULL;
switch (c) {
case 'N': // Newest RTT(msec)
val = msec2float(host[at].last); break;
case 'B': // Min/Best RTT(msec)
val = msec2float(host[at].best); break;
case 'A': // Average RTT(msec)
val = host[at].avg; break;
case 'W': // Max/Worst RTT(msec)
val = msec2float(host[at].worst); break;
case 'G': // Geometric Mean
val = host[at].mean; break;
case 'L': { // Loss Ratio
int known = host[at].sent - (int)host[at].transit; // transit ? 1 : 0;
val = known ? (100 - (100.0 * host[at].recv / known)) : 0;
suffix = "%";
} break;
case 'V': { // Standard Deviation
int re = host[at].recv - 1;
val = (re > 0) ? sqrt(host[at].var / re) : 0;
} break;
case 'J': // Current Jitter
val = host[at].jitter; break;
case 'M': // Jitter Mean/Avg
val = host[at].javg; break;
case 'X': // Worst Jitter
val = host[at].jworst; break;
case 'I': // Interarrival Jitter
val = host[at].jinta; break;
default: return NULL;
}
snprintf(elemstr, sizeof(elemstr), "%.*f%s", val2len(val), val, suffix ? suffix : "");
return elemstr;
}
int net_max(void) {
int max = 0;
for (int at = 0; at < maxTTL; at++) {
if (addr_equal(&CURRENT_IP(at), remote_ipaddr)) {
max = at + 1;
if (endpoint_mode)
fstTTL = max;
break;
} else if (addr_exist(&CURRENT_IP(at))) {
max = at + 2;
if (endpoint_mode)
fstTTL = max - 1; // -1: show previous known hop
}
}
if (max > maxTTL)
max = maxTTL;
return max;
}
inline int net_min(void) { return (fstTTL - 1); }
inline void net_end_transit(void) { for (int at = 0; at < MAXHOST; at++) host[at].transit = false; }
int net_send_batch(void) {
if (batch_at < fstTTL) {
// Randomize bit-pattern and packet-size if requested
bitpattern = (cbitpattern < 0) ? RANDUNIFORM(256) : cbitpattern;
if (cpacketsize < 0) {
int base = -cpacketsize - MINPACKET;
packetsize = base ? (MINPACKET + RANDUNIFORM(base)) : MINPACKET;
} else
packetsize = cpacketsize;
}
{ // Send packet if needed
bool ping = true;
if (cache_mode)
if (host[batch_at].up && (host[batch_at].seen > 0))
if ((time(NULL) - host[batch_at].seen) <= cache_timeout)
ping = false;
if (ping && !( (mtrtype == IPPROTO_TCP) ?
net_send_tcp(batch_at) : net_send_icmp_udp(batch_at) ))
LOG_RE(-1, "failed");
}
{ // Calculate rc for caller
int n_unknown = 0;
for (int at = fstTTL - 1; at < batch_at; at++) {
if (!addr_exist(&CURRENT_IP(at)))
n_unknown++;
if (addr_equal(&CURRENT_IP(at), remote_ipaddr))
n_unknown = MAXHOST; // Make sure we drop into "we should restart"
}
if (addr_equal(&CURRENT_IP(batch_at), remote_ipaddr) // success in reaching target
|| (n_unknown > MAX_UNKNOWN_HOSTS) // fail in consecuitive MAX_UNKNOWN_HOSTS
|| (batch_at >= (maxTTL - 1)) // or reach limit
|| (batch_at >= stopper)) { // or learnt unreachable
numhosts = batch_at + 1;
batch_at = fstTTL - 1;
LOGMSG("stop at hop #%d", numhosts);
return 1;
}
}
batch_at++;
return 0;
}
static void net_sock_close(void) {
CLOSE(sendsock4);
CLOSE(recvsock4);
#ifdef ENABLE_IPV6
CLOSE(sendsock6_icmp);
CLOSE(sendsock6_udp);
CLOSE(recvsock6);
#endif
}
#ifdef LIBCAP
#include <sys/capability.h>
void set_rawcap_flag(cap_flag_value_t flag) {
static const cap_value_t cap_net_raw = CAP_NET_RAW;
cap_t proc = cap_get_proc();
if (proc) {
cap_flag_value_t perm = flag;
cap_get_flag(proc, cap_net_raw, CAP_PERMITTED, &perm);
if (perm == flag) {
if (cap_set_flag(proc, CAP_EFFECTIVE, 1, &cap_net_raw, flag) < 0) WARN("cap_set_flag");
else if (cap_set_proc(proc) < 0) WARN("cap_set_proc");
}
cap_free(proc);
} else WARN("cap_get_proc");
}
#define RAWCAP_ON set_rawcap_flag(CAP_SET)
#define RAWCAP_OFF set_rawcap_flag(CAP_CLEAR)
#else
#define RAWCAP_ON
#define RAWCAP_OFF
#endif
static int net_socket(int domain, int type, int proto, const char *what) {
RAWCAP_ON;
int sock = socket(domain, type, proto);
if (sock < 0) warn("%s: %s", __func__, what); else /*summ*/ sum_sock[0]++;
RAWCAP_OFF;
if (sock < 0) net_sock_close();
return sock;
}
bool net_open(void) {
// mandatory ipv4
RAWCAP_ON;
sendsock4 = socket(AF_INET, SOCK_RAW, IPPROTO_RAW);
RAWCAP_OFF;
if (sendsock4 < 0) // backup
if ((sendsock4 = net_socket(AF_INET, SOCK_RAW, IPPROTO_ICMP, "send socket")) < 0)
return false;
if ((recvsock4 = net_socket(AF_INET, SOCK_RAW, IPPROTO_ICMP, "recv socket")) < 0)
return false;
#ifdef ENABLE_IPV6
// optional ipv6: to not fail
RAWCAP_ON;
recvsock6 = socket(AF_INET6, SOCK_RAW, IPPROTO_ICMPV6);