/* $OpenBSD: check_icmp.c,v 1.37 2014/10/25 03:23:49 lteo Exp $ */ /* * Copyright (c) 2006 Pierre-Yves Ritschard * * Permission to use, copy, modify, and distribute this software for any * purpose with or without fee is hereby granted, provided that the above * copyright notice and this permission notice appear in all copies. * * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include "relayd.h" void icmp_setup(struct relayd *, struct ctl_icmp_event *, int); void check_icmp_add(struct ctl_icmp_event *, int, struct timeval *, void (*)(int, short, void *)); int icmp_checks_done(struct ctl_icmp_event *); void icmp_checks_timeout(struct ctl_icmp_event *, enum host_error); void send_icmp(int, short, void *); void recv_icmp(int, short, void *); int in_cksum(u_short *, int); void icmp_setup(struct relayd *env, struct ctl_icmp_event *cie, int af) { int proto = IPPROTO_ICMP; if (af == AF_INET6) proto = IPPROTO_ICMPV6; if ((cie->s = socket(af, SOCK_RAW, proto)) < 0) fatal("icmp_setup: socket"); socket_set_blockmode(cie->s, BM_NONBLOCK); cie->env = env; cie->af = af; } void icmp_init(struct relayd *env) { icmp_setup(env, &env->sc_icmp_send, AF_INET); icmp_setup(env, &env->sc_icmp_recv, AF_INET); icmp_setup(env, &env->sc_icmp6_send, AF_INET6); icmp_setup(env, &env->sc_icmp6_recv, AF_INET6); env->sc_id = getpid() & 0xffff; } void schedule_icmp(struct relayd *env, struct host *host) { host->last_up = host->up; host->flags &= ~(F_CHECK_SENT|F_CHECK_DONE); if (((struct sockaddr *)&host->conf.ss)->sa_family == AF_INET) env->sc_has_icmp = 1; else env->sc_has_icmp6 = 1; } void check_icmp_add(struct ctl_icmp_event *cie, int flags, struct timeval *start, void (*fn)(int, short, void *)) { struct timeval tv; if (start != NULL) bcopy(start, &cie->tv_start, sizeof(cie->tv_start)); bcopy(&cie->env->sc_timeout, &tv, sizeof(tv)); getmonotime(&cie->tv_start); event_del(&cie->ev); event_set(&cie->ev, cie->s, EV_TIMEOUT|flags, fn, cie); event_add(&cie->ev, &tv); } void check_icmp(struct relayd *env, struct timeval *tv) { if (env->sc_has_icmp) { check_icmp_add(&env->sc_icmp_recv, EV_READ, tv, recv_icmp); check_icmp_add(&env->sc_icmp_send, EV_WRITE, tv, send_icmp); } if (env->sc_has_icmp6) { check_icmp_add(&env->sc_icmp6_recv, EV_READ, tv, recv_icmp); check_icmp_add(&env->sc_icmp6_send, EV_WRITE, tv, send_icmp); } } int icmp_checks_done(struct ctl_icmp_event *cie) { struct table *table; struct host *host; TAILQ_FOREACH(table, cie->env->sc_tables, entry) { if (table->conf.flags & F_DISABLE || table->conf.check != CHECK_ICMP) continue; TAILQ_FOREACH(host, &table->hosts, entry) { if (((struct sockaddr *)&host->conf.ss)->sa_family != cie->af) continue; if (!(host->flags & F_CHECK_DONE)) return (0); } } return (1); } void icmp_checks_timeout(struct ctl_icmp_event *cie, enum host_error he) { struct table *table; struct host *host; TAILQ_FOREACH(table, cie->env->sc_tables, entry) { if (table->conf.flags & F_DISABLE || table->conf.check != CHECK_ICMP) continue; TAILQ_FOREACH(host, &table->hosts, entry) { if (((struct sockaddr *)&host->conf.ss)->sa_family != cie->af) continue; if (!(host->flags & (F_CHECK_DONE|F_DISABLE))) { host->up = HOST_DOWN; hce_notify_done(host, he); } } } } void send_icmp(int s, short event, void *arg) { struct ctl_icmp_event *cie = arg; struct table *table; struct host *host; struct sockaddr *to; struct icmp *icp; struct icmp6_hdr *icp6; ssize_t r; u_char packet[ICMP_BUF_SIZE]; socklen_t slen; int i = 0, ttl, mib[4]; size_t len; u_int32_t id; if (event == EV_TIMEOUT) { icmp_checks_timeout(cie, HCE_ICMP_WRITE_TIMEOUT); return; } bzero(&packet, sizeof(packet)); icp = (struct icmp *)packet; icp6 = (struct icmp6_hdr *)packet; if (cie->af == AF_INET) { icp->icmp_type = ICMP_ECHO; icp->icmp_code = 0; icp->icmp_id = htons(cie->env->sc_id); icp->icmp_cksum = 0; slen = sizeof(struct sockaddr_in); } else { icp6->icmp6_type = ICMP6_ECHO_REQUEST; icp6->icmp6_code = 0; icp6->icmp6_cksum = 0; icp6->icmp6_id = htons(cie->env->sc_id); slen = sizeof(struct sockaddr_in6); } TAILQ_FOREACH(table, cie->env->sc_tables, entry) { if (table->conf.check != CHECK_ICMP || table->conf.flags & F_DISABLE) continue; TAILQ_FOREACH(host, &table->hosts, entry) { if (host->flags & (F_DISABLE | F_CHECK_SENT) || host->conf.parentid) continue; if (((struct sockaddr *)&host->conf.ss)->sa_family != cie->af) continue; i++; to = (struct sockaddr *)&host->conf.ss; id = htonl(host->conf.id); if (cie->af == AF_INET) { icp->icmp_seq = htons(i); icp->icmp_cksum = 0; icp->icmp_mask = id; icp->icmp_cksum = in_cksum((u_short *)icp, sizeof(packet)); } else { icp6->icmp6_seq = htons(i); icp6->icmp6_cksum = 0; memcpy(packet + sizeof(*icp6), &id, sizeof(id)); icp6->icmp6_cksum = in_cksum((u_short *)icp6, sizeof(packet)); } if ((ttl = host->conf.ttl) > 0) (void)setsockopt(s, IPPROTO_IP, IP_TTL, &host->conf.ttl, sizeof(int)); else { /* Revert to default TTL */ mib[0] = CTL_NET; mib[1] = cie->af; mib[2] = IPPROTO_IP; mib[3] = IPCTL_DEFTTL; len = sizeof(ttl); if (sysctl(mib, 4, &ttl, &len, NULL, 0) == 0) (void)setsockopt(s, IPPROTO_IP, IP_TTL, &ttl, sizeof(int)); } r = sendto(s, packet, sizeof(packet), 0, to, slen); if (r == -1) { if (errno == EAGAIN || errno == EINTR) goto retry; host->flags |= F_CHECK_SENT|F_CHECK_DONE; host->up = HOST_DOWN; } else if (r != sizeof(packet)) goto retry; host->flags |= F_CHECK_SENT; } } return; retry: event_again(&cie->ev, s, EV_TIMEOUT|EV_WRITE, send_icmp, &cie->tv_start, &cie->env->sc_timeout, cie); } void recv_icmp(int s, short event, void *arg) { struct ctl_icmp_event *cie = arg; u_char packet[ICMP_BUF_SIZE]; socklen_t slen; struct sockaddr_storage ss; struct icmp *icp; struct icmp6_hdr *icp6; u_int16_t icpid; struct host *host; ssize_t r; u_int32_t id; if (event == EV_TIMEOUT) { icmp_checks_timeout(cie, HCE_ICMP_READ_TIMEOUT); return; } bzero(&packet, sizeof(packet)); bzero(&ss, sizeof(ss)); slen = sizeof(ss); r = recvfrom(s, packet, sizeof(packet), 0, (struct sockaddr *)&ss, &slen); if (r == -1 || r != ICMP_BUF_SIZE) { if (r == -1 && errno != EAGAIN && errno != EINTR) log_debug("%s: receive error", __func__); goto retry; } if (cie->af == AF_INET) { icp = (struct icmp *)(packet + sizeof(struct ip)); icpid = ntohs(icp->icmp_id); id = icp->icmp_mask; } else { icp6 = (struct icmp6_hdr *)packet; icpid = ntohs(icp6->icmp6_id); memcpy(&id, packet + sizeof(*icp6), sizeof(id)); } if (icpid != cie->env->sc_id) goto retry; id = ntohl(id); host = host_find(cie->env, id); if (host == NULL) { log_warn("%s: ping for unknown host received", __func__); goto retry; } if (bcmp(&ss, &host->conf.ss, slen)) { log_warnx("%s: forged icmp packet?", __func__); goto retry; } host->up = HOST_UP; host->flags |= F_CHECK_DONE; hce_notify_done(host, HCE_ICMP_OK); if (icmp_checks_done(cie)) return; retry: event_again(&cie->ev, s, EV_TIMEOUT|EV_READ, recv_icmp, &cie->tv_start, &cie->env->sc_timeout, cie); } /* in_cksum from ping.c -- * Checksum routine for Internet Protocol family headers (C Version) * * Copyright (c) 1989, 1993 * The Regents of the University of California. All rights reserved. * * This code is derived from software contributed to Berkeley by * Mike Muuss. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. Neither the name of the University nor the names of its contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. */ int in_cksum(u_short *addr, int len) { int nleft = len; u_short *w = addr; int sum = 0; u_short answer = 0; /* * Our algorithm is simple, using a 32 bit accumulator (sum), we add * sequential 16 bit words to it, and at the end, fold back all the * carry bits from the top 16 bits into the lower 16 bits. */ while (nleft > 1) { sum += *w++; nleft -= 2; } /* mop up an odd byte, if necessary */ if (nleft == 1) { *(u_char *)(&answer) = *(u_char *)w ; sum += answer; } /* add back carry outs from top 16 bits to low 16 bits */ sum = (sum >> 16) + (sum & 0xffff); /* add hi 16 to low 16 */ sum += (sum >> 16); /* add carry */ answer = ~sum; /* truncate to 16 bits */ return (answer); }