X-Git-Url: http://git.sameswireless.fr/l2tpns.git/blobdiff_plain/74967abeca9a78ad4a1b7a953fd67988cfb67895..eecccf0b30d5ec19960633d1b62cf5f9936d1f16:/bgp.c diff --git a/bgp.c b/bgp.c index 2d411e7..36894be 100644 --- a/bgp.c +++ b/bgp.c @@ -10,8 +10,6 @@ * nor RFC2385 (which requires a kernel patch on 2.4 kernels). */ -char const *cvs_id_bgp = "$Id: bgp.c,v 1.6 2004/11/11 03:07:42 bodea Exp $"; - #include #include #include @@ -22,6 +20,7 @@ char const *cvs_id_bgp = "$Id: bgp.c,v 1.6 2004/11/11 03:07:42 bodea Exp $"; #include #include #include +#include #include "l2tpns.h" #include "bgp.h" @@ -29,12 +28,15 @@ char const *cvs_id_bgp = "$Id: bgp.c,v 1.6 2004/11/11 03:07:42 bodea Exp $"; static void bgp_clear(struct bgp_peer *peer); static void bgp_set_retry(struct bgp_peer *peer); -static void bgp_cidr(in_addr_t ip, in_addr_t mask, struct bgp_ip_prefix *pfx); static struct bgp_route_list *bgp_insert_route(struct bgp_route_list *head, struct bgp_route_list *new); +static struct bgp_route6_list *bgp_insert_route6(struct bgp_route6_list *head, + struct bgp_route6_list *new); +static void bgp_process_timers(struct bgp_peer *peer); static void bgp_free_routes(struct bgp_route_list *routes); -static char const *bgp_msg_type_str(u8 type); +static void bgp_free_routes6(struct bgp_route6_list *routes); +static char const *bgp_msg_type_str(uint8_t type); static int bgp_connect(struct bgp_peer *peer); static int bgp_handle_connect(struct bgp_peer *peer); static int bgp_write(struct bgp_peer *peer); @@ -43,10 +45,13 @@ static int bgp_handle_input(struct bgp_peer *peer); static int bgp_send_open(struct bgp_peer *peer); static int bgp_send_keepalive(struct bgp_peer *peer); static int bgp_send_update(struct bgp_peer *peer); -static int bgp_send_notification(struct bgp_peer *peer, u8 code, u8 subcode); +static int bgp_send_update6(struct bgp_peer *peer); +static int bgp_send_notification(struct bgp_peer *peer, uint8_t code, + uint8_t subcode); -static u16 our_as; +static uint16_t our_as; static struct bgp_route_list *bgp_routes = 0; +static struct bgp_route6_list *bgp_routes6 = 0; int bgp_configured = 0; struct bgp_peer *bgp_peers = 0; @@ -69,11 +74,15 @@ int bgp_setup(int as) if (!((peer->outbuf = malloc(sizeof(*peer->outbuf))) && (peer->inbuf = malloc(sizeof(*peer->inbuf))))) { - LOG(0, 0, 0, 0, "Can't allocate buffers for bgp peer (%s)\n", + LOG(0, 0, 0, "Can't allocate buffers for bgp peer (%s)\n", strerror(errno)); return 0; } + + peer->edata.type = FD_TYPE_BGP; + peer->edata.index = i; + peer->events = 0; } if (as < 1) @@ -83,13 +92,15 @@ int bgp_setup(int as) return 0; bgp_routes = 0; + bgp_routes6 = 0; bgp_configured = 0; /* set by bgp_start */ return 1; } /* start connection with a peer */ -int bgp_start(struct bgp_peer *peer, char *name, int as, int keepalive, int hold, int enable) +int bgp_start(struct bgp_peer *peer, char *name, int as, int keepalive, + int hold, struct in_addr update_source, int enable) { struct hostent *h; int ibgp; @@ -98,8 +109,8 @@ int bgp_start(struct bgp_peer *peer, char *name, int as, int keepalive, int hold char path_attrs[64]; char *p = path_attrs; in_addr_t ip; - u32 metric = htonl(BGP_METRIC); - u32 no_export = htonl(BGP_COMMUNITY_NO_EXPORT); + uint32_t metric = htonl(BGP_METRIC); + uint32_t no_export = htonl(BGP_COMMUNITY_NO_EXPORT); if (!our_as) return 0; @@ -111,13 +122,14 @@ int bgp_start(struct bgp_peer *peer, char *name, int as, int keepalive, int hold if (!(h = gethostbyname(name)) || h->h_addrtype != AF_INET) { - LOG(0, 0, 0, 0, "Can't get address for BGP peer %s (%s)\n", + LOG(0, 0, 0, "Can't get address for BGP peer %s (%s)\n", name, h ? "no address" : hstrerror(h_errno)); return 0; } memcpy(&peer->addr, h->h_addr, sizeof(peer->addr)); + peer->source_addr = update_source.s_addr; peer->as = as > 0 ? as : our_as; ibgp = peer->as == our_as; @@ -170,9 +182,9 @@ int bgp_start(struct bgp_peer *peer, char *name, int as, int keepalive, int hold { /* just our AS */ struct { - u8 type; - u8 len; - u16 value; + uint8_t type; + uint8_t len; + uint16_t value; } as_path = { BGP_PATH_ATTR_CODE_AS_PATH_AS_SEQUENCE, 1, @@ -185,15 +197,6 @@ int bgp_start(struct bgp_peer *peer, char *name, int as, int keepalive, int hold ADD_ATTRIBUTE(); - /* NEXT_HOP */ - a.flags = BGP_PATH_ATTR_FLAG_TRANS; - a.code = BGP_PATH_ATTR_CODE_NEXT_HOP; - ip = my_address; /* we're it */ - a.data.s.len = sizeof(ip); - memcpy(a.data.s.value, &ip, sizeof(ip)); - - ADD_ATTRIBUTE(); - /* MULTI_EXIT_DISC */ a.flags = BGP_PATH_ATTR_FLAG_OPTIONAL; a.code = BGP_PATH_ATTR_CODE_MULTI_EXIT_DISC; @@ -204,7 +207,7 @@ int bgp_start(struct bgp_peer *peer, char *name, int as, int keepalive, int hold if (ibgp) { - u32 local_pref = htonl(BGP_LOCAL_PREF); + uint32_t local_pref = htonl(BGP_LOCAL_PREF); /* LOCAL_PREF */ a.flags = BGP_PATH_ATTR_FLAG_TRANS; @@ -223,9 +226,28 @@ int bgp_start(struct bgp_peer *peer, char *name, int as, int keepalive, int hold ADD_ATTRIBUTE(); + /* remember the len before adding NEXT_HOP */ + peer->path_attr_len_without_nexthop = peer->path_attr_len; + + /* NEXT_HOP */ + a.flags = BGP_PATH_ATTR_FLAG_TRANS; + a.code = BGP_PATH_ATTR_CODE_NEXT_HOP; + if (config->nexthop_address) + { + ip = config->nexthop_address; + } + else + { + ip = my_address; /* we're it */ + } + a.data.s.len = sizeof(ip); + memcpy(a.data.s.value, &ip, sizeof(ip)); + + ADD_ATTRIBUTE(); + if (!(peer->path_attrs = malloc(peer->path_attr_len))) { - LOG(0, 0, 0, 0, "Can't allocate path_attrs for %s (%s)\n", + LOG(0, 0, 0, "Can't allocate path_attrs for %s (%s)\n", name, strerror(errno)); return 0; @@ -233,7 +255,54 @@ int bgp_start(struct bgp_peer *peer, char *name, int as, int keepalive, int hold memcpy(peer->path_attrs, path_attrs, peer->path_attr_len); - LOG(4, 0, 0, 0, "Initiating BGP connection to %s (routing %s)\n", + /* multiprotocol attributes initialization */ + if (config->ipv6_prefix.s6_addr[0]) + { + struct bgp_attr_mp_reach_nlri_partial mp_reach_nlri_partial; + struct bgp_attr_mp_unreach_nlri_partial mp_unreach_nlri_partial; + + a.flags = BGP_PATH_ATTR_FLAG_OPTIONAL; + a.code = BGP_PATH_ATTR_CODE_MP_REACH_NLRI; + a.data.s.len = 0; /* will be set on UPDATE */ + + mp_reach_nlri_partial.afi = htons(BGP_MP_AFI_IPv6); + mp_reach_nlri_partial.safi = BGP_MP_SAFI_UNICAST; + mp_reach_nlri_partial.reserved = 0; + mp_reach_nlri_partial.next_hop_len = 16; + + /* use the defined nexthop6, or our address in ipv6_prefix */ + if (config->nexthop6_address.s6_addr[0]) + memcpy(&mp_reach_nlri_partial.next_hop, + &config->nexthop6_address.s6_addr, 16); + else + { + /* our address is ipv6prefix::1 */ + memcpy(&mp_reach_nlri_partial.next_hop, + &config->ipv6_prefix.s6_addr, 16); + mp_reach_nlri_partial.next_hop[15] = 1; + } + + memcpy(&a.data.s.value, &mp_reach_nlri_partial, + sizeof(struct bgp_attr_mp_reach_nlri_partial)); + memcpy(&peer->mp_reach_nlri_partial, &a, + BGP_PATH_ATTR_MP_REACH_NLRI_PARTIAL_SIZE); + + a.flags = BGP_PATH_ATTR_FLAG_OPTIONAL | BGP_PATH_ATTR_FLAG_EXTLEN; + a.code = BGP_PATH_ATTR_CODE_MP_UNREACH_NLRI; + a.data.e.len = 0; /* will be set on UPDATE */ + + mp_unreach_nlri_partial.afi = htons(BGP_MP_AFI_IPv6); + mp_unreach_nlri_partial.safi = BGP_MP_SAFI_UNICAST; + + memcpy(&a.data.e.value, &mp_unreach_nlri_partial, + sizeof(struct bgp_attr_mp_unreach_nlri_partial)); + memcpy(&peer->mp_unreach_nlri_partial, &a, + BGP_PATH_ATTR_MP_UNREACH_NLRI_PARTIAL_SIZE); + } + + peer->mp_handling = HandlingUnknown; + + LOG(4, 0, 0, "Initiating BGP connection to %s (routing %s)\n", name, enable ? "enabled" : "suspended"); /* we have at least one peer configured */ @@ -261,6 +330,8 @@ static void bgp_clear(struct bgp_peer *peer) bgp_free_routes(peer->routes); peer->routes = 0; + bgp_free_routes6(peer->routes6); + peer->routes6 = 0; peer->outbuf->packet.header.len = 0; peer->outbuf->done = 0; @@ -268,13 +339,14 @@ static void bgp_clear(struct bgp_peer *peer) peer->inbuf->done = 0; peer->cli_flag = 0; + peer->events = 0; if (peer->state != peer->next_state) { peer->state = peer->next_state; peer->state_time = time_now; - LOG(4, 0, 0, 0, "BGP peer %s: state %s\n", peer->name, + LOG(4, 0, 0, "BGP peer %s: state %s\n", peer->name, bgp_state_str(peer->next_state)); } } @@ -282,14 +354,14 @@ static void bgp_clear(struct bgp_peer *peer) /* initiate a clean shutdown */ void bgp_stop(struct bgp_peer *peer) { - LOG(4, 0, 0, 0, "Terminating BGP connection to %s\n", peer->name); + LOG(4, 0, 0, "Terminating BGP connection to %s\n", peer->name); bgp_send_notification(peer, BGP_ERR_CEASE, 0); } /* drop connection (if any) and set state to Disabled */ void bgp_halt(struct bgp_peer *peer) { - LOG(4, 0, 0, 0, "Aborting BGP connection to %s\n", peer->name); + LOG(4, 0, 0, "Aborting BGP connection to %s\n", peer->name); peer->next_state = Disabled; bgp_clear(peer); } @@ -320,32 +392,39 @@ static void bgp_set_retry(struct bgp_peer *peer) bgp_halt(peer); /* give up */ } -/* convert ip/mask to CIDR notation */ -static void bgp_cidr(in_addr_t ip, in_addr_t mask, struct bgp_ip_prefix *pfx) +/* insert route into list; sorted */ +static struct bgp_route_list *bgp_insert_route(struct bgp_route_list *head, + struct bgp_route_list *new) { - int i; - u32 b; - - /* convert to prefix notation */ - pfx->len = 32; - pfx->prefix = ip; + struct bgp_route_list *p = head; + struct bgp_route_list *e = 0; - if (!mask) /* bogus */ - mask = 0xffffffff; + while (p && memcmp(&p->dest, &new->dest, sizeof(p->dest)) < 0) + { + e = p; + p = p->next; + } - for (i = 0; i < 32 && ((b = ntohl(1 << i)), !(mask & b)); i++) + if (e) + { + new->next = e->next; + e->next = new; + } + else { - pfx->len--; - pfx->prefix &= ~b; + new->next = head; + head = new; } + + return head; } -/* insert route into list; sorted */ -static struct bgp_route_list *bgp_insert_route(struct bgp_route_list *head, - struct bgp_route_list *new) +/* insert route6 into list; sorted */ +static struct bgp_route6_list *bgp_insert_route6(struct bgp_route6_list *head, + struct bgp_route6_list *new) { - struct bgp_route_list *p = head; - struct bgp_route_list *e = 0; + struct bgp_route6_list *p = head; + struct bgp_route6_list *e = 0; while (p && memcmp(&p->dest, &new->dest, sizeof(p->dest)) < 0) { @@ -374,13 +453,14 @@ static struct bgp_route_list *bgp_insert_route(struct bgp_route_list *head, * that if that route is later deleted we don't have to be concerned * about adding back the more specific one). */ -int bgp_add_route(in_addr_t ip, in_addr_t mask) +int bgp_add_route(in_addr_t ip, int prefixlen) { struct bgp_route_list *r = bgp_routes; struct bgp_route_list add; int i; - bgp_cidr(ip, mask, &add.dest); + add.dest.prefix = ip; + add.dest.len = prefixlen; add.next = 0; /* check for duplicate */ @@ -399,8 +479,8 @@ int bgp_add_route(in_addr_t ip, in_addr_t mask) /* insert into route list; sorted */ if (!(r = malloc(sizeof(*r)))) { - LOG(0, 0, 0, 0, "Can't allocate route for %s/%d (%s)\n", - inet_toa(add.dest.prefix), add.dest.len, strerror(errno)); + LOG(0, 0, 0, "Can't allocate route for %s/%d (%s)\n", + fmtaddr(add.dest.prefix, 0), add.dest.len, strerror(errno)); return 0; } @@ -413,21 +493,75 @@ int bgp_add_route(in_addr_t ip, in_addr_t mask) if (bgp_peers[i].state == Established) bgp_peers[i].update_routes = 1; - LOG(4, 0, 0, 0, "Registered BGP route %s/%d\n", inet_toa(add.dest.prefix), - add.dest.len); + LOG(4, 0, 0, "Registered BGP route %s/%d\n", + fmtaddr(add.dest.prefix, 0), add.dest.len); + + return 1; +} + +/* add route to list for peers */ +/* + * Note: same provisions as above + */ +int bgp_add_route6(struct in6_addr ip, int prefixlen) +{ + struct bgp_route6_list *r = bgp_routes6; + struct bgp_route6_list add; + int i; + char ipv6addr[INET6_ADDRSTRLEN]; + + memcpy(&add.dest.prefix, &ip.s6_addr, 16); + add.dest.len = prefixlen; + add.next = 0; + + /* check for duplicate */ + while (r) + { + i = memcmp(&r->dest, &add.dest, sizeof(r->dest)); + if (!i) + return 1; /* already covered */ + + if (i > 0) + break; + + r = r->next; + } + + /* insert into route list; sorted */ + if (!(r = malloc(sizeof(*r)))) + { + LOG(0, 0, 0, "Can't allocate route for %s/%d (%s)\n", + inet_ntop(AF_INET6, &ip, ipv6addr, INET6_ADDRSTRLEN), add.dest.len, + strerror(errno)); + + return 0; + } + + memcpy(r, &add, sizeof(*r)); + bgp_routes6 = bgp_insert_route6(bgp_routes6, r); + + /* flag established peers for update */ + for (i = 0; i < BGP_NUM_PEERS; i++) + if (bgp_peers[i].state == Established + && bgp_peers[i].mp_handling == HandleIPv6Routes) + bgp_peers[i].update_routes6 = 1; + + LOG(4, 0, 0, "Registered BGP route %s/%d\n", + inet_ntop(AF_INET6, &ip, ipv6addr, INET6_ADDRSTRLEN), add.dest.len); return 1; } /* remove route from list for peers */ -int bgp_del_route(in_addr_t ip, in_addr_t mask) +int bgp_del_route(in_addr_t ip, int prefixlen) { struct bgp_route_list *r = bgp_routes; struct bgp_route_list *e = 0; struct bgp_route_list del; int i; - bgp_cidr(ip, mask, &del.dest); + del.dest.prefix = ip; + del.dest.len = prefixlen; del.next = 0; /* find entry in routes list and remove */ @@ -462,8 +596,60 @@ int bgp_del_route(in_addr_t ip, in_addr_t mask) if (bgp_peers[i].state == Established) bgp_peers[i].update_routes = 1; - LOG(4, 0, 0, 0, "Removed BGP route %s/%d\n", inet_toa(del.dest.prefix), - del.dest.len); + LOG(4, 0, 0, "Removed BGP route %s/%d\n", + fmtaddr(del.dest.prefix, 0), del.dest.len); + + return 1; +} + +/* remove route from list for peers */ +int bgp_del_route6(struct in6_addr ip, int prefixlen) +{ + struct bgp_route6_list *r = bgp_routes6; + struct bgp_route6_list *e = 0; + struct bgp_route6_list del; + int i; + char ipv6addr[INET6_ADDRSTRLEN]; + + memcpy(&del.dest.prefix, &ip.s6_addr, 16); + del.dest.len = prefixlen; + del.next = 0; + + /* find entry in routes list and remove */ + while (r) + { + i = memcmp(&r->dest, &del.dest, sizeof(r->dest)); + if (!i) + { + if (e) + e->next = r->next; + else + bgp_routes6 = r->next; + + free(r); + break; + } + + e = r; + + if (i > 0) + r = 0; /* stop */ + else + r = r->next; + } + + /* not found */ + if (!r) + return 1; + + /* flag established peers for update */ + for (i = 0; i < BGP_NUM_PEERS; i++) + if (bgp_peers[i].state == Established + && bgp_peers[i].mp_handling == HandleIPv6Routes) + bgp_peers[i].update_routes6 = 1; + + LOG(4, 0, 0, "Removed BGP route %s/%d\n", + inet_ntop(AF_INET6, &ip, ipv6addr, INET6_ADDRSTRLEN), del.dest.len); return 1; } @@ -482,117 +668,181 @@ void bgp_enable_routing(int enable) bgp_peers[i].update_routes = 1; } - LOG(4, 0, 0, 0, "%s BGP routing\n", enable ? "Enabled" : "Suspended"); + LOG(4, 0, 0, "%s BGP routing\n", enable ? "Enabled" : "Suspended"); } -/* return a bitmask indicating if the socket should be added to the - read set (1) and or write set (2) for select */ -int bgp_select_state(struct bgp_peer *peer) +#ifdef HAVE_EPOLL +# include +#else +# include "fake_epoll.h" +#endif + +/* return a bitmask of the events required to poll this peer's fd */ +int bgp_set_poll() { - int flags = 0; + int i; if (!bgp_configured) return 0; - if (peer->state == Disabled || peer->state == Idle) - return 0; + for (i = 0; i < BGP_NUM_PEERS; i++) + { + struct bgp_peer *peer = &bgp_peers[i]; + int events = 0; - if (peer->inbuf->done < BGP_MAX_PACKET_SIZE) - flags |= 1; + if (peer->state == Disabled || peer->state == Idle) + continue; + + if (peer->inbuf->done < BGP_MAX_PACKET_SIZE) + events |= EPOLLIN; + + if (peer->state == Connect || /* connection in progress */ + peer->update_routes || /* routing updates */ + peer->outbuf->packet.header.len) /* pending output */ + events |= EPOLLOUT; - if (peer->state == Connect || /* connection in progress */ - peer->update_routes || /* routing updates */ - peer->outbuf->packet.header.len) /* pending output */ - flags |= 2; + if (peer->events != events) + { + struct epoll_event ev; - return flags; + ev.events = peer->events = events; + ev.data.ptr = &peer->edata; + epoll_ctl(epollfd, EPOLL_CTL_MOD, peer->sock, &ev); + } + } + + return 1; } -/* process bgp peer */ -int bgp_process(struct bgp_peer *peer, int readable, int writable) +/* process bgp events/timers */ +int bgp_process(uint32_t events[]) { + int i; + if (!bgp_configured) return 0; - if (*peer->name && peer->cli_flag == BGP_CLI_RESTART) - return bgp_restart(peer); + for (i = 0; i < BGP_NUM_PEERS; i++) + { + struct bgp_peer *peer = &bgp_peers[i]; - if (peer->state == Disabled) - return 1; + if (*peer->name && peer->cli_flag == BGP_CLI_RESTART) + { + bgp_restart(peer); + continue; + } - if (peer->cli_flag) - { - switch (peer->cli_flag) + if (peer->state == Disabled) + continue; + + if (peer->cli_flag) { - case BGP_CLI_SUSPEND: - if (peer->routing) + switch (peer->cli_flag) { - peer->routing = 0; - if (peer->state == Established) - peer->update_routes = 1; - } + case BGP_CLI_SUSPEND: + if (peer->routing) + { + peer->routing = 0; + if (peer->state == Established) + peer->update_routes = 1; + } - break; + break; - case BGP_CLI_ENABLE: - if (!peer->routing) - { - peer->routing = 1; - if (peer->state == Established) - peer->update_routes = 1; + case BGP_CLI_ENABLE: + if (!peer->routing) + { + peer->routing = 1; + if (peer->state == Established) + peer->update_routes = 1; + } + + break; } - break; + peer->cli_flag = 0; } - peer->cli_flag = 0; - } + /* handle empty/fill of buffers */ + if (events[i] & EPOLLOUT) + { + int r = 1; + if (peer->state == Connect) + r = bgp_handle_connect(peer); + else if (peer->outbuf->packet.header.len) + r = bgp_write(peer); - /* handle empty/fill of buffers */ - if (writable) - { - int r = 1; - if (peer->state == Connect) - r = bgp_handle_connect(peer); - else if (peer->outbuf->packet.header.len) - r = bgp_write(peer); + if (!r) + continue; + } - if (!r) - return 0; - } + if (events[i] & (EPOLLIN|EPOLLHUP)) + { + if (!bgp_read(peer)) + continue; + } - if (readable) - { - if (!bgp_read(peer)) - return 0; - } + /* process input buffer contents */ + while (peer->inbuf->done >= sizeof(peer->inbuf->packet.header) + && !peer->outbuf->packet.header.len) /* may need to queue a response */ + { + if (bgp_handle_input(peer) < 0) + continue; + } - /* process input buffer contents */ - while (peer->inbuf->done >= sizeof(peer->inbuf->packet.header) - && !peer->outbuf->packet.header.len) /* may need to queue a response */ - { - if (bgp_handle_input(peer) < 0) - return 0; + /* process pending updates */ + if (peer->update_routes + && !peer->outbuf->packet.header.len) /* ditto */ + { + if (!bgp_send_update(peer)) + continue; + } + + /* process pending IPv6 updates */ + if (peer->update_routes6 + && !peer->outbuf->packet.header.len) /* ditto */ + { + if (!bgp_send_update6(peer)) + continue; + } + + /* process timers */ + bgp_process_timers(peer); } - /* process pending updates */ - if (peer->update_routes - && !peer->outbuf->packet.header.len) /* ditto */ + return 1; +} + +/* process bgp timers only */ +void bgp_process_peers_timers() +{ + int i; + + if (!bgp_configured) + return; + + for (i = 0; i < BGP_NUM_PEERS; i++) { - if (!bgp_send_update(peer)) - return 0; + struct bgp_peer *peer = &bgp_peers[i]; + + if (peer->state == Disabled) + continue; + + bgp_process_timers(peer); } +} - /* process timers */ +static void bgp_process_timers(struct bgp_peer *peer) +{ if (peer->state == Established) { if (time_now > peer->expire_time) { - LOG(1, 0, 0, 0, "No message from BGP peer %s in %ds\n", + LOG(1, 0, 0, "No message from BGP peer %s in %ds\n", peer->name, peer->hold); bgp_send_notification(peer, BGP_ERR_HOLD_TIMER_EXP, 0); - return 0; + return; } if (time_now > peer->keepalive_time && !peer->outbuf->packet.header.len) @@ -601,17 +851,15 @@ int bgp_process(struct bgp_peer *peer, int readable, int writable) else if (peer->state == Idle) { if (time_now > peer->retry_time) - return bgp_connect(peer); + bgp_connect(peer); } else if (time_now > peer->state_time + BGP_STATE_TIME) { - LOG(1, 0, 0, 0, "%s timer expired for BGP peer %s\n", + LOG(1, 0, 0, "%s timer expired for BGP peer %s\n", bgp_state_str(peer->state), peer->name); - return bgp_restart(peer); + bgp_restart(peer); } - - return 1; } static void bgp_free_routes(struct bgp_route_list *routes) @@ -625,6 +873,17 @@ static void bgp_free_routes(struct bgp_route_list *routes) } } +static void bgp_free_routes6(struct bgp_route6_list *routes) +{ + struct bgp_route6_list *tmp; + + while ((tmp = routes)) + { + routes = tmp->next; + free(tmp); + } +} + char const *bgp_state_str(enum bgp_state state) { switch (state) @@ -641,7 +900,7 @@ char const *bgp_state_str(enum bgp_state state) return "?"; } -static char const *bgp_msg_type_str(u8 type) +static char const *bgp_msg_type_str(uint8_t type) { switch (type) { @@ -659,13 +918,15 @@ static int bgp_connect(struct bgp_peer *peer) { static int bgp_port = 0; struct sockaddr_in addr; + struct sockaddr_in source_addr; + struct epoll_event ev; if (!bgp_port) { struct servent *serv; if (!(serv = getservbyname("bgp", "tcp"))) { - LOG(0, 0, 0, 0, "Can't get bgp service (%s)\n", strerror(errno)); + LOG(0, 0, 0, "Can't get bgp service (%s)\n", strerror(errno)); return 0; } @@ -674,16 +935,34 @@ static int bgp_connect(struct bgp_peer *peer) if ((peer->sock = socket(PF_INET, SOCK_STREAM, IPPROTO_TCP)) < 0) { - LOG(0, 0, 0, 0, "Can't create a socket for BGP peer %s (%s)\n", + LOG(0, 0, 0, "Can't create a socket for BGP peer %s (%s)\n", peer->name, strerror(errno)); peer->state = peer->next_state = Disabled; return 0; } + /* add to poll set */ + ev.events = peer->events = EPOLLOUT; + ev.data.ptr = &peer->edata; + epoll_ctl(epollfd, EPOLL_CTL_ADD, peer->sock, &ev); + /* set to non-blocking */ fcntl(peer->sock, F_SETFL, fcntl(peer->sock, F_GETFL, 0) | O_NONBLOCK); + /* set source address */ + memset(&source_addr, 0, sizeof(source_addr)); + source_addr.sin_family = AF_INET; + source_addr.sin_addr.s_addr = peer->source_addr; /* defaults to INADDR_ANY */ + if (bind(peer->sock, (struct sockaddr *) &source_addr, sizeof(source_addr)) < 0) + { + LOG(1, 0, 0, "Can't set source address to %s: %s\n", + inet_ntoa(source_addr.sin_addr), strerror(errno)); + + bgp_set_retry(peer); + return 0; + } + /* try connect */ memset(&addr, 0, sizeof(addr)); addr.sin_family = AF_INET; @@ -697,7 +976,7 @@ static int bgp_connect(struct bgp_peer *peer) if (errno != EINPROGRESS) { - LOG(1, 0, 0, 0, "Can't connect to BGP peer %s (%s)\n", + LOG(1, 0, 0, "Can't connect to BGP peer %s (%s)\n", inet_ntoa(addr.sin_addr), strerror(errno)); bgp_set_retry(peer); @@ -707,7 +986,7 @@ static int bgp_connect(struct bgp_peer *peer) peer->state = Connect; peer->state_time = time_now; - LOG(4, 0, 0, 0, "BGP peer %s: state Connect\n", peer->name); + LOG(4, 0, 0, "BGP peer %s: state Connect\n", peer->name); return 1; } @@ -715,7 +994,7 @@ static int bgp_connect(struct bgp_peer *peer) peer->state_time = time_now; peer->retry_time = peer->retry_count = 0; - LOG(4, 0, 0, 0, "BGP peer %s: state Active\n", inet_ntoa(addr.sin_addr)); + LOG(4, 0, 0, "BGP peer %s: state Active\n", inet_ntoa(addr.sin_addr)); return bgp_send_open(peer); } @@ -724,11 +1003,11 @@ static int bgp_connect(struct bgp_peer *peer) static int bgp_handle_connect(struct bgp_peer *peer) { int err = 0; - int len = sizeof(int); + socklen_t len = sizeof(int); getsockopt(peer->sock, SOL_SOCKET, SO_ERROR, &err, &len); if (err) { - LOG(1, 0, 0, 0, "Can't connect to BGP peer %s (%s)\n", peer->name, + LOG(1, 0, 0, "Can't connect to BGP peer %s (%s)\n", peer->name, strerror(err)); bgp_set_retry(peer); @@ -738,7 +1017,7 @@ static int bgp_handle_connect(struct bgp_peer *peer) peer->state = Active; peer->state_time = time_now; - LOG(4, 0, 0, 0, "BGP peer %s: state Active\n", peer->name); + LOG(4, 0, 0, "BGP peer %s: state Active\n", peer->name); return bgp_send_open(peer); } @@ -759,9 +1038,9 @@ static int bgp_write(struct bgp_peer *peer) return 1; if (errno == EPIPE) - LOG(1, 0, 0, 0, "Connection to BGP peer %s closed\n", peer->name); + LOG(1, 0, 0, "Connection to BGP peer %s closed\n", peer->name); else - LOG(1, 0, 0, 0, "Can't write to BGP peer %s (%s)\n", peer->name, + LOG(1, 0, 0, "Can't write to BGP peer %s (%s)\n", peer->name, strerror(errno)); bgp_set_retry(peer); @@ -774,7 +1053,7 @@ static int bgp_write(struct bgp_peer *peer) return 1; } - LOG(4, 0, 0, 0, "Sent %s to BGP peer %s\n", + LOG(4, 0, 0, "Sent %s to BGP peer %s\n", bgp_msg_type_str(peer->outbuf->packet.header.type), peer->name); peer->outbuf->packet.header.len = 0; @@ -794,7 +1073,7 @@ static int bgp_write(struct bgp_peer *peer) peer->state = peer->next_state; peer->state_time = time_now; - LOG(4, 0, 0, 0, "BGP peer %s: state %s\n", peer->name, + LOG(4, 0, 0, "BGP peer %s: state %s\n", peer->name, bgp_state_str(peer->state)); } @@ -811,7 +1090,7 @@ static int bgp_read(struct bgp_peer *peer) { if (!r) { - LOG(1, 0, 0, 0, "Connection to BGP peer %s closed\n", peer->name); + LOG(1, 0, 0, "Connection to BGP peer %s closed\n", peer->name); } else { @@ -821,7 +1100,7 @@ static int bgp_read(struct bgp_peer *peer) if (errno == EAGAIN) return 1; - LOG(1, 0, 0, 0, "Can't read from BGP peer %s (%s)\n", peer->name, + LOG(1, 0, 0, "Can't read from BGP peer %s (%s)\n", peer->name, strerror(errno)); } @@ -841,7 +1120,7 @@ static int bgp_handle_input(struct bgp_peer *peer) if (len > BGP_MAX_PACKET_SIZE) { - LOG(1, 0, 0, 0, "Bad header length from BGP %s\n", peer->name); + LOG(1, 0, 0, "Bad header length from BGP %s\n", peer->name); bgp_send_notification(peer, BGP_ERR_HEADER, BGP_ERR_HDR_BAD_LEN); return 0; } @@ -849,7 +1128,7 @@ static int bgp_handle_input(struct bgp_peer *peer) if (peer->inbuf->done < len) return 0; - LOG(4, 0, 0, 0, "Received %s from BGP peer %s\n", + LOG(4, 0, 0, "Received %s from BGP peer %s\n", bgp_msg_type_str(p->header.type), peer->name); switch (p->header.type) @@ -857,13 +1136,20 @@ static int bgp_handle_input(struct bgp_peer *peer) case BGP_MSG_OPEN: { struct bgp_data_open data; + int hold; int i; + off_t param_offset, capability_offset; + struct bgp_opt_param *param; + uint8_t capabilities_len; + char *capabilities = NULL; + struct bgp_capability *capability; + struct bgp_mp_cap_param *mp_cap; for (i = 0; i < sizeof(p->header.marker); i++) { if ((unsigned char) p->header.marker[i] != 0xff) { - LOG(1, 0, 0, 0, "Invalid marker from BGP peer %s\n", + LOG(1, 0, 0, "Invalid marker from BGP peer %s\n", peer->name); bgp_send_notification(peer, BGP_ERR_HEADER, @@ -875,7 +1161,7 @@ static int bgp_handle_input(struct bgp_peer *peer) if (peer->state != OpenSent) { - LOG(1, 0, 0, 0, "OPEN from BGP peer %s in %s state\n", + LOG(1, 0, 0, "OPEN from BGP peer %s in %s state\n", peer->name, bgp_state_str(peer->state)); bgp_send_notification(peer, BGP_ERR_FSM, 0); @@ -886,7 +1172,7 @@ static int bgp_handle_input(struct bgp_peer *peer) if (data.version != BGP_VERSION) { - LOG(1, 0, 0, 0, "Bad version (%d) sent by BGP peer %s\n", + LOG(1, 0, 0, "Bad version (%d) sent by BGP peer %s\n", (int) data.version, peer->name); bgp_send_notification(peer, BGP_ERR_OPEN, BGP_ERR_OPN_VERSION); @@ -895,7 +1181,7 @@ static int bgp_handle_input(struct bgp_peer *peer) if (ntohs(data.as) != peer->as) { - LOG(1, 0, 0, 0, "Bad AS sent by BGP peer %s (got %d, " + LOG(1, 0, 0, "Bad AS sent by BGP peer %s (got %d, " "expected %d)\n", peer->name, (int) htons(data.as), (int) peer->as); @@ -903,23 +1189,110 @@ static int bgp_handle_input(struct bgp_peer *peer) return 0; } - if ((peer->hold = ntohs(data.hold_time)) < 3) + if ((hold = ntohs(data.hold_time)) < 3) { - LOG(1, 0, 0, 0, "Bad hold time (%d) from BGP peer %s\n", - peer->hold, peer->name); + LOG(1, 0, 0, "Bad hold time (%d) from BGP peer %s\n", + hold, peer->name); bgp_send_notification(peer, BGP_ERR_OPEN, BGP_ERR_OPN_HOLD_TIME); return 0; } + /* pick lowest hold time */ + if (hold < peer->hold) + peer->hold = hold; + /* adjust our keepalive based on negotiated hold value */ if (peer->keepalive * 3 > peer->hold) peer->keepalive = peer->hold / 3; + /* check for optional parameters */ + /* 2 is for the size of type + len (both uint8_t) */ + for (param_offset = 0; + param_offset < data.opt_len; + param_offset += 2 + param->len) + { + param = (struct bgp_opt_param *)((char *)&data.opt_params + param_offset); + + /* sensible check */ + if (data.opt_len - param_offset < 2 + || param->len > data.opt_len - param_offset - 2) + { + LOG(1, 0, 0, "Malformed Optional Parameter list from BGP peer %s\n", + peer->name); + + bgp_send_notification(peer, BGP_ERR_OPEN, BGP_ERR_UNSPEC); + return 0; + } + + /* we know only one parameter type */ + if (param->type != BGP_PARAM_TYPE_CAPABILITY) + { + LOG(1, 0, 0, "Unsupported Optional Parameter type %d from BGP peer %s\n", + param->type, peer->name); + + bgp_send_notification(peer, BGP_ERR_OPEN, BGP_ERR_OPN_UNSUP_PARAM); + return 0; + } + + capabilities_len = param->len; + capabilities = (char *)¶m->value; + + /* look for BGP multiprotocol capability */ + for (capability_offset = 0; + capability_offset < capabilities_len; + capability_offset += 2 + capability->len) + { + capability = (struct bgp_capability *)(capabilities + capability_offset); + + /* sensible check */ + if (capabilities_len - capability_offset < 2 + || capability->len > capabilities_len - capability_offset - 2) + { + LOG(1, 0, 0, "Malformed Capabilities list from BGP peer %s\n", + peer->name); + + bgp_send_notification(peer, BGP_ERR_OPEN, BGP_ERR_UNSPEC); + return 0; + } + + /* we only know one capability code */ + if (capability->code != BGP_CAP_CODE_MP + && capability->len != sizeof(struct bgp_mp_cap_param)) + { + LOG(4, 0, 0, "Unsupported Capability code %d from BGP peer %s\n", + capability->code, peer->name); + + /* we don't terminate, still; we just jump to the next one */ + continue; + } + + mp_cap = (struct bgp_mp_cap_param *)&capability->value; + /* the only tuple we support */ + if (ntohs(mp_cap->afi) != BGP_MP_AFI_IPv6 && mp_cap->safi != BGP_MP_SAFI_UNICAST) + { + LOG(4, 0, 0, "Unsupported multiprotocol AFI %d and SAFI %d from BGP peer %s\n", + mp_cap->afi, mp_cap->safi, peer->name); + + /* we don't terminate, still; we just jump to the next one */ + continue; + } + + /* yes it can! */ + peer->mp_handling = HandleIPv6Routes; + } + } + + if (peer->mp_handling != HandleIPv6Routes) + { + peer->mp_handling = DoesntHandleIPv6Routes; + if (config->ipv6_prefix.s6_addr[0]) + LOG(1, 0, 0, "Warning: BGP peer %s doesn't handle IPv6 prefixes updates\n", + peer->name); + } + /* next transition requires an exchange of keepalives */ bgp_send_keepalive(peer); - - /* FIXME: may need to check for optional params */ } break; @@ -934,7 +1307,7 @@ static int bgp_handle_input(struct bgp_peer *peer) peer->retry_count = 0; peer->retry_time = 0; - LOG(4, 0, 0, 0, "BGP peer %s: state Established\n", peer->name); + LOG(4, 0, 0, "BGP peer %s: state Established\n", peer->name); } break; @@ -947,13 +1320,33 @@ static int bgp_handle_input(struct bgp_peer *peer) if (notification->error_code == BGP_ERR_CEASE) { - LOG(4, 0, 0, 0, "BGP peer %s sent CEASE\n", peer->name); - bgp_halt(peer); + LOG(4, 0, 0, "BGP peer %s sent CEASE\n", peer->name); + bgp_set_retry(peer); return 0; } + if (notification->error_code == BGP_ERR_OPEN + && notification->error_subcode == BGP_ERR_OPN_UNSUP_PARAM) + { + LOG(4, 0, 0, "BGP peer %s doesn't support BGP Capabilities\n", peer->name); + peer->mp_handling = DoesntHandleIPv6Routes; + bgp_set_retry(peer); + return 0; + } + + if (notification->error_code == BGP_ERR_OPEN + && notification->error_subcode == BGP_ERR_OPN_UNSUP_CAP) + { + /* the only capability we advertise is this one, so upon receiving + an "unsupported capability" message, we disable IPv6 routes for + this peer */ + LOG(4, 0, 0, "BGP peer %s doesn't support IPv6 routes advertisement\n", peer->name); + peer->mp_handling = DoesntHandleIPv6Routes; + break; + } + /* FIXME: should handle more notifications */ - LOG(4, 0, 0, 0, "BGP peer %s sent unhandled NOTIFICATION %d\n", + LOG(4, 0, 0, "BGP peer %s sent unhandled NOTIFICATION %d\n", peer->name, (int) notification->error_code); } @@ -982,7 +1375,10 @@ static int bgp_handle_input(struct bgp_peer *peer) static int bgp_send_open(struct bgp_peer *peer) { struct bgp_data_open data; - u16 len = sizeof(peer->outbuf->packet.header); + struct bgp_mp_cap_param mp_ipv6 = { htons(BGP_MP_AFI_IPv6), 0, BGP_MP_SAFI_UNICAST }; + struct bgp_capability cap_mp_ipv6; + struct bgp_opt_param param_cap_mp_ipv6; + uint16_t len = sizeof(peer->outbuf->packet.header); memset(peer->outbuf->packet.header.marker, 0xff, sizeof(peer->outbuf->packet.header.marker)); @@ -992,11 +1388,35 @@ static int bgp_send_open(struct bgp_peer *peer) data.version = BGP_VERSION; data.as = htons(our_as); data.hold_time = htons(peer->hold); - data.identifier = my_address; - data.opt_len = 0; + /* use the source IP we use as identifier, if available */ + if (peer->source_addr != INADDR_ANY) + data.identifier = peer->source_addr; + else + data.identifier = my_address; - memcpy(peer->outbuf->packet.data, &data, BGP_DATA_OPEN_SIZE); - len += BGP_DATA_OPEN_SIZE; + /* if we know peer doesn't support MP (mp_handling == DoesntHandleIPv6Routes) + then don't add this parameter */ + if (config->ipv6_prefix.s6_addr[0] + && (peer->mp_handling == HandlingUnknown + || peer->mp_handling == HandleIPv6Routes)) + { + /* construct the param and capability */ + cap_mp_ipv6.code = BGP_CAP_CODE_MP; + cap_mp_ipv6.len = sizeof(mp_ipv6); + memcpy(&cap_mp_ipv6.value, &mp_ipv6, cap_mp_ipv6.len); + + param_cap_mp_ipv6.type = BGP_PARAM_TYPE_CAPABILITY; + param_cap_mp_ipv6.len = 2 + sizeof(mp_ipv6); + memcpy(¶m_cap_mp_ipv6.value, &cap_mp_ipv6, param_cap_mp_ipv6.len); + + data.opt_len = 2 + param_cap_mp_ipv6.len; + memcpy(&data.opt_params, ¶m_cap_mp_ipv6, data.opt_len); + } + else + data.opt_len = 0; + + memcpy(peer->outbuf->packet.data, &data, BGP_DATA_OPEN_SIZE + data.opt_len); + len += BGP_DATA_OPEN_SIZE + data.opt_len; peer->outbuf->packet.header.len = htons(len); peer->outbuf->done = 0; @@ -1024,9 +1444,9 @@ static int bgp_send_keepalive(struct bgp_peer *peer) /* send/buffer UPDATE message */ static int bgp_send_update(struct bgp_peer *peer) { - u16 unf_len = 0; - u16 attr_len; - u16 len = sizeof(peer->outbuf->packet.header); + uint16_t unf_len = 0; + uint16_t attr_len; + uint16_t len = sizeof(peer->outbuf->packet.header); struct bgp_route_list *have = peer->routes; struct bgp_route_list *want = peer->routing ? bgp_routes : 0; struct bgp_route_list *e = 0; @@ -1072,8 +1492,8 @@ static int bgp_send_update(struct bgp_peer *peer) unf_len += s; len += s; - LOG(5, 0, 0, 0, "Withdrawing route %s/%d from BGP peer %s\n", - inet_toa(tmp->dest.prefix), tmp->dest.len, peer->name); + LOG(5, 0, 0, "Withdrawing route %s/%d from BGP peer %s\n", + fmtaddr(tmp->dest.prefix, 0), tmp->dest.len, peer->name); free(tmp); @@ -1122,8 +1542,8 @@ static int bgp_send_update(struct bgp_peer *peer) { if (!(e = malloc(sizeof(*e)))) { - LOG(0, 0, 0, 0, "Can't allocate route for %s/%d (%s)\n", - inet_toa(add->dest.prefix), add->dest.len, strerror(errno)); + LOG(0, 0, 0, "Can't allocate route for %s/%d (%s)\n", + fmtaddr(add->dest.prefix, 0), add->dest.len, strerror(errno)); return 0; } @@ -1146,8 +1566,8 @@ static int bgp_send_update(struct bgp_peer *peer) data += s; len += s; - LOG(5, 0, 0, 0, "Advertising route %s/%d to BGP peer %s\n", - inet_toa(add->dest.prefix), add->dest.len, peer->name); + LOG(5, 0, 0, "Advertising route %s/%d to BGP peer %s\n", + fmtaddr(add->dest.prefix, 0), add->dest.len, peer->name); } else { @@ -1163,11 +1583,185 @@ static int bgp_send_update(struct bgp_peer *peer) return bgp_write(peer); } +/* send/buffer UPDATE message for IPv6 routes */ +static int bgp_send_update6(struct bgp_peer *peer) +{ + uint16_t attr_len; + uint16_t unreach_len = 0; + char *unreach_len_pos; + uint8_t reach_len; + uint16_t len = sizeof(peer->outbuf->packet.header); + struct bgp_route6_list *have = peer->routes6; + struct bgp_route6_list *want = peer->routing ? bgp_routes6 : 0; + struct bgp_route6_list *e = 0; + struct bgp_route6_list *add = 0; + int s; + char ipv6addr[INET6_ADDRSTRLEN]; + + char *data = (char *) &peer->outbuf->packet.data; + + /* need leave room for attr_len, bgp_path_attrs and one prefix */ + char *max = (char *) &peer->outbuf->packet.data + + sizeof(peer->outbuf->packet.data) + - sizeof(attr_len) - peer->path_attr_len_without_nexthop + - BGP_PATH_ATTR_MP_REACH_NLRI_PARTIAL_SIZE - sizeof(struct bgp_ip6_prefix); + + memset(peer->outbuf->packet.header.marker, 0xff, + sizeof(peer->outbuf->packet.header.marker)); + + peer->outbuf->packet.header.type = BGP_MSG_UPDATE; + + /* insert non-MP unfeasible routes length */ + memcpy(data, &unreach_len, sizeof(unreach_len)); + /* skip over it and attr_len too; it will be filled when known */ + data += sizeof(unreach_len) + sizeof(attr_len); + len += sizeof(unreach_len) + sizeof(attr_len); + + /* copy usual attributes */ + memcpy(data, peer->path_attrs, peer->path_attr_len_without_nexthop); + data += peer->path_attr_len_without_nexthop; + attr_len = peer->path_attr_len_without_nexthop; + + /* copy MP unreachable NLRI heading */ + memcpy(data, peer->mp_unreach_nlri_partial, + BGP_PATH_ATTR_MP_UNREACH_NLRI_PARTIAL_SIZE); + /* remember where to update this attr len */ + unreach_len_pos = data + 2; + data += BGP_PATH_ATTR_MP_UNREACH_NLRI_PARTIAL_SIZE; + attr_len += BGP_PATH_ATTR_MP_UNREACH_NLRI_PARTIAL_SIZE; + + peer->update_routes6 = 0; /* tentatively clear */ + + /* find differences */ + while ((have || want) && data < (max - sizeof(struct bgp_ip6_prefix))) + { + if (have) + s = want + ? memcmp(&have->dest, &want->dest, sizeof(have->dest)) + : -1; + else + s = 1; + + if (s < 0) /* found one to delete */ + { + struct bgp_route6_list *tmp = have; + have = have->next; + + s = BGP_IP_PREFIX_SIZE(tmp->dest); + memcpy(data, &tmp->dest, s); + data += s; + unreach_len += s; + attr_len += s; + + LOG(5, 0, 0, "Withdrawing route %s/%d from BGP peer %s\n", + inet_ntop(AF_INET6, &tmp->dest.prefix, ipv6addr, INET6_ADDRSTRLEN), + tmp->dest.len, peer->name); + + free(tmp); + + if (e) + e->next = have; + else + peer->routes6 = have; + } + else + { + if (!s) /* same */ + { + e = have; /* stash the last found to relink above */ + have = have->next; + want = want->next; + } + else if (s > 0) /* addition reqd. */ + { + if (add) + { + peer->update_routes6 = 1; /* only one add per packet */ + if (!have) + break; + } + else + add = want; + + if (want) + want = want->next; + } + } + } + + if (have || want) + peer->update_routes6 = 1; /* more to do */ + + /* anything changed? */ + if (!(unreach_len || add)) + return 1; + + if (unreach_len) + { + /* go back and insert MP unreach_len */ + unreach_len += sizeof(struct bgp_attr_mp_unreach_nlri_partial); + unreach_len = htons(unreach_len); + memcpy(unreach_len_pos, &unreach_len, sizeof(unreach_len)); + } + else + { + /* we can remove this attribute, then */ + data -= BGP_PATH_ATTR_MP_UNREACH_NLRI_PARTIAL_SIZE; + attr_len -= BGP_PATH_ATTR_MP_UNREACH_NLRI_PARTIAL_SIZE; + } + + if (add) + { + if (!(e = malloc(sizeof(*e)))) + { + LOG(0, 0, 0, "Can't allocate route for %s/%d (%s)\n", + inet_ntop(AF_INET6, &add->dest.prefix, ipv6addr, INET6_ADDRSTRLEN), + add->dest.len, strerror(errno)); + + return 0; + } + + memcpy(e, add, sizeof(*e)); + e->next = 0; + peer->routes6 = bgp_insert_route6(peer->routes6, e); + + /* copy MP reachable NLRI heading */ + memcpy(data, peer->mp_reach_nlri_partial, + BGP_PATH_ATTR_MP_REACH_NLRI_PARTIAL_SIZE); + /* with proper len */ + reach_len = BGP_IP_PREFIX_SIZE(add->dest); + data[2] = sizeof(struct bgp_attr_mp_reach_nlri_partial) + reach_len; + data += BGP_PATH_ATTR_MP_REACH_NLRI_PARTIAL_SIZE; + attr_len += BGP_PATH_ATTR_MP_REACH_NLRI_PARTIAL_SIZE; + + memcpy(data, &add->dest, reach_len); + data += reach_len; + attr_len += reach_len; + + LOG(5, 0, 0, "Advertising route %s/%d to BGP peer %s\n", + inet_ntop(AF_INET6, &add->dest.prefix, ipv6addr, INET6_ADDRSTRLEN), + add->dest.len, peer->name); + } + + /* update len with attributes we added */ + len += attr_len; + + /* go back and insert attr_len */ + attr_len = htons(attr_len); + memcpy((char *)&peer->outbuf->packet.data + 2, &attr_len, sizeof(attr_len)); + + peer->outbuf->packet.header.len = htons(len); + peer->outbuf->done = 0; + + return bgp_write(peer); +} + /* send/buffer NOTIFICATION message */ -static int bgp_send_notification(struct bgp_peer *peer, u8 code, u8 subcode) +static int bgp_send_notification(struct bgp_peer *peer, uint8_t code, + uint8_t subcode) { struct bgp_data_notification data; - u16 len = 0; + uint16_t len = 0; data.error_code = code; len += sizeof(data.error_code);