- Bird-users - Mailing Lists

One possible way of supporting MPLS labels in bird 1.6
by Lennert Buytenhek 17 Aug '17

17 Aug '17

Hi! Included below is the patch we're using on top of bird 1.6 for kernel MPLS label injection support. This is not a very generic solution, and it shouldn't be merged, and bird 2.0 probably does this in a better way (though I haven't looked), and it's somewhat ugly, etc etc, but it works for us in production for its intended purpose. #include <disclaimer.h> The basic idea is that we add a variable 'mpls_label', which when set on a route, indicates that that MPLS label should be used when sending traffic for this prefix to the given gateway. Example using static routes: protocol static { route 200.0.0.0/24 via 172.17.76.1 { mpls_label 666; }; } protocol static { route 200.0.0.0/24 via 172.17.76.2 { mpls_label 667; }; } protocol static { route 200.0.0.0/24 via 172.17.76.3 { mpls_label 668; }; } protocol kernel { import filter { reject; }; export filter { if net = 200.0.0.0/24 then accept; else reject; }; merge paths yes; } This will give you: # birdc show route BIRD 1.6.3 ready. 200.0.0.0/24 via 172.17.76.1 (MPLS label 666) on br0 [static1 04:05:18] * (200) via 172.17.76.3 (MPLS label 668) on br0 [static3 04:05:18] (200) via 172.17.76.2 (MPLS label 667) on br0 [static2 04:05:18] (200) # And: $ ip route show 200.0.0.0/24 200.0.0.0/24 proto bird nexthop encap mpls 666 via 172.17.76.1 dev br0 weight 1 nexthop encap mpls 667 via 172.17.76.2 dev br0 weight 1 nexthop encap mpls 668 via 172.17.76.3 dev br0 weight 1 $ Learning routes back from the kernel also works, e.g. with this config: protocol kernel { import filter { print "Importing ", net; accept; }; export filter { print "Exporting ", net; reject; }; merge paths yes; learn yes; } And: # ip route add 200.0.0.0/24 \ nexthop encap mpls 667 via 172.17.76.2 \ nexthop encap mpls 668 via 172.17.76.3 We get: # birdc show route 200.0.0.0/24 BIRD 1.6.3 ready. 200.0.0.0/24 multipath [kernel1 04:09:51] * (10) via 172.17.76.2 (MPLS label 667) on br0 weight 1 via 172.17.76.3 (MPLS label 668) on br0 weight 1 # Finally, an example involving BGP. Configuration for router "A": router id 172.17.76.1; log stderr all; debug protocols all; protocol device { scan time 3600; } protocol static { route 200.0.0.0/24 via 172.17.76.1; } protocol static { route 200.0.0.0/24 via 172.17.76.2; } protocol static { route 200.0.0.0/24 via 172.17.76.3; } protocol bgp b { local 172.17.76.1 as 1; neighbor 172.17.76.10 as 10; direct; export filter { if source = RTS_STATIC then accept; reject; }; import filter { reject; }; add paths tx; } And the configuration for router "B": router id 172.17.76.10; log stderr all; debug protocols all; protocol device { scan time 3600; } protocol kernel { import filter { reject; }; export filter { if source = RTS_BGP then { if gw = 172.17.76.1 then { mpls_label = 666; accept; } if gw = 172.17.76.2 then { mpls_label = 667; accept; } if gw = 172.17.76.3 then { mpls_label = 668; accept; } } reject; }; merge paths yes; } protocol bgp a { local 172.17.76.10 as 10; neighbor 172.17.76.1 as 1; direct; export filter { reject; }; import filter { accept; }; add paths rx; } This gives me on router "B", as expected: # ip route show 200.0.0.0/24 200.0.0.0/24 proto bird nexthop encap mpls 666 via 172.17.76.1 dev eth0 weight 1 nexthop encap mpls 667 via 172.17.76.2 dev eth0 weight 1 nexthop encap mpls 668 via 172.17.76.3 dev eth0 weight 1 # Notes: * The mpls_label attribute is a member of struct rta instead of a sysdep/linux specific attribute that lives on the eattr list. I had to do it this way because the multipath route merging code isn't set up to deal with platform(Linux)-specific attributes nicely. * mpls_label is different and is treated differently from all the other route attributes (except RTA_GATEWAY) in that none of those attributes can differ between multipath route nexthops, while mpls_label can. * The netlink serialization code in bird 1.6 has a bug whereby if you try to serialize an attribute that is not a multiple of 4 bytes long (such as RTA_ENCAP_TYPE, which is needed for MPLS and has a payload of only 2 bytes) it will round up the size in the netlink header (which is wrong -- that field should reflect the length _without_ padding) but then send out a netlink message without the padding appended if this was the last attribute in the message (which is also wrong). Also, if you do try to serialize an attribute that is not a multiple of 4 bytes long, then between 1 and 3 bytes of uninitialized bird stack memory get leaked over netlink. (The netlink attributes that upstream bird cares about are all multiples of 4 bytes long, so this won't actually surface in current bird upstream.) * This patch doesn't implement BGP Labeled Unicast or anything like that -- we currently signal MPLS labels using BGP extended communities. Again, I don't see this being merged in any shape or form, but feedback is always appreciated. Cheers, Lennert diff --git a/filter/config.Y b/filter/config.Y index 5ea83f81..56ef598b 100644 --- a/filter/config.Y +++ b/filter/config.Y @@ -335,7 +335,7 @@ CF_KEYWORDS(FUNCTION, PRINT, PRINTN, UNSET, RETURN, SET, STRING, BGPMASK, BGPPATH, CLIST, ECLIST, LCLIST, IF, THEN, ELSE, CASE, TRUE, FALSE, RT, RO, UNKNOWN, GENERIC, - FROM, GW, NET, MASK, PROTO, SOURCE, SCOPE, CAST, DEST, IFNAME, IFINDEX, + FROM, GW, NET, MASK, PROTO, SOURCE, SCOPE, CAST, DEST, IFNAME, IFINDEX, MPLS_LABEL, PREFERENCE, LEN, DEFINED, @@ -786,6 +786,7 @@ static_attr: | DEST { $$ = f_new_inst(); $$->aux = T_ENUM_RTD; $$->a2.i = SA_DEST; $$->a1.i = 1; } | IFNAME { $$ = f_new_inst(); $$->aux = T_STRING; $$->a2.i = SA_IFNAME; } | IFINDEX { $$ = f_new_inst(); $$->aux = T_INT; $$->a2.i = SA_IFINDEX; } + | MPLS_LABEL { $$ = f_new_inst(); $$->aux = T_INT; $$->a2.i = SA_MPLS_LABEL; $$->a1.i = 1; } ; term: diff --git a/filter/filter.c b/filter/filter.c index f18970e0..773bf61d 100644 --- a/filter/filter.c +++ b/filter/filter.c @@ -928,6 +928,7 @@ interpret(struct f_inst *what) case SA_DEST: res.val.i = rta->dest; break; case SA_IFNAME: res.val.s = rta->iface ? rta->iface->name : ""; break; case SA_IFINDEX: res.val.i = rta->iface ? rta->iface->index : 0; break; + case SA_MPLS_LABEL: res.val.i = rta->mpls_label; break; default: bug("Invalid static attribute access (%x)", res.type); @@ -959,6 +960,7 @@ interpret(struct f_inst *what) rta->dest = RTD_ROUTER; rta->gw = ip; + rta->mpls_label = 0; rta->iface = n->iface; rta->nexthops = NULL; rta->hostentry = NULL; @@ -976,11 +978,16 @@ interpret(struct f_inst *what) rta->dest = i; rta->gw = IPA_NONE; + rta->mpls_label = 0; rta->iface = NULL; rta->nexthops = NULL; rta->hostentry = NULL; break; + case SA_MPLS_LABEL: + rta->mpls_label = v1.val.i; + break; + default: bug("Invalid static attribute access (%x)", res.type); } diff --git a/filter/filter.h b/filter/filter.h index 049ceb76..dfcb5b7a 100644 --- a/filter/filter.h +++ b/filter/filter.h @@ -200,6 +200,7 @@ void val_format(struct f_val v, buffer *buf); #define SA_DEST 8 #define SA_IFNAME 9 #define SA_IFINDEX 10 +#define SA_MPLS_LABEL 11 struct f_tree { diff --git a/nest/route.h b/nest/route.h index 383f4def..2defb29f 100644 --- a/nest/route.h +++ b/nest/route.h @@ -332,6 +332,7 @@ void rt_show(struct rt_show_data *); /* Multipath next-hop */ struct mpnh { ip_addr gw; /* Next hop */ + u32 mpls_label; struct iface *iface; /* Outgoing interface */ struct mpnh *next; byte weight; @@ -359,6 +360,7 @@ typedef struct rta { u16 hash_key; /* Hash over important fields */ u32 igp_metric; /* IGP metric to next hop (for iBGP routes) */ ip_addr gw; /* Next hop */ + u32 mpls_label; ip_addr from; /* Advertising router */ struct hostentry *hostentry; /* Hostentry for recursive next-hops */ struct iface *iface; /* Outgoing interface */ diff --git a/nest/rt-attr.c b/nest/rt-attr.c index edf27d44..8d2e517f 100644 --- a/nest/rt-attr.c +++ b/nest/rt-attr.c @@ -200,7 +200,7 @@ mpnh_hash(struct mpnh *x) { uint h = 0; for (; x; x = x->next) - h ^= ipa_hash(x->gw); + h ^= ipa_hash(x->gw) ^ x->mpls_label; return h; } @@ -209,7 +209,7 @@ int mpnh__same(struct mpnh *x, struct mpnh *y) { for (; x && y; x = x->next, y = y->next) - if (!ipa_equal(x->gw, y->gw) || (x->iface != y->iface) || (x->weight != y->weight)) + if (!ipa_equal(x->gw, y->gw) || (x->mpls_label != y->mpls_label) || (x->iface != y->iface) || (x->weight != y->weight)) return 0; return x == y; @@ -230,6 +230,11 @@ mpnh_compare_node(struct mpnh *x, struct mpnh *y) if (r) return r; + if (x->mpls_label < y->mpls_label) + return -1; + else if (x->mpls_label > y->mpls_label) + return 1; + r = ipa_compare(x->gw, y->gw); if (r) return r; @@ -242,6 +247,7 @@ mpnh_copy_node(const struct mpnh *src, linpool *lp) { struct mpnh *n = lp_alloc(lp, sizeof(struct mpnh)); n->gw = src->gw; + n->mpls_label = src->mpls_label; n->iface = src->iface; n->next = NULL; n->weight = src->weight; @@ -341,6 +347,7 @@ mpnh_copy(struct mpnh *o) { struct mpnh *n = sl_alloc(mpnh_slab); n->gw = o->gw; + n->mpls_label = o->mpls_label; n->iface = o->iface; n->next = NULL; n->weight = o->weight; @@ -1051,7 +1058,7 @@ rta_alloc_hash(void) static inline uint rta_hash(rta *a) { - return (((uint) (uintptr_t) a->src) ^ ipa_hash(a->gw) ^ + return (((uint) (uintptr_t) a->src) ^ ipa_hash(a->gw) ^ a->mpls_label ^ mpnh_hash(a->nexthops) ^ ea_hash(a->eattrs)) & 0xffff; } @@ -1066,6 +1073,7 @@ rta_same(rta *x, rta *y) x->flags == y->flags && x->igp_metric == y->igp_metric && ipa_equal(x->gw, y->gw) && + x->mpls_label == y->mpls_label && ipa_equal(x->from, y->from) && x->iface == y->iface && x->hostentry == y->hostentry && @@ -1214,7 +1222,11 @@ rta_dump(rta *a) debug(" !CACHED"); debug(" <-%I", a->from); if (a->dest == RTD_ROUTER) - debug(" ->%I", a->gw); + { + debug(" ->%I", a->gw); + if (a->mpls_label) + debug(" (MPLS label %d)", a->mpls_label); + } if (a->dest == RTD_DEVICE || a->dest == RTD_ROUTER) debug(" [%s]", a->iface ? a->iface->name : "???" ); if (a->eattrs) diff --git a/nest/rt-table.c b/nest/rt-table.c index c6e48c38..4e6954f0 100644 --- a/nest/rt-table.c +++ b/nest/rt-table.c @@ -594,7 +594,7 @@ rt_notify_accepted(struct announce_hook *ah, net *net, rte *new_changed, rte *ol static struct mpnh * mpnh_merge_rta(struct mpnh *nhs, rta *a, linpool *pool, int max) { - struct mpnh nh = { .gw = a->gw, .iface = a->iface }; + struct mpnh nh = { .gw = a->gw, .mpls_label = a->mpls_label, .iface = a->iface }; struct mpnh *nh2 = (a->dest == RTD_MULTIPATH) ? a->nexthops : &nh; return mpnh_merge(nhs, nh2, 1, 0, max, pool); } @@ -1696,6 +1696,7 @@ rta_next_hop_outdated(rta *a) return a->dest != RTD_UNREACHABLE; return (a->iface != he->src->iface) || !ipa_equal(a->gw, he->gw) || + (a->mpls_label != he->src->mpls_label) || (a->dest != he->dest) || (a->igp_metric != he->igp_metric) || !mpnh_same(a->nexthops, he->src->nexthops); } @@ -1706,6 +1707,7 @@ rta_apply_hostentry(rta *a, struct hostentry *he) a->hostentry = he; a->iface = he->src ? he->src->iface : NULL; a->gw = he->gw; + a->mpls_label = he->src ? he->src->mpls_label : 0; a->dest = he->dest; a->igp_metric = he->igp_metric; a->nexthops = he->src ? he->src->nexthops : NULL; @@ -2302,6 +2304,7 @@ rt_update_hostentry(rtable *tab, struct hostentry *he) { /* The host is reachable through some route entry */ he->gw = a->gw; + // @@@ a->mpls_label; he->dest = a->dest; } @@ -2379,12 +2382,19 @@ rt_format_via(rte *e) { rta *a = e->attrs; + byte info[256]; + + if (a->mpls_label) + bsprintf(info, " (MPLS label %d)", a->mpls_label); + else + info[0] = 0; + /* Max text length w/o IP addr and interface name is 16 */ static byte via[STD_ADDRESS_P_LENGTH+sizeof(a->iface->name)+16]; switch (a->dest) { - case RTD_ROUTER: bsprintf(via, "via %I on %s", a->gw, a->iface->name); break; + case RTD_ROUTER: bsprintf(via, "via %I%s on %s", a->gw, info, a->iface->name); break; case RTD_DEVICE: bsprintf(via, "dev %s", a->iface->name); break; case RTD_BLACKHOLE: bsprintf(via, "blackhole"); break; case RTD_UNREACHABLE: bsprintf(via, "unreachable"); break; @@ -2429,7 +2439,13 @@ rt_show_rte(struct cli *c, byte *ia, rte *e, struct rt_show_data *d, ea_list *tm cli_printf(c, -1007, "%-18s %s [%s %s%s]%s%s", ia, rt_format_via(e), a->src->proto->name, tm, from, primary ? (sync_error ? " !" : " *") : "", info); for (nh = a->nexthops; nh; nh = nh->next) - cli_printf(c, -1007, "\tvia %I on %s weight %d", nh->gw, nh->iface->name, nh->weight + 1); + { + if (nh->mpls_label) + bsprintf(info, " (MPLS label %d)", nh->mpls_label); + else + info[0] = 0; + cli_printf(c, -1007, "\tvia %I%s on %s weight %d", nh->gw, info, nh->iface->name, nh->weight + 1); + } if (d->verbose) rta_show(c, a, tmpa); } diff --git a/proto/bgp/packets.c b/proto/bgp/packets.c index cd3f3f4f..a23b081c 100644 --- a/proto/bgp/packets.c +++ b/proto/bgp/packets.c @@ -1179,6 +1179,7 @@ bgp_set_next_hop(struct bgp_proto *p, rta *a) a->dest = RTD_ROUTER; a->gw = ng->addr; + a->mpls_label = 0; a->iface = ng->iface; a->hostentry = NULL; a->igp_metric = 0; diff --git a/proto/ospf/rt.c b/proto/ospf/rt.c index 368e3d05..73ed474e 100644 --- a/proto/ospf/rt.c +++ b/proto/ospf/rt.c @@ -47,6 +47,7 @@ new_nexthop(struct ospf_proto *p, ip_addr gw, struct iface *iface, byte weight) { struct mpnh *nh = lp_alloc(p->nhpool, sizeof(struct mpnh)); nh->gw = gw; + nh->mpls_label = 0; nh->iface = iface; nh->next = NULL; nh->weight = weight; @@ -1893,6 +1894,7 @@ ort_changed(ort *nf, rta *nr) (nf->n.tag != nf->old_tag) || (nf->n.rid != nf->old_rid) || (nr->source != or->source) || (nr->dest != or->dest) || (nr->iface != or->iface) || !ipa_equal(nr->gw, or->gw) || + (nr->mpls_label != or->mpls_label) || !mpnh_same(nr->nexthops, or->nexthops); } @@ -1956,6 +1958,7 @@ again1: a0.dest = RTD_ROUTER; a0.iface = nf->n.nhs->iface; a0.gw = nf->n.nhs->gw; + a0.mpls_label = 0; } else { diff --git a/proto/rip/rip.c b/proto/rip/rip.c index 7b380097..81821dcd 100644 --- a/proto/rip/rip.c +++ b/proto/rip/rip.c @@ -182,6 +182,7 @@ rip_announce_rte(struct rip_proto *p, struct rip_entry *en) struct mpnh *nh = alloca(sizeof(struct mpnh)); nh->gw = rt->next_hop; + nh->mpls_label = 0; nh->iface = rt->from->nbr->iface; nh->weight = rt->from->ifa->cf->ecmp_weight; mpnh_insert(&nhs, nh); @@ -199,6 +200,7 @@ rip_announce_rte(struct rip_proto *p, struct rip_entry *en) /* Unipath route */ a0.dest = RTD_ROUTER; a0.gw = rt->next_hop; + a0.mpls_label = 0; a0.iface = rt->from->nbr->iface; a0.from = rt->from->nbr->addr; } diff --git a/proto/static/config.Y b/proto/static/config.Y index 182721b3..9d2a9205 100644 --- a/proto/static/config.Y +++ b/proto/static/config.Y @@ -78,6 +78,9 @@ stat_multipath1: | stat_multipath1 BFD bool { this_srt_nh->use_bfd = $3; cf_check_bfd($3); } + | stat_multipath1 MPLS_LABEL expr { + this_srt_nh->mpls_label = $3; + } ; stat_multipath: @@ -115,6 +118,7 @@ stat_route: stat_route_item: cmd { *this_srt_last_cmd = $1; this_srt_last_cmd = &($1->next); } | BFD bool ';' { this_srt->use_bfd = $2; cf_check_bfd($2); } + | MPLS_LABEL expr ';' { this_srt->mpls_label = $2; } ; stat_route_opts: diff --git a/proto/static/static.c b/proto/static/static.c index 849067b9..36a0bd60 100644 --- a/proto/static/static.c +++ b/proto/static/static.c @@ -75,6 +75,7 @@ static_install(struct proto *p, struct static_route *r, struct iface *ifa) a.cast = RTC_UNICAST; a.dest = r->dest; a.gw = r->via; + a.mpls_label = r->mpls_label; a.iface = ifa; if (r->dest == RTD_MULTIPATH) @@ -87,6 +88,7 @@ static_install(struct proto *p, struct static_route *r, struct iface *ifa) { struct mpnh *nh = alloca(sizeof(struct mpnh)); nh->gw = r2->via; + nh->mpls_label = r2->mpls_label; nh->iface = r2->neigh->iface; nh->weight = r2->masklen; /* really */ mpnh_insert(&nhs, nh); @@ -98,6 +100,7 @@ static_install(struct proto *p, struct static_route *r, struct iface *ifa) /* Fallback to unipath route for exactly one nexthop */ a.dest = RTD_ROUTER; a.gw = nhs->gw; + a.mpls_label = nhs->mpls_label; a.iface = nhs->iface; } else @@ -128,11 +131,17 @@ static void static_remove(struct proto *p, struct static_route *r) { net *n; + byte info[256]; if (!r->installed) return; - DBG("Removing static route %I/%d via %I\n", r->net, r->masklen, r->via); + if (r->mpls_label) + bsprintf(info, " (MPLS label %d)", r->mpls_label); + else + info[0] = 0; + + DBG("Removing static route %I/%d via %I%s\n", r->net, r->masklen, r->via, info); n = net_find(p->table, r->net, r->masklen); rte_update(p, n, NULL); r->installed = 0; @@ -389,7 +398,10 @@ static_dump_rt(struct static_route *r) switch (r->dest) { case RTD_ROUTER: - debug("via %I\n", r->via); + debug("via %I", r->via); + if (r->mpls_label) + debug(" (MPLS label %d)", r->mpls_label); + debug("\n", r->via); break; case RTD_DEVICE: debug("dev %s\n", r->if_name); @@ -474,7 +486,7 @@ static_same_dest(struct static_route *x, struct static_route *y) switch (x->dest) { case RTD_ROUTER: - return ipa_equal(x->via, y->via) && (x->via_if == y->via_if); + return ipa_equal(x->via, y->via) && (x->mpls_label == y->mpls_label) && (x->via_if == y->via_if); case RTD_DEVICE: return !strcmp(x->if_name, y->if_name); @@ -483,7 +495,7 @@ static_same_dest(struct static_route *x, struct static_route *y) for (x = x->mp_next, y = y->mp_next; x && y; x = x->mp_next, y = y->mp_next) - if (!ipa_equal(x->via, y->via) || (x->via_if != y->via_if) || (x->use_bfd != y->use_bfd)) + if (!ipa_equal(x->via, y->via) || (x->mpls_label != y->mpls_label) || (x->via_if != y->via_if) || (x->use_bfd != y->use_bfd)) return 0; return !x && !y; @@ -645,6 +657,7 @@ static void static_show_rt(struct static_route *r) { byte via[STD_ADDRESS_P_LENGTH + 16]; + byte info[256]; switch (r->dest) { @@ -657,14 +670,24 @@ static_show_rt(struct static_route *r) case RTDX_RECURSIVE: bsprintf(via, "recursive %I", r->via); break; default: bsprintf(via, "???"); } - cli_msg(-1009, "%I/%d %s%s%s", r->net, r->masklen, via, + if (r->dest == RTD_ROUTER && r->mpls_label) + bsprintf(info, " (MPLS label %d)", r->mpls_label); + else + info[0] = 0; + cli_msg(-1009, "%I/%d %s%s%s%s", r->net, r->masklen, via, info, r->bfd_req ? " (bfd)" : "", r->installed ? "" : " (dormant)"); struct static_route *r2; if (r->dest == RTD_MULTIPATH) for (r2 = r->mp_next; r2; r2 = r2->mp_next) - cli_msg(-1009, "\tvia %I%J weight %d%s%s", r2->via, r2->via_if, r2->masklen + 1, /* really */ - r2->bfd_req ? " (bfd)" : "", r2->installed ? "" : " (dormant)"); + { + if (r->dest == RTD_ROUTER && r->mpls_label) + bsprintf(info, " (MPLS label %d)", r->mpls_label); + else + info[0] = 0; + cli_msg(-1009, "\tvia %I%J%s weight %d%s%s", r2->via, r2->via_if, info, r2->masklen + 1, /* really */ + r2->bfd_req ? " (bfd)" : "", r2->installed ? "" : " (dormant)"); + } } void diff --git a/proto/static/static.h b/proto/static/static.h index 6b047234..5811a888 100644 --- a/proto/static/static.h +++ b/proto/static/static.h @@ -30,6 +30,7 @@ struct static_route { int masklen; /* Mask length */ int dest; /* Destination type (RTD_*) */ ip_addr via; /* Destination router */ + u32 mpls_label; struct iface *via_if; /* Destination iface, for link-local vias */ struct neighbor *neigh; byte *if_name; /* Name for RTD_DEVICE routes */ diff --git a/sysdep/linux/netlink.c b/sysdep/linux/netlink.c index 22313f43..16e8605f 100644 --- a/sysdep/linux/netlink.c +++ b/sysdep/linux/netlink.c @@ -49,6 +49,22 @@ #define RTA_TABLE 15 #endif +#ifndef RTA_ENCAP_TYPE +#define RTA_ENCAP_TYPE 21 +#endif + +#ifndef RTA_ENCAP +#define RTA_ENCAP 22 +#endif + +#ifndef MPLS_IPTUNNEL_DST +#define MPLS_IPTUNNEL_DST 1 +#endif + +#ifndef LWTUNNEL_ENCAP_MPLS +#define LWTUNNEL_ENCAP_MPLS 1 +#endif + #ifdef IPV6 #define krt_ecmp6(X) 1 @@ -318,10 +334,12 @@ static struct nl_want_attrs ifa_attr_want6[BIRD_IFA_MAX] = { #endif -#define BIRD_RTA_MAX (RTA_TABLE+1) +#define BIRD_RTA_MAX (RTA_ENCAP + 1) static struct nl_want_attrs mpnh_attr_want4[BIRD_RTA_MAX] = { [RTA_GATEWAY] = { 1, 1, sizeof(ip4_addr) }, + [RTA_ENCAP_TYPE]= { 1, 1, sizeof(u16) }, + [RTA_ENCAP] = { 1, 1, RTA_LENGTH(4) }, }; #ifndef IPV6 @@ -335,6 +353,8 @@ static struct nl_want_attrs rtm_attr_want4[BIRD_RTA_MAX] = { [RTA_MULTIPATH] = { 1, 0, 0 }, [RTA_FLOW] = { 1, 1, sizeof(u32) }, [RTA_TABLE] = { 1, 1, sizeof(u32) }, + [RTA_ENCAP_TYPE]= { 1, 1, sizeof(u16) }, + [RTA_ENCAP] = { 1, 1, RTA_LENGTH(4) }, }; #else static struct nl_want_attrs rtm_attr_want6[BIRD_RTA_MAX] = { @@ -347,6 +367,8 @@ static struct nl_want_attrs rtm_attr_want6[BIRD_RTA_MAX] = { [RTA_METRICS] = { 1, 0, 0 }, [RTA_FLOW] = { 1, 1, sizeof(u32) }, [RTA_TABLE] = { 1, 1, sizeof(u32) }, + [RTA_ENCAP_TYPE]= { 1, 1, sizeof(u16) }, + [RTA_ENCAP] = { 1, 1, RTA_LENGTH(4) }, }; #endif @@ -380,6 +402,9 @@ nl_parse_attrs(struct rtattr *a, struct nl_want_attrs *want, struct rtattr **k, return 1; } +static inline u16 rta_get_u16(struct rtattr *a) +{ return *(u16 *) RTA_DATA(a); } + static inline u32 rta_get_u32(struct rtattr *a) { return *(u32 *) RTA_DATA(a); } @@ -396,17 +421,23 @@ nl_add_attr(struct nlmsghdr *h, uint bufsize, uint code, const void *data, uint uint pos = NLMSG_ALIGN(h->nlmsg_len); uint len = RTA_LENGTH(dlen); + if (pos != h->nlmsg_len) + bug("nl_add_attr: packet buffer misaligned"); + if (pos + len > bufsize) bug("nl_add_attr: packet buffer overflow"); struct rtattr *a = (struct rtattr *)((char *)h + pos); a->rta_type = code; a->rta_len = len; - h->nlmsg_len = pos + len; + h->nlmsg_len = NLMSG_ALIGN(pos + len); if (dlen > 0) memcpy(RTA_DATA(a), data, dlen); + if (h->nlmsg_len != pos + len) + memset(RTA_DATA(a) + len, 0, h->nlmsg_len - (pos + len)); + return a; } @@ -432,7 +463,16 @@ nl_open_attr(struct nlmsghdr *h, uint bufsize, uint code) static inline void nl_close_attr(struct nlmsghdr *h, struct rtattr *a) { - a->rta_len = (void *)h + NLMSG_ALIGN(h->nlmsg_len) - (void *)a; + uint pos; + + a->rta_len = (void *)h + h->nlmsg_len - (void *)a; + + pos = NLMSG_ALIGN(h->nlmsg_len); + if (pos != h->nlmsg_len) + { + memset((void *)h + h->nlmsg_len, 0, pos - h->nlmsg_len); + h->nlmsg_len = pos; + } } static inline struct rtnexthop * @@ -441,10 +481,13 @@ nl_open_nexthop(struct nlmsghdr *h, uint bufsize) uint pos = NLMSG_ALIGN(h->nlmsg_len); uint len = RTNH_LENGTH(0); + if (pos != h->nlmsg_len) + bug("nl_open_nexthop: packet buffer misaligned"); + if (pos + len > bufsize) bug("nl_open_nexthop: packet buffer overflow"); - h->nlmsg_len = pos + len; + h->nlmsg_len = NLMSG_ALIGN(pos + len); return (void *)h + pos; } @@ -452,7 +495,16 @@ nl_open_nexthop(struct nlmsghdr *h, uint bufsize) static inline void nl_close_nexthop(struct nlmsghdr *h, struct rtnexthop *nh) { - nh->rtnh_len = (void *)h + NLMSG_ALIGN(h->nlmsg_len) - (void *)nh; + uint pos; + + nh->rtnh_len = (void *)h + h->nlmsg_len - (void *)nh; + + pos = NLMSG_ALIGN(h->nlmsg_len); + if (pos != h->nlmsg_len) + { + memset((void *)h + h->nlmsg_len, 0, pos - h->nlmsg_len); + h->nlmsg_len = pos; + } } static void @@ -470,6 +522,23 @@ nl_add_multipath(struct nlmsghdr *h, unsigned bufsize, struct mpnh *nh) nl_add_attr_ipa(h, bufsize, RTA_GATEWAY, nh->gw); + if (nh->mpls_label) + { + struct rta_encap { + struct rtattr header; + u32 mpls_label; + } encap; + u16 et; + + encap.header.rta_len = sizeof(encap); + encap.header.rta_type = MPLS_IPTUNNEL_DST; + encap.mpls_label = htonl(((nh->mpls_label & 0xfffff) << 12) | 0x100); + nl_add_attr(h, bufsize, RTA_ENCAP, &encap, sizeof(encap)); + + et = LWTUNNEL_ENCAP_MPLS; + nl_add_attr(h, bufsize, RTA_ENCAP_TYPE, &et, sizeof(et)); + } + nl_close_nexthop(h, rtnh); } @@ -529,6 +598,19 @@ nl_parse_multipath(struct krt_proto *p, struct rtattr *ra) else return NULL; + rv->mpls_label = 0; + if (a[RTA_ENCAP_TYPE] && rta_get_u16(a[RTA_ENCAP_TYPE]) == LWTUNNEL_ENCAP_MPLS) + { + struct rtattr *encap = a[RTA_ENCAP]; + + if (encap != NULL) + { + struct rtattr *enc2 = RTA_DATA(encap); + if (enc2->rta_type == MPLS_IPTUNNEL_DST) + rv->mpls_label = (ntohl(rta_get_u32(enc2)) >> 12) & 0xfffff; + } + } + len -= NLMSG_ALIGN(nh->rtnh_len); nh = RTNH_NEXT(nh); } @@ -870,12 +952,12 @@ nh_bufsize(struct mpnh *nh) { int rv = 0; for (; nh != NULL; nh = nh->next) - rv += RTNH_LENGTH(RTA_LENGTH(sizeof(ip_addr))); + rv += RTNH_LENGTH(RTA_LENGTH(sizeof(ip_addr)) + RTA_LENGTH(RTA_LENGTH(4)) + RTA_LENGTH(2)); return rv; } static int -nl_send_route(struct krt_proto *p, rte *e, struct ea_list *eattrs, int op, int dest, ip_addr gw, struct iface *iface) +nl_send_route(struct krt_proto *p, rte *e, struct ea_list *eattrs, int op, int dest, ip_addr gw, u32 mpls_label, struct iface *iface) { eattr *ea; net *net = e->net; @@ -893,7 +975,7 @@ nl_send_route(struct krt_proto *p, rte *e, struct ea_list *eattrs, int op, int d bzero(&r.h, sizeof(r.h)); bzero(&r.r, sizeof(r.r)); r.h.nlmsg_type = op ? RTM_NEWROUTE : RTM_DELROUTE; - r.h.nlmsg_len = NLMSG_LENGTH(sizeof(struct rtmsg)); + r.h.nlmsg_len = NLMSG_ALIGN(NLMSG_LENGTH(sizeof(struct rtmsg))); r.h.nlmsg_flags = op | NLM_F_REQUEST | NLM_F_ACK; r.r.rtm_family = BIRD_AF; @@ -964,6 +1046,22 @@ dest: r.r.rtm_type = RTN_UNICAST; nl_add_attr_u32(&r.h, sizeof(r), RTA_OIF, iface->index); nl_add_attr_ipa(&r.h, sizeof(r), RTA_GATEWAY, gw); + if (mpls_label) + { + struct rta_encap { + struct rtattr header; + u32 mpls_label; + } encap; + u16 et; + + encap.header.rta_len = sizeof(encap); + encap.header.rta_type = MPLS_IPTUNNEL_DST; + encap.mpls_label = htonl(((mpls_label & 0xfffff) << 12) | 0x100); + nl_add_attr(&r.h, sizeof(r), RTA_ENCAP, &encap, sizeof(encap)); + + et = LWTUNNEL_ENCAP_MPLS; + nl_add_attr(&r.h, sizeof(r), RTA_ENCAP_TYPE, &et, sizeof(et)); + } break; case RTD_DEVICE: r.r.rtm_type = RTN_UNICAST; @@ -1002,17 +1100,17 @@ nl_add_rte(struct krt_proto *p, rte *e, struct ea_list *eattrs) { struct mpnh *nh = a->nexthops; - err = nl_send_route(p, e, eattrs, NL_OP_ADD, RTD_ROUTER, nh->gw, nh->iface); + err = nl_send_route(p, e, eattrs, NL_OP_ADD, RTD_ROUTER, nh->gw, nh->mpls_label, nh->iface); if (err < 0) return err; for (nh = nh->next; nh; nh = nh->next) - err += nl_send_route(p, e, eattrs, NL_OP_APPEND, RTD_ROUTER, nh->gw, nh->iface); + err += nl_send_route(p, e, eattrs, NL_OP_APPEND, RTD_ROUTER, nh->gw, nh->mpls_label, nh->iface); return err; } - return nl_send_route(p, e, eattrs, NL_OP_ADD, a->dest, a->gw, a->iface); + return nl_send_route(p, e, eattrs, NL_OP_ADD, a->dest, a->gw, a->mpls_label, a->iface); } static inline int @@ -1022,7 +1120,7 @@ nl_delete_rte(struct krt_proto *p, rte *e, struct ea_list *eattrs) /* For IPv6, we just repeatedly request DELETE until we get error */ do - err = nl_send_route(p, e, eattrs, NL_OP_DELETE, RTD_NONE, IPA_NONE, NULL); + err = nl_send_route(p, e, eattrs, NL_OP_DELETE, RTD_NONE, IPA_NONE, 0, NULL); while (krt_ecmp6(p) && !err); return err; @@ -1058,11 +1156,12 @@ krt_replace_rte(struct krt_proto *p, net *n, rte *new, rte *old, struct ea_list static inline struct mpnh * -nl_alloc_mpnh(struct nl_parse_state *s, ip_addr gw, struct iface *iface, byte weight) +nl_alloc_mpnh(struct nl_parse_state *s, ip_addr gw, u32 mpls_label, struct iface *iface, byte weight) { struct mpnh *nh = lp_alloc(s->pool, sizeof(struct mpnh)); nh->gw = gw; + nh->mpls_label = mpls_label; nh->iface = iface; nh->next = NULL; nh->weight = weight; @@ -1268,6 +1367,7 @@ nl_parse_route(struct nl_parse_state *s, struct nlmsghdr *h) ra->dest = RTD_ROUTER; memcpy(&ra->gw, RTA_DATA(a[RTA_GATEWAY]), sizeof(ra->gw)); ipa_ntoh(ra->gw); + ra->mpls_label = 0; #ifdef IPV6 /* Silently skip strange 6to4 routes */ @@ -1290,6 +1390,18 @@ nl_parse_route(struct nl_parse_state *s, struct nlmsghdr *h) def_scope = RT_SCOPE_LINK; } + if (a[RTA_ENCAP_TYPE] && rta_get_u16(a[RTA_ENCAP_TYPE]) == LWTUNNEL_ENCAP_MPLS) + { + struct rtattr *encap = a[RTA_ENCAP]; + + if (encap != NULL) + { + struct rtattr *enc2 = RTA_DATA(encap); + if (enc2->rta_type == MPLS_IPTUNNEL_DST) + ra->mpls_label = (ntohl(rta_get_u32(enc2)) >> 12) & 0xfffff; + } + } + break; case RTN_BLACKHOLE: ra->dest = RTD_BLACKHOLE; @@ -1409,10 +1521,10 @@ nl_parse_route(struct nl_parse_state *s, struct nlmsghdr *h) if (a->dest != RTD_MULTIPATH) { a->dest = RTD_MULTIPATH; - a->nexthops = nl_alloc_mpnh(s, a->gw, a->iface, 0); + a->nexthops = nl_alloc_mpnh(s, a->gw, a->mpls_label, a->iface, 0); } - mpnh_insert(&a->nexthops, nl_alloc_mpnh(s, ra->gw, ra->iface, 0)); + mpnh_insert(&a->nexthops, nl_alloc_mpnh(s, ra->gw, ra->mpls_label, ra->iface, 0)); } } diff --git a/sysdep/unix/krt.c b/sysdep/unix/krt.c index 07a55c0d..2d4c4029 100644 --- a/sysdep/unix/krt.c +++ b/sysdep/unix/krt.c @@ -657,7 +657,7 @@ krt_same_dest(rte *k, rte *e) switch (ka->dest) { case RTD_ROUTER: - return ipa_equal(ka->gw, ea->gw); + return ipa_equal(ka->gw, ea->gw) && (ka->mpls_label == ea->mpls_label); case RTD_DEVICE: return !strcmp(ka->iface->name, ea->iface->name); case RTD_MULTIPATH:

3 3

EBGP Multihop TTL vs. TCP accept()
by Matthew Walster 16 Aug '17

16 Aug '17

Hello, I've just spent a while trying to debug a BGP session not coming up with bird, it turns out I had mistakenly set the multihop setting to "1" instead of "2" -- my fault. However, the reason it took me so long to realise this error was because the TCP session is established, with default connection parameters, THEN the TTL limit is assigned to the skb. As far as I'm aware, with TCP sockets in both Linux and BSD, the connections are already established when accept() returns a new connection. As I understand it, that's why inbound connections can't be filtered by the incoming source address either until the TCP session is already established, which usually means iptables/pf rules on the box synced with whoever the configured peers are. Which tends to lead to log spam when [known security company] does yet another scan of global address space for open BGP neighbours and fires a bunch of alerts... This may be a case of "that's just what you have to live with" but I was wondering if there's an alternative? Obviously libwrap0 (tcp-wrappers) won't help here because that also just processes a rule-set after accept() -- it's presumably something that needs to be addressed in the kernel. In any case, you can't set the TTL on a listen() for the SYN-ACK that is returned on a SYNC as far as I'm aware, unless you modify the default TTL for all TCP connections, you can't even set it with "ip route" or "route" on a per-destination basis. How do other people handle this situation? Do they create iptables/pf/whatever rules dynamically generated from their bird.conf for neighbor access control? How about ebgp multihop -- or does everyone just set it to 16/64 and forget about it? Or do people largely not care? M

2 1

AUTOMATIC INCOMING FILTERING
by Janvier Rwakagabo 10 Aug '17

10 Aug '17

Hi All Has anyone automated prefix filtering, for example if a peer acquire a new prefix to be received automatically may be any IRR, you can share the working configuration. Find on the attachment the manual prefix list filtering I am using, Regards, Janvier R.

2 1

bird: memory exhausted
by Job Snijders 10 Aug '17

10 Aug '17

Dear all, It appears I'm hitting some kind of limit in yacc stuff (on 1.6.3-1+trusty+1): router# bird -c rpki-match-roa-ipv4.conf -p bird: rpki-match-roa-ipv4.conf, line 4998: memory exhausted The error message is generated by generated code: $ grep -r 'memory exhausted' * Binary file bird matches obj/conf/cf-parse.tab.c: yyoverflow (YY_("memory exhausted"), obj/conf/cf-parse.tab.c: yyerror (YY_("memory exhausted")); Binary file obj/conf/all.o matches Binary file obj/conf/cf-parse.tab.o matches The config file can be found here: http://instituut.net/~job/rpki-match-roa-ipv4.conf.txt Any ideas how to increase the memory available to things on the stack? Kind regards, Job

3 3

Re: Unexpected behavior of static routes
by Alexander Demenshin 10 Aug '17

10 Aug '17

On 2017-05-10 13:16, Ondrej Zajicek wrote: > But such design would bring plenty of issues w.r.t. multiple routing > tables - may next hops resolve just in the same routing table or also > in > another routing table? Well, at least in Linux device/direct routes may exist in any table, so where is the difference between device/direct routes and arbitrary next-hop route? How I would implement this: - Try to find if next-hop is routable via any of existing devices (including explicitly routed networks - as in my case); - If it is routable (which also means that device has to be up), install it (so it can be exported to the kernel tables). > What if the device route is available, but not exported to kernel? You mean if it is configured but not exported? Then it should be ignored, obviously, though this could happen only in few cases: - device does not exists (yet) - device is not up (kernel will not accept any routes in this case) - protocol is disabled If no kernel export is configured at all then the route does not make any sense anyway. > Static onlink routes is something that probably could be implemented > and > it is a good idea. This could solve one of the most typical scenarios - keep static (local) routes up for specified set of devices/gateways - once they become available. What is interesting, quagga behaves exactly in the same way as bird, and only cisco (well, may be other routers too) does what is (logically) expected - and it allows Linux-like onlink routes, which specify both device and gateway address, and those routes are linked to device state (i.e. exported and used when device is active). --- Best regards, Alexander.

3 3

gateway recursive
by Mikhail Mayorov 10 Aug '17

10 Aug '17

Hi all! Please help me with recursion to caclulate bgp_next_hop using IGP. I have 2 router "A" and "B" connected to each other via iBGP. Each of the routers has connection to the outside world in different AS and get full view. "A" has a eBGP session to the neighbor "A2" and "B" has a eBGP session to the neighbor "B2". I configure "gateway recursive; direct;" on iBGP. In this case, all the routes from the A2 got to B via A are as unreachable. A IP: 95.174.97.1/26 A2 IP: 188.43.24.90/30 B: 95.174.97.33/26 B2: 87.245.238.204/31 This is debug from birdcl router B: bird ~ # birdcl 'show route for 37.231.192.1' table master all BIRD 1.6.3 ready. 37.231.192.0/22 via 87.245.238.204 on vlan91-retn [retn_ipv4 07.08.2017 11:06:57] * (100) [AS47589i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: 9002 15412 9155 47589 BGP.next_hop: 87.245.238.204 BGP.med: 999 BGP.local_pref: 100 BGP.community: (9002,9002) (9002,64679) (49037,0) (49037,10001) unreachable [twix_ipv4 07.08.2017 17:17:17 from 95.174.97.1] (100) [AS47589i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: 20485 15412 9155 47589 BGP.next_hop: 188.43.24.90 BGP.med: 999 BGP.local_pref: 100 BGP.community: (20485,22007) (20485,22307) (49037,0) bird ~ # birdcl 'show route for 188.43.24.90' table master all BIRD 1.6.3 ready. 188.43.24.88/30 via 95.174.97.1 on vlan8 [twix_ipv4 07.08.2017 17:17:17] * (100) [i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: BGP.next_hop: 95.174.97.1 BGP.med: 0 BGP.local_pref: 500 BGP.community: (49037,49037) bird ~ # birdcl 'show route for 95.174.97.1' table master all BIRD 1.6.3 ready. 95.174.97.0/26 dev vlan8 [direct1 07.08.2017 00:20:09] * (240) Type: device unicast univ via 95.174.97.1 on vlan8 [twix_ipv4 07.08.2017 17:17:17] (100) [i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: BGP.next_hop: 95.174.97.1 BGP.med: 0 BGP.local_pref: 500 BGP.community: (49037,49037) -- Regards, Mikhail V. Majorov

2 2

Error: Hold timer expired
by Marek Królikowski 09 Aug '17

09 Aug '17

Hello guys, I buy VPS with BGP and try setup BGP session but got strange problem. VPS support send me this information to setup session: "Please peer with 185.144.XXX.YYY, you will need to enable eBGP multihop. Our ASN is XXXX. The IPv6 peer address is 2a04:9dc0:XX:YYY::1" My bird configuration looks like this: --------------------------------------- router id 85.204.XXX.YYY; # debug protocols all; protocol device { scan time 10; } protocol static { route 2a04:9dc0:XX:YYY::1/128 via "eth0"; } protocol kernel { device routes; import none; scan time 15; export filter { if net = 2a04:9dc0:XX:YYY::1/128 then reject; if source = 2a04:9dc0:XX:YYY::1/128 then reject; accept; }; } protocol bgp Romania { description "Romania link local"; local as YYYYY; source address 2a04:9dc0:X:YY::f0; graceful restart on; default bgp_local_pref 200; multihop 2; neighbor 2a04:9dc0:XX:YYY::1 as XXXX; import filter { if net = 2a04:9dc0:XX:YYY::1/128 then reject; if source = 2a04:9dc0:XX:YYY::1/128 then reject; if bgp_path.len > 49 then reject; accept; }; export filter { if net = 2a07:1c44:3511::/48 then accept; reject; }; } --------------------------------------- In logs got it: Aug 8 16:21:41 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 16:24:02 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 16:27:07 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 16:35:58 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 16:37:25 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 16:39:27 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 16:41:57 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 16:56:19 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 17:00:29 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 17:06:10 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 17:09:58 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 17:13:01 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 17:15:08 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 17:18:15 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 17:22:58 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 17:29:17 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 17:31:49 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 17:35:17 ro-bgp bird6: Romania: Error: Hold timer expired Aug 8 17:40:08 ro-bgp bird6: Romania: Error: Hold timer expired root@bgp:~# birdc6 show protocols all Romania BIRD 1.6.3 ready. name proto table state since info Romania BGP master up 16:22:37 Established Description: Romania link local Preference: 100 Input filter: (unnamed) Output filter: (unnamed) Routes: 11560 imported, 5 exported, 11560 preferred Route change stats: received rejected filtered ignored accepted Import updates: 11560 0 0 0 11560 Import withdraws: 0 0 --- 0 0 Export updates: 11586 11560 21 --- 5 Export withdraws: 0 --- --- --- 0 BGP state: Established Neighbor address: 2a04:9dc0:XX:YYY::1 Neighbor AS: XXXX Neighbor ID: 185.144.XXXX Neighbor caps: refresh restart-aware AS4 Session: external multihop AS4 Source address: 2a04:9dc0:X:YY::f0; Hold timer: 27/90 Keepalive timer: 23/30 root@bgp:~# birdc6 show protocols all Romania BIRD 1.6.3 ready. name proto table state since info Romania BGP master start 16:24:05 Idle BGP Error: Hold timer expired Description: Romania link local Preference: 100 Input filter: (unnamed) Output filter: (unnamed) Routes: 0 imported, 0 exported, 0 preferred Route change stats: received rejected filtered ignored accepted Import updates: 0 0 0 0 0 Import withdraws: 0 0 --- 0 0 Export updates: 0 0 0 --- 0 Export withdraws: 0 --- --- --- 0 BGP state: Idle Neighbor address: 2a04:9dc0:XX:YYY::1 Neighbor AS: XXXX Error wait: 57/120 Last error: BGP Error: Hold timer expired Best Regards Marek Krolikowski

2 1

BIRD BGP and VRF - Cannot assign requested address
by Clément Guivy 08 Aug '17

08 Aug '17

Hello, I am trying to set up 2 BIRD routers as AS border routers (on Debian 9). Installation is as follows : |BIRD router 1|-----IBGP-----|BIRD router 2| | | EBGP EBGP | | |Hardware router 1|---IBGP---|Hardware router 2| | (downstream) | | (downstream) | (later the BIRD routers would also be connected to our upstreams ISPs, but for the sake of simplicity let's just consider this part of the setup) I plan to use both ipv4 and ipv6 and also, on BIRD routers both the forwarding and the BGP peerings are to take place in a linux vrf named "internet" (using l3mdev, not namespaces or custom ip rules). At the BIRD level I also set up a table named "internet". Now, to my problem : the EBGP peerings with the downstreams routers are UP, but on the other hand the IBGP between the two BIRD routers won't come up, I get the following error messages : On ipv4 peering, the service complains that : ibgp_internet: Socket error: bind: Cannot assign requested address On ipv6 peering : the service won't complain but using birdc6 I can see that the protocol says : Socket: Network is unreachable Both routers have almost identical config (apart from the IP addresses of course), here is the conf on BIRD router 1 : ------------------------------------------------------ ---------------------bird.conf------------------------ ------------------------------------------------------ table internet; router id 10.206.81.36; define my_as=205555; protocol device { } protocol kernel { kernel table 10; table internet; metric 64; import all; export all; scan time 10; } protocol static default_route_to_bgp { import all; table internet; route 0.0.0.0/0 reject; } template bgp template_base_bgp { med metric; } protocol bgp ibgp_internet from template_base_bgp { table internet; source address 10.206.81.81; local as my_as; neighbor 10.206.81.82 as my_as; import all; export all; next hop self; } protocol bgp downstream_internet from template_base_bgp { table internet; export where proto = "default_route_to_bgp"; source address 10.206.81.90; local as my_as; neighbor 10.206.81.89 as 65206; } ------------------------------------------------------ ------------------end of bird.conf-------------------- ------------------------------------------------------ ------------------------------------------------------ ---------------------bird6.conf----------------------- ------------------------------------------------------ table internet; router id 10.206.81.36; define my_as=205555; protocol device { } protocol kernel { kernel table 10; table internet; metric 64; import all; export all; scan time 10; } template bgp template_base_bgp { med metric; } protocol bgp ibgp_internet from template_base_bgp { table internet; source address 2a0b:95c0:0000:0101::1; local as my_as; neighbor 2a0b:95c0:0000:0101::2 as my_as; import all; export all; next hop self; } protocol static default_route_to_bgp { import all; table internet; route ::/0 reject; } protocol bgp downstream_internet from template_base_bgp { table internet; export where proto = "default_route_to_bgp"; source address 2a0b:95c0:0000:0102::2; local as my_as; neighbor 2a0b:95c0:0000:0102::1 as 65206; } ------------------------------------------------------ -----------------end of bird6.conf-------------------- ------------------------------------------------------ The network interfaces seems to be properly UP and attached to the internet VRF, here is the IBGP one : # ip addr show eth1.3 7: eth1.3@eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master internet state UP group default qlen 1000 link/ether 94:18:82:ab:8b:ec brd ff:ff:ff:ff:ff:ff inet 10.206.81.81/29 brd 10.206.81.87 scope global eth1.3 valid_lft forever preferred_lft forever inet6 2a0b:95c0:0:101::1/64 scope global valid_lft forever preferred_lft forever inet6 fe80::9618:82ff:feab:8bec/64 scope link valid_lft forever preferred_lft forever and here is the EBGP one (the one which has the functional peering) : # ip addr show ens1.14 8: ens1.14@ens1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue master internet state UP group default qlen 1000 link/ether 9c:dc:71:45:5f:e0 brd ff:ff:ff:ff:ff:ff inet 10.206.81.90/29 brd 10.206.81.95 scope global ens1.14 valid_lft forever preferred_lft forever inet6 2a0b:95c0:0:102::2/64 scope global valid_lft forever preferred_lft forever inet6 fe80::9edc:71ff:fe45:5fe0/64 scope link valid_lft forever preferred_lft forever Now the "internet" kernel routing table, we can see that the local and broadcast routes are attached to it too : # ip route show table internet unreachable default proto bird metric 64 broadcast 10.206.81.80 dev eth1.3 proto kernel scope link src 10.206.81.81 10.206.81.80/29 dev eth1.3 proto kernel scope link src 10.206.81.81 local 10.206.81.81 dev eth1.3 proto kernel scope host src 10.206.81.81 broadcast 10.206.81.87 dev eth1.3 proto kernel scope link src 10.206.81.81 broadcast 10.206.81.88 dev ens1.14 proto kernel scope link src 10.206.81.90 10.206.81.88/29 dev ens1.14 proto kernel scope link src 10.206.81.90 local 10.206.81.90 dev ens1.14 proto kernel scope host src 10.206.81.90 broadcast 10.206.81.95 dev ens1.14 proto kernel scope link src 10.206.81.90 Finally, netstat shows that the BGP service is listening on * as it is supposed to. We also see the established EBGP downstream sessions : # netstat -an | grep 179 tcp 0 0 0.0.0.0:179 0.0.0.0:* LISTEN tcp 0 0 10.206.81.90:179 10.206.81.89:9266 ESTABLISHED tcp6 0 0 :::179 :::* LISTEN tcp6 0 0 2a0b:95c0:0:102::2:179 2a0b:95c0:0:102:::64247 ESTABLISHED and tcp_l3mdev_accept is enabled in the kernel config : # sysctl net.ipv4.tcp_l3mdev_accept net.ipv4.tcp_l3mdev_accept = 1 So overall it seems like BIRD fails to bind to one interface, this is possibly linked to the kernel VRF setup but I can't see why it doesn't work when the peering to downstream routers using the same VRF do work properly. Any help would be greatly appreciated. Thanks. clément

2 9

Hold time expired error
by Ajai Kumar 06 Aug '17

06 Aug '17

Dear All, We are facing problem with few peers in our IX. Frequently they get Hold Timer Expired Error and they are not able to ping our route server IP. After shut no shut this problem resolves for sometime. One log message appended bgp_hold_timeout:4690: NOTIFICATION sent to X.X.X.X (External AS 132953): code 4 (Hold Timer Expired Error), Reason: holdtime expired for X.X.X.X (External AS 132953), socket buffer sndcc: 57 rcvcc: 0 TCP state: 4, snd_una: 728368567 snd_nxt: 728368624 snd_wnd: 15744 rcv_nxt: 429172199 rcv_adv: 429188583, hold timer out 90s, hold timer remain 0s Requesting for help pls. Regards, Ajai Kumar -- (M) +91-9868477444 Skype ID:erajay P-mail: joinajay1 at gmail.com ................................. Please don't print this email unless you really need to. This will preserve trees on our planet.

4 5

Unable to connect to server control socket (/run/bird/bird.ctl): No such file or directory
by Janvier Rwakagabo 05 Aug '17

05 Aug '17

Hi all Need your help, I have installed properly bird, Edit bird.conf file with my configuration, then when I restart the service bird it give this error #/etc/init.d/bird restart stop: Unknown instance: start: Job failed to start by typing #birdc it give below error #birdc Unable to connect to server control socket (/run/bird/bird.ctl): No such file or directory I checked under /run/bird/ folder there is no bird.ctl only bird6.ctl is present I deleted bird.conf and restaured default bird.conf the issue was solved, by editing again bird.conf and restarting the service, bird.ctl is droped again. Can anyone assist me to solve this issue. Thanks, Regards, Janvier R.

2 2