Softreconfig out support. On config reload filter changes of outgoing rules

will propagte directly to the neighbors. There is no need to restart bgpd
in that case. Currently not optimal but a good start. "get it in" henning@

1 files changed, 66 insertions, 12 deletions

diff --git a/usr.sbin/bgpd/rde.c b/usr.sbin/bgpd/rde.c
index d2dcd3559ca..14eda22d516 100644
--- a/usr.sbin/bgpd/rde.c
+++ b/usr.sbin/bgpd/rde.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: rde.c,v 1.176 2005/11/01 14:37:16 claudio Exp $ */
+/*	$OpenBSD: rde.c,v 1.177 2005/11/01 15:21:54 claudio Exp $ */
 
 /*
  * Copyright (c) 2003, 2004 Henning Brauer <henning@openbsd.org>
@@ -65,6 +65,7 @@ void		 rde_dump_upcall(struct pt_entry *, void *);
 void		 rde_dump_as(struct filter_as *, pid_t);
 void		 rde_dump_prefix_upcall(struct pt_entry *, void *);
 void		 rde_dump_prefix(struct ctl_show_rib_prefix *, pid_t);
+void		 rde_softreconfig_out(struct pt_entry *, void *);
 void		 rde_up_dump_upcall(struct pt_entry *, void *);
 void		 rde_update_queue_runner(void);
 void		 rde_update6_queue_runner(void);
@@ -520,7 +521,8 @@ rde_dispatch_imsg_parent(struct imsgbuf *ibuf)
 			nconf = NULL;
 			parent_set = NULL;
 			prefix_network_clean(&peerself, reloadtime);
-			/* soft reconfig out */
+			
+			pt_dump(rde_softreconfig_out, NULL, AF_UNSPEC);
 			while ((r = TAILQ_FIRST(rules_l)) != NULL) {
 				TAILQ_REMOVE(rules_l, r, entry);
 				filterset_free(&r->set);
@@ -706,7 +708,7 @@ rde_update_dispatch(struct imsg *imsg)
 		}
 
 		/* input filter */
-		if (rde_filter(peer, NULL, &prefix, prefixlen, peer,
+		if (rde_filter(rules_l, peer, NULL, &prefix, prefixlen, peer,
 		    DIR_IN) == ACTION_DENY)
 			continue;
 
@@ -762,8 +764,8 @@ rde_update_dispatch(struct imsg *imsg)
 				mplen -= pos;
 
 				/* input filter */
-				if (rde_filter(peer, NULL, &prefix, prefixlen,
-				    peer, DIR_IN) == ACTION_DENY)
+				if (rde_filter(rules_l, peer, NULL, &prefix,
+				    prefixlen, peer, DIR_IN) == ACTION_DENY)
 					continue;
 
 				rde_update_log("withdraw", peer, NULL,
@@ -827,7 +829,7 @@ rde_update_dispatch(struct imsg *imsg)
 		 */
 		fasp = path_copy(asp);
 		/* input filter */
-		if (rde_filter(peer, fasp, &prefix, prefixlen, peer,
+		if (rde_filter(rules_l, peer, fasp, &prefix, prefixlen, peer,
 		    DIR_IN) == ACTION_DENY) {
 			path_put(fasp);
 			continue;
@@ -908,8 +910,8 @@ rde_update_dispatch(struct imsg *imsg)
 
 				fasp = path_copy(asp);
 				/* input filter */
-				if (rde_filter(peer, fasp, &prefix, prefixlen,
-				    peer, DIR_IN) == ACTION_DENY) {
+				if (rde_filter(rules_l, peer, fasp, &prefix,
+				    prefixlen, peer, DIR_IN) == ACTION_DENY) {
 					path_put(fasp);
 					continue;
 				}
@@ -1777,13 +1779,65 @@ rde_send_nexthop(struct bgpd_addr *next, int valid)
 u_char	queue_buf[4096];
 
 void
+rde_softreconfig_out(struct pt_entry *pt, void *ptr)
+{
+	struct prefix		*p = pt->active;
+	struct rde_peer		*peer;
+	struct rde_aspath	*oasp, *nasp;
+	enum filter_actions	 oa, na;
+	struct bgpd_addr	 addr;
+	
+	if (p == NULL)
+		return;
+
+	pt_getaddr(pt, &addr);
+	LIST_FOREACH(peer, &peerlist, peer_l) {
+		if (up_test_update(peer, p) != 1)
+			continue;
+
+		/* copy attributes for output filter */
+		oasp = path_copy(p->aspath);
+		nasp = path_copy(p->aspath);
+
+		oa = rde_filter(rules_l, peer, oasp, &addr, pt->prefixlen,
+		    p->peer, DIR_OUT);
+		na = rde_filter(newrules, peer, nasp, &addr, pt->prefixlen,
+		    p->peer, DIR_OUT);
+
+		if (oa == ACTION_DENY && na == ACTION_DENY)
+			/* nothing todo */
+			goto done;
+		if (oa == ACTION_DENY && na == ACTION_ALLOW) {
+			/* send update */
+			up_generate(peer, nasp, &addr, pt->prefixlen);
+			goto done;
+		}
+		if (oa == ACTION_ALLOW && na == ACTION_DENY) {
+			/* send withdraw */
+			up_generate(peer, NULL, &addr, pt->prefixlen);
+			goto done;
+		}
+		if (oa == ACTION_ALLOW && na == ACTION_ALLOW) {
+			if (path_compare(nasp, oasp) == 0) 
+				goto done;
+			/* send update */
+			up_generate(peer, nasp, &addr, pt->prefixlen);
+		}
+
+done:
+		path_put(oasp);
+		path_put(nasp);
+	}
+}
+
+void
 rde_up_dump_upcall(struct pt_entry *pt, void *ptr)
 {
 	struct rde_peer		*peer = ptr;
 
 	if (pt->active == NULL)
 		return;
-	up_generate_updates(peer, pt->active, NULL);
+	up_generate_updates(rules_l, peer, pt->active, NULL);
 }
 
 void
@@ -1803,7 +1857,7 @@ rde_generate_updates(struct prefix *new, struct prefix *old)
 	LIST_FOREACH(peer, &peerlist, peer_l) {
 		if (peer->state != PEER_UP)
 			continue;
-		up_generate_updates(peer, new, old);
+		up_generate_updates(rules_l, peer, new, old);
 	}
 }
 
@@ -2150,7 +2204,7 @@ peer_dump(u_int32_t id, u_int16_t afi, u_int8_t safi)
 		    safi == SAFI_BOTH) {
 			if (peer->conf.announce_type ==
 			    ANNOUNCE_DEFAULT_ROUTE)
-				up_generate_default(peer, AF_INET);
+				up_generate_default(rules_l, peer, AF_INET);
 			else
 				pt_dump(rde_up_dump_upcall, peer, AF_INET);
 		}
@@ -2159,7 +2213,7 @@ peer_dump(u_int32_t id, u_int16_t afi, u_int8_t safi)
 		    safi == SAFI_BOTH) {
 			if (peer->conf.announce_type ==
 			    ANNOUNCE_DEFAULT_ROUTE)
-				up_generate_default(peer, AF_INET6);
+				up_generate_default(rules_l, peer, AF_INET6);
 			else
 				pt_dump(rde_up_dump_upcall, peer, AF_INET6);
 		}