Merge pull request #17652 from pguibert6WIND/topotest_bgp_evpn_rt5

bgpd, tests: bgp_evpn_rt5, add test with match evpn vni command

bgpd/bgp_route.c

@@ -4923,6 +4923,7 @@ void bgp_update(struct peer *peer, const struct prefix *p, uint32_t addpath_id,
 	bool force_evpn_import = false;
 	safi_t orig_safi = safi;
 	struct bgp_labels bgp_labels = {};
+	struct bgp_route_evpn *p_evpn = evpn;
 	uint8_t i;
 
 	if (frrtrace_enabled(frr_bgp, process_update)) {
@@ -4964,11 +4965,9 @@ void bgp_update(struct peer *peer, const struct prefix *p, uint32_t addpath_id,
 		 * will not be interned. In which case, it is ok to update the
 		 * attr->evpn_overlay, so that, this can be stored in adj_in.
 		 */
-		if (evpn) {
-			if (afi == AFI_L2VPN)
-				bgp_attr_set_evpn_overlay(attr, evpn);
-			else
-				evpn_overlay_free(evpn);
+		if (evpn && afi == AFI_L2VPN) {
+			bgp_attr_set_evpn_overlay(attr, evpn);
+			p_evpn = NULL;
 		}
 		bgp_adj_in_set(dest, peer, attr, addpath_id, &bgp_labels);
 	}
@@ -5141,11 +5140,9 @@ void bgp_update(struct peer *peer, const struct prefix *p, uint32_t addpath_id,
 	 * attr->evpn_overlay with evpn directly. Instead memcpy
 	 * evpn to new_atr.evpn_overlay before it is interned.
 	 */
-	if (soft_reconfig && evpn) {
-		if (afi == AFI_L2VPN)
-			bgp_attr_set_evpn_overlay(&new_attr, evpn);
-		else
-			evpn_overlay_free(evpn);
+	if (soft_reconfig && evpn && afi == AFI_L2VPN) {
+		bgp_attr_set_evpn_overlay(&new_attr, evpn);
+		p_evpn = NULL;
 	}
 
 	/* Apply incoming route-map.
@@ -5314,7 +5311,8 @@ void bgp_update(struct peer *peer, const struct prefix *p, uint32_t addpath_id,
 
 			bgp_dest_unlock_node(dest);
 			bgp_attr_unintern(&attr_new);
-
+			if (p_evpn)
+				evpn_overlay_free(p_evpn);
 			return;
 		}
 
@@ -5479,6 +5477,8 @@ void bgp_update(struct peer *peer, const struct prefix *p, uint32_t addpath_id,
 			ret = bgp_damp_update(pi, dest, afi, safi);
 			if (ret == BGP_DAMP_SUPPRESSED) {
 				bgp_dest_unlock_node(dest);
+				if (p_evpn)
+					evpn_overlay_free(p_evpn);
 				return;
 			}
 		}
@@ -5565,6 +5565,8 @@ void bgp_update(struct peer *peer, const struct prefix *p, uint32_t addpath_id,
 					   type, sub_type, NULL);
 		}
 #endif
+		if (p_evpn)
+			evpn_overlay_free(p_evpn);
 		return;
 	} // End of implicit withdraw
 
@@ -5659,6 +5661,8 @@ void bgp_update(struct peer *peer, const struct prefix *p, uint32_t addpath_id,
 	}
 #endif
 
+	if (p_evpn)
+		evpn_overlay_free(p_evpn);
 	return;
 
 /* This BGP update is filtered.  Log the reason then update BGP
@@ -5722,6 +5726,8 @@ void bgp_update(struct peer *peer, const struct prefix *p, uint32_t addpath_id,
 	}
 #endif
 
+	if (p_evpn)
+		evpn_overlay_free(p_evpn);
 	return;
 }
 

bgpd/bgp_vty.c

@@ -1499,13 +1499,12 @@ DEFUN_NOSH (router_bgp,
 	int idx_asn = 2;
 	int idx_view_vrf = 3;
 	int idx_vrf = 4;
-	int is_new_bgp = 0;
 	int idx_asnotation = 3;
 	int idx_asnotation_kind = 4;
 	enum asnotation_mode asnotation = ASNOTATION_UNDEFINED;
 	int ret;
 	as_t as;
-	struct bgp *bgp;
+	struct bgp *bgp = NULL;
 	const char *name = NULL;
 	enum bgp_instance_type inst_type;
 
@@ -1567,35 +1566,40 @@ DEFUN_NOSH (router_bgp,
 				asnotation = ASNOTATION_PLAIN;
 		}
 
-		if (inst_type == BGP_INSTANCE_TYPE_DEFAULT)
-			is_new_bgp = (bgp_lookup(as, name) == NULL);
-
-		ret = bgp_get_vty(&bgp, &as, name, inst_type,
-				  argv[idx_asn]->arg, asnotation);
+		ret = bgp_lookup_by_as_name_type(&bgp, &as, argv[idx_asn]->arg, asnotation, name,
+						 inst_type, true);
+		if (bgp && ret == BGP_INSTANCE_EXISTS)
+			ret = CMD_SUCCESS;
+		else if (bgp == NULL && ret == CMD_SUCCESS)
+			/* SUCCESS and bgp is NULL */
+			ret = bgp_get_vty(&bgp, &as, name, inst_type, argv[idx_asn]->arg,
+					  asnotation);
 		switch (ret) {
 		case BGP_ERR_AS_MISMATCH:
 			vty_out(vty, "BGP is already running; AS is %s\n",
-				bgp->as_pretty);
+				bgp ? bgp->as_pretty : "unknown");
 			return CMD_WARNING_CONFIG_FAILED;
 		case BGP_ERR_INSTANCE_MISMATCH:
 			vty_out(vty,
 				"BGP instance name and AS number mismatch\n");
-			vty_out(vty,
-				"BGP instance is already running; AS is %s\n",
-				bgp->as_pretty);
+			vty_out(vty, "BGP instance is already running; AS is %s\n",
+				bgp ? bgp->as_pretty : "unknown");
 			return CMD_WARNING_CONFIG_FAILED;
 		}
 
+		if (!bgp) {
+			vty_out(vty, "BGP instance not found\n");
+			return CMD_WARNING_CONFIG_FAILED;
+		}
 		/*
 		 * If we just instantiated the default instance, complete
 		 * any pending VRF-VPN leaking that was configured via
 		 * earlier "router bgp X vrf FOO" blocks.
 		 */
-		if (is_new_bgp && inst_type == BGP_INSTANCE_TYPE_DEFAULT)
+		if (inst_type == BGP_INSTANCE_TYPE_DEFAULT)
 			vpn_leak_postchange_all();
 
-		if (inst_type == BGP_INSTANCE_TYPE_VRF ||
-		    IS_BGP_INSTANCE_HIDDEN(bgp)) {
+		if (inst_type == BGP_INSTANCE_TYPE_VRF || IS_BGP_INSTANCE_HIDDEN(bgp)) {
 			bgp_vpn_leak_export(bgp);
 			UNSET_FLAG(bgp->flags, BGP_FLAG_INSTANCE_HIDDEN);
 			UNSET_FLAG(bgp->flags, BGP_FLAG_DELETE_IN_PROGRESS);
@@ -10559,7 +10563,7 @@ DEFPY(bgp_imexport_vrf, bgp_imexport_vrf_cmd,
 		SET_FLAG(bgp_default->flags, BGP_FLAG_INSTANCE_HIDDEN);
 	}
 
-	vrf_bgp = bgp_lookup_by_name(import_name);
+	vrf_bgp = bgp_lookup_by_name_filter(import_name, false);
 	if (!vrf_bgp) {
 		if (strcmp(import_name, VRF_DEFAULT_NAME) == 0) {
 			vrf_bgp = bgp_default;

bgpd/bgpd.c

@@ -3634,13 +3634,13 @@ struct bgp *bgp_lookup(as_t as, const char *name)
 }
 
 /* Lookup BGP structure by view name. */
-struct bgp *bgp_lookup_by_name(const char *name)
+struct bgp *bgp_lookup_by_name_filter(const char *name, bool filter_auto)
 {
 	struct bgp *bgp;
 	struct listnode *node, *nnode;
 
 	for (ALL_LIST_ELEMENTS(bm->bgp, node, nnode, bgp)) {
-		if (CHECK_FLAG(bgp->vrf_flags, BGP_VRF_AUTO))
+		if (filter_auto && CHECK_FLAG(bgp->vrf_flags, BGP_VRF_AUTO))
 			continue;
 		if ((bgp->name == NULL && name == NULL)
 		    || (bgp->name && name && strcmp(bgp->name, name) == 0))
@@ -3649,6 +3649,11 @@ struct bgp *bgp_lookup_by_name(const char *name)
 	return NULL;
 }
 
+struct bgp *bgp_lookup_by_name(const char *name)
+{
+	return bgp_lookup_by_name_filter(name, true);
+}
+
 /* Lookup BGP instance based on VRF id. */
 /* Note: Only to be used for incoming messages from Zebra. */
 struct bgp *bgp_lookup_by_vrf_id(vrf_id_t vrf_id)
@@ -3734,10 +3739,9 @@ int bgp_handle_socket(struct bgp *bgp, struct vrf *vrf, vrf_id_t old_vrf_id,
 		return bgp_check_main_socket(create, bgp);
 }
 
-int bgp_lookup_by_as_name_type(struct bgp **bgp_val, as_t *as,
-			       const char *as_pretty,
+int bgp_lookup_by_as_name_type(struct bgp **bgp_val, as_t *as, const char *as_pretty,
 			       enum asnotation_mode asnotation, const char *name,
-			       enum bgp_instance_type inst_type)
+			       enum bgp_instance_type inst_type, bool force_config)
 {
 	struct bgp *bgp;
 	struct peer *peer = NULL;
@@ -3746,7 +3750,7 @@ int bgp_lookup_by_as_name_type(struct bgp **bgp_val, as_t *as,
 
 	/* Multiple instance check. */
 	if (name)
-		bgp = bgp_lookup_by_name(name);
+		bgp = bgp_lookup_by_name_filter(name, !force_config);
 	else
 		bgp = bgp_get_default();
 
@@ -3756,15 +3760,16 @@ int bgp_lookup_by_as_name_type(struct bgp **bgp_val, as_t *as,
 		/* Handle AS number change */
 		if (bgp->as != *as) {
 			if (hidden || CHECK_FLAG(bgp->vrf_flags, BGP_VRF_AUTO)) {
-				if (hidden) {
+				if (force_config == false && hidden) {
 					bgp_create(as, name, inst_type,
 						   as_pretty, asnotation, bgp,
 						   hidden);
 					UNSET_FLAG(bgp->flags,
 						   BGP_FLAG_INSTANCE_HIDDEN);
 				} else {
 					bgp->as = *as;
-					UNSET_FLAG(bgp->vrf_flags, BGP_VRF_AUTO);
+					if (force_config == false)
+						UNSET_FLAG(bgp->vrf_flags, BGP_VRF_AUTO);
 				}
 
 				/* Set all peer's local AS with this ASN */
@@ -3801,8 +3806,7 @@ int bgp_get(struct bgp **bgp_val, as_t *as, const char *name,
 	struct vrf *vrf = NULL;
 	int ret = 0;
 
-	ret = bgp_lookup_by_as_name_type(bgp_val, as, as_pretty, asnotation,
-					 name, inst_type);
+	ret = bgp_lookup_by_as_name_type(bgp_val, as, as_pretty, asnotation, name, inst_type, false);
 	if (ret || *bgp_val)
 		return ret;
 

bgpd/bgpd.h

@@ -2285,6 +2285,7 @@ extern void bgp_zclient_reset(void);
 extern struct bgp *bgp_get_default(void);
 extern struct bgp *bgp_lookup(as_t, const char *);
 extern struct bgp *bgp_lookup_by_name(const char *);
+extern struct bgp *bgp_lookup_by_name_filter(const char *name, bool filter_auto);
 extern struct bgp *bgp_lookup_by_vrf_id(vrf_id_t);
 extern struct bgp *bgp_get_evpn(void);
 extern void bgp_set_evpn(struct bgp *bgp);
@@ -2859,11 +2860,9 @@ extern struct peer *peer_new(struct bgp *bgp);
 
 extern struct peer *peer_lookup_in_view(struct vty *vty, struct bgp *bgp,
 					const char *ip_str, bool use_json);
-extern int bgp_lookup_by_as_name_type(struct bgp **bgp_val, as_t *as,
-				      const char *as_pretty,
-				      enum asnotation_mode asnotation,
-				      const char *name,
-				      enum bgp_instance_type inst_type);
+extern int bgp_lookup_by_as_name_type(struct bgp **bgp_val, as_t *as, const char *as_pretty,
+				      enum asnotation_mode asnotation, const char *name,
+				      enum bgp_instance_type inst_type, bool force_config);
 
 /* Hooks */
 DECLARE_HOOK(bgp_vrf_status_changed, (struct bgp *bgp, struct interface *ifp),

tests/topotests/bgp_bmp/test_bgp_bmp_1.py

@@ -78,7 +78,7 @@ def setup_module(mod):
             "tcpdump -nni r1-eth0 -s 0 -w {} &".format(pcap_file), stdout=None
         )
 
-    for rname, router in tgen.routers().items():
+    for _, (rname, router) in enumerate(tgen.routers().items(), 1):
         logger.info("Loading router %s" % rname)
         router.load_frr_config(
             os.path.join(CWD, "{}/frr.conf".format(rname)),

tests/topotests/bgp_evpn_rt5/r1/bgp_l2vpn_evpn_routes.json

@@ -2,8 +2,8 @@
     "bgpLocalRouterId":"192.168.100.21",
     "defaultLocPrf":100,
     "localAS":65000,
-    "192.168.101.41:2":{
-        "rd":"192.168.101.41:2",
+    "65000:201":{
+        "rd":"65000:201",
         "[5]:[0]:[32]:[192.168.101.41]":{
             "prefix":"[5]:[0]:[32]:[192.168.101.41]",
             "prefixLen":352,
@@ -65,8 +65,8 @@
             ]
         }
     },
-    "192.168.102.21:2":{
-        "rd":"192.168.102.21:2",
+    "65000:101":{
+        "rd":"65000:101",
         "[5]:[0]:[32]:[192.168.102.21]":{
             "prefix":"[5]:[0]:[32]:[192.168.102.21]",
             "prefixLen":352,

tests/topotests/bgp_evpn_rt5/r1/bgp_l2vpn_evpn_routes_all.json

@@ -2,8 +2,8 @@
     "bgpLocalRouterId":"192.168.100.21",
     "defaultLocPrf":100,
     "localAS":65000,
-    "192.168.101.41:2":{
-        "rd":"192.168.101.41:2",
+    "65000:201":{
+        "rd":"65000:201",
         "[5]:[0]:[32]:[192.168.101.41]":{
             "prefix":"[5]:[0]:[32]:[192.168.101.41]",
             "prefixLen":352,
@@ -125,8 +125,8 @@
             ]
         }
     },
-    "192.168.102.21:2":{
-        "rd":"192.168.102.21:2",
+    "65000:101":{
+        "rd":"65000:101",
         "[5]:[0]:[32]:[192.168.102.21]":{
             "prefix":"[5]:[0]:[32]:[192.168.102.21]",
             "prefixLen":352,

tests/topotests/bgp_evpn_rt5/r1/bgpd.conf → tests/topotests/bgp_evpn_rt5/r1/frr.conf

@@ -1,15 +1,36 @@
+! debug zebra vxlan
+! debug zebra kernel
+! debug zebra dplane
+! debug zebra rib
 ! debug bgp neighbor-events
 ! debug bgp updates
 ! debug bgp zebra
+vrf r1-vrf-101
+ vni 101
+ exit-vrf
+!
+interface r1-eth0
+ ip address 192.168.100.21/24
+!
+interface loop101 vrf r1-vrf-101
+ ip address 192.168.102.21/32
+ ipv6 address fd00::1/128
+!
 router bgp 65000
  bgp router-id 192.168.100.21
  bgp log-neighbor-changes
  no bgp default ipv4-unicast
+ no bgp ebgp-requires-policy
  neighbor 192.168.100.41 remote-as 65000
  neighbor 192.168.100.41 capability extended-nexthop
+ neighbor 192.168.100.61 remote-as 65500
+ neighbor 192.168.100.61 capability extended-nexthop
  !
  address-family l2vpn evpn
   neighbor 192.168.100.41 activate
+  neighbor 192.168.100.41 route-map rmap_r1 in
+  neighbor 192.168.100.61 activate
+  neighbor 192.168.100.61 route-map rmap_r3 in
   advertise-all-vni
  exit-address-family
 !
@@ -24,7 +45,17 @@ router bgp 65000 vrf r1-vrf-101
   network fd00::1/128
  exit-address-family
  address-family l2vpn evpn
+  rd 65000:101
+  route-target both 65:101
   advertise ipv4 unicast
   advertise ipv6 unicast
  exit-address-family
  !
+route-map rmap_r3 deny 1
+ match evpn vni 102
+exit
+route-map rmap_r1 permit 1
+ match evpn vni 101
+exit
+
+