zebra, bgp: Support type-5 routes with asymmetric routing

Asymmetric routing is an ideal choice when all VLANs are cfged on all leafs.
It simplifies the routing configuration and
eliminates potential need for advertising subnet routes.
However, we need to reach the Internet or global destinations
or to do subnet-based routing between PODs or DCs.
This requires EVPN type-5 routes but those routes require L3 VNI configuration.

This task is to support EVPN type-5 routes for prefix-based routing in
conjunction with asymmetric routing within the POD/DC.
It is done by providing an option to use the L3 VNI only for prefix routes,
so that type-2 routes (host routes) will only use the L2 VNI.

Signed-off-by: Mitesh Kanjariya <mitesh@cumulusnetworks.com>

bgpd/bgp_evpn.c

@@ -666,10 +666,11 @@ static void build_evpn_route_extcomm(struct bgpevpn *vpn, struct attr *attr,
 	for (ALL_LIST_ELEMENTS(vpn->export_rtl, node, nnode, ecom))
 		attr->ecommunity = ecommunity_merge(attr->ecommunity, ecom);
 
-	/* Add the export RTs for L3VNI - currently only supported for IPV4 host
-	 * routes
+	/*
+	 * only attach l3-vni export rts for ipv4 address family and if we are
+	 * advertising both the labels in type-2 routes
 	 */
-	if (afi == AFI_IP) {
+	if (afi == AFI_IP && CHECK_FLAG(vpn->flags, VNI_FLAG_USE_TWO_LABELS)) {
 		vrf_export_rtl = bgpevpn_get_vrf_export_rtl(vpn);
 		if (vrf_export_rtl && !list_isempty(vrf_export_rtl)) {
 			for (ALL_LIST_ELEMENTS(vrf_export_rtl, node, nnode,
@@ -690,7 +691,12 @@ static void build_evpn_route_extcomm(struct bgpevpn *vpn, struct attr *attr,
 			ecommunity_merge(attr->ecommunity, &ecom_sticky);
 	}
 
-	if (afi == AFI_IP && !is_zero_mac(&attr->rmac)) {
+	/*
+	 * only attach l3-vni rmac for ipv4 address family and if we are
+	 * advertising both the labels in type-2 routes
+	 */
+	if (afi == AFI_IP && !is_zero_mac(&attr->rmac) &&
+	    CHECK_FLAG(vpn->flags, VNI_FLAG_USE_TWO_LABELS)) {
 		memset(&ecom_rmac, 0, sizeof(ecom_rmac));
 		encode_rmac_extcomm(&eval_rmac, &attr->rmac);
 		ecom_rmac.size = 1;
@@ -1189,8 +1195,13 @@ static int update_evpn_route_entry(struct bgp *bgp, struct bgpevpn *vpn,
 
 		/* The VNI goes into the 'label' field of the route */
 		vni2label(vpn->vni, &label[0]);
-		/* Type-2 routes may carry a second VNI - the L3-VNI */
-		if (evp->prefix.route_type == BGP_EVPN_MAC_IP_ROUTE) {
+
+		/* Type-2 routes may carry a second VNI - the L3-VNI.
+		 * Only attach second label if we are advertising two labels for
+		 * type-2 routes.
+		 */
+		if (evp->prefix.route_type == BGP_EVPN_MAC_IP_ROUTE &&
+		    CHECK_FLAG(vpn->flags, VNI_FLAG_USE_TWO_LABELS)) {
 			vni_t l3vni;
 
 			l3vni = bgpevpn_get_l3vni(vpn);
@@ -1209,6 +1220,24 @@ static int update_evpn_route_entry(struct bgp *bgp, struct bgpevpn *vpn,
 		    && !CHECK_FLAG(tmp_ri->flags, BGP_INFO_REMOVED))
 			route_change = 0;
 		else {
+			/*
+			 * The attributes have changed, type-2 routes needs to
+			 * be advertised with right labels.
+			 */
+			vni2label(vpn->vni, &label[0]);
+			if (evp->prefix.route_type == BGP_EVPN_MAC_IP_ROUTE &&
+			    CHECK_FLAG(vpn->flags, VNI_FLAG_USE_TWO_LABELS)) {
+				vni_t l3vni;
+
+				l3vni = bgpevpn_get_l3vni(vpn);
+				if (l3vni) {
+					vni2label(l3vni, &label[1]);
+					num_labels++;
+				}
+			}
+			memcpy(&tmp_ri->extra->label, label, sizeof(label));
+			tmp_ri->extra->num_labels = num_labels;
+
 			/* The attribute has changed. */
 			/* Add (or update) attribute to hash. */
 			attr_new = bgp_attr_intern(attr);
@@ -4214,7 +4243,8 @@ static void link_l2vni_hash_to_l3vni(struct hash_backet *backet,
 int bgp_evpn_local_l3vni_add(vni_t l3vni,
 			     vrf_id_t vrf_id,
 			     struct ethaddr *rmac,
-			     struct in_addr originator_ip)
+			     struct in_addr originator_ip,
+			     int filter)
 {
 	struct bgp *bgp_vrf = NULL; /* bgp VRF instance */
 	struct bgp *bgp_def = NULL; /* default bgp instance */
@@ -4266,6 +4296,10 @@ int bgp_evpn_local_l3vni_add(vni_t l3vni,
 	/* set the originator ip */
 	bgp_vrf->originator_ip = originator_ip;
 
+	/* set the right filter - are we using l3vni only for prefix routes? */
+	if (filter)
+		SET_FLAG(bgp_vrf->vrf_flags, BGP_VRF_L3VNI_PREFIX_ROUTES_ONLY);
+
 	/* auto derive RD/RT */
 	if (!CHECK_FLAG(bgp_vrf->vrf_flags, BGP_VRF_IMPORT_RT_CFGD))
 		evpn_auto_rt_import_add_for_vrf(bgp_vrf);
@@ -4279,9 +4313,12 @@ int bgp_evpn_local_l3vni_add(vni_t l3vni,
 			link_l2vni_hash_to_l3vni,
 		     bgp_vrf);
 
-	/* updates all corresponding local mac-ip routes */
-	for (ALL_LIST_ELEMENTS_RO(bgp_vrf->l2vnis, node, vpn))
-		update_routes_for_vni(bgp_def, vpn);
+	/* Only update all corresponding type-2 routes if we are advertising two
+	 * labels along with type-2 routes
+	 */
+	if (!filter)
+		for (ALL_LIST_ELEMENTS_RO(bgp_vrf->l2vnis, node, vpn))
+			update_routes_for_vni(bgp_def, vpn);
 
 	/* advertise type-5 routes if needed */
 	update_advertise_vrf_routes(bgp_vrf);
@@ -4340,9 +4377,12 @@ int bgp_evpn_local_l3vni_del(vni_t l3vni,
 	}
 
 	/* update all corresponding local mac-ip routes */
-	for (ALL_LIST_ELEMENTS_RO(bgp_vrf->l2vnis, node, vpn))
-		update_routes_for_vni(bgp_def, vpn);
-
+	if (!CHECK_FLAG(bgp_vrf->vrf_flags, BGP_VRF_L3VNI_PREFIX_ROUTES_ONLY)) {
+		for (ALL_LIST_ELEMENTS_RO(bgp_vrf->l2vnis, node, vpn)) {
+			UNSET_FLAG(vpn->flags, VNI_FLAG_USE_TWO_LABELS);
+			update_routes_for_vni(bgp_def, vpn);
+		}
+	}
 
 	/* Delete the instance if it was autocreated */
 	if (CHECK_FLAG(bgp_vrf->vrf_flags, BGP_VRF_AUTO))

bgpd/bgp_evpn.h

@@ -91,7 +91,8 @@ extern int bgp_evpn_local_macip_add(struct bgp *bgp, vni_t vni,
 				    u_char flags);
 extern int bgp_evpn_local_l3vni_add(vni_t vni, vrf_id_t vrf_id,
 				    struct ethaddr *rmac,
-				    struct in_addr originator_ip);
+				    struct in_addr originator_ip,
+				    int filter);
 extern int bgp_evpn_local_l3vni_del(vni_t vni, vrf_id_t vrf_id);
 extern int bgp_evpn_local_vni_del(struct bgp *bgp, vni_t vni);
 extern int bgp_evpn_local_vni_add(struct bgp *bgp, vni_t vni,

bgpd/bgp_evpn_private.h

@@ -61,6 +61,8 @@ struct bgpevpn {
 #define VNI_FLAG_RD_CFGD           0x4  /* RD is user configured. */
 #define VNI_FLAG_IMPRT_CFGD        0x8  /* Import RT is user configured */
 #define VNI_FLAG_EXPRT_CFGD        0x10 /* Export RT is user configured */
+#define VNI_FLAG_USE_TWO_LABELS    0x20 /* Attach both L2-VNI and L3-VNI if
+					   needed for this VPN */
 
 	/* Flag to indicate if we are advertising the g/w mac ip for this VNI*/
 	u_int8_t advertise_gw_macip;
@@ -179,19 +181,30 @@ static inline void bgpevpn_unlink_from_l3vni(struct bgpevpn *vpn)
 	struct bgp *bgp_vrf = NULL;
 
 	bgp_vrf = bgp_lookup_by_vrf_id(vpn->tenant_vrf_id);
-	if (!bgp_vrf || !bgp_vrf->l2vnis)
+	if (!bgp_vrf)
 		return;
-	listnode_delete(bgp_vrf->l2vnis, vpn);
+
+	UNSET_FLAG(vpn->flags, VNI_FLAG_USE_TWO_LABELS);
+
+	if (bgp_vrf->l2vnis)
+		listnode_delete(bgp_vrf->l2vnis, vpn);
 }
 
 static inline void bgpevpn_link_to_l3vni(struct bgpevpn *vpn)
 {
 	struct bgp *bgp_vrf = NULL;
 
 	bgp_vrf = bgp_lookup_by_vrf_id(vpn->tenant_vrf_id);
-	if (!bgp_vrf || !bgp_vrf->l2vnis)
+	if (!bgp_vrf)
 		return;
-	listnode_add_sort(bgp_vrf->l2vnis, vpn);
+
+	/* check if we are advertising two labels for this vpn */
+	if (!CHECK_FLAG(bgp_vrf->vrf_flags,
+		       BGP_VRF_L3VNI_PREFIX_ROUTES_ONLY))
+		SET_FLAG(vpn->flags, VNI_FLAG_USE_TWO_LABELS);
+
+	if (bgp_vrf->l2vnis)
+		listnode_add_sort(bgp_vrf->l2vnis, vpn);
 }
 
 static inline int is_vni_configured(struct bgpevpn *vpn)

bgpd/bgp_evpn_vty.c

@@ -436,6 +436,10 @@ static void display_vni(struct vty *vty, struct bgpevpn *vpn, json_object *json)
 				       inet_ntoa(vpn->originator_ip));
 		json_object_string_add(json, "advertiseGatewayMacip",
 				       vpn->advertise_gw_macip ? "Yes" : "No");
+		json_object_int_add(json, "numAdvLabels",
+				    CHECK_FLAG(vpn->flags,
+					       VNI_FLAG_USE_TWO_LABELS) ?
+					BGP_MAX_LABELS : BGP_MAX_LABELS - 1);
 	} else {
 		vty_out(vty, "VNI: %d", vpn->vni);
 		if (is_vni_live(vpn))
@@ -451,6 +455,9 @@ static void display_vni(struct vty *vty, struct bgpevpn *vpn, json_object *json)
 			inet_ntoa(vpn->originator_ip));
 		vty_out(vty, "  Advertise-gw-macip : %s\n",
 			vpn->advertise_gw_macip ? "Yes" : "No");
+		vty_out(vty, "  Number of advertised labels %d\n",
+			CHECK_FLAG(vpn->flags, VNI_FLAG_USE_TWO_LABELS) ?
+				BGP_MAX_LABELS : BGP_MAX_LABELS - 1);
 	}
 
 	if (!json)
@@ -3897,6 +3904,10 @@ DEFUN (show_bgp_vrf_l3vni_info,
 		vty_out(vty, "  L3-VNI: %u\n", bgp->l3vni);
 		vty_out(vty, "  Rmac: %s\n",
 			prefix_mac2str(&bgp->rmac, buf, sizeof(buf)));
+		vty_out(vty, "  VNI Filter: %s\n",
+			CHECK_FLAG(bgp->vrf_flags,
+				   BGP_VRF_L3VNI_PREFIX_ROUTES_ONLY) ?
+				"prefix-routes-only" : "none");
 		vty_out(vty, "  L2-VNI List:\n");
 		vty_out(vty, "    ");
 		for (ALL_LIST_ELEMENTS_RO(bgp->l2vnis, node, vpn))
@@ -3922,6 +3933,10 @@ DEFUN (show_bgp_vrf_l3vni_info,
 		json_object_string_add(json, "rmac",
 				       prefix_mac2str(&bgp->rmac, buf,
 						      sizeof(buf)));
+		json_object_string_add(json, "vniFilter",
+				       CHECK_FLAG(bgp->vrf_flags,
+						  BGP_VRF_L3VNI_PREFIX_ROUTES_ONLY)
+				       ? "prefix-routes-only" : "none");
 		/* list of l2vnis */
 		for (ALL_LIST_ELEMENTS_RO(bgp->l2vnis, node, vpn))
 			json_object_array_add(json_vnis,

bgpd/bgp_zebra.c

@@ -1731,6 +1731,7 @@ static void bgp_zebra_connected(struct zclient *zclient)
 static int bgp_zebra_process_local_l3vni(int cmd, struct zclient *zclient,
 					 zebra_size_t length, vrf_id_t vrf_id)
 {
+	int filter = 0;
 	char buf[ETHER_ADDR_STRLEN];
 	vni_t l3vni = 0;
 	struct ethaddr rmac;
@@ -1744,17 +1745,20 @@ static int bgp_zebra_process_local_l3vni(int cmd, struct zclient *zclient,
 	if (cmd == ZEBRA_L3VNI_ADD) {
 		stream_get(&rmac, s, sizeof(struct ethaddr));
 		originator_ip.s_addr = stream_get_ipv4(s);
+		stream_get(&filter, s, sizeof(int));
 	}
 
 	if (BGP_DEBUG(zebra, ZEBRA))
-		zlog_debug("Rx L3-VNI %s VRF %s VNI %u RMAC %s",
+		zlog_debug("Rx L3-VNI %s VRF %s VNI %u RMAC %s filter %s",
 			   (cmd == ZEBRA_L3VNI_ADD) ? "add" : "del",
 			   vrf_id_to_name(vrf_id),
 			   l3vni,
-			   prefix_mac2str(&rmac, buf, sizeof(buf)));
+			   prefix_mac2str(&rmac, buf, sizeof(buf)),
+			   filter ? "prefix-routes-only" : "none");
 
 	if (cmd == ZEBRA_L3VNI_ADD)
-		bgp_evpn_local_l3vni_add(l3vni, vrf_id, &rmac, originator_ip);
+		bgp_evpn_local_l3vni_add(l3vni, vrf_id, &rmac, originator_ip,
+					 filter);
 	else
 		bgp_evpn_local_l3vni_del(l3vni, vrf_id);
 

bgpd/bgpd.h

@@ -435,6 +435,7 @@ struct bgp {
 #define BGP_VRF_IMPORT_RT_CFGD              (1 << 3)
 #define BGP_VRF_EXPORT_RT_CFGD              (1 << 4)
 #define BGP_VRF_RD_CFGD                     (1 << 5)
+#define BGP_VRF_L3VNI_PREFIX_ROUTES_ONLY    (1 << 6)
 
 	/* unique ID for auto derivation of RD for this vrf */
 	uint16_t vrf_rd_id;

zebra/zebra_vrf.c

@@ -598,7 +598,10 @@ static int vrf_config_write(struct vty *vty)
 		if (vrf_is_user_cfged(vrf)) {
 			vty_out(vty, "vrf %s\n", zvrf_name(zvrf));
 			if (zvrf->l3vni)
-				vty_out(vty, " vni %u\n", zvrf->l3vni);
+				vty_out(vty, " vni %u%s\n",
+					zvrf->l3vni,
+					is_l3vni_for_prefix_routes_only(zvrf->l3vni) ?
+					" prefix-routes-only" :"");
 			vty_out(vty, "!\n");
 		}
 

zebra/zebra_vty.c

@@ -2338,20 +2338,26 @@ DEFUN (show_vrf,
 
 DEFUN (default_vrf_vni_mapping,
        default_vrf_vni_mapping_cmd,
-       "vni " CMD_VNI_RANGE,
+       "vni " CMD_VNI_RANGE "[prefix-routes-only]",
        "VNI corresponding to the DEFAULT VRF\n"
-       "VNI-ID\n")
+       "VNI-ID\n"
+       "Prefix routes only \n")
 {
 	int ret = 0;
 	char err[ERR_STR_SZ];
 	struct zebra_vrf *zvrf = NULL;
 	vni_t vni = strtoul(argv[1]->arg, NULL, 10);
+	int filter = 0;
 
 	zvrf = vrf_info_lookup(VRF_DEFAULT);
 	if (!zvrf)
 		return CMD_WARNING;
 
-	ret = zebra_vxlan_process_vrf_vni_cmd(zvrf, vni, err, ERR_STR_SZ, 1);
+	if (argc == 3)
+		filter = 1;
+
+	ret = zebra_vxlan_process_vrf_vni_cmd(zvrf, vni, err, ERR_STR_SZ,
+					      filter, 1);
 	if (ret != 0) {
 		vty_out(vty, "%s\n", err);
 		return CMD_WARNING;
@@ -2376,7 +2382,7 @@ DEFUN (no_default_vrf_vni_mapping,
 	if (!zvrf)
 		return CMD_WARNING;
 
-	ret = zebra_vxlan_process_vrf_vni_cmd(zvrf, vni, err, ERR_STR_SZ, 0);
+	ret = zebra_vxlan_process_vrf_vni_cmd(zvrf, vni, err, ERR_STR_SZ, 0, 0);
 	if (ret != 0) {
 		vty_out(vty, "%s\n", err);
 		return CMD_WARNING;
@@ -2387,11 +2393,13 @@ DEFUN (no_default_vrf_vni_mapping,
 
 DEFUN (vrf_vni_mapping,
        vrf_vni_mapping_cmd,
-       "vni " CMD_VNI_RANGE,
+       "vni " CMD_VNI_RANGE "[prefix-routes-only]",
        "VNI corresponding to tenant VRF\n"
-       "VNI-ID\n")
+       "VNI-ID\n"
+       "prefix-routes-only\n")
 {
 	int ret = 0;
+	int filter = 0;
 
 	ZEBRA_DECLVAR_CONTEXT(vrf, zvrf);
 	vni_t vni = strtoul(argv[1]->arg, NULL, 10);
@@ -2400,9 +2408,13 @@ DEFUN (vrf_vni_mapping,
 	assert(vrf);
 	assert(zvrf);
 
+	if (argc == 3)
+		filter = 1;
+
 	/* Mark as having FRR configuration */
 	vrf_set_user_cfged(vrf);
-	ret = zebra_vxlan_process_vrf_vni_cmd(zvrf, vni, err, ERR_STR_SZ, 1);
+	ret = zebra_vxlan_process_vrf_vni_cmd(zvrf, vni, err, ERR_STR_SZ,
+					      filter, 1);
 	if (ret != 0) {
 		vty_out(vty, "%s\n", err);
 		return CMD_WARNING;
@@ -2427,7 +2439,7 @@ DEFUN (no_vrf_vni_mapping,
 	assert(vrf);
 	assert(zvrf);
 
-	ret = zebra_vxlan_process_vrf_vni_cmd(zvrf, vni, err, ERR_STR_SZ, 0);
+	ret = zebra_vxlan_process_vrf_vni_cmd(zvrf, vni, err, ERR_STR_SZ, 0, 0);
 	if (ret != 0) {
 		vty_out(vty, "%s\n", err);
 		return CMD_WARNING;