#!/usr/bin/env python3 import random import socket import unittest from scapy.contrib.mpls import MPLS from scapy.contrib.gtp import GTP_U_Header from scapy.layers.inet import IP, UDP, TCP, ICMP, icmptypes, icmpcodes from scapy.layers.inet6 import IPv6 from scapy.layers.l2 import Ether, Dot1Q, ARP from scapy.packet import Raw from six import moves from framework import VppTestCase from asfframework import VppTestRunner, tag_fixme_vpp_workers from util import ppp from vpp_ip_route import ( VppIpRoute, VppRoutePath, VppIpMRoute, VppMRoutePath, VppMplsIpBind, VppMplsTable, VppIpTable, FibPathType, find_route, VppIpInterfaceAddress, find_route_in_dump, find_mroute_in_dump, ) from vpp_ip import VppIpPuntPolicer, VppIpPuntRedirect, VppIpPathMtu from vpp_sub_interface import VppSubInterface, VppDot1QSubint, VppDot1ADSubint from vpp_papi import vpp_papi, VppEnum from vpp_neighbor import VppNeighbor from vpp_lo_interface import VppLoInterface from vpp_policer import VppPolicer, PolicerAction from config import config NUM_PKTS = 67 class TestIPv4(VppTestCase): """IPv4 Test Case""" @classmethod def setUpClass(cls): super(TestIPv4, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPv4, cls).tearDownClass() def setUp(self): """ Perform test setup before test case. **Config:** - create 3 pg interfaces - untagged pg0 interface - Dot1Q subinterface on pg1 - Dot1AD subinterface on pg2 - setup interfaces: - put it into UP state - set IPv4 addresses - resolve neighbor address using ARP - configure 200 fib entries :ivar list interfaces: pg interfaces and subinterfaces. :ivar dict flows: IPv4 packet flows in test. """ super(TestIPv4, self).setUp() # create 3 pg interfaces self.create_pg_interfaces(range(3)) # create 2 subinterfaces for pg1 and pg2 self.sub_interfaces = [ VppDot1QSubint(self, self.pg1, 100), VppDot1ADSubint(self, self.pg2, 200, 300, 400), ] # packet flows mapping pg0 -> pg1.sub, pg2.sub, etc. self.flows = dict() self.flows[self.pg0] = [self.pg1.sub_if, self.pg2.sub_if] self.flows[self.pg1.sub_if] = [self.pg0, self.pg2.sub_if] self.flows[self.pg2.sub_if] = [self.pg0, self.pg1.sub_if] # packet sizes self.pg_if_packet_sizes = [64, 1500, 9020] self.interfaces = list(self.pg_interfaces) self.interfaces.extend(self.sub_interfaces) # setup all interfaces for i in self.interfaces: i.admin_up() i.config_ip4() i.resolve_arp() # config 2M FIB entries def tearDown(self): """Run standard test teardown and log ``show ip arp``.""" super(TestIPv4, self).tearDown() def show_commands_at_teardown(self): self.logger.info(self.vapi.cli("show ip4 neighbors")) # info(self.vapi.cli("show ip fib")) # many entries def modify_packet(self, src_if, packet_size, pkt): """Add load, set destination IP and extend packet to required packet size for defined interface. :param VppInterface src_if: Interface to create packet for. :param int packet_size: Required packet size. :param Scapy pkt: Packet to be modified. """ dst_if_idx = int(packet_size / 10 % 2) dst_if = self.flows[src_if][dst_if_idx] info = self.create_packet_info(src_if, dst_if) payload = self.info_to_payload(info) p = pkt / Raw(payload) p[IP].dst = dst_if.remote_ip4 info.data = p.copy() if isinstance(src_if, VppSubInterface): p = src_if.add_dot1_layer(p) self.extend_packet(p, packet_size) return p def create_stream(self, src_if): """Create input packet stream for defined interface. :param VppInterface src_if: Interface to create packet stream for. """ hdr_ext = 4 if isinstance(src_if, VppSubInterface) else 0 pkt_tmpl = ( Ether(dst=src_if.local_mac, src=src_if.remote_mac) / IP(src=src_if.remote_ip4) / UDP(sport=1234, dport=1234) ) pkts = [ self.modify_packet(src_if, i, pkt_tmpl) for i in moves.range( self.pg_if_packet_sizes[0], self.pg_if_packet_sizes[1], 10 ) ] pkts_b = [ self.modify_packet(src_if, i, pkt_tmpl) for i in moves.range( self.pg_if_packet_sizes[1] + hdr_ext, self.pg_if_packet_sizes[2] + hdr_ext, 50, ) ] pkts.extend(pkts_b) return pkts def verify_capture(self, dst_if, capture): """Verify captured input packet stream for defined interface. :param VppInterface dst_if: Interface to verify captured packet stream for. :param list capture: Captured packet stream. """ self.logger.info("Verifying capture on interface %s" % dst_if.name) last_info = dict() for i in self.interfaces: last_info[i.sw_if_index] = None is_sub_if = False dst_sw_if_index = dst_if.sw_if_index if hasattr(dst_if, "parent"): is_sub_if = True for packet in capture: if is_sub_if: # Check VLAN tags and Ethernet header packet = dst_if.remove_dot1_layer(packet) self.assertTrue(Dot1Q not in packet) try: ip = packet[IP] udp = packet[UDP] payload_info = self.payload_to_info(packet[Raw]) packet_index = payload_info.index self.assertEqual(payload_info.dst, dst_sw_if_index) self.logger.debug( "Got packet on port %s: src=%u (id=%u)" % (dst_if.name, payload_info.src, packet_index) ) next_info = self.get_next_packet_info_for_interface2( payload_info.src, dst_sw_if_index, last_info[payload_info.src] ) last_info[payload_info.src] = next_info self.assertTrue(next_info is not None) self.assertEqual(packet_index, next_info.index) saved_packet = next_info.data # Check standard fields self.assertEqual(ip.src, saved_packet[IP].src) self.assertEqual(ip.dst, saved_packet[IP].dst) self.assertEqual(udp.sport, saved_packet[UDP].sport) self.assertEqual(udp.dport, saved_packet[UDP].dport) except: self.logger.error(ppp("Unexpected or invalid packet:", packet)) raise for i in self.interfaces: remaining_packet = self.get_next_packet_info_for_interface2( i.sw_if_index, dst_sw_if_index, last_info[i.sw_if_index] ) self.assertTrue( remaining_packet is None, "Interface %s: Packet expected from interface %s " "didn't arrive" % (dst_if.name, i.name), ) def test_fib(self): """IPv4 FIB test Test scenario: - Create IPv4 stream for pg0 interface - Create IPv4 tagged streams for pg1's and pg2's sub-interface. - Send and verify received packets on each interface. """ pkts = self.create_stream(self.pg0) self.pg0.add_stream(pkts) for i in self.sub_interfaces: pkts = self.create_stream(i) i.parent.add_stream(pkts) self.pg_enable_capture(self.pg_interfaces) self.pg_start() pkts = self.pg0.get_capture() self.verify_capture(self.pg0, pkts) for i in self.sub_interfaces: pkts = i.parent.get_capture() self.verify_capture(i, pkts) class TestIPv4RouteLookup(VppTestCase): """IPv4 Route Lookup Test Case""" routes = [] tables = [] def route_lookup(self, prefix, exact, table_id=0): return self.vapi.api( self.vapi.papi.ip_route_lookup, { "table_id": table_id, "exact": exact, "prefix": prefix, }, ) @classmethod def setUpClass(cls): super(TestIPv4RouteLookup, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPv4RouteLookup, cls).tearDownClass() def setUp(self): super(TestIPv4RouteLookup, self).setUp() drop_nh = VppRoutePath( "127.0.0.1", 0xFFFFFFFF, type=FibPathType.FIB_PATH_TYPE_DROP ) # Add 3 routes r = VppIpRoute(self, "1.1.0.0", 16, [drop_nh]) r.add_vpp_config() self.routes.append(r) r = VppIpRoute(self, "1.1.1.0", 24, [drop_nh]) r.add_vpp_config() self.routes.append(r) r = VppIpRoute(self, "1.1.1.1", 32, [drop_nh]) r.add_vpp_config() self.routes.append(r) custom_vrf = VppIpTable(self, 200) custom_vrf.add_vpp_config() self.tables.append(custom_vrf) r = VppIpRoute(self, "2.2.0.0", 16, [drop_nh], 200) r.add_vpp_config() self.routes.append(r) r = VppIpRoute(self, "2.2.2.0", 24, [drop_nh], 200) r.add_vpp_config() self.routes.append(r) r = VppIpRoute(self, "2.2.2.2", 32, [drop_nh], 200) r.add_vpp_config() self.routes.append(r) def tearDown(self): # Remove the routes we added for r in self.routes: r.remove_vpp_config() for vrf in self.tables: vrf.remove_vpp_config() super(TestIPv4RouteLookup, self).tearDown() def test_exact_match(self): # Verify we find the host route prefix = "1.1.1.1/32" result = self.route_lookup(prefix, True) assert prefix == str(result.route.prefix) # Verify we find a middle prefix route prefix = "1.1.1.0/24" result = self.route_lookup(prefix, True) assert prefix == str(result.route.prefix) # Verify we do not find an available LPM. with self.vapi.assert_negative_api_retval(): self.route_lookup("1.1.1.2/32", True) # Verify we find the host route prefix = "2.2.2.2/32" result = self.route_lookup(prefix, True, 200) assert prefix == str(result.route.prefix) # Verify we find a middle prefix route prefix = "2.2.2.0/24" result = self.route_lookup(prefix, True, 200) assert prefix == str(result.route.prefix) # Verify we do not find an available LPM. with self.vapi.assert_negative_api_retval(): self.route_lookup("2.2.2.1/32", True, 200) def test_longest_prefix_match(self): # verify we find lpm lpm_prefix = "1.1.1.0/24" result = self.route_lookup("1.1.1.2/32", False) assert lpm_prefix == str(result.route.prefix) # Verify we find the exact when not requested result = self.route_lookup(lpm_prefix, False) assert lpm_prefix == str(result.route.prefix) # verify we find lpm lpm_prefix = "2.2.2.0/24" result = self.route_lookup("2.2.2.1/32", False, 200) assert lpm_prefix == str(result.route.prefix) # Verify we find the exact when not requested result = self.route_lookup(lpm_prefix, False, 200) assert lpm_prefix == str(result.route.prefix) # Can't seem to delete the default route so no negative LPM test. class TestIPv4IfAddrRoute(VppTestCase): """IPv4 Interface Addr Route Test Case""" @classmethod def setUpClass(cls): super(TestIPv4IfAddrRoute, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPv4IfAddrRoute, cls).tearDownClass() def setUp(self): super(TestIPv4IfAddrRoute, self).setUp() # create 1 pg interface self.create_pg_interfaces(range(1)) for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() def tearDown(self): super(TestIPv4IfAddrRoute, self).tearDown() for i in self.pg_interfaces: i.unconfig_ip4() i.admin_down() def test_ipv4_ifaddrs_same_prefix(self): """IPv4 Interface Addresses Same Prefix test Test scenario: - Verify no route in FIB for prefix 10.10.10.0/24 - Configure IPv4 address 10.10.10.10/24 on an interface - Verify route in FIB for prefix 10.10.10.0/24 - Configure IPv4 address 10.10.10.20/24 on an interface - Delete 10.10.10.10/24 from interface - Verify route in FIB for prefix 10.10.10.0/24 - Delete 10.10.10.20/24 from interface - Verify no route in FIB for prefix 10.10.10.0/24 """ # create two addresses, verify route not present if_addr1 = VppIpInterfaceAddress(self, self.pg0, "10.10.10.10", 24) if_addr2 = VppIpInterfaceAddress(self, self.pg0, "10.10.10.20", 24) self.assertFalse(if_addr1.query_vpp_config()) # 10.10.10.10/24 self.assertFalse(find_route(self, "10.10.10.10", 32)) self.assertFalse(find_route(self, "10.10.10.20", 32)) self.assertFalse(find_route(self, "10.10.10.255", 32)) self.assertFalse(find_route(self, "10.10.10.0", 32)) # configure first address, verify route present if_addr1.add_vpp_config() self.assertTrue(if_addr1.query_vpp_config()) # 10.10.10.10/24 self.assertTrue(find_route(self, "10.10.10.10", 32)) self.assertFalse(find_route(self, "10.10.10.20", 32)) self.assertTrue(find_route(self, "10.10.10.255", 32)) self.assertTrue(find_route(self, "10.10.10.0", 32)) # configure second address, delete first, verify route not removed if_addr2.add_vpp_config() if_addr1.remove_vpp_config() self.assertFalse(if_addr1.query_vpp_config()) # 10.10.10.10/24 self.assertTrue(if_addr2.query_vpp_config()) # 10.10.10.20/24 self.assertFalse(find_route(self, "10.10.10.10", 32)) self.assertTrue(find_route(self, "10.10.10.20", 32)) self.assertTrue(find_route(self, "10.10.10.255", 32)) self.assertTrue(find_route(self, "10.10.10.0", 32)) # delete second address, verify route removed if_addr2.remove_vpp_config() self.assertFalse(if_addr2.query_vpp_config()) # 10.10.10.20/24 self.assertFalse(find_route(self, "10.10.10.10", 32)) self.assertFalse(find_route(self, "10.10.10.20", 32)) self.assertFalse(find_route(self, "10.10.10.255", 32)) self.assertFalse(find_route(self, "10.10.10.0", 32)) def test_ipv4_ifaddr_route(self): """IPv4 Interface Address Route test Test scenario: - Create loopback - Configure IPv4 address on loopback - Verify that address is not in the FIB - Bring loopback up - Verify that address is in the FIB now - Bring loopback down - Verify that address is not in the FIB anymore - Bring loopback up - Configure IPv4 address on loopback - Verify that address is in the FIB now """ # create a loopback and configure IPv4 loopbacks = self.create_loopback_interfaces(1) lo_if = self.lo_interfaces[0] lo_if.local_ip4_prefix_len = 32 lo_if.config_ip4() # The intf was down when addr was added -> entry not in FIB fib4_dump = self.vapi.ip_route_dump(0) self.assertFalse(lo_if.is_ip4_entry_in_fib_dump(fib4_dump)) # When intf is brought up, entry is added lo_if.admin_up() fib4_dump = self.vapi.ip_route_dump(0) self.assertTrue(lo_if.is_ip4_entry_in_fib_dump(fib4_dump)) # When intf is brought down, entry is removed lo_if.admin_down() fib4_dump = self.vapi.ip_route_dump(0) self.assertFalse(lo_if.is_ip4_entry_in_fib_dump(fib4_dump)) # Remove addr, bring up interface, re-add -> entry in FIB lo_if.unconfig_ip4() lo_if.admin_up() lo_if.config_ip4() fib4_dump = self.vapi.ip_route_dump(0) self.assertTrue(lo_if.is_ip4_entry_in_fib_dump(fib4_dump)) def test_ipv4_ifaddr_del(self): """Delete an interface address that does not exist""" loopbacks = self.create_loopback_interfaces(1) lo = self.lo_interfaces[0] lo.config_ip4() lo.admin_up() # # try and remove pg0's subnet from lo # with self.vapi.assert_negative_api_retval(): self.vapi.sw_interface_add_del_address( sw_if_index=lo.sw_if_index, prefix=self.pg0.local_ip4_prefix, is_add=0 ) @unittest.skipIf( "ping" in config.excluded_plugins, "Exclude tests requiring Ping plugin" ) class TestICMPEcho(VppTestCase): """ICMP Echo Test Case""" @classmethod def setUpClass(cls): super(TestICMPEcho, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestICMPEcho, cls).tearDownClass() def setUp(self): super(TestICMPEcho, self).setUp() # create 1 pg interface self.create_pg_interfaces(range(1)) for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() def tearDown(self): super(TestICMPEcho, self).tearDown() for i in self.pg_interfaces: i.unconfig_ip4() i.admin_down() def test_icmp_echo(self): """VPP replies to ICMP Echo Request Test scenario: - Receive ICMP Echo Request message on pg0 interface. - Check outgoing ICMP Echo Reply message on pg0 interface. """ icmp_id = 0xB icmp_seq = 5 icmp_load = b"\x0a" * 18 p_echo_request = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg0.local_ip4) / ICMP(id=icmp_id, seq=icmp_seq) / Raw(load=icmp_load) ) self.pg0.add_stream(p_echo_request) self.pg_enable_capture(self.pg_interfaces) self.pg_start() rx = self.pg0.get_capture(1) rx = rx[0] ether = rx[Ether] ipv4 = rx[IP] icmp = rx[ICMP] self.assertEqual(ether.src, self.pg0.local_mac) self.assertEqual(ether.dst, self.pg0.remote_mac) self.assertEqual(ipv4.src, self.pg0.local_ip4) self.assertEqual(ipv4.dst, self.pg0.remote_ip4) self.assertEqual(icmptypes[icmp.type], "echo-reply") self.assertEqual(icmp.id, icmp_id) self.assertEqual(icmp.seq, icmp_seq) self.assertEqual(icmp[Raw].load, icmp_load) class TestIPv4FibCrud(VppTestCase): """FIB - add/update/delete - ip4 routes Test scenario: - add 1k, - del 100, - add new 1k, - del 1.5k ..note:: Python API is too slow to add many routes, needs replacement. """ def config_fib_many_to_one(self, start_dest_addr, next_hop_addr, count, start=0): """ :param start_dest_addr: :param next_hop_addr: :param count: :return list: added ips with 32 prefix """ routes = [] for i in range(count): r = VppIpRoute( self, start_dest_addr % (i + start), 32, [VppRoutePath(next_hop_addr, 0xFFFFFFFF)], ) r.add_vpp_config() routes.append(r) return routes def unconfig_fib_many_to_one(self, start_dest_addr, next_hop_addr, count, start=0): routes = [] for i in range(count): r = VppIpRoute( self, start_dest_addr % (i + start), 32, [VppRoutePath(next_hop_addr, 0xFFFFFFFF)], ) r.remove_vpp_config() routes.append(r) return routes def create_stream(self, src_if, dst_if, routes, count): pkts = [] for _ in range(count): dst_addr = random.choice(routes).prefix.network_address info = self.create_packet_info(src_if, dst_if) payload = self.info_to_payload(info) p = ( Ether(dst=src_if.local_mac, src=src_if.remote_mac) / IP(src=src_if.remote_ip4, dst=str(dst_addr)) / UDP(sport=1234, dport=1234) / Raw(payload) ) info.data = p.copy() self.extend_packet(p, random.choice(self.pg_if_packet_sizes)) pkts.append(p) return pkts def _find_ip_match(self, find_in, pkt): for p in find_in: if self.payload_to_info(p[Raw]) == self.payload_to_info(pkt[Raw]): if p[IP].src != pkt[IP].src: break if p[IP].dst != pkt[IP].dst: break if p[UDP].sport != pkt[UDP].sport: break if p[UDP].dport != pkt[UDP].dport: break return p return None def verify_capture(self, dst_interface, received_pkts, expected_pkts): self.assertEqual(len(received_pkts), len(expected_pkts)) to_verify = list(expected_pkts) for p in received_pkts: self.assertEqual(p.src, dst_interface.local_mac) self.assertEqual(p.dst, dst_interface.remote_mac) x = self._find_ip_match(to_verify, p) to_verify.remove(x) self.assertListEqual(to_verify, []) def verify_route_dump(self, routes): for r in routes: self.assertTrue( find_route(self, r.prefix.network_address, r.prefix.prefixlen) ) def verify_not_in_route_dump(self, routes): for r in routes: self.assertFalse( find_route(self, r.prefix.network_address, r.prefix.prefixlen) ) @classmethod def setUpClass(cls): """ #. Create and initialize 3 pg interfaces. #. initialize class attributes configured_routes and deleted_routes to store information between tests. """ super(TestIPv4FibCrud, cls).setUpClass() try: # create 3 pg interfaces cls.create_pg_interfaces(range(3)) cls.interfaces = list(cls.pg_interfaces) # setup all interfaces for i in cls.interfaces: i.admin_up() i.config_ip4() i.resolve_arp() cls.configured_routes = [] cls.deleted_routes = [] cls.pg_if_packet_sizes = [64, 512, 1518, 9018] except Exception: super(TestIPv4FibCrud, cls).tearDownClass() raise @classmethod def tearDownClass(cls): super(TestIPv4FibCrud, cls).tearDownClass() def setUp(self): super(TestIPv4FibCrud, self).setUp() self.reset_packet_infos() self.configured_routes = [] self.deleted_routes = [] def test_1_add_routes(self): """Add 1k routes""" # add 100 routes check with traffic script. self.configured_routes.extend( self.config_fib_many_to_one("10.0.0.%d", self.pg0.remote_ip4, 100) ) self.verify_route_dump(self.configured_routes) self.stream_1 = self.create_stream( self.pg1, self.pg0, self.configured_routes, 100 ) self.stream_2 = self.create_stream( self.pg2, self.pg0, self.configured_routes, 100 ) self.pg1.add_stream(self.stream_1) self.pg2.add_stream(self.stream_2) self.pg_enable_capture(self.pg_interfaces) self.pg_start() pkts = self.pg0.get_capture(len(self.stream_1) + len(self.stream_2)) self.verify_capture(self.pg0, pkts, self.stream_1 + self.stream_2) def test_2_del_routes(self): """Delete 100 routes - delete 10 routes check with traffic script. """ # config 1M FIB entries self.configured_routes.extend( self.config_fib_many_to_one("10.0.0.%d", self.pg0.remote_ip4, 100) ) self.deleted_routes.extend( self.unconfig_fib_many_to_one( "10.0.0.%d", self.pg0.remote_ip4, 10, start=10 ) ) for x in self.deleted_routes: self.configured_routes.remove(x) self.verify_route_dump(self.configured_routes) self.stream_1 = self.create_stream( self.pg1, self.pg0, self.configured_routes, 100 ) self.stream_2 = self.create_stream( self.pg2, self.pg0, self.configured_routes, 100 ) self.stream_3 = self.create_stream(self.pg1, self.pg0, self.deleted_routes, 100) self.stream_4 = self.create_stream(self.pg2, self.pg0, self.deleted_routes, 100) self.pg1.add_stream(self.stream_1 + self.stream_3) self.pg2.add_stream(self.stream_2 + self.stream_4) self.pg_enable_capture(self.pg_interfaces) self.pg_start() pkts = self.pg0.get_capture(len(self.stream_1) + len(self.stream_2)) self.verify_capture(self.pg0, pkts, self.stream_1 + self.stream_2) def test_3_add_new_routes(self): """Add 1k routes - re-add 5 routes check with traffic script. - add 100 routes check with traffic script. """ # config 1M FIB entries self.configured_routes.extend( self.config_fib_many_to_one("10.0.0.%d", self.pg0.remote_ip4, 100) ) self.deleted_routes.extend( self.unconfig_fib_many_to_one( "10.0.0.%d", self.pg0.remote_ip4, 10, start=10 ) ) for x in self.deleted_routes: self.configured_routes.remove(x) tmp = self.config_fib_many_to_one("10.0.0.%d", self.pg0.remote_ip4, 5, start=10) self.configured_routes.extend(tmp) for x in tmp: self.deleted_routes.remove(x) self.configured_routes.extend( self.config_fib_many_to_one("10.0.1.%d", self.pg0.remote_ip4, 100) ) self.verify_route_dump(self.configured_routes) self.stream_1 = self.create_stream( self.pg1, self.pg0, self.configured_routes, 300 ) self.stream_2 = self.create_stream( self.pg2, self.pg0, self.configured_routes, 300 ) self.stream_3 = self.create_stream(self.pg1, self.pg0, self.deleted_routes, 100) self.stream_4 = self.create_stream(self.pg2, self.pg0, self.deleted_routes, 100) self.pg1.add_stream(self.stream_1 + self.stream_3) self.pg2.add_stream(self.stream_2 + self.stream_4) self.pg_enable_capture(self.pg_interfaces) self.pg_start() pkts = self.pg0.get_capture(len(self.stream_1) + len(self.stream_2)) self.verify_capture(self.pg0, pkts, self.stream_1 + self.stream_2) # delete 5 routes check with traffic script. # add 100 routes check with traffic script. self.deleted_routes.extend( self.unconfig_fib_many_to_one("10.0.0.%d", self.pg0.remote_ip4, 15) ) self.deleted_routes.extend( self.unconfig_fib_many_to_one("10.0.0.%d", self.pg0.remote_ip4, 85) ) self.deleted_routes.extend( self.unconfig_fib_many_to_one("10.0.1.%d", self.pg0.remote_ip4, 100) ) self.verify_not_in_route_dump(self.deleted_routes) class TestIPNull(VppTestCase): """IPv4 routes via NULL""" @classmethod def setUpClass(cls): super(TestIPNull, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPNull, cls).tearDownClass() def setUp(self): super(TestIPNull, self).setUp() # create 2 pg interfaces self.create_pg_interfaces(range(2)) for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() def tearDown(self): super(TestIPNull, self).tearDown() for i in self.pg_interfaces: i.unconfig_ip4() i.admin_down() def test_ip_null(self): """IP NULL route""" # # A route via IP NULL that will reply with ICMP unreachables # ip_unreach = VppIpRoute( self, "10.0.0.1", 32, [ VppRoutePath( "0.0.0.0", 0xFFFFFFFF, type=FibPathType.FIB_PATH_TYPE_ICMP_UNREACH ) ], ) ip_unreach.add_vpp_config() p_unreach = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst="10.0.0.1") / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) self.pg0.add_stream(p_unreach) self.pg_enable_capture(self.pg_interfaces) self.pg_start() rx = self.pg0.get_capture(1) rx = rx[0] icmp = rx[ICMP] self.assertEqual(icmptypes[icmp.type], "dest-unreach") self.assertEqual(icmpcodes[icmp.type][icmp.code], "host-unreachable") self.assertEqual(icmp.src, self.pg0.remote_ip4) self.assertEqual(icmp.dst, "10.0.0.1") # # ICMP replies are rate limited. so sit and spin. # self.sleep(1) # # A route via IP NULL that will reply with ICMP prohibited # ip_prohibit = VppIpRoute( self, "10.0.0.2", 32, [ VppRoutePath( "0.0.0.0", 0xFFFFFFFF, type=FibPathType.FIB_PATH_TYPE_ICMP_PROHIBIT ) ], ) ip_prohibit.add_vpp_config() p_prohibit = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst="10.0.0.2") / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) self.pg0.add_stream(p_prohibit) self.pg_enable_capture(self.pg_interfaces) self.pg_start() rx = self.pg0.get_capture(1) rx = rx[0] icmp = rx[ICMP] self.assertEqual(icmptypes[icmp.type], "dest-unreach") self.assertEqual(icmpcodes[icmp.type][icmp.code], "host-prohibited") self.assertEqual(icmp.src, self.pg0.remote_ip4) self.assertEqual(icmp.dst, "10.0.0.2") def test_ip_drop(self): """IP Drop Routes""" p = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst="1.1.1.1") / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) r1 = VppIpRoute( self, "1.1.1.0", 24, [VppRoutePath(self.pg1.remote_ip4, self.pg1.sw_if_index)], ) r1.add_vpp_config() rx = self.send_and_expect(self.pg0, p * NUM_PKTS, self.pg1) # # insert a more specific as a drop # r2 = VppIpRoute( self, "1.1.1.1", 32, [VppRoutePath("0.0.0.0", 0xFFFFFFFF, type=FibPathType.FIB_PATH_TYPE_DROP)], ) r2.add_vpp_config() self.send_and_assert_no_replies(self.pg0, p * NUM_PKTS, "Drop Route") r2.remove_vpp_config() rx = self.send_and_expect(self.pg0, p * NUM_PKTS, self.pg1) t = VppIpTable(self, 2, False) t.add_vpp_config() r3 = VppIpRoute( self, "1.1.1.0", 31, [VppRoutePath("0.0.0.0", 0xFFFFFFFF, type=FibPathType.FIB_PATH_TYPE_DROP)], table_id=2, ) r3.add_vpp_config() r3.remove_vpp_config() t.remove_vpp_config() class TestIPDisabled(VppTestCase): """IPv4 disabled""" @classmethod def setUpClass(cls): super(TestIPDisabled, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPDisabled, cls).tearDownClass() def setUp(self): super(TestIPDisabled, self).setUp() # create 2 pg interfaces self.create_pg_interfaces(range(2)) # PG0 is IP enalbed self.pg0.admin_up() self.pg0.config_ip4() self.pg0.resolve_arp() # PG 1 is not IP enabled self.pg1.admin_up() def tearDown(self): super(TestIPDisabled, self).tearDown() for i in self.pg_interfaces: i.unconfig_ip4() i.admin_down() def test_ip_disabled(self): """IP Disabled""" MRouteItfFlags = VppEnum.vl_api_mfib_itf_flags_t MRouteEntryFlags = VppEnum.vl_api_mfib_entry_flags_t # # An (S,G). # one accepting interface, pg0, 2 forwarding interfaces # route_232_1_1_1 = VppIpMRoute( self, "0.0.0.0", "232.1.1.1", 32, MRouteEntryFlags.MFIB_API_ENTRY_FLAG_NONE, [ VppMRoutePath( self.pg1.sw_if_index, MRouteItfFlags.MFIB_API_ITF_FLAG_ACCEPT ), VppMRoutePath( self.pg0.sw_if_index, MRouteItfFlags.MFIB_API_ITF_FLAG_FORWARD ), ], ) route_232_1_1_1.add_vpp_config() pu = ( Ether(src=self.pg1.remote_mac, dst=self.pg1.local_mac) / IP(src="10.10.10.10", dst=self.pg0.remote_ip4) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) pm = ( Ether(src=self.pg1.remote_mac, dst=self.pg1.local_mac) / IP(src="10.10.10.10", dst="232.1.1.1") / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) # # PG1 does not forward IP traffic # self.send_and_assert_no_replies(self.pg1, pu, "IP disabled") self.send_and_assert_no_replies(self.pg1, pm, "IP disabled") # # IP enable PG1 # self.pg1.config_ip4() # # Now we get packets through # self.pg1.add_stream(pu) self.pg_enable_capture(self.pg_interfaces) self.pg_start() rx = self.pg0.get_capture(1) self.pg1.add_stream(pm) self.pg_enable_capture(self.pg_interfaces) self.pg_start() rx = self.pg0.get_capture(1) # # Disable PG1 # self.pg1.unconfig_ip4() # # PG1 does not forward IP traffic # self.send_and_assert_no_replies(self.pg1, pu, "IP disabled") self.send_and_assert_no_replies(self.pg1, pm, "IP disabled") class TestIPSubNets(VppTestCase): """IPv4 Subnets""" @classmethod def setUpClass(cls): super(TestIPSubNets, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPSubNets, cls).tearDownClass() def setUp(self): super(TestIPSubNets, self).setUp() # create a 2 pg interfaces self.create_pg_interfaces(range(2)) # pg0 we will use to experiment self.pg0.admin_up() # pg1 is setup normally self.pg1.admin_up() self.pg1.config_ip4() self.pg1.resolve_arp() def tearDown(self): super(TestIPSubNets, self).tearDown() for i in self.pg_interfaces: i.admin_down() def test_ip_sub_nets(self): """IP Sub Nets""" # # Configure a covering route to forward so we know # when we are dropping # cover_route = VppIpRoute( self, "10.0.0.0", 8, [VppRoutePath(self.pg1.remote_ip4, self.pg1.sw_if_index)], ) cover_route.add_vpp_config() p = ( Ether(src=self.pg1.remote_mac, dst=self.pg1.local_mac) / IP(dst="10.10.10.10", src=self.pg0.local_ip4) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) self.pg1.add_stream(p) self.pg_enable_capture(self.pg_interfaces) self.pg_start() rx = self.pg1.get_capture(1) # # Configure some non-/24 subnets on an IP interface # ip_addr_n = socket.inet_pton(socket.AF_INET, "10.10.10.10") self.vapi.sw_interface_add_del_address( sw_if_index=self.pg0.sw_if_index, prefix="10.10.10.10/16" ) pn = ( Ether(src=self.pg1.remote_mac, dst=self.pg1.local_mac) / IP(dst="10.10.0.0", src=self.pg0.local_ip4) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) pb = ( Ether(src=self.pg1.remote_mac, dst=self.pg1.local_mac) / IP(dst="10.10.255.255", src=self.pg0.local_ip4) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) self.send_and_assert_no_replies(self.pg1, pn, "IP Network address") self.send_and_assert_no_replies(self.pg1, pb, "IP Broadcast address") # remove the sub-net and we are forwarding via the cover again self.vapi.sw_interface_add_del_address( sw_if_index=self.pg0.sw_if_index, prefix="10.10.10.10/16", is_add=0 ) self.pg1.add_stream(pn) self.pg_enable_capture(self.pg_interfaces) self.pg_start() rx = self.pg1.get_capture(1) self.pg1.add_stream(pb) self.pg_enable_capture(self.pg_interfaces) self.pg_start() rx = self.pg1.get_capture(1) # # A /31 is a special case where the 'other-side' is an attached host # packets to that peer generate ARP requests # ip_addr_n = socket.inet_pton(socket.AF_INET, "10.10.10.10") self.vapi.sw_interface_add_del_address( sw_if_index=self.pg0.sw_if_index, prefix="10.10.10.10/31" ) pn = ( Ether(src=self.pg1.remote_mac, dst=self.pg1.local_mac) / IP(dst="10.10.10.11", src=self.pg0.local_ip4) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) self.pg1.add_stream(pn) self.pg_enable_capture(self.pg_interfaces) self.pg_start() rx = self.pg0.get_capture(1) rx[ARP] # remove the sub-net and we are forwarding via the cover again self.vapi.sw_interface_add_del_address( sw_if_index=self.pg0.sw_if_index, prefix="10.10.10.10/31", is_add=0 ) self.pg1.add_stream(pn) self.pg_enable_capture(self.pg_interfaces) self.pg_start() rx = self.pg1.get_capture(1) class TestIPLoadBalance(VppTestCase): """IPv4 Load-Balancing""" @classmethod def setUpClass(cls): super(TestIPLoadBalance, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPLoadBalance, cls).tearDownClass() def setUp(self): super(TestIPLoadBalance, self).setUp() self.create_pg_interfaces(range(5)) mpls_tbl = VppMplsTable(self, 0) mpls_tbl.add_vpp_config() for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() i.enable_mpls() def tearDown(self): for i in self.pg_interfaces: i.disable_mpls() i.unconfig_ip4() i.admin_down() super(TestIPLoadBalance, self).tearDown() def total_len(self, rxs): n = 0 for rx in rxs: n += len(rx) return n def test_ip_load_balance(self): """IP Load-Balancing""" fhc = VppEnum.vl_api_ip_flow_hash_config_t fhcv2 = VppEnum.vl_api_ip_flow_hash_config_v2_t af = VppEnum.vl_api_address_family_t # # An array of packets that differ only in the destination port # port_ip_pkts = [] port_mpls_pkts = [] port_gtp_pkts = [] # # An array of packets that differ only in the source address # src_ip_pkts = [] src_mpls_pkts = [] src_gtp_pkts = [] for ii in range(NUM_PKTS): internal_src_ip_hdr = IP(dst="10.0.0.1", src="20.0.0.1") port_ip_hdr = ( internal_src_ip_hdr / UDP(sport=1234, dport=1234 + ii) / Raw(b"\xa5" * 100) ) port_ip_pkts.append( (Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / port_ip_hdr) ) port_mpls_pkts.append( ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / MPLS(label=66, ttl=2) / port_ip_hdr ) ) port_gtp_pkts.append( ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / internal_src_ip_hdr / UDP(sport=2152, dport=2152, chksum=0) / GTP_U_Header(gtp_type="g_pdu", teid=200) / Raw(b"\xa5" * 100) ) ) src_ip_hdr = ( IP(dst="10.0.0.1", src="20.0.0.%d" % ii) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) src_ip_pkts.append( (Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / src_ip_hdr) ) src_mpls_pkts.append( ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / MPLS(label=66, ttl=2) / src_ip_hdr ) ) src_gtp_pkts.append( ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(dst="10.0.0.1", src="20.0.0.1") / UDP(sport=2152, dport=2152, chksum=0) / GTP_U_Header(gtp_type="g_pdu", teid=ii) / Raw(b"\xa5" * 100) ) ) route_10_0_0_1 = VppIpRoute( self, "10.0.0.1", 32, [ VppRoutePath(self.pg1.remote_ip4, self.pg1.sw_if_index), VppRoutePath(self.pg2.remote_ip4, self.pg2.sw_if_index), ], ) route_10_0_0_1.add_vpp_config() binding = VppMplsIpBind(self, 66, "10.0.0.1", 32) binding.add_vpp_config() # # inject the packet on pg0 - expect load-balancing across the 2 paths # - since the default hash config is to use IP src,dst and port # src,dst # We are not going to ensure equal amounts of packets across each link, # since the hash algorithm is statistical and therefore this can never # be guaranteed. But with 64 different packets we do expect some # balancing. So instead just ensure there is traffic on each link. # rx = self.send_and_expect_load_balancing( self.pg0, port_ip_pkts, [self.pg1, self.pg2] ) n_ip_pg0 = len(rx[0]) self.send_and_expect_load_balancing(self.pg0, src_ip_pkts, [self.pg1, self.pg2]) self.send_and_expect_load_balancing( self.pg0, port_mpls_pkts, [self.pg1, self.pg2] ) rx = self.send_and_expect_load_balancing( self.pg0, src_mpls_pkts, [self.pg1, self.pg2] ) n_mpls_pg0 = len(rx[0]) # # change the router ID and expect the distribution changes # self.vapi.set_ip_flow_hash_router_id(router_id=0x11111111) rx = self.send_and_expect_load_balancing( self.pg0, port_ip_pkts, [self.pg1, self.pg2] ) self.assertNotEqual(n_ip_pg0, len(rx[0])) rx = self.send_and_expect_load_balancing( self.pg0, src_mpls_pkts, [self.pg1, self.pg2] ) self.assertNotEqual(n_mpls_pg0, len(rx[0])) # # change the flow hash config so it's only IP src,dst # - now only the stream with differing source address will # load-balance # self.vapi.set_ip_flow_hash_v2( af=af.ADDRESS_IP4, table_id=0, flow_hash_config=( fhc.IP_API_FLOW_HASH_SRC_IP | fhc.IP_API_FLOW_HASH_DST_IP | fhc.IP_API_FLOW_HASH_PROTO ), ) self.send_and_expect_load_balancing(self.pg0, src_ip_pkts, [self.pg1, self.pg2]) self.send_and_expect_load_balancing( self.pg0, src_mpls_pkts, [self.pg1, self.pg2] ) self.send_and_expect_only(self.pg0, port_ip_pkts, self.pg2) # # this case gtp v1 teid key LB # self.vapi.set_ip_flow_hash_v3( af=af.ADDRESS_IP4, table_id=0, flow_hash_config=( fhcv2.IP_API_V2_FLOW_HASH_SRC_IP | fhcv2.IP_API_V2_FLOW_HASH_PROTO | fhcv2.IP_API_V2_FLOW_HASH_GTPV1_TEID ), ) self.logger.info(self.vapi.cli("show ip fib")) self.send_and_expect_load_balancing( self.pg0, src_gtp_pkts, [self.pg1, self.pg2] ) self.send_and_expect_only(self.pg0, port_gtp_pkts, self.pg2) # # change the flow hash config back to defaults # self.vapi.set_ip_flow_hash(vrf_id=0, src=1, dst=1, proto=1, sport=1, dport=1) # # Recursive prefixes # - testing that 2 stages of load-balancing occurs and there is no # polarisation (i.e. only 2 of 4 paths are used) # port_pkts = [] src_pkts = [] for ii in range(257): port_pkts.append( ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(dst="1.1.1.1", src="20.0.0.1") / UDP(sport=1234, dport=1234 + ii) / Raw(b"\xa5" * 100) ) ) src_pkts.append( ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(dst="1.1.1.1", src="20.0.0.%d" % ii) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) ) route_10_0_0_2 = VppIpRoute( self, "10.0.0.2", 32, [ VppRoutePath(self.pg3.remote_ip4, self.pg3.sw_if_index), VppRoutePath(self.pg4.remote_ip4, self.pg4.sw_if_index), ], ) route_10_0_0_2.add_vpp_config() route_1_1_1_1 = VppIpRoute( self, "1.1.1.1", 32, [ VppRoutePath("10.0.0.2", 0xFFFFFFFF), VppRoutePath("10.0.0.1", 0xFFFFFFFF), ], ) route_1_1_1_1.add_vpp_config() # # inject the packet on pg0 - expect load-balancing across all 4 paths # self.vapi.cli("clear trace") self.send_and_expect_load_balancing( self.pg0, port_pkts, [self.pg1, self.pg2, self.pg3, self.pg4] ) self.send_and_expect_load_balancing( self.pg0, src_pkts, [self.pg1, self.pg2, self.pg3, self.pg4] ) # # bring down pg1 expect LB to adjust to use only those that are up # self.pg1.link_down() rx = self.send_and_expect_load_balancing( self.pg0, src_pkts, [self.pg2, self.pg3, self.pg4] ) self.assertEqual(len(src_pkts), self.total_len(rx)) # # bring down pg2 expect LB to adjust to use only those that are up # self.pg2.link_down() rx = self.send_and_expect_load_balancing( self.pg0, src_pkts, [self.pg3, self.pg4] ) self.assertEqual(len(src_pkts), self.total_len(rx)) # # bring the links back up - expect LB over all again # self.pg1.link_up() self.pg2.link_up() rx = self.send_and_expect_load_balancing( self.pg0, src_pkts, [self.pg1, self.pg2, self.pg3, self.pg4] ) self.assertEqual(len(src_pkts), self.total_len(rx)) # # The same link-up/down but this time admin state # self.pg1.admin_down() self.pg2.admin_down() rx = self.send_and_expect_load_balancing( self.pg0, src_pkts, [self.pg3, self.pg4] ) self.assertEqual(len(src_pkts), self.total_len(rx)) self.pg1.admin_up() self.pg2.admin_up() self.pg1.resolve_arp() self.pg2.resolve_arp() rx = self.send_and_expect_load_balancing( self.pg0, src_pkts, [self.pg1, self.pg2, self.pg3, self.pg4] ) self.assertEqual(len(src_pkts), self.total_len(rx)) # # Recursive prefixes # - testing that 2 stages of load-balancing, no choices # port_pkts = [] for ii in range(257): port_pkts.append( ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(dst="1.1.1.2", src="20.0.0.2") / UDP(sport=1234, dport=1234 + ii) / Raw(b"\xa5" * 100) ) ) route_10_0_0_3 = VppIpRoute( self, "10.0.0.3", 32, [VppRoutePath(self.pg3.remote_ip4, self.pg3.sw_if_index)], ) route_10_0_0_3.add_vpp_config() route_1_1_1_2 = VppIpRoute( self, "1.1.1.2", 32, [VppRoutePath("10.0.0.3", 0xFFFFFFFF)] ) route_1_1_1_2.add_vpp_config() # # inject the packet on pg0 - rx only on via routes output interface # self.vapi.cli("clear trace") self.send_and_expect_only(self.pg0, port_pkts, self.pg3) # # Add a LB route in the presence of a down link - expect no # packets over the down link # self.pg3.link_down() route_10_0_0_3 = VppIpRoute( self, "10.0.0.3", 32, [ VppRoutePath(self.pg3.remote_ip4, self.pg3.sw_if_index), VppRoutePath(self.pg4.remote_ip4, self.pg4.sw_if_index), ], ) route_10_0_0_3.add_vpp_config() port_pkts = [] for ii in range(257): port_pkts.append( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(dst="10.0.0.3", src="20.0.0.2") / UDP(sport=1234, dport=1234 + ii) / Raw(b"\xa5" * 100) ) self.send_and_expect_only(self.pg0, port_pkts, self.pg4) # bring the link back up self.pg3.link_up() rx = self.send_and_expect_load_balancing( self.pg0, port_pkts, [self.pg3, self.pg4] ) self.assertEqual(len(src_pkts), self.total_len(rx)) class TestIPVlan0(VppTestCase): """IPv4 VLAN-0""" @classmethod def setUpClass(cls): super(TestIPVlan0, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPVlan0, cls).tearDownClass() def setUp(self): super(TestIPVlan0, self).setUp() self.create_pg_interfaces(range(2)) mpls_tbl = VppMplsTable(self, 0) mpls_tbl.add_vpp_config() for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() i.enable_mpls() def tearDown(self): for i in self.pg_interfaces: i.disable_mpls() i.unconfig_ip4() i.admin_down() super(TestIPVlan0, self).tearDown() def test_ip_vlan_0(self): """IP VLAN-0""" pkts = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / Dot1Q(vlan=0) / IP(dst=self.pg1.remote_ip4, src=self.pg0.remote_ip4) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) * NUM_PKTS # # Expect that packets sent on VLAN-0 are forwarded on the # main interface. # self.send_and_expect(self.pg0, pkts, self.pg1) class IPPuntSetup(object): """Setup for IPv4 Punt Police/Redirect""" def punt_setup(self): self.create_pg_interfaces(range(4)) for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() # use UDP packet that have a port we need to explicitly # register to get punted. pt_l4 = VppEnum.vl_api_punt_type_t.PUNT_API_TYPE_L4 af_ip4 = VppEnum.vl_api_address_family_t.ADDRESS_IP4 udp_proto = VppEnum.vl_api_ip_proto_t.IP_API_PROTO_UDP punt_udp = { "type": pt_l4, "punt": { "l4": { "af": af_ip4, "protocol": udp_proto, "port": 1234, } }, } self.vapi.set_punt(is_add=1, punt=punt_udp) af_ip6 = VppEnum.vl_api_address_family_t.ADDRESS_IP6 punt_udp = { "type": pt_l4, "punt": { "l4": { "af": af_ip6, "protocol": udp_proto, "port": 1236, } }, } self.vapi.set_punt(is_add=1, punt=punt_udp) self.pkt = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg0.local_ip4) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) def punt_teardown(self): for i in self.pg_interfaces: i.unconfig_ip4() i.admin_down() class TestIPPunt(IPPuntSetup, VppTestCase): """IPv4 Punt Police/Redirect""" def setUp(self): super().setUp() super().punt_setup() def tearDown(self): super().punt_teardown() super().tearDown() def test_ip_punt_api_validation(self): """IP punt API parameter validation""" nh_addr = self.pg1.remote_ip4 punt = { "rx_sw_if_index": self.pg0.sw_if_index, "af": VppEnum.vl_api_address_family_t.ADDRESS_IP4, "n_paths": 1000000, "paths": [], } with self.assertRaises(vpp_papi.VPPIOError): self.vapi.add_del_ip_punt_redirect_v2(punt=punt, is_add=True) punt = { "rx_sw_if_index": self.pg0.sw_if_index, "af": VppEnum.vl_api_address_family_t.ADDRESS_IP4, "n_paths": 0, "paths": [], } self.vapi.add_del_ip_punt_redirect_v2(punt=punt, is_add=True) def test_ip_punt(self): """IP punt police and redirect""" pkts = self.pkt * 1025 # # Configure a punt redirect via pg1. # nh_addr = self.pg1.remote_ip4 ip_punt_redirect = VppIpPuntRedirect( self, self.pg0.sw_if_index, self.pg1.sw_if_index, nh_addr ) ip_punt_redirect.add_vpp_config() self.send_and_expect(self.pg0, pkts, self.pg1) # # add a policer # policer = VppPolicer(self, "ip4-punt", 400, 0, 10, 0, rate_type=1) policer.add_vpp_config() ip_punt_policer = VppIpPuntPolicer(self, policer.policer_index) ip_punt_policer.add_vpp_config() self.vapi.cli("clear trace") self.pg0.add_stream(pkts) self.pg_enable_capture(self.pg_interfaces) self.pg_start() # # the number of packet received should be greater than 0, # but not equal to the number sent, since some were policed # rx = self.pg1._get_capture(1) stats = policer.get_stats() # Single rate policer - expect conform, violate but no exceed self.assertGreater(stats["conform_packets"], 0) self.assertEqual(stats["exceed_packets"], 0) self.assertGreater(stats["violate_packets"], 0) self.assertGreater(len(rx), 0) self.assertLess(len(rx), len(pkts)) # # remove the policer. back to full rx # ip_punt_policer.remove_vpp_config() policer.remove_vpp_config() self.send_and_expect(self.pg0, pkts, self.pg1) # # remove the redirect. expect full drop. # ip_punt_redirect.remove_vpp_config() self.send_and_assert_no_replies(self.pg0, pkts, "IP no punt config") # # Add a redirect that is not input port selective # ip_punt_redirect = VppIpPuntRedirect( self, 0xFFFFFFFF, self.pg1.sw_if_index, nh_addr ) ip_punt_redirect.add_vpp_config() self.send_and_expect(self.pg0, pkts, self.pg1) ip_punt_redirect.remove_vpp_config() def test_ip_punt_vrf(self): """IP punt/local with VRFs""" # use a punt redirect to test if for-us packets are accepted pkts = self.pkt * 1025 vlans_pg0 = [VppDot1QSubint(self, self.pg0, v) for v in range(100, 104)] vlans_pg1 = [VppDot1QSubint(self, self.pg1, v) for v in range(100, 104)] tbl4 = [VppIpTable(self, v).add_vpp_config() for v in range(100, 104)] tbl6 = [VppIpTable(self, v, True).add_vpp_config() for v in range(100, 104)] for v in vlans_pg0 + vlans_pg1: v.admin_up() v.set_table_ip4(v.vlan) v.set_table_ip6(v.vlan) v.config_ip4() v.config_ip6() v.resolve_arp() v.resolve_ndp() [ VppIpPuntRedirect( self, vlans_pg0[i].sw_if_index, vlans_pg1[i].sw_if_index, vlans_pg1[i].remote_ip4, ).add_vpp_config() for i in range(4) ] [ VppIpPuntRedirect( self, vlans_pg0[i].sw_if_index, vlans_pg1[i].sw_if_index, vlans_pg1[i].remote_ip6, ).add_vpp_config() for i in range(4) ] pkts = [ ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / Dot1Q(vlan=i.vlan) / IP(src=i.remote_ip4, dst=i.local_ip4) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) for i in vlans_pg0 ] self.send_and_expect(self.pg0, pkts, self.pg1) # # IPv4 # # we reject packets for source addresses in the wrong vlan/VRF pkts = [ ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / Dot1Q(vlan=i.vlan) / IP(src="1.1.1.1", dst=i.local_ip4) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) for i in vlans_pg0 ] # single and dual loop self.send_and_assert_no_replies(self.pg0, [pkts[0]]) self.send_and_assert_no_replies(self.pg0, pkts) self.assert_error_counter_equal("/err/ip4-local/src_lookup_miss", len(pkts) + 1) # using the same source in different tables, should reject # for the table that the source is not present in # the first packet in the stream is drop pkts = [ ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / Dot1Q(vlan=i.vlan) / IP(src=vlans_pg0[0].remote_ip4, dst=i.local_ip4) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) for i in vlans_pg0 ] # single loop accept and drop # followed by both in the same frame/loop self.send_and_expect(self.pg0, [pkts[0]], self.pg1) self.send_and_assert_no_replies(self.pg0, [pkts[1]]) self.send_and_expect(self.pg0, pkts * 4, self.pg1, n_rx=4) # using the same source in different tables, should reject # for the table that the source is not present in # the first packet in the stream is accept pkts = [ ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / Dot1Q(vlan=i.vlan) / IP(src=vlans_pg0[3].remote_ip4, dst=i.local_ip4) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) for i in vlans_pg0 ] # single loop accept and drop # followed by both in the same frame/loop self.send_and_expect(self.pg0, [pkts[3]], self.pg1) self.send_and_assert_no_replies(self.pg0, [pkts[1]]) self.send_and_expect(self.pg0, pkts * 4, self.pg1, n_rx=4) # # IPv6 # # we reject packets for source addresses in the wrong vlan/VRF pkts = [ ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / Dot1Q(vlan=i.vlan) / IPv6(src="1::1", dst=i.local_ip6) / UDP(sport=1236, dport=1236) / Raw(b"\xa5" * 100) ) for i in vlans_pg0 ] # single and dual loop self.send_and_assert_no_replies(self.pg0, [pkts[0]]) self.send_and_assert_no_replies(self.pg0, pkts) self.assert_error_counter_equal("/err/ip6-input/src_lookup_miss", len(pkts) + 1) # using the same source in different tables, should reject # for the table that the source is not present in # the first packet in the stream is drop pkts = [ ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / Dot1Q(vlan=i.vlan) / IPv6(src=vlans_pg0[0].remote_ip6, dst=i.local_ip6) / UDP(sport=1236, dport=1236) / Raw(b"\xa5" * 100) ) for i in vlans_pg0 ] # single loop accept and drop # followed by both in the same frame/loop self.send_and_expect(self.pg0, [pkts[0]], self.pg1) self.send_and_assert_no_replies(self.pg0, [pkts[1]]) self.send_and_expect(self.pg0, pkts * 4, self.pg1, n_rx=4) # using the same source in different tables, should reject # for the table that the source is not present in # the first packet in the stream is accept pkts = [ ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / Dot1Q(vlan=i.vlan) / IPv6(src=vlans_pg0[3].remote_ip6, dst=i.local_ip6) / UDP(sport=1236, dport=1236) / Raw(b"\xa5" * 100) ) for i in vlans_pg0 ] # single loop accept and drop # followed by both in the same frame/loop self.send_and_expect(self.pg0, [pkts[3]], self.pg1) self.send_and_assert_no_replies(self.pg0, [pkts[1]]) self.send_and_expect(self.pg0, pkts * 4, self.pg1, n_rx=4) for v in vlans_pg0 + vlans_pg1: v.unconfig_ip4() v.unconfig_ip6() v.set_table_ip4(0) v.set_table_ip6(0) def test_ip_punt_dump(self): """IP4 punt redirect dump""" # # Configure a punt redirects # nh_address = self.pg3.remote_ip4 ipr_03 = VppIpPuntRedirect( self, self.pg0.sw_if_index, self.pg3.sw_if_index, nh_address ) ipr_13 = VppIpPuntRedirect( self, self.pg1.sw_if_index, self.pg3.sw_if_index, nh_address ) ipr_23 = VppIpPuntRedirect( self, self.pg2.sw_if_index, self.pg3.sw_if_index, "0.0.0.0" ) ipr_03.add_vpp_config() ipr_13.add_vpp_config() ipr_23.add_vpp_config() # # Dump pg0 punt redirects # self.assertTrue(ipr_03.query_vpp_config()) self.assertTrue(ipr_13.query_vpp_config()) self.assertTrue(ipr_23.query_vpp_config()) # # Dump punt redirects for all interfaces # punts = self.vapi.ip_punt_redirect_dump(0xFFFFFFFF) self.assertEqual(len(punts), 3) for p in punts: self.assertEqual(p.punt.tx_sw_if_index, self.pg3.sw_if_index) self.assertNotEqual(punts[1].punt.nh, self.pg3.remote_ip4) self.assertEqual(str(punts[2].punt.nh), "0.0.0.0") class TestIPPuntHandoff(IPPuntSetup, VppTestCase): """IPv4 Punt Policer thread handoff""" vpp_worker_count = 2 def setUp(self): super(TestIPPuntHandoff, self).setUp() super(TestIPPuntHandoff, self).punt_setup() def tearDown(self): super(TestIPPuntHandoff, self).punt_teardown() super(TestIPPuntHandoff, self).tearDown() def test_ip_punt_policer_handoff(self): """IP4 punt policer thread handoff""" pkts = self.pkt * NUM_PKTS # # Configure a punt redirect via pg1. # nh_addr = self.pg1.remote_ip4 ip_punt_redirect = VppIpPuntRedirect( self, self.pg0.sw_if_index, self.pg1.sw_if_index, nh_addr ) ip_punt_redirect.add_vpp_config() action_tx = PolicerAction( VppEnum.vl_api_sse2_qos_action_type_t.SSE2_QOS_ACTION_API_TRANSMIT, 0 ) # # This policer drops no packets, we are just # testing that they get to the right thread. # policer = VppPolicer( self, "ip4-punt", 400, 0, 10, 0, 1, 0, 0, False, action_tx, action_tx, action_tx, ) policer.add_vpp_config() ip_punt_policer = VppIpPuntPolicer(self, policer.policer_index) ip_punt_policer.add_vpp_config() for worker in [0, 1]: self.send_and_expect(self.pg0, pkts, self.pg1, worker=worker) self.logger.debug(self.vapi.cli("show trace max 100")) # Combined stats, all threads stats = policer.get_stats() # Single rate policer - expect conform, violate but no exceed self.assertGreater(stats["conform_packets"], 0) self.assertEqual(stats["exceed_packets"], 0) self.assertGreater(stats["violate_packets"], 0) # Worker 0, should have done all the policing stats0 = policer.get_stats(worker=0) self.assertEqual(stats, stats0) # Worker 1, should have handed everything off stats1 = policer.get_stats(worker=1) self.assertEqual(stats1["conform_packets"], 0) self.assertEqual(stats1["exceed_packets"], 0) self.assertEqual(stats1["violate_packets"], 0) # Bind the policer to worker 1 and repeat policer.bind_vpp_config(1, True) for worker in [0, 1]: self.send_and_expect(self.pg0, pkts, self.pg1, worker=worker) self.logger.debug(self.vapi.cli("show trace max 100")) # The 2 workers should now have policed the same amount stats = policer.get_stats() stats0 = policer.get_stats(worker=0) stats1 = policer.get_stats(worker=1) self.assertGreater(stats0["conform_packets"], 0) self.assertEqual(stats0["exceed_packets"], 0) self.assertGreater(stats0["violate_packets"], 0) self.assertGreater(stats1["conform_packets"], 0) self.assertEqual(stats1["exceed_packets"], 0) self.assertGreater(stats1["violate_packets"], 0) self.assertEqual( stats0["conform_packets"] + stats1["conform_packets"], stats["conform_packets"], ) self.assertEqual( stats0["violate_packets"] + stats1["violate_packets"], stats["violate_packets"], ) # Unbind the policer and repeat policer.bind_vpp_config(1, False) for worker in [0, 1]: self.send_and_expect(self.pg0, pkts, self.pg1, worker=worker) self.logger.debug(self.vapi.cli("show trace max 100")) # The policer should auto-bind to worker 0 when packets arrive stats = policer.get_stats() stats0new = policer.get_stats(worker=0) stats1new = policer.get_stats(worker=1) self.assertGreater(stats0new["conform_packets"], stats0["conform_packets"]) self.assertEqual(stats0new["exceed_packets"], 0) self.assertGreater(stats0new["violate_packets"], stats0["violate_packets"]) self.assertEqual(stats1, stats1new) # # Clean up # ip_punt_policer.remove_vpp_config() policer.remove_vpp_config() ip_punt_redirect.remove_vpp_config() class TestIPDeag(VppTestCase): """IPv4 Deaggregate Routes""" @classmethod def setUpClass(cls): super(TestIPDeag, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPDeag, cls).tearDownClass() def setUp(self): super(TestIPDeag, self).setUp() self.create_pg_interfaces(range(3)) for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() def tearDown(self): super(TestIPDeag, self).tearDown() for i in self.pg_interfaces: i.unconfig_ip4() i.admin_down() def test_ip_deag(self): """IP Deag Routes""" # # Create a table to be used for: # 1 - another destination address lookup # 2 - a source address lookup # table_dst = VppIpTable(self, 1) table_src = VppIpTable(self, 2) table_dst.add_vpp_config() table_src.add_vpp_config() # # Add a route in the default table to point to a deag/ # second lookup in each of these tables # route_to_dst = VppIpRoute( self, "1.1.1.1", 32, [VppRoutePath("0.0.0.0", 0xFFFFFFFF, nh_table_id=1)] ) route_to_src = VppIpRoute( self, "1.1.1.2", 32, [ VppRoutePath( "0.0.0.0", 0xFFFFFFFF, nh_table_id=2, type=FibPathType.FIB_PATH_TYPE_SOURCE_LOOKUP, ) ], ) route_to_dst.add_vpp_config() route_to_src.add_vpp_config() # # packets to these destination are dropped, since they'll # hit the respective default routes in the second table # p_dst = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src="5.5.5.5", dst="1.1.1.1") / TCP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) p_src = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src="2.2.2.2", dst="1.1.1.2") / TCP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) pkts_dst = p_dst * 257 pkts_src = p_src * 257 self.send_and_assert_no_replies(self.pg0, pkts_dst, "IP in dst table") self.send_and_assert_no_replies(self.pg0, pkts_src, "IP in src table") # # add a route in the dst table to forward via pg1 # route_in_dst = VppIpRoute( self, "1.1.1.1", 32, [VppRoutePath(self.pg1.remote_ip4, self.pg1.sw_if_index)], table_id=1, ) route_in_dst.add_vpp_config() self.send_and_expect(self.pg0, pkts_dst, self.pg1) # # add a route in the src table to forward via pg2 # route_in_src = VppIpRoute( self, "2.2.2.2", 32, [VppRoutePath(self.pg2.remote_ip4, self.pg2.sw_if_index)], table_id=2, ) route_in_src.add_vpp_config() self.send_and_expect(self.pg0, pkts_src, self.pg2) # # loop in the lookup DP # route_loop = VppIpRoute( self, "2.2.2.3", 32, [VppRoutePath("0.0.0.0", 0xFFFFFFFF, nh_table_id=0)] ) route_loop.add_vpp_config() p_l = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src="2.2.2.4", dst="2.2.2.3") / TCP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) self.send_and_assert_no_replies(self.pg0, p_l * 257, "IP lookup loop") class TestIPInput(VppTestCase): """IPv4 Input Exceptions""" @classmethod def setUpClass(cls): super(TestIPInput, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPInput, cls).tearDownClass() def setUp(self): super(TestIPInput, self).setUp() self.create_pg_interfaces(range(2)) for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() def tearDown(self): super(TestIPInput, self).tearDown() for i in self.pg_interfaces: i.unconfig_ip4() i.admin_down() def test_ip_input(self): """IP Input Exceptions""" # i can't find a way in scapy to construct an IP packet # with a length less than the IP header length # # Packet too short - this is forwarded # p_short = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg1.remote_ip4, len=40) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) rx = self.send_and_expect(self.pg0, p_short * NUM_PKTS, self.pg1) # # Packet too long - this is dropped # p_long = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg1.remote_ip4, len=400) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) rx = self.send_and_assert_no_replies(self.pg0, p_long * NUM_PKTS, "too long") # # bad chksum - this is dropped # p_chksum = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg1.remote_ip4, chksum=400) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) rx = self.send_and_assert_no_replies( self.pg0, p_chksum * NUM_PKTS, "bad checksum" ) # # bad version - this is dropped # p_ver = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg1.remote_ip4, version=3) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) rx = self.send_and_assert_no_replies( self.pg0, p_ver * NUM_PKTS, "funky version" ) # # fragment offset 1 - this is dropped # p_frag = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg1.remote_ip4, frag=1) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) rx = self.send_and_assert_no_replies(self.pg0, p_frag * NUM_PKTS, "frag offset") # # TTL expired packet # p_ttl = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg1.remote_ip4, ttl=1) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) rxs = self.send_and_expect_some(self.pg0, p_ttl * NUM_PKTS, self.pg0) for rx in rxs: icmp = rx[ICMP] self.assertEqual(icmptypes[icmp.type], "time-exceeded") self.assertEqual(icmpcodes[icmp.type][icmp.code], "ttl-zero-during-transit") self.assertEqual(icmp.src, self.pg0.remote_ip4) self.assertEqual(icmp.dst, self.pg1.remote_ip4) # # MTU exceeded # p_mtu = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg1.remote_ip4, ttl=10, flags="DF") / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 2000) ) self.vapi.sw_interface_set_mtu(self.pg1.sw_if_index, [1500, 0, 0, 0]) rxs = self.send_and_expect_some(self.pg0, p_mtu * NUM_PKTS, self.pg0) for rx in rxs: icmp = rx[ICMP] self.assertEqual(icmptypes[icmp.type], "dest-unreach") self.assertEqual(icmpcodes[icmp.type][icmp.code], "fragmentation-needed") self.assertEqual(icmp.nexthopmtu, 1500) self.assertEqual(icmp.src, self.pg0.remote_ip4) self.assertEqual(icmp.dst, self.pg1.remote_ip4) self.vapi.sw_interface_set_mtu(self.pg1.sw_if_index, [2500, 0, 0, 0]) rx = self.send_and_expect(self.pg0, p_mtu * NUM_PKTS, self.pg1) # Reset MTU for subsequent tests self.vapi.sw_interface_set_mtu(self.pg1.sw_if_index, [9000, 0, 0, 0]) # # source address 0.0.0.0 and 25.255.255.255 and for-us # p_s0 = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src="0.0.0.0", dst=self.pg0.local_ip4) / ICMP(id=4, seq=4) / Raw(load=b"\x0a" * 18) ) rx = self.send_and_assert_no_replies(self.pg0, p_s0 * 17) p_s0 = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src="255.255.255.255", dst=self.pg0.local_ip4) / ICMP(id=4, seq=4) / Raw(load=b"\x0a" * 18) ) rx = self.send_and_assert_no_replies(self.pg0, p_s0 * 17) class TestIPDirectedBroadcast(VppTestCase): """IPv4 Directed Broadcast""" @classmethod def setUpClass(cls): super(TestIPDirectedBroadcast, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPDirectedBroadcast, cls).tearDownClass() def setUp(self): super(TestIPDirectedBroadcast, self).setUp() self.create_pg_interfaces(range(2)) for i in self.pg_interfaces: i.admin_up() def tearDown(self): super(TestIPDirectedBroadcast, self).tearDown() for i in self.pg_interfaces: i.admin_down() def test_ip_input(self): """IP Directed Broadcast""" # # set the directed broadcast on pg0 first, then config IP4 addresses # for pg1 directed broadcast is always disabled self.vapi.sw_interface_set_ip_directed_broadcast(self.pg0.sw_if_index, 1) p0 = ( Ether(src=self.pg1.remote_mac, dst=self.pg1.local_mac) / IP(src="1.1.1.1", dst=self.pg0._local_ip4_bcast) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 2000) ) p1 = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src="1.1.1.1", dst=self.pg1._local_ip4_bcast) / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 2000) ) self.pg0.config_ip4() self.pg0.resolve_arp() self.pg1.config_ip4() self.pg1.resolve_arp() # # test packet is L2 broadcast # rx = self.send_and_expect(self.pg1, p0 * NUM_PKTS, self.pg0) self.assertTrue(rx[0][Ether].dst, "ff:ff:ff:ff:ff:ff") self.send_and_assert_no_replies( self.pg0, p1 * NUM_PKTS, "directed broadcast disabled" ) # # toggle directed broadcast on pg0 # self.vapi.sw_interface_set_ip_directed_broadcast(self.pg0.sw_if_index, 0) self.send_and_assert_no_replies( self.pg1, p0 * NUM_PKTS, "directed broadcast disabled" ) self.vapi.sw_interface_set_ip_directed_broadcast(self.pg0.sw_if_index, 1) rx = self.send_and_expect(self.pg1, p0 * NUM_PKTS, self.pg0) self.pg0.unconfig_ip4() self.pg1.unconfig_ip4() class TestIPLPM(VppTestCase): """IPv4 longest Prefix Match""" @classmethod def setUpClass(cls): super(TestIPLPM, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPLPM, cls).tearDownClass() def setUp(self): super(TestIPLPM, self).setUp() self.create_pg_interfaces(range(4)) for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() def tearDown(self): super(TestIPLPM, self).tearDown() for i in self.pg_interfaces: i.admin_down() i.unconfig_ip4() def test_ip_lpm(self): """IP longest Prefix Match""" s_24 = VppIpRoute( self, "10.1.2.0", 24, [VppRoutePath(self.pg1.remote_ip4, self.pg1.sw_if_index)], ) s_24.add_vpp_config() s_8 = VppIpRoute( self, "10.0.0.0", 8, [VppRoutePath(self.pg2.remote_ip4, self.pg2.sw_if_index)], ) s_8.add_vpp_config() p_8 = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src="1.1.1.1", dst="10.1.1.1") / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 2000) ) p_24 = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src="1.1.1.1", dst="10.1.2.1") / UDP(sport=1234, dport=1234) / Raw(b"\xa5" * 2000) ) self.logger.info(self.vapi.cli("sh ip fib mtrie")) rx = self.send_and_expect(self.pg0, p_8 * NUM_PKTS, self.pg2) rx = self.send_and_expect(self.pg0, p_24 * NUM_PKTS, self.pg1) @tag_fixme_vpp_workers class TestIPv4Frag(VppTestCase): """IPv4 fragmentation""" @classmethod def setUpClass(cls): super(TestIPv4Frag, cls).setUpClass() cls.create_pg_interfaces([0, 1]) cls.src_if = cls.pg0 cls.dst_if = cls.pg1 # setup all interfaces for i in cls.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() @classmethod def tearDownClass(cls): super(TestIPv4Frag, cls).tearDownClass() def test_frag_large_packets(self): """Fragmentation of large packets""" self.vapi.cli("adjacency counters enable") p = ( Ether(dst=self.src_if.local_mac, src=self.src_if.remote_mac) / IP(src=self.src_if.remote_ip4, dst=self.dst_if.remote_ip4) / UDP(sport=1234, dport=5678) / Raw() ) self.extend_packet(p, 6000, "abcde") saved_payload = p[Raw].load nbr = VppNeighbor( self, self.dst_if.sw_if_index, self.dst_if.remote_mac, self.dst_if.remote_ip4, ).add_vpp_config() # Force fragmentation by setting MTU of output interface # lower than packet size self.vapi.sw_interface_set_mtu(self.dst_if.sw_if_index, [5000, 0, 0, 0]) self.pg_enable_capture() self.src_if.add_stream(p) self.pg_start() # Expecting 3 fragments because size of created fragments currently # cannot be larger then VPP buffer size (which is 2048) packets = self.dst_if.get_capture(3) # we should show 3 packets thru the neighbor self.assertEqual(3, nbr.get_stats()["packets"]) # Assume VPP sends the fragments in order payload = b"" for p in packets: payload_offset = p.frag * 8 if payload_offset > 0: payload_offset -= 8 # UDP header is not in payload self.assert_equal(payload_offset, len(payload)) payload += p[Raw].load self.assert_equal(payload, saved_payload, "payload") class TestIPReplace(VppTestCase): """IPv4 Table Replace""" @classmethod def setUpClass(cls): super(TestIPReplace, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPReplace, cls).tearDownClass() def setUp(self): super(TestIPReplace, self).setUp() self.create_pg_interfaces(range(4)) table_id = 1 self.tables = [] for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() i.generate_remote_hosts(2) self.tables.append(VppIpTable(self, table_id).add_vpp_config()) table_id += 1 def tearDown(self): super(TestIPReplace, self).tearDown() for i in self.pg_interfaces: i.admin_down() i.unconfig_ip4() def test_replace(self): """IP Table Replace""" MRouteItfFlags = VppEnum.vl_api_mfib_itf_flags_t MRouteEntryFlags = VppEnum.vl_api_mfib_entry_flags_t N_ROUTES = 20 links = [self.pg0, self.pg1, self.pg2, self.pg3] routes = [[], [], [], []] # load up the tables with some routes for ii, t in enumerate(self.tables): for jj in range(N_ROUTES): uni = VppIpRoute( self, "10.0.0.%d" % jj, 32, [ VppRoutePath( links[ii].remote_hosts[0].ip4, links[ii].sw_if_index ), VppRoutePath( links[ii].remote_hosts[1].ip4, links[ii].sw_if_index ), ], table_id=t.table_id, ).add_vpp_config() multi = VppIpMRoute( self, "0.0.0.0", "239.0.0.%d" % jj, 32, MRouteEntryFlags.MFIB_API_ENTRY_FLAG_NONE, [ VppMRoutePath( self.pg0.sw_if_index, MRouteItfFlags.MFIB_API_ITF_FLAG_ACCEPT, ), VppMRoutePath( self.pg1.sw_if_index, MRouteItfFlags.MFIB_API_ITF_FLAG_FORWARD, ), VppMRoutePath( self.pg2.sw_if_index, MRouteItfFlags.MFIB_API_ITF_FLAG_FORWARD, ), VppMRoutePath( self.pg3.sw_if_index, MRouteItfFlags.MFIB_API_ITF_FLAG_FORWARD, ), ], table_id=t.table_id, ).add_vpp_config() routes[ii].append({"uni": uni, "multi": multi}) # # replace the tables a few times # for kk in range(3): # replace_begin each table for t in self.tables: t.replace_begin() # all the routes are still there for ii, t in enumerate(self.tables): dump = t.dump() mdump = t.mdump() for r in routes[ii]: self.assertTrue(find_route_in_dump(dump, r["uni"], t)) self.assertTrue(find_mroute_in_dump(mdump, r["multi"], t)) # redownload the even numbered routes for ii, t in enumerate(self.tables): for jj in range(0, N_ROUTES, 2): routes[ii][jj]["uni"].add_vpp_config() routes[ii][jj]["multi"].add_vpp_config() # signal each table replace_end for t in self.tables: t.replace_end() # we should find the even routes, but not the odd for ii, t in enumerate(self.tables): dump = t.dump() mdump = t.mdump() for jj in range(0, N_ROUTES, 2): self.assertTrue(find_route_in_dump(dump, routes[ii][jj]["uni"], t)) self.assertTrue( find_mroute_in_dump(mdump, routes[ii][jj]["multi"], t) ) for jj in range(1, N_ROUTES - 1, 2): self.assertFalse(find_route_in_dump(dump, routes[ii][jj]["uni"], t)) self.assertFalse( find_mroute_in_dump(mdump, routes[ii][jj]["multi"], t) ) # reload all the routes for ii, t in enumerate(self.tables): for r in routes[ii]: r["uni"].add_vpp_config() r["multi"].add_vpp_config() # all the routes are still there for ii, t in enumerate(self.tables): dump = t.dump() mdump = t.mdump() for r in routes[ii]: self.assertTrue(find_route_in_dump(dump, r["uni"], t)) self.assertTrue(find_mroute_in_dump(mdump, r["multi"], t)) # # finally flush the tables for good measure # for t in self.tables: t.flush() self.assertEqual(len(t.dump()), 5) self.assertEqual(len(t.mdump()), 3) class TestIPCover(VppTestCase): """IPv4 Table Cover""" @classmethod def setUpClass(cls): super(TestIPCover, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIPCover, cls).tearDownClass() def setUp(self): super(TestIPCover, self).setUp() self.create_pg_interfaces(range(4)) table_id = 1 self.tables = [] for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() i.generate_remote_hosts(2) self.tables.append(VppIpTable(self, table_id).add_vpp_config()) table_id += 1 def tearDown(self): super(TestIPCover, self).tearDown() for i in self.pg_interfaces: i.admin_down() i.unconfig_ip4() def test_cover(self): """IP Table Cover""" # add a loop back with a /32 prefix lo = VppLoInterface(self) lo.admin_up() a = VppIpInterfaceAddress(self, lo, "127.0.0.1", 32).add_vpp_config() # add a neighbour that matches the loopback's /32 nbr = VppNeighbor( self, lo.sw_if_index, lo.remote_mac, "127.0.0.1" ).add_vpp_config() # add the default route which will be the cover for /32 r = VppIpRoute( self, "0.0.0.0", 0, [VppRoutePath("127.0.0.1", lo.sw_if_index)], register=False, ).add_vpp_config() # add/remove/add a longer mask cover r8 = VppIpRoute( self, "127.0.0.0", 8, [VppRoutePath("127.0.0.1", lo.sw_if_index)] ).add_vpp_config() r8.remove_vpp_config() r8.add_vpp_config() r8.remove_vpp_config() # remove the default route r.remove_vpp_config() # remove the interface prefix a.remove_vpp_config() class TestIP4Replace(VppTestCase): """IPv4 Interface Address Replace""" @classmethod def setUpClass(cls): super(TestIP4Replace, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIP4Replace, cls).tearDownClass() def setUp(self): super(TestIP4Replace, self).setUp() self.create_pg_interfaces(range(4)) for i in self.pg_interfaces: i.admin_up() def tearDown(self): super(TestIP4Replace, self).tearDown() for i in self.pg_interfaces: i.admin_down() def get_n_pfxs(self, intf): return len(self.vapi.ip_address_dump(intf.sw_if_index)) def test_replace(self): """IP interface address replace""" intf_pfxs = [[], [], [], []] # add prefixes to each of the interfaces for i in range(len(self.pg_interfaces)): intf = self.pg_interfaces[i] # 172.16.x.1/24 addr = "172.16.%d.1" % intf.sw_if_index a = VppIpInterfaceAddress(self, intf, addr, 24).add_vpp_config() intf_pfxs[i].append(a) # 172.16.x.2/24 - a different address in the same subnet as above addr = "172.16.%d.2" % intf.sw_if_index a = VppIpInterfaceAddress(self, intf, addr, 24).add_vpp_config() intf_pfxs[i].append(a) # 172.15.x.2/24 - a different address and subnet addr = "172.15.%d.2" % intf.sw_if_index a = VppIpInterfaceAddress(self, intf, addr, 24).add_vpp_config() intf_pfxs[i].append(a) # a dump should n_address in it for intf in self.pg_interfaces: self.assertEqual(self.get_n_pfxs(intf), 3) # # remove all the address thru a replace # self.vapi.sw_interface_address_replace_begin() self.vapi.sw_interface_address_replace_end() for intf in self.pg_interfaces: self.assertEqual(self.get_n_pfxs(intf), 0) # # add all the interface addresses back # for p in intf_pfxs: for v in p: v.add_vpp_config() for intf in self.pg_interfaces: self.assertEqual(self.get_n_pfxs(intf), 3) # # replace again, but this time update/re-add the address on the first # two interfaces # self.vapi.sw_interface_address_replace_begin() for p in intf_pfxs[:2]: for v in p: v.add_vpp_config() self.vapi.sw_interface_address_replace_end() # on the first two the address still exist, # on the other two they do not for intf in self.pg_interfaces[:2]: self.assertEqual(self.get_n_pfxs(intf), 3) for p in intf_pfxs[:2]: for v in p: self.assertTrue(v.query_vpp_config()) for intf in self.pg_interfaces[2:]: self.assertEqual(self.get_n_pfxs(intf), 0) # # add all the interface addresses back on the last two # for p in intf_pfxs[2:]: for v in p: v.add_vpp_config() for intf in self.pg_interfaces: self.assertEqual(self.get_n_pfxs(intf), 3) # # replace again, this time add different prefixes on all the interfaces # self.vapi.sw_interface_address_replace_begin() pfxs = [] for intf in self.pg_interfaces: # 172.18.x.1/24 addr = "172.18.%d.1" % intf.sw_if_index pfxs.append(VppIpInterfaceAddress(self, intf, addr, 24).add_vpp_config()) self.vapi.sw_interface_address_replace_end() # only .18 should exist on each interface for intf in self.pg_interfaces: self.assertEqual(self.get_n_pfxs(intf), 1) for pfx in pfxs: self.assertTrue(pfx.query_vpp_config()) # # remove everything # self.vapi.sw_interface_address_replace_begin() self.vapi.sw_interface_address_replace_end() for intf in self.pg_interfaces: self.assertEqual(self.get_n_pfxs(intf), 0) # # add prefixes to each interface. post-begin add the prefix from # interface X onto interface Y. this would normally be an error # since it would generate a 'duplicate address' warning. but in # this case, since what is newly downloaded is sane, it's ok # for intf in self.pg_interfaces: # 172.18.x.1/24 addr = "172.18.%d.1" % intf.sw_if_index VppIpInterfaceAddress(self, intf, addr, 24).add_vpp_config() self.vapi.sw_interface_address_replace_begin() pfxs = [] for intf in self.pg_interfaces: # 172.18.x.1/24 addr = "172.18.%d.1" % (intf.sw_if_index + 1) pfxs.append(VppIpInterfaceAddress(self, intf, addr, 24).add_vpp_config()) self.vapi.sw_interface_address_replace_end() self.logger.info(self.vapi.cli("sh int addr")) for intf in self.pg_interfaces: self.assertEqual(self.get_n_pfxs(intf), 1) for pfx in pfxs: self.assertTrue(pfx.query_vpp_config()) class TestIPv4PathMTU(VppTestCase): """IPv4 Path MTU""" @classmethod def setUpClass(cls): super(TestIPv4PathMTU, cls).setUpClass() cls.create_pg_interfaces(range(2)) # setup all interfaces for i in cls.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() @classmethod def tearDownClass(cls): super(TestIPv4PathMTU, cls).tearDownClass() def test_path_mtu(self): """Path MTU""" # # The goal here is not to test that fragmentation works correctly, # that's done elsewhere, the intent is to ensure that the Path MTU # settings are honoured. # self.vapi.cli("adjacency counters enable") # set the interface MTU to a reasonable value self.vapi.sw_interface_set_mtu(self.pg1.sw_if_index, [1800, 0, 0, 0]) self.pg1.generate_remote_hosts(4) p_2k = ( Ether(dst=self.pg0.local_mac, src=self.pg0.remote_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg1.remote_ip4) / UDP(sport=1234, dport=5678) / Raw(b"0xa" * 640) ) p_1k = ( Ether(dst=self.pg0.local_mac, src=self.pg0.remote_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg1.remote_ip4) / UDP(sport=1234, dport=5678) / Raw(b"0xa" * 320) ) nbr = VppNeighbor( self, self.pg1.sw_if_index, self.pg1.remote_mac, self.pg1.remote_ip4 ).add_vpp_config() # this is now the interface MTU frags self.send_and_expect(self.pg0, [p_2k], self.pg1, n_rx=2) self.send_and_expect(self.pg0, [p_1k], self.pg1) # drop the path MTU for this neighbour to below the interface MTU # expect more frags pmtu = VppIpPathMtu(self, self.pg1.remote_ip4, 900).add_vpp_config() self.send_and_expect(self.pg0, [p_2k], self.pg1, n_rx=3) self.send_and_expect(self.pg0, [p_1k], self.pg1, n_rx=2) # print/format the adj delegate self.logger.info(self.vapi.cli("sh adj 5")) # increase the path MTU to more than the interface # expect to use the interface MTU pmtu.modify(8192) self.send_and_expect(self.pg0, [p_2k], self.pg1, n_rx=2) self.send_and_expect(self.pg0, [p_1k], self.pg1) # go back to an MTU from the path # wrap the call around mark-n-sweep to enusre updates clear stale self.vapi.ip_path_mtu_replace_begin() pmtu.modify(900) self.vapi.ip_path_mtu_replace_end() self.send_and_expect(self.pg0, [p_2k], self.pg1, n_rx=3) self.send_and_expect(self.pg0, [p_1k], self.pg1, n_rx=2) # raise the interface's MTU # should still use that of the path self.vapi.sw_interface_set_mtu(self.pg1.sw_if_index, [2000, 0, 0, 0]) self.send_and_expect(self.pg0, [p_2k], self.pg1, n_rx=3) self.send_and_expect(self.pg0, [p_1k], self.pg1, n_rx=2) # set path high and interface low pmtu.modify(2000) self.vapi.sw_interface_set_mtu(self.pg1.sw_if_index, [900, 0, 0, 0]) self.send_and_expect(self.pg0, [p_2k], self.pg1, n_rx=3) self.send_and_expect(self.pg0, [p_1k], self.pg1, n_rx=2) # remove the path MTU using the mark-n-sweep semantics self.vapi.sw_interface_set_mtu(self.pg1.sw_if_index, [1800, 0, 0, 0]) self.vapi.ip_path_mtu_replace_begin() self.vapi.ip_path_mtu_replace_end() self.send_and_expect(self.pg0, [p_2k], self.pg1, n_rx=2) self.send_and_expect(self.pg0, [p_1k], self.pg1) # # set path MTU for a neighbour that doesn't exist, yet # pmtu2 = VppIpPathMtu(self, self.pg1.remote_hosts[2].ip4, 900).add_vpp_config() p_2k = ( Ether(dst=self.pg0.local_mac, src=self.pg0.remote_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg1.remote_hosts[2].ip4) / UDP(sport=1234, dport=5678) / Raw(b"0xa" * 640) ) p_1k = ( Ether(dst=self.pg0.local_mac, src=self.pg0.remote_mac) / IP(src=self.pg0.remote_ip4, dst=self.pg1.remote_hosts[2].ip4) / UDP(sport=1234, dport=5678) / Raw(b"0xa" * 320) ) nbr2 = VppNeighbor( self, self.pg1.sw_if_index, self.pg1.remote_hosts[2].mac, self.pg1.remote_hosts[2].ip4, ).add_vpp_config() # should frag to the path MTU self.send_and_expect(self.pg0, [p_2k], self.pg1, n_rx=3) self.send_and_expect(self.pg0, [p_1k], self.pg1, n_rx=2) # remove and re-add the neighbour nbr2.remove_vpp_config() nbr2.add_vpp_config() # should frag to the path MTU self.send_and_expect(self.pg0, [p_2k], self.pg1, n_rx=3) self.send_and_expect(self.pg0, [p_1k], self.pg1, n_rx=2) # # set PMTUs for many peers # N_HOSTS = 16 self.pg1.generate_remote_hosts(16) self.pg1.configure_ipv4_neighbors() for h in range(N_HOSTS): pmtu = VppIpPathMtu(self, self.pg1.remote_hosts[h].ip4, 900) pmtu.add_vpp_config() self.assertTrue(pmtu.query_vpp_config()) self.logger.info(self.vapi.cli("sh ip pmtu")) dump = list(self.vapi.vpp.details_iter(self.vapi.ip_path_mtu_get)) self.assertEqual(N_HOSTS, len(dump)) for h in range(N_HOSTS): p_2k[IP].dst = self.pg1.remote_hosts[h].ip4 p_1k[IP].dst = self.pg1.remote_hosts[h].ip4 # should frag to the path MTU self.send_and_expect(self.pg0, [p_2k], self.pg1, n_rx=3) self.send_and_expect(self.pg0, [p_1k], self.pg1, n_rx=2) class TestIPv4ItfRebind(VppTestCase): """IPv4 Interface Bind w/ attached routes""" def setUp(self): super(TestIPv4ItfRebind, self).setUp() self.create_pg_interfaces(range(3)) def tearDown(self): super(TestIPv4ItfRebind, self).tearDown() def test_rebind(self): """Import to no import""" TABLE_ID = 1 tbl = VppIpTable(self, TABLE_ID).add_vpp_config() self.pg1.set_table_ip4(TABLE_ID) for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() # add an attached route via an pg0 # in a different table. this prefix should import rt = VppIpRoute( self, self.pg0.local_ip4, 24, [VppRoutePath("0.0.0.0", self.pg0.sw_if_index)], table_id=TABLE_ID, ).add_vpp_config() p = ( Ether(dst=self.pg1.local_mac, src=self.pg1.remote_mac) / IP(src=self.pg1.remote_ip4, dst=self.pg0.remote_ip4) / UDP(sport=1234, dport=5678) / Raw(b"0xa" * 640) ) rx = self.send_and_expect(self.pg1, [p], self.pg0) self.assertFalse(rx[0].haslayer(ARP)) # then bind pg0 to a new table # so the prefix no longer imports self.pg0.unconfig_ip4() self.pg0.set_table_ip4(TABLE_ID) self.pg0.config_ip4() self.pg0.resolve_arp() rx = self.send_and_expect(self.pg1, [p], self.pg0) self.assertFalse(rx[0].haslayer(ARP)) # revert back to imported self.pg0.unconfig_ip4() self.pg0.set_table_ip4(0) self.pg0.config_ip4() self.pg0.resolve_arp() rx = self.send_and_expect(self.pg1, [p], self.pg0) self.assertFalse(rx[0].haslayer(ARP)) # cleanup for i in self.pg_interfaces: i.unconfig_ip4() i.set_table_ip4(0) i.admin_down() rt.remove_vpp_config() tbl.remove_vpp_config() def test_delete(self): """Swap import tables""" TABLE_ID1 = 1 tbl1_4 = VppIpTable(self, TABLE_ID1).add_vpp_config() tbl1_6 = VppIpTable(self, TABLE_ID1, True).add_vpp_config() TABLE_ID2 = 2 tbl2_4 = VppIpTable(self, TABLE_ID2).add_vpp_config() tbl2_6 = VppIpTable(self, TABLE_ID2, True).add_vpp_config() # table mappings self.pg1.set_table_ip4(TABLE_ID1) self.pg1.set_table_ip6(TABLE_ID1) self.pg2.set_table_ip4(TABLE_ID2) self.pg2.set_table_ip6(TABLE_ID2) for i in self.pg_interfaces: i.admin_up() i.config_ip4() i.resolve_arp() # add an attached route in the default table via pg0 # this should import to table 1 rt4 = VppIpRoute( self, self.pg1.local_ip4, 24, [VppRoutePath("0.0.0.0", self.pg1.sw_if_index)], ).add_vpp_config() rt6 = VppIpRoute( self, self.pg1.local_ip6, 64, [VppRoutePath("0.0.0.0", self.pg1.sw_if_index)], ).add_vpp_config() p1 = ( Ether(dst=self.pg0.local_mac, src=self.pg0.remote_mac) / IP(src=self.pg1.remote_ip4, dst=self.pg1.remote_ip4) / UDP(sport=1234, dport=5678) / Raw(b"0xa" * 640) ) # inject into table 0 rx = self.send_and_expect(self.pg0, [p1], self.pg1) self.assertFalse(rx[0].haslayer(ARP)) # swap the attached interface to table 2 self.pg1.unconfig_ip4() self.pg1.unconfig_ip6() self.pg1.set_table_ip4(TABLE_ID2) self.pg1.set_table_ip6(TABLE_ID2) self.pg1.config_ip4() self.pg1.config_ip6() self.pg1.resolve_arp() # delete table 1 tbl1_4.flush() tbl1_6.flush() tbl1_4.remove_vpp_config() tbl1_6.remove_vpp_config() rx = self.send_and_expect(self.pg0, [p1], self.pg1) self.assertFalse(rx[0].haslayer(ARP)) for i in self.pg_interfaces: i.unconfig_ip4() i.unconfig_ip6() i.set_table_ip4(0) i.set_table_ip6(0) i.admin_down() class TestIP4InterfaceRx(VppTestCase): """IPv4 Interface Receive""" @classmethod def setUpClass(cls): super(TestIP4InterfaceRx, cls).setUpClass() @classmethod def tearDownClass(cls): super(TestIP4InterfaceRx, cls).tearDownClass() def setUp(self): super(TestIP4InterfaceRx, self).setUp() self.create_pg_interfaces(range(3)) table_id = 0 for i in self.pg_interfaces: i.admin_up() if table_id != 0: table = VppIpTable(self, table_id) table.add_vpp_config() i.set_table_ip4(table_id) i.config_ip4() i.resolve_arp() table_id += 1 def tearDown(self): for i in self.pg_interfaces: i.unconfig_ip4() i.admin_down() i.set_table_ip4(0) super(TestIP4InterfaceRx, self).tearDown() def test_interface_rx(self): """IPv4 Interface Receive""" # # add a route in the default table to receive ... # route_to_dst = VppIpRoute( self, "1.1.1.0", 24, [ VppRoutePath( "0.0.0.0", self.pg1.sw_if_index, type=FibPathType.FIB_PATH_TYPE_INTERFACE_RX, ) ], ) route_to_dst.add_vpp_config() # # packets to these destination are dropped, since they'll # hit the respective default routes in table 1 # p_dst = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src="5.5.5.5", dst="1.1.1.1") / TCP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) pkts_dst = p_dst * 10 self.send_and_assert_no_replies(self.pg0, pkts_dst, "IP in table 1") # # add a route in the dst table to forward via pg1 # route_in_dst = VppIpRoute( self, "1.1.1.1", 32, [VppRoutePath(self.pg1.remote_ip4, self.pg1.sw_if_index)], table_id=1, ) route_in_dst.add_vpp_config() self.send_and_expect(self.pg0, pkts_dst, self.pg1) # # add a route in the default table to receive ... # route_to_dst = VppIpRoute( self, "1.1.1.0", 24, [ VppRoutePath( "0.0.0.0", self.pg2.sw_if_index, type=FibPathType.FIB_PATH_TYPE_INTERFACE_RX, ) ], table_id=1, ) route_to_dst.add_vpp_config() # # packets to these destination are dropped, since they'll # hit the respective default routes in table 2 # p_dst = ( Ether(src=self.pg0.remote_mac, dst=self.pg0.local_mac) / IP(src="6.6.6.6", dst="1.1.1.2") / TCP(sport=1234, dport=1234) / Raw(b"\xa5" * 100) ) pkts_dst = p_dst * 10 self.send_and_assert_no_replies(self.pg0, pkts_dst, "IP in table 2") # # add a route in the table 2 to forward via pg2 # route_in_dst = VppIpRoute( self, "1.1.1.2", 32, [VppRoutePath(self.pg2.remote_ip4, self.pg2.sw_if_index)], table_id=2, ) route_in_dst.add_vpp_config() self.send_and_expect(self.pg0, pkts_dst, self.pg2) if __name__ == "__main__": unittest.main(testRunner=VppTestRunner)