[RFC PATCH 1/3] selftests: Add VRF icmp error route lookup test

[RFC PATCH 1/3] selftests: Add VRF icmp error route lookup test

[Mathieu Desnoyers]

From: Michael Jeanson <mjeanson@efficios.com>

The objective is to check that the incoming vrf routing table is selected
to send an ICMP error back to the source when the ttl of a packet reaches 1
while it is forwarded between different vrfs.

The first test sends a ping with a ttl of 1 from h1 to h2 and parses the
output of the command to check that a ttl expired error is received.

[This may be flaky, I'm open to suggestions of a more robust approch.]

The second test runs traceroute from h1 to h2 and parses the output to check
for a hop on r1.

Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Cc: David Ahern <dsa@cumulusnetworks.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: netdev@vger.kernel.org
---
 tools/testing/selftests/net/Makefile          |   1 +
 .../selftests/net/vrf_icmp_error_route.sh     | 461 ++++++++++++++++++
 2 files changed, 462 insertions(+)
 create mode 100755 tools/testing/selftests/net/vrf_icmp_error_route.sh

diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index 895ec992b2f1..2fc72bc2908c 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -17,6 +17,7 @@ TEST_PROGS += route_localnet.sh
 TEST_PROGS += reuseaddr_ports_exhausted.sh
 TEST_PROGS += txtimestamp.sh
 TEST_PROGS += vrf-xfrm-tests.sh
+TEST_PROGS += vrf_icmp_error_route.sh
 TEST_PROGS_EXTENDED := in_netns.sh
 TEST_GEN_FILES =  socket nettest
 TEST_GEN_FILES += psock_fanout psock_tpacket msg_zerocopy reuseport_addr_any
diff --git a/tools/testing/selftests/net/vrf_icmp_error_route.sh b/tools/testing/selftests/net/vrf_icmp_error_route.sh
new file mode 100755
index 000000000000..faeedf9f9e1c
--- /dev/null
+++ b/tools/testing/selftests/net/vrf_icmp_error_route.sh
@@ -0,0 +1,461 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# Copyright (c) 2019 David Ahern <dsahern@gmail.com>. All rights reserved.
+# Copyright (c) 2020 Michael Jeanson <mjeanson@efficios.com>. All rights reserved.
+#
+# redirect test
+#
+#                     blue         red
+#                     .253 +----+ .253
+#                     +----| r1 |----+
+#                     |    +----+    |
+# +----+              |              |              +----+
+# | h1 |--------------+              +--------------| h2 |
+# +----+ .1           |              |           .2 +----+
+#         172.16.1/24 |    +----+    | 172.16.2/24
+#    2001:db8:16:1/64 +----| r2 |----+ 2001:db8:16:2/64
+#                     .254 +----+ .254
+#
+#
+# Route from h1 to h2 goes through r1, incoming vrf blue has a route to the
+# outgoing vrf red for the n2 network but red doesn't have a route back to n1.
+# Route from h2 to h1 goes through r2.
+#
+# The objective is to check that the incoming vrf routing table is selected
+# to send an ICMP error back to the source when the ttl of a packet reaches 1
+# while it is forwarded between different vrfs.
+#
+# The first test sends a ping with a ttl of 1 from h1 to h2 and parses the
+# output of the command to check that a ttl expired error is received.
+#
+# The second test runs traceroute from h1 to h2 and parses the output to check
+# for a hop on r1.
+#
+# Requires CONFIG_NET_VRF, CONFIG_VETH, CONFIG_BRIDGE and CONFIG_NET_NS.
+
+VERBOSE=0
+PAUSE_ON_FAIL=no
+
+H1_N1_IP=172.16.1.1
+R1_N1_IP=172.16.1.253
+R2_N1_IP=172.16.1.254
+
+H1_N1_IP6=2001:db8:16:1::1
+R1_N1_IP6=2001:db8:16:1::253
+R2_N1_IP6=2001:db8:16:1::254
+
+H2_N2=172.16.2.0/24
+H2_N2_6=2001:db8:16:2::/64
+
+H2_N2_IP=172.16.2.2
+R1_N2_IP=172.16.2.253
+R2_N2_IP=172.16.2.254
+
+H2_N2_IP6=2001:db8:16:2::2
+R1_N2_IP6=2001:db8:16:2::253
+R2_N2_IP6=2001:db8:16:2::254
+
+################################################################################
+# helpers
+
+log_section()
+{
+	echo
+	echo "###########################################################################"
+	echo "$*"
+	echo "###########################################################################"
+	echo
+}
+
+log_test()
+{
+	local rc=$1
+	local expected=$2
+	local msg="$3"
+
+	if [ "${rc}" -eq "${expected}" ]; then
+		printf "TEST: %-60s  [ OK ]\n" "${msg}"
+		nsuccess=$((nsuccess+1))
+	else
+		ret=1
+		nfail=$((nfail+1))
+		printf "TEST: %-60s  [FAIL]\n" "${msg}"
+		if [ "${PAUSE_ON_FAIL}" = "yes" ]; then
+			echo
+			echo "hit enter to continue, 'q' to quit"
+			read -r a
+			[ "$a" = "q" ] && exit 1
+		fi
+	fi
+}
+
+log_debug()
+{
+	if [ "$VERBOSE" = "1" ]; then
+		echo "$*"
+	fi
+}
+
+run_cmd()
+{
+	local cmd="$*"
+	local out
+	local rc
+
+	if [ "$VERBOSE" = "1" ]; then
+		echo "COMMAND: $cmd"
+	fi
+
+	out=$(eval $cmd 2>&1)
+	rc=$?
+	if [ "$VERBOSE" = "1" ] && [ -n "$out" ]; then
+		echo "$out"
+	fi
+
+	[ "$VERBOSE" = "1" ] && echo
+
+	return $rc
+}
+
+get_linklocal()
+{
+	local ns=$1
+	local dev=$2
+	local addr
+
+	addr=$(ip -netns "$ns" -6 -br addr show dev "${dev}" | \
+	awk '{
+		for (i = 3; i <= NF; ++i) {
+			if ($i ~ /^fe80/)
+				print $i
+		}
+	}'
+	)
+	addr=${addr/\/*}
+
+	[ -z "$addr" ] && return 1
+
+	echo "$addr"
+
+	return 0
+}
+
+################################################################################
+# setup and teardown
+
+cleanup()
+{
+	local ns
+
+	setup=0
+
+	for ns in h1 h2 r1 r2; do
+		ip netns del $ns 2>/dev/null
+	done
+}
+
+setup_vrf()
+{
+	local ns=$1
+
+	ip -netns "${ns}" ru del pref 0
+	ip -netns "${ns}" ru add pref 32765 from all lookup local
+	ip -netns "${ns}" -6 ru del pref 0
+	ip -netns "${ns}" -6 ru add pref 32765 from all lookup local
+}
+
+create_vrf()
+{
+	local ns=$1
+	local vrf=$2
+	local table=$3
+
+	ip -netns "${ns}" link add "${vrf}" type vrf table "${table}"
+	ip -netns "${ns}" link set "${vrf}" up
+	ip -netns "${ns}" route add vrf "${vrf}" unreachable default metric 8192
+	ip -netns "${ns}" -6 route add vrf "${vrf}" unreachable default metric 8192
+
+	ip -netns "${ns}" addr add 127.0.0.1/8 dev "${vrf}"
+	ip -netns "${ns}" -6 addr add ::1 dev "${vrf}" nodad
+}
+
+setup()
+{
+	local ns
+
+	if [ "${setup}" -eq 1 ]; then
+		return 0
+	fi
+
+	# make sure we are starting with a clean slate
+	cleanup
+
+	setup=1
+
+	#
+	# create nodes as namespaces
+	#
+	for ns in h1 h2 r1 r2; do
+		ip netns add $ns
+		ip -netns $ns li set lo up
+
+		case "${ns}" in
+		h[12]) ip netns exec $ns sysctl -q -w net.ipv6.conf.all.forwarding=0
+		       ip netns exec $ns sysctl -q -w net.ipv6.conf.all.keep_addr_on_down=1
+			;;
+		r[12]) ip netns exec $ns sysctl -q -w net.ipv4.ip_forward=1
+		       ip netns exec $ns sysctl -q -w net.ipv6.conf.all.forwarding=1
+		esac
+	done
+
+	#
+	# create interconnects
+	#
+	ip -netns h1 li add eth0 type veth peer name r1h1
+	ip -netns h1 li set r1h1 netns r1 name eth0 up
+
+	ip -netns h1 li add eth1 type veth peer name r2h1
+	ip -netns h1 li set r2h1 netns r2 name eth0 up
+
+	ip -netns h2 li add eth0 type veth peer name r1h2
+	ip -netns h2 li set r1h2 netns r1 name eth1 up
+
+	ip -netns h2 li add eth1 type veth peer name r2h2
+	ip -netns h2 li set r2h2 netns r2 name eth1 up
+
+	#
+	# h1
+	#
+	ip -netns h1 li add br0 type bridge
+	ip -netns h1 li set br0 up
+	ip -netns h1 addr add dev br0 ${H1_N1_IP}/24
+	ip -netns h1 -6 addr add dev br0 ${H1_N1_IP6}/64 nodad
+	ip -netns h1 li set eth0 master br0 up
+	ip -netns h1 li set eth1 master br0 up
+
+	# h1 to h2 via r1
+	ip -netns h1    ro add ${H2_N2} via ${R1_N1_IP} dev br0
+	ip -netns h1 -6 ro add ${H2_N2_6} via "${R1_N1_IP6}" dev br0
+
+	#
+	# h2
+	#
+	ip -netns h2 li add br0 type bridge
+	ip -netns h2 li set br0 up
+	ip -netns h2 addr add dev br0 ${H2_N2_IP}/24
+	ip -netns h2 -6 addr add dev br0 ${H2_N2_IP6}/64 nodad
+	ip -netns h2 li set eth0 master br0 up
+	ip -netns h2 li set eth1 master br0 up
+
+	ip -netns h2 ro add default via ${R2_N2_IP} dev br0
+	ip -netns h2 -6 ro add default via ${R2_N2_IP6} dev br0
+
+	#
+	# r1
+	#
+	setup_vrf r1
+	create_vrf r1 blue 1101
+	create_vrf r1 red 1102
+	ip -netns r1 li set eth0 vrf blue up
+	ip -netns r1 li set eth1 vrf red up
+	ip -netns r1 addr add dev eth0 ${R1_N1_IP}/24
+	ip -netns r1 -6 addr add dev eth0 ${R1_N1_IP6}/64 nodad
+	ip -netns r1 addr add dev eth1 ${R1_N2_IP}/24
+	ip -netns r1 -6 addr add dev eth1 ${R1_N2_IP6}/64 nodad
+
+	# Route leak from blue to red
+	ip -netns r1 route add vrf blue ${H2_N2} dev red
+	ip -netns r1 -6 route add vrf blue ${H2_N2_6} dev red
+
+	#
+	# r2
+	#
+	ip -netns r2 addr add dev eth0 ${R2_N1_IP}/24
+	ip -netns r2 -6 addr add dev eth0 ${R2_N1_IP6}/64 nodad
+	ip -netns r2 addr add dev eth1 ${R2_N2_IP}/24
+	ip -netns r2 -6 addr add dev eth1 ${R2_N2_IP6}/64 nodad
+
+	# Wait for ip config to settle
+	sleep 2
+}
+
+check_connectivity4()
+{
+	ip netns exec h1 ping -c1 -w1 ${H2_N2_IP} >/dev/null 2>&1
+}
+
+check_connectivity6()
+{
+	ip netns exec h1 "${ping6}" -c1 -w1 ${H2_N2_IP6} >/dev/null 2>&1
+}
+
+ipv4_traceroute()
+{
+	log_section "IPv4: VRF ICMP error route lookup traceroute"
+
+	if [ ! -x "$(command -v traceroute)" ]; then
+		echo "SKIP: Could not run IPV4 test without traceroute"
+		return
+	fi
+
+	setup
+
+	# verify connectivity
+	if ! check_connectivity4; then
+		echo "Error: Basic connectivity is broken"
+		ret=1
+		return
+	fi
+
+	if [ "$VERBOSE" = "1" ]; then
+		run_cmd ip netns exec h1 traceroute ${H2_N2_IP}
+	fi
+
+	ip netns exec h1 traceroute ${H2_N2_IP} | grep -q "${R1_N1_IP}"
+	log_test $? 0 "Traceroute reports a hop on r1"
+}
+
+ipv6_traceroute()
+{
+	log_section "IPv6: VRF ICMP error route lookup traceroute"
+
+	if [ ! -x "$(command -v traceroute6)" ]; then
+		echo "SKIP: Could not run IPV6 test without traceroute6"
+		return
+	fi
+
+	setup
+
+	# verify connectivity
+	if ! check_connectivity6; then
+		echo "Error: Basic connectivity is broken"
+		ret=1
+		return
+	fi
+
+	if [ "$VERBOSE" = "1" ]; then
+		run_cmd ip netns exec h1 traceroute6 ${H2_N2_IP6}
+	fi
+
+	ip netns exec h1 traceroute6 ${H2_N2_IP6} | grep -q "${R1_N1_IP6}"
+	log_test $? 0 "Traceroute6 reports a hop on r1"
+}
+
+ipv4_ping()
+{
+	log_section "IPv4: VRF ICMP error route lookup ping"
+
+	setup
+
+	# verify connectivity
+	if ! check_connectivity4; then
+		echo "Error: Basic connectivity is broken"
+		ret=1
+		return
+	fi
+
+	if [ "$VERBOSE" = "1" ]; then
+		echo "Command to check for ICMP ttl exceeded:"
+		run_cmd ip netns exec h1 ping -t1 -c1 -W2 ${H2_N2_IP}
+	fi
+
+	ip netns exec h1 ping -t1 -c1 -W2 ${H2_N2_IP} | grep -q "Time to live exceeded"
+	log_test $? 0 "Ping received ICMP ttl exceeded"
+}
+
+ipv6_ping()
+{
+	log_section "IPv6: VRF ICMP error route lookup ping"
+
+	setup
+
+	# verify connectivity
+	if ! check_connectivity6; then
+		echo "Error: Basic connectivity is broken"
+		ret=1
+		return
+	fi
+
+	if [ "$VERBOSE" = "1" ]; then
+		echo "Command to check for ICMP ttl exceeded:"
+		run_cmd ip netns exec h1 "${ping6}" -t1 -c1 -W2 ${H2_N2_IP6}
+	fi
+
+	ip netns exec h1 "${ping6}" -t1 -c1 -W2 ${H2_N2_IP6} | grep -q "Time exceeded: Hop limit"
+	log_test $? 0 "Ping received ICMP ttl exceeded"
+}
+################################################################################
+# usage
+
+usage()
+{
+        cat <<EOF
+usage: ${0##*/} OPTS
+
+	-4          IPv4 tests only
+	-6          IPv6 tests only
+	-p          Pause on fail
+	-v          verbose mode (show commands and output)
+EOF
+}
+
+################################################################################
+# main
+
+# Some systems don't have a ping6 binary anymore
+command -v ping6 > /dev/null 2>&1 && ping6=$(command -v ping6) || ping6=$(command -v ping)
+
+TESTS_IPV4="ipv4_ping ipv4_traceroute"
+TESTS_IPV6="ipv6_ping ipv6_traceroute"
+
+ret=0
+nsuccess=0
+nfail=0
+setup=0
+
+while getopts :46pvh o
+do
+	case $o in
+		4) TESTS=ipv4;;
+		6) TESTS=ipv6;;
+                p) PAUSE_ON_FAIL=yes;;
+                v) VERBOSE=1;;
+		h) usage; exit 0;;
+                *) usage; exit 1;;
+	esac
+done
+
+#
+# show user test config
+#
+if [ -z "$TESTS" ]; then
+        TESTS="$TESTS_IPV4 $TESTS_IPV6"
+elif [ "$TESTS" = "ipv4" ]; then
+        TESTS="$TESTS_IPV4"
+elif [ "$TESTS" = "ipv6" ]; then
+        TESTS="$TESTS_IPV6"
+fi
+
+for t in $TESTS
+do
+	case $t in
+	ipv4_ping|ping)             ipv4_ping;;
+	ipv4_traceroute|traceroute) ipv4_traceroute;;
+
+	ipv6_ping|ping)             ipv6_ping;;
+	ipv6_traceroute|traceroute) ipv6_traceroute;;
+
+	# setup namespaces and config, but do not run any tests
+	setup)                      setup; exit 0;;
+
+	help)                       echo "Test names: $TESTS"; exit 0;;
+	esac
+done
+
+cleanup
+
+printf "\nTests passed: %3d\n" ${nsuccess}
+printf "Tests failed: %3d\n"   ${nfail}
+
+exit $ret
-- 
2.17.1

[RFC PATCH 2/3] ipv4/icmp: l3mdev: Perform icmp error route lookup on source device routing table

[Mathieu Desnoyers]

As per RFC792, ICMP errors should be sent to the source host.

However, in configurations with Virtual Routing and Forwarding tables,
looking up which routing table to use is currently done by using the
destination net_device.

commit 9d1a6c4ea43e ("net: icmp_route_lookup should use rt dev to
determine L3 domain") changes the interface passed to
l3mdev_master_ifindex() and inet_addr_type_dev_table() from skb_in->dev
to skb_dst(skb_in)->dev. This effectively uses the destination device
rather than the source device for choosing which routing table should be
used to lookup where to send the ICMP error.

Therefore, if the source and destination interfaces are within separate
VRFs, or one in the global routing table and the other in a VRF, looking
up the source host in the destination interface's routing table will
fail if the destination interface's routing table contains no route to
the source host.

One observable effect of this issue is that traceroute does not work in
the following cases:

- Route leaking between global routing table and VRF
- Route leaking between VRFs

Preferably use the source device routing table when sending ICMPv6 error
messages. If no source device is set, fall-back on the destination
device routing table.

Fixes: 9d1a6c4ea43e ("net: icmp_route_lookup should use rt dev to determine L3 domain")
Link: https://tools.ietf.org/html/rfc792
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: David Ahern <dsa@cumulusnetworks.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: netdev@vger.kernel.org
---
 net/ipv4/icmp.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index e30515f89802..029e12d35b19 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -465,6 +465,7 @@ static struct rtable *icmp_route_lookup(struct net *net,
 					int type, int code,
 					struct icmp_bxm *param)
 {
+	struct net_device *route_lookup_dev = NULL;
 	struct rtable *rt, *rt2;
 	struct flowi4 fl4_dec;
 	int err;
@@ -479,7 +480,17 @@ static struct rtable *icmp_route_lookup(struct net *net,
 	fl4->flowi4_proto = IPPROTO_ICMP;
 	fl4->fl4_icmp_type = type;
 	fl4->fl4_icmp_code = code;
-	fl4->flowi4_oif = l3mdev_master_ifindex(skb_dst(skb_in)->dev);
+	/*
+	 * The device used for looking up which routing table to use is
+	 * preferably the source whenever it is set, which should ensure
+	 * the icmp error can be sent to the source host, else fallback
+	 * on the destination device.
+	 */
+	if (skb_in->dev)
+		route_lookup_dev = skb_in->dev;
+	else if (skb_dst(skb_in))
+		route_lookup_dev = skb_dst(skb_in)->dev;
+	fl4->flowi4_oif = l3mdev_master_ifindex(route_lookup_dev);
 
 	security_skb_classify_flow(skb_in, flowi4_to_flowi(fl4));
 	rt = ip_route_output_key_hash(net, fl4, skb_in);
@@ -503,7 +514,7 @@ static struct rtable *icmp_route_lookup(struct net *net,
 	if (err)
 		goto relookup_failed;
 
-	if (inet_addr_type_dev_table(net, skb_dst(skb_in)->dev,
+	if (inet_addr_type_dev_table(net, route_lookup_dev,
 				     fl4_dec.saddr) == RTN_LOCAL) {
 		rt2 = __ip_route_output_key(net, &fl4_dec);
 		if (IS_ERR(rt2))
-- 
2.17.1

[RFC PATCH 3/3] ipv6/icmp: l3mdev: Perform icmp error route lookup on source device routing table

[Mathieu Desnoyers]

As per RFC4443, the destination address field for ICMPv6 error messages
is copied from the source address field of the invoking packet.

In configurations with Virtual Routing and Forwarding tables, looking up
which routing table to use for sending ICMPv6 error messages is
currently done by using the destination net_device.

If the source and destination interfaces are within separate VRFs, or
one in the global routing table and the other in a VRF, looking up the
source address of the invoking packet in the destination interface's
routing table will fail if the destination interface's routing table
contains no route to the invoking packet's source address.

One observable effect of this issue is that traceroute6 does not work in
the following cases:

- Route leaking between global routing table and VRF
- Route leaking between VRFs

Preferably use the source device routing table when sending ICMPv6 error
messages. If no source device is set, fall-back on the destination
device routing table.

Link: https://tools.ietf.org/html/rfc4443
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: David Ahern <dsa@cumulusnetworks.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: netdev@vger.kernel.org
---
 net/ipv6/icmp.c       | 15 +++++++++++++--
 net/ipv6/ip6_output.c |  2 --
 2 files changed, 13 insertions(+), 4 deletions(-)

diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 9df8737ae0d3..beee93065688 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -501,8 +501,19 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info,
 	if (__ipv6_addr_needs_scope_id(addr_type)) {
 		iif = icmp6_iif(skb);
 	} else {
-		dst = skb_dst(skb);
-		iif = l3mdev_master_ifindex(dst ? dst->dev : skb->dev);
+		struct net_device *route_lookup_dev = NULL;
+
+		/*
+		 * The device used for looking up which routing table to use is
+		 * preferably the source whenever it is set, which should
+		 * ensure the icmp error can be sent to the source host, else
+		 * fallback on the destination device.
+		 */
+		if (skb->dev)
+			route_lookup_dev = skb->dev;
+		else if (skb_dst(skb))
+			route_lookup_dev = skb_dst(skb)->dev;
+		iif = l3mdev_master_ifindex(route_lookup_dev);
 	}
 
 	/*
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 8a8c2d0cfcc8..00243d7d276c 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -468,8 +468,6 @@ int ip6_forward(struct sk_buff *skb)
 	 *	check and decrement ttl
 	 */
 	if (hdr->hop_limit <= 1) {
-		/* Force OUTPUT device used as source address */
-		skb->dev = dst->dev;
 		icmpv6_send(skb, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, 0);
 		__IP6_INC_STATS(net, idev, IPSTATS_MIB_INHDRERRORS);
 
-- 
2.17.1

Re: [RFC PATCH 1/3] selftests: Add VRF icmp error route lookup test

[David Ahern]

On 7/29/20 3:12 PM, Mathieu Desnoyers wrote:
> From: Michael Jeanson <mjeanson@efficios.com>
> 
> The objective is to check that the incoming vrf routing table is selected
> to send an ICMP error back to the source when the ttl of a packet reaches 1
> while it is forwarded between different vrfs.
> 
> The first test sends a ping with a ttl of 1 from h1 to h2 and parses the
> output of the command to check that a ttl expired error is received.
> 
> [This may be flaky, I'm open to suggestions of a more robust approch.]
> 
> The second test runs traceroute from h1 to h2 and parses the output to check
> for a hop on r1.
> 
> Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
> Cc: David Ahern <dsa@cumulusnetworks.com>

Update the address to dsahern@kernel.org


> Cc: David S. Miller <davem@davemloft.net>
> Cc: netdev@vger.kernel.org
> ---
>  tools/testing/selftests/net/Makefile          |   1 +
>  .../selftests/net/vrf_icmp_error_route.sh     | 461 ++++++++++++++++++
>  2 files changed, 462 insertions(+)
>  create mode 100755 tools/testing/selftests/net/vrf_icmp_error_route.sh
> 

Test seems fine to me. you copied icmp_redirect.sh which is fine but
please clean up comments and functions not needed for this test.

[PATCH] selftests: Add VRF icmp error route lookup test

[Michael Jeanson]

The objective is to check that the incoming vrf routing table is selected
to send an ICMP error back to the source when the ttl of a packet reaches 1
while it is forwarded between different vrfs.

The first test sends a ping with a ttl of 1 from h1 to h2 and parses the
output of the command to check that a ttl expired error is received.

[This may be flaky, I'm open to suggestions of a more robust approch.]

The second test runs traceroute from h1 to h2 and parses the output to
check for a hop on r1.

Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
Cc: David Ahern <dsahern@kernel.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: netdev@vger.kernel.org
---
 tools/testing/selftests/net/Makefile          |   1 +
 .../selftests/net/vrf_icmp_error_route.sh     | 429 ++++++++++++++++++
 2 files changed, 430 insertions(+)
 create mode 100755 tools/testing/selftests/net/vrf_icmp_error_route.sh

diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index 895ec992b2f1..2fc72bc2908c 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -17,6 +17,7 @@ TEST_PROGS += route_localnet.sh
 TEST_PROGS += reuseaddr_ports_exhausted.sh
 TEST_PROGS += txtimestamp.sh
 TEST_PROGS += vrf-xfrm-tests.sh
+TEST_PROGS += vrf_icmp_error_route.sh
 TEST_PROGS_EXTENDED := in_netns.sh
 TEST_GEN_FILES =  socket nettest
 TEST_GEN_FILES += psock_fanout psock_tpacket msg_zerocopy reuseport_addr_any
diff --git a/tools/testing/selftests/net/vrf_icmp_error_route.sh b/tools/testing/selftests/net/vrf_icmp_error_route.sh
new file mode 100755
index 000000000000..0b15a886bf5b
--- /dev/null
+++ b/tools/testing/selftests/net/vrf_icmp_error_route.sh
@@ -0,0 +1,429 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+#
+# Copyright (c) 2019 David Ahern <dsahern@gmail.com>. All rights reserved.
+# Copyright (c) 2020 Michael Jeanson <mjeanson@efficios.com>. All rights reserved.
+#
+#                     blue         red
+#                     .253 +----+ .253
+#                     +----| r1 |----+
+#                     |    +----+    |
+# +----+              |              |              +----+
+# | h1 |--------------+              +--------------| h2 |
+# +----+ .1           |              |           .2 +----+
+#         172.16.1/24 |    +----+    | 172.16.2/24
+#    2001:db8:16:1/64 +----| r2 |----+ 2001:db8:16:2/64
+#                     .254 +----+ .254
+#
+#
+# Route from h1 to h2 goes through r1, incoming vrf blue has a route to the
+# outgoing vrf red for the n2 network but red doesn't have a route back to n1.
+# Route from h2 to h1 goes through r2.
+#
+# The objective is to check that the incoming vrf routing table is selected
+# to send an ICMP error back to the source when the ttl of a packet reaches 1
+# while it is forwarded between different vrfs.
+#
+# The first test sends a ping with a ttl of 1 from h1 to h2 and parses the
+# output of the command to check that a ttl expired error is received.
+#
+# The second test runs traceroute from h1 to h2 and parses the output to check
+# for a hop on r1.
+#
+# Requires CONFIG_NET_VRF, CONFIG_VETH, CONFIG_BRIDGE and CONFIG_NET_NS.
+
+VERBOSE=0
+PAUSE_ON_FAIL=no
+
+H1_N1_IP=172.16.1.1
+R1_N1_IP=172.16.1.253
+R2_N1_IP=172.16.1.254
+
+H1_N1_IP6=2001:db8:16:1::1
+R1_N1_IP6=2001:db8:16:1::253
+R2_N1_IP6=2001:db8:16:1::254
+
+H2_N2=172.16.2.0/24
+H2_N2_6=2001:db8:16:2::/64
+
+H2_N2_IP=172.16.2.2
+R1_N2_IP=172.16.2.253
+R2_N2_IP=172.16.2.254
+
+H2_N2_IP6=2001:db8:16:2::2
+R1_N2_IP6=2001:db8:16:2::253
+R2_N2_IP6=2001:db8:16:2::254
+
+################################################################################
+# helpers
+
+log_section()
+{
+	echo
+	echo "###########################################################################"
+	echo "$*"
+	echo "###########################################################################"
+	echo
+}
+
+log_test()
+{
+	local rc=$1
+	local expected=$2
+	local msg="$3"
+
+	if [ "${rc}" -eq "${expected}" ]; then
+		printf "TEST: %-60s  [ OK ]\n" "${msg}"
+		nsuccess=$((nsuccess+1))
+	else
+		ret=1
+		nfail=$((nfail+1))
+		printf "TEST: %-60s  [FAIL]\n" "${msg}"
+		if [ "${PAUSE_ON_FAIL}" = "yes" ]; then
+			echo
+			echo "hit enter to continue, 'q' to quit"
+			read -r a
+			[ "$a" = "q" ] && exit 1
+		fi
+	fi
+}
+
+run_cmd()
+{
+	local cmd="$*"
+	local out
+	local rc
+
+	if [ "$VERBOSE" = "1" ]; then
+		echo "COMMAND: $cmd"
+	fi
+
+	out=$(eval $cmd 2>&1)
+	rc=$?
+	if [ "$VERBOSE" = "1" ] && [ -n "$out" ]; then
+		echo "$out"
+	fi
+
+	[ "$VERBOSE" = "1" ] && echo
+
+	return $rc
+}
+
+################################################################################
+# setup and teardown
+
+cleanup()
+{
+	local ns
+
+	setup=0
+
+	for ns in h1 h2 r1 r2; do
+		ip netns del $ns 2>/dev/null
+	done
+}
+
+setup_vrf()
+{
+	local ns=$1
+
+	ip -netns "${ns}" ru del pref 0
+	ip -netns "${ns}" ru add pref 32765 from all lookup local
+	ip -netns "${ns}" -6 ru del pref 0
+	ip -netns "${ns}" -6 ru add pref 32765 from all lookup local
+}
+
+create_vrf()
+{
+	local ns=$1
+	local vrf=$2
+	local table=$3
+
+	ip -netns "${ns}" link add "${vrf}" type vrf table "${table}"
+	ip -netns "${ns}" link set "${vrf}" up
+	ip -netns "${ns}" route add vrf "${vrf}" unreachable default metric 8192
+	ip -netns "${ns}" -6 route add vrf "${vrf}" unreachable default metric 8192
+
+	ip -netns "${ns}" addr add 127.0.0.1/8 dev "${vrf}"
+	ip -netns "${ns}" -6 addr add ::1 dev "${vrf}" nodad
+}
+
+setup()
+{
+	local ns
+
+	if [ "${setup}" -eq 1 ]; then
+		return 0
+	fi
+
+	# make sure we are starting with a clean slate
+	cleanup
+
+	setup=1
+
+	#
+	# create nodes as namespaces
+	#
+	for ns in h1 h2 r1 r2; do
+		ip netns add $ns
+		ip -netns $ns li set lo up
+
+		case "${ns}" in
+		h[12]) ip netns exec $ns sysctl -q -w net.ipv6.conf.all.forwarding=0
+		       ip netns exec $ns sysctl -q -w net.ipv6.conf.all.keep_addr_on_down=1
+			;;
+		r[12]) ip netns exec $ns sysctl -q -w net.ipv4.ip_forward=1
+		       ip netns exec $ns sysctl -q -w net.ipv6.conf.all.forwarding=1
+		esac
+	done
+
+	#
+	# create interconnects
+	#
+	ip -netns h1 li add eth0 type veth peer name r1h1
+	ip -netns h1 li set r1h1 netns r1 name eth0 up
+
+	ip -netns h1 li add eth1 type veth peer name r2h1
+	ip -netns h1 li set r2h1 netns r2 name eth0 up
+
+	ip -netns h2 li add eth0 type veth peer name r1h2
+	ip -netns h2 li set r1h2 netns r1 name eth1 up
+
+	ip -netns h2 li add eth1 type veth peer name r2h2
+	ip -netns h2 li set r2h2 netns r2 name eth1 up
+
+	#
+	# h1
+	#
+	ip -netns h1 li add br0 type bridge
+	ip -netns h1 li set br0 up
+	ip -netns h1 addr add dev br0 ${H1_N1_IP}/24
+	ip -netns h1 -6 addr add dev br0 ${H1_N1_IP6}/64 nodad
+	ip -netns h1 li set eth0 master br0 up
+	ip -netns h1 li set eth1 master br0 up
+
+	# h1 to h2 via r1
+	ip -netns h1    ro add ${H2_N2} via ${R1_N1_IP} dev br0
+	ip -netns h1 -6 ro add ${H2_N2_6} via "${R1_N1_IP6}" dev br0
+
+	#
+	# h2
+	#
+	ip -netns h2 li add br0 type bridge
+	ip -netns h2 li set br0 up
+	ip -netns h2 addr add dev br0 ${H2_N2_IP}/24
+	ip -netns h2 -6 addr add dev br0 ${H2_N2_IP6}/64 nodad
+	ip -netns h2 li set eth0 master br0 up
+	ip -netns h2 li set eth1 master br0 up
+
+	ip -netns h2 ro add default via ${R2_N2_IP} dev br0
+	ip -netns h2 -6 ro add default via ${R2_N2_IP6} dev br0
+
+	#
+	# r1
+	#
+	setup_vrf r1
+	create_vrf r1 blue 1101
+	create_vrf r1 red 1102
+	ip -netns r1 li set eth0 vrf blue up
+	ip -netns r1 li set eth1 vrf red up
+	ip -netns r1 addr add dev eth0 ${R1_N1_IP}/24
+	ip -netns r1 -6 addr add dev eth0 ${R1_N1_IP6}/64 nodad
+	ip -netns r1 addr add dev eth1 ${R1_N2_IP}/24
+	ip -netns r1 -6 addr add dev eth1 ${R1_N2_IP6}/64 nodad
+
+	# Route leak from blue to red
+	ip -netns r1 route add vrf blue ${H2_N2} dev red
+	ip -netns r1 -6 route add vrf blue ${H2_N2_6} dev red
+
+	#
+	# r2
+	#
+	ip -netns r2 addr add dev eth0 ${R2_N1_IP}/24
+	ip -netns r2 -6 addr add dev eth0 ${R2_N1_IP6}/64 nodad
+	ip -netns r2 addr add dev eth1 ${R2_N2_IP}/24
+	ip -netns r2 -6 addr add dev eth1 ${R2_N2_IP6}/64 nodad
+
+	# Wait for ip config to settle
+	sleep 2
+}
+
+check_connectivity4()
+{
+	ip netns exec h1 ping -c1 -w1 ${H2_N2_IP} >/dev/null 2>&1
+}
+
+check_connectivity6()
+{
+	ip netns exec h1 "${ping6}" -c1 -w1 ${H2_N2_IP6} >/dev/null 2>&1
+}
+
+ipv4_traceroute()
+{
+	log_section "IPv4: VRF ICMP error route lookup traceroute"
+
+	if [ ! -x "$(command -v traceroute)" ]; then
+		echo "SKIP: Could not run IPV4 test without traceroute"
+		return
+	fi
+
+	setup
+
+	# verify connectivity
+	if ! check_connectivity4; then
+		echo "Error: Basic connectivity is broken"
+		ret=1
+		return
+	fi
+
+	if [ "$VERBOSE" = "1" ]; then
+		run_cmd ip netns exec h1 traceroute ${H2_N2_IP}
+	fi
+
+	ip netns exec h1 traceroute ${H2_N2_IP} | grep -q "${R1_N1_IP}"
+	log_test $? 0 "Traceroute reports a hop on r1"
+}
+
+ipv6_traceroute()
+{
+	log_section "IPv6: VRF ICMP error route lookup traceroute"
+
+	if [ ! -x "$(command -v traceroute6)" ]; then
+		echo "SKIP: Could not run IPV6 test without traceroute6"
+		return
+	fi
+
+	setup
+
+	# verify connectivity
+	if ! check_connectivity6; then
+		echo "Error: Basic connectivity is broken"
+		ret=1
+		return
+	fi
+
+	if [ "$VERBOSE" = "1" ]; then
+		run_cmd ip netns exec h1 traceroute6 ${H2_N2_IP6}
+	fi
+
+	ip netns exec h1 traceroute6 ${H2_N2_IP6} | grep -q "${R1_N1_IP6}"
+	log_test $? 0 "Traceroute6 reports a hop on r1"
+}
+
+ipv4_ping()
+{
+	log_section "IPv4: VRF ICMP error route lookup ping"
+
+	setup
+
+	# verify connectivity
+	if ! check_connectivity4; then
+		echo "Error: Basic connectivity is broken"
+		ret=1
+		return
+	fi
+
+	if [ "$VERBOSE" = "1" ]; then
+		echo "Command to check for ICMP ttl exceeded:"
+		run_cmd ip netns exec h1 ping -t1 -c1 -W2 ${H2_N2_IP}
+	fi
+
+	ip netns exec h1 ping -t1 -c1 -W2 ${H2_N2_IP} | grep -q "Time to live exceeded"
+	log_test $? 0 "Ping received ICMP ttl exceeded"
+}
+
+ipv6_ping()
+{
+	log_section "IPv6: VRF ICMP error route lookup ping"
+
+	setup
+
+	# verify connectivity
+	if ! check_connectivity6; then
+		echo "Error: Basic connectivity is broken"
+		ret=1
+		return
+	fi
+
+	if [ "$VERBOSE" = "1" ]; then
+		echo "Command to check for ICMP ttl exceeded:"
+		run_cmd ip netns exec h1 "${ping6}" -t1 -c1 -W2 ${H2_N2_IP6}
+	fi
+
+	ip netns exec h1 "${ping6}" -t1 -c1 -W2 ${H2_N2_IP6} | grep -q "Time exceeded: Hop limit"
+	log_test $? 0 "Ping received ICMP ttl exceeded"
+}
+################################################################################
+# usage
+
+usage()
+{
+        cat <<EOF
+usage: ${0##*/} OPTS
+
+	-4          IPv4 tests only
+	-6          IPv6 tests only
+	-p          Pause on fail
+	-v          verbose mode (show commands and output)
+EOF
+}
+
+################################################################################
+# main
+
+# Some systems don't have a ping6 binary anymore
+command -v ping6 > /dev/null 2>&1 && ping6=$(command -v ping6) || ping6=$(command -v ping)
+
+TESTS_IPV4="ipv4_ping ipv4_traceroute"
+TESTS_IPV6="ipv6_ping ipv6_traceroute"
+
+ret=0
+nsuccess=0
+nfail=0
+setup=0
+
+while getopts :46pvh o
+do
+	case $o in
+		4) TESTS=ipv4;;
+		6) TESTS=ipv6;;
+                p) PAUSE_ON_FAIL=yes;;
+                v) VERBOSE=1;;
+		h) usage; exit 0;;
+                *) usage; exit 1;;
+	esac
+done
+
+#
+# show user test config
+#
+if [ -z "$TESTS" ]; then
+        TESTS="$TESTS_IPV4 $TESTS_IPV6"
+elif [ "$TESTS" = "ipv4" ]; then
+        TESTS="$TESTS_IPV4"
+elif [ "$TESTS" = "ipv6" ]; then
+        TESTS="$TESTS_IPV6"
+fi
+
+for t in $TESTS
+do
+	case $t in
+	ipv4_ping|ping)             ipv4_ping;;
+	ipv4_traceroute|traceroute) ipv4_traceroute;;
+
+	ipv6_ping|ping)             ipv6_ping;;
+	ipv6_traceroute|traceroute) ipv6_traceroute;;
+
+	# setup namespaces and config, but do not run any tests
+	setup)                      setup; exit 0;;
+
+	help)                       echo "Test names: $TESTS"; exit 0;;
+	esac
+done
+
+cleanup
+
+printf "\nTests passed: %3d\n" ${nsuccess}
+printf "Tests failed: %3d\n"   ${nfail}
+
+exit $ret
-- 
2.27.0

Re: [PATCH] selftests: Add VRF icmp error route lookup test

[David Miller]

From: Michael Jeanson <mjeanson@efficios.com>
Date: Thu,  6 Aug 2020 14:51:21 -0400

> The objective is to check that the incoming vrf routing table is selected
> to send an ICMP error back to the source when the ttl of a packet reaches 1
> while it is forwarded between different vrfs.
> 
> The first test sends a ping with a ttl of 1 from h1 to h2 and parses the
> output of the command to check that a ttl expired error is received.
> 
> [This may be flaky, I'm open to suggestions of a more robust approch.]
> 
> The second test runs traceroute from h1 to h2 and parses the output to
> check for a hop on r1.
> 
> Signed-off-by: Michael Jeanson <mjeanson@efficios.com>

This patch does not apply cleanly to the current net tree.

Re: [PATCH] selftests: Add VRF icmp error route lookup test

[David Ahern]

On 8/11/20 11:28 AM, David Miller wrote:
> From: Michael Jeanson <mjeanson@efficios.com>
> Date: Thu,  6 Aug 2020 14:51:21 -0400
> 
>> The objective is to check that the incoming vrf routing table is selected
>> to send an ICMP error back to the source when the ttl of a packet reaches 1
>> while it is forwarded between different vrfs.
>>
>> The first test sends a ping with a ttl of 1 from h1 to h2 and parses the
>> output of the command to check that a ttl expired error is received.
>>
>> [This may be flaky, I'm open to suggestions of a more robust approch.]
>>
>> The second test runs traceroute from h1 to h2 and parses the output to
>> check for a hop on r1.
>>
>> Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
> 
> This patch does not apply cleanly to the current net tree.
> 

It is also out of context since the tests fail on current net and net-next.

The tests along with the patches that fix the problem should be sent
together.

Re: [PATCH] selftests: Add VRF icmp error route lookup test

[Mathieu Desnoyers]

----- On Aug 11, 2020, at 2:57 PM, David Ahern dsahern@gmail.com wrote:

> On 8/11/20 11:28 AM, David Miller wrote:
>> From: Michael Jeanson <mjeanson@efficios.com>
>> Date: Thu,  6 Aug 2020 14:51:21 -0400
>> 
>>> The objective is to check that the incoming vrf routing table is selected
>>> to send an ICMP error back to the source when the ttl of a packet reaches 1
>>> while it is forwarded between different vrfs.
>>>
>>> The first test sends a ping with a ttl of 1 from h1 to h2 and parses the
>>> output of the command to check that a ttl expired error is received.
>>>
>>> [This may be flaky, I'm open to suggestions of a more robust approch.]
>>>
>>> The second test runs traceroute from h1 to h2 and parses the output to
>>> check for a hop on r1.
>>>
>>> Signed-off-by: Michael Jeanson <mjeanson@efficios.com>
>> 
>> This patch does not apply cleanly to the current net tree.
>> 
> 
> It is also out of context since the tests fail on current net and net-next.
> 
> The tests along with the patches that fix the problem should be sent
> together.

One thing I am missing before this series can be considered for upstreaming
is an Acked-by of the 2 fixes for ipv4 and ipv6 from you, as maintainer
of l3mdev, if you think the approach I am taking with those fixes makes sense.

Thanks,

Mathieu

-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com

Re: [PATCH] selftests: Add VRF icmp error route lookup test

[David Ahern]

On 8/11/20 1:11 PM, Mathieu Desnoyers wrote:
> One thing I am missing before this series can be considered for upstreaming
> is an Acked-by of the 2 fixes for ipv4 and ipv6 from you, as maintainer
> of l3mdev, if you think the approach I am taking with those fixes makes sense.

Send the set, and I will review as vrf/l3mdev maintainer. I need working
tests and patches to see the before and after.

Re: [PATCH] selftests: Add VRF icmp error route lookup test

[Mathieu Desnoyers]

----- On Aug 11, 2020, at 3:14 PM, David Ahern dsahern@gmail.com wrote:

> On 8/11/20 1:11 PM, Mathieu Desnoyers wrote:
>> One thing I am missing before this series can be considered for upstreaming
>> is an Acked-by of the 2 fixes for ipv4 and ipv6 from you, as maintainer
>> of l3mdev, if you think the approach I am taking with those fixes makes sense.
> 
> Send the set, and I will review as vrf/l3mdev maintainer. I need working
> tests and patches to see the before and after.

Allright, I'm rebasing on the net tree as we speak, and the patches will
be on their way shortly.

Thanks,

Mathieu


-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com