My debug scripts/programs used to debug multicast routing in Linux kernel and overall

.gitignore

@@ -2,3 +2,4 @@ A
 B
 C
 D
+pim-dm_*.yaml

namespace_demo/linux-dbg

+#!/bin/bash
+
+~/src/wendelin/wendelin.core/t/qemu-runlinux -v \
+	~/src/linux/obj-qemu_debug/arch/x86/boot/bzImage \
+	./linux-dbg.sh

namespace_demo/linux-dbg.sh

+#!/bin/bash
+
+mount -t tmpfs none t
+cd t
+../topo-mr-debug.sh test_local

namespace_demo/topo-mr-debug.sh

+#!/bin/bash -e
+# topo mr-debug creates simple A-B topology as a playground for multicast routing debugging
+# on A routes are setup in fully static way - without smcrouted
+# on B smcrouted is running
+#
+# run as e.g. ./topo-mr-debug test_locallocal
+
+# whether to use use mcast_rx/mcast_tx interfaces instead of lo
+# see
+#
+#   https://lab.nexedi.com/kirr/linux/commit/25bcc76020c5 and
+#   https://lab.nexedi.com/kirr/linux/commit/43eadf90a7a6
+#
+# for why lo does not work out of the box.
+use_mcast_rxtx_instead_of_lo=y
+
+# whether to use IPv4 or IPv6 for tests
+IPvX=6
+tx=tx$IPvX
+rx=rx$IPvX
+txrx=txrx$IPvX
+
+
+# ----------------------------------------
+
+. $(dirname $0)/lib.sh
+
+if [ $# = 0 ]; then
+    echo "Usage: $0 test_<...>"
+    exit 1
+fi
+
+xnewhost A
+xnewhost B
+xlink A B
+
+# Bshow ... displays ... and runs it inside B.
+Bshow() {
+    echo -e "\n>>>" "$@"
+    xnsenter B -- "$@"
+}
+
+
+if [ $use_mcast_rxtx_instead_of_lo = n ]; then
+    mrx=lo
+    mtx=lo
+else
+    mrx=mcast-rx
+    mtx=mcast-tx
+
+    # mrx <-> mtx  veth pair
+    xnsenter B -- ip link add $mrx type veth peer name $mtx
+    xnsenter B -- ip link set $mrx up
+    xnsenter B -- ip link set $mtx up
+
+    # programs from local host will be sending multicast via mtx (having default route for multicast addresses)
+    # those packets will be delivered back to local delivery via mrx <- mtx
+    # tell IPv4 routing not to reject them
+    # NOTE IPv6 does not need this
+    xnsenter B -- sysctl net.ipv4.conf.mcast-rx.accept_local=1
+fi
+
+
+xnsenter A -- ip addr add 1.0.0.1/24 dev a-b
+xnsenter A -- ip addr add 1::1/16    dev a-b
+# A: hardcoded route to send multicast to (no need for IPv6 - it gets setup'ed automatically via A-B MLD)
+xnsenter A -- ip r add 224.0.0.0/4   dev a-b
+
+xnsenter B -- ip addr add 2.0.0.2/24 dev $mtx scope global
+xnsenter B -- ip addr add 2::2/16    dev $mtx scope global
+
+
+# sleep a bit so that IPv6 address is activated for real on $mtx
+# if we don't delay - vvv `ip -6 route ... dev $mtx src ...` fails with "Error: Invalid source address".
+#
+# XXX isn't it a kernel bug that we have to delay?
+# XXX and why delay 3s? (with 1s it always fails, with 2s it still fails sometimes)
+sleep 3
+
+
+# from-external multicast traffic goes to lo
+# from-us       multicast traffic is sent to the world via lo too
+
+# add explicit rule to route global multicast traffic to lo instead of relying on ad-hoc hack inside Linux kernel:
+# https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/net/ipv4/route.c?id=v5.18-rc5-28-ga7391ad35724#n2747
+# (see "Apparently, routing tables are wrong. Assume, that the destination is on link." there)
+# (present starting from very long ago: Linux 2.1.68 year=1997)
+#
+# As we are establishing route for multicast, also explicitly specify preferred
+# source address with which to send from-us multicast traffic with.
+#
+# For table local pref ... - see IPv6 case vvv for the rationale. Probably for
+# IPv4 this is not strictly required, but we still do it just in case.
+xnsenter B -- ip r add multicast 224.0.0.0/4 oif $mtx dev $mtx scope global \
+                        table local pref high metric 1 \
+                        src 2.0.0.2
+
+# for IPv6 similar explicit mcast route -> lo is required, because IPv6 routing does not have in-kernel ad-hoc that IPv4 has.
+# NOTE Linux, unless patched, turns such lo-routes into rejects, so it won't work without patch to the kernel:
+# https://lab.nexedi.com/kirr/linux/commit/25bcc76020c5
+#
+# Put this route into "local" table with highest possible preference to force
+# the kernel to use it even if a neighbour node advertises via MLD its multicast capabilities.
+# Upon receiving MLD messages the kernel creates `pref medium` routes in `table
+# local`, so if we don't adjust "priority" of our route, it will become ignored.
+#
+# Also explicitly specify source address.
+xnsenter B -- ip r add multicast ff1e::/16   oif $mtx dev $mtx scope global \
+                        table local pref high metric 1 \
+                        src 2::2 \
+
+
+
+# debug dump of setup routing
+Bdebug() {
+    Bshow ip link
+    Bshow ip addr
+    Bshow ip rule
+    Bshow ip route show table all ;
+#   Bshow ip r get 2.0.0.17
+    Bshow ip -d r get 224.1.1.1
+    Bshow ip -d r get 224.1.1.1 fibmatch
+    Bshow ip -d r get 224.1.1.1 oif $mtx
+    Bshow ip -d r get 224.1.1.1 oif $mtx fibmatch   # NOTE: fails without explicit multicast -> lo route
+
+    Bshow ip -6 r show table local
+    Bshow ip -6 r show table main
+    Bshow ip -6 r show cache
+    Bshow ip -d r get ff1e::1
+    Bshow ip -d r get ff1e::1 fibmatch
+    Bshow ip -d r get ff1e::1   oif $mtx
+    Bshow ip -d r get ff1e::1   oif $mtx fibmatch
+    Bshow ip -6 r show cache
+    Bshow ip mr
+    Bshow ip -6 mr
+}
+#Bdebug
+
+
+echo
+echo '---------------------------'
+echo
+
+# B: start and setup mrouted
+Bstartmrouted() {
+    xnsenter B -- smcrouted -n -l debug &
+    sleep 1
+    xnsenter B -- smcroutectl add b-a  1.0.0.1 224.0.0.0/4 $mtx
+    xnsenter B -- smcroutectl add b-a  1::1    ff1e::/16   $mtx
+    # multicast originating from local host is forwarded to external interfaces
+    # NOTE when we are sending via mtx, the kernel does not notify mrouted about
+    # mtx - only about packets on mrx - so we hook into mrx instead of directly
+    # into mtx.
+    xnsenter B -- smcroutectl add $mrx 2.0.0.2 224.0.0.0/4 b-a
+    xnsenter B -- smcroutectl add $mrx 2::2    ff1e::/16   b-a
+
+    #(sleep 4 && Bmrtdump ) &
+}
+
+# debug dump from inside about what happens with (m)routing
+Bmrtdump() {
+    Bshow smcroutectl
+    Bshow ip rule
+    Bshow ip route show table all
+    Bshow ip mr
+    Bshow ip r get 224.1.1.1 oif b-a
+    Bshow ip r get 224.1.1.1 oif $mtx ;
+    #Bshow ip r get 224.1.1.1 from 1.0.0.1 oif lo       # Network is unreachable
+    Bshow ip r get 224.1.1.1 from 1.0.0.1 iif b-a
+    #Bshow ip r get 224.1.1.1 iif b-a                   Invalid argument
+    #Bshow ip r get 224.1.1.1                           Network is unreachable
+    #Bshow ip r get 224.1.1.1 fibmatch                  Network is unreachable
+    Bshow ip -6 mr
+    Bshow ip -6 r get ff1e::1 fibmatch
+    Bshow ip -6 r get ff1e::1
+    #Bshow cat /proc/net/route
+    #Bshow cat /proc/net/fib_trie
+    #Bshow cat /proc/net/ipv6_route
+}
+
+# local->local delivery
+test_locallocal() {
+    # Bstartmrouted - not started
+
+    #xnsenter B -- tshark -lni $mtx &
+    xnsenter B -- $(dirname $0)/tmcast.py $txrx B
+}
+
+# test ext->local delivery
+test_extlocal() {
+    Bstartmrouted
+
+    ( sleep 2 && echo -e "\n\n\n\n\n" && xnsenter A -- $(dirname $0)/tmcast.py $tx A ) &
+    xnsenter B -- $(dirname $0)/tmcast.py $rx
+}
+
+# test ext<-local
+test_localext() {
+    Bstartmrouted
+
+    ( echo -e "\n\n\n\n\n" && xnsenter A -- $(dirname $0)/tmcast.py $rx ) &
+    sleep 2; xnsenter B -- $(dirname $0)/tmcast.py $tx B
+}
+
+
+# eval commandline arguments
+"$*"