On some distributions, the rp_filter is automatically set (=1) by
default on a netdev basis (also on VRFs).
In an SRv6 End.DT46 behavior, decapsulated IPv4 packets are routed using
the table associated with the VRF bound to that tunnel. During lookup
operations, the rp_filter can lead to packet loss when activated on the
VRF.
Therefore, we chose to make this selftest more robust by explicitly
disabling the rp_filter during tests (as it is automatically set by some
Linux distributions).
Fixes: 03a0b567a03d ("selftests: seg6: add selftest for SRv6 End.DT46 Behavior")
Reported-by: Hangbin Liu <[email protected]>
Signed-off-by: Andrea Mayer <[email protected]>
Tested-by: Hangbin Liu <[email protected]>
---
.../testing/selftests/net/srv6_end_dt46_l3vpn_test.sh | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/tools/testing/selftests/net/srv6_end_dt46_l3vpn_test.sh b/tools/testing/selftests/net/srv6_end_dt46_l3vpn_test.sh
index aebaab8ce44c..441eededa031 100755
--- a/tools/testing/selftests/net/srv6_end_dt46_l3vpn_test.sh
+++ b/tools/testing/selftests/net/srv6_end_dt46_l3vpn_test.sh
@@ -292,6 +292,11 @@ setup_hs()
ip netns exec ${hsname} sysctl -wq net.ipv6.conf.all.accept_dad=0
ip netns exec ${hsname} sysctl -wq net.ipv6.conf.default.accept_dad=0
+ # disable the rp_filter otherwise the kernel gets confused about how
+ # to route decap ipv4 packets.
+ ip netns exec ${rtname} sysctl -wq net.ipv4.conf.all.rp_filter=0
+ ip netns exec ${rtname} sysctl -wq net.ipv4.conf.default.rp_filter=0
+
ip -netns ${hsname} link add veth0 type veth peer name ${rtveth}
ip -netns ${hsname} link set ${rtveth} netns ${rtname}
ip -netns ${hsname} addr add ${IPv6_HS_NETWORK}::${hs}/64 dev veth0 nodad
@@ -316,11 +321,6 @@ setup_hs()
ip netns exec ${rtname} sysctl -wq net.ipv6.conf.${rtveth}.proxy_ndp=1
ip netns exec ${rtname} sysctl -wq net.ipv4.conf.${rtveth}.proxy_arp=1
- # disable the rp_filter otherwise the kernel gets confused about how
- # to route decap ipv4 packets.
- ip netns exec ${rtname} sysctl -wq net.ipv4.conf.all.rp_filter=0
- ip netns exec ${rtname} sysctl -wq net.ipv4.conf.${rtveth}.rp_filter=0
-
ip netns exec ${rtname} sh -c "echo 1 > /proc/sys/net/vrf/strict_mode"
}
--
2.20.1
On 4/27/23 3:49 AM, Andrea Mayer wrote:
> On some distributions, the rp_filter is automatically set (=1) by
> default on a netdev basis (also on VRFs).
> In an SRv6 End.DT46 behavior, decapsulated IPv4 packets are routed using
> the table associated with the VRF bound to that tunnel. During lookup
> operations, the rp_filter can lead to packet loss when activated on the
> VRF.
> Therefore, we chose to make this selftest more robust by explicitly
> disabling the rp_filter during tests (as it is automatically set by some
> Linux distributions).
>
> Fixes: 03a0b567a03d ("selftests: seg6: add selftest for SRv6 End.DT46 Behavior")
> Reported-by: Hangbin Liu <[email protected]>
> Signed-off-by: Andrea Mayer <[email protected]>
> Tested-by: Hangbin Liu <[email protected]>
> ---
> .../testing/selftests/net/srv6_end_dt46_l3vpn_test.sh | 10 +++++-----
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
Reviewed-by: David Ahern <[email protected]>
Hello:
This patch was applied to netdev/net.git (main)
by David S. Miller <[email protected]>:
On Thu, 27 Apr 2023 11:49:23 +0200 you wrote:
> On some distributions, the rp_filter is automatically set (=1) by
> default on a netdev basis (also on VRFs).
> In an SRv6 End.DT46 behavior, decapsulated IPv4 packets are routed using
> the table associated with the VRF bound to that tunnel. During lookup
> operations, the rp_filter can lead to packet loss when activated on the
> VRF.
> Therefore, we chose to make this selftest more robust by explicitly
> disabling the rp_filter during tests (as it is automatically set by some
> Linux distributions).
>
> [...]
Here is the summary with links:
- [net] selftests: srv6: make srv6_end_dt46_l3vpn_test more robust
https://git.kernel.org/netdev/net/c/46ef24c60f8e
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html