diff mbox

[net] ipv6: Fix IPv6 packet loss in scenarios involving roaming + snooping switches

Message ID 20170203071103.25976-1-linus.luessing@c0d3.blue
State Accepted, archived
Delegated to: David Miller
Headers show

Commit Message

Linus Lüssing Feb. 3, 2017, 7:11 a.m. UTC
When for instance a mobile Linux device roams from one access point to
another with both APs sharing the same broadcast domain and a
multicast snooping switch in between:

1)    (c) <~~~> (AP1) <--[SSW]--> (AP2)

2)              (AP1) <--[SSW]--> (AP2) <~~~> (c)

Then currently IPv6 multicast packets will get lost for (c) until an
MLD Querier sends its next query message. The packet loss occurs
because upon roaming the Linux host so far stayed silent regarding
MLD and the snooping switch will therefore be unaware of the
multicast topology change for a while.

This patch fixes this by always resending MLD reports when an interface
change happens, for instance from NO-CARRIER to CARRIER state.

Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue>

---

Initial problem report was sent to the bridge mailing list a while ago:
- https://lists.linuxfoundation.org/pipermail/bridge/2015-September/009754.html

The RFCs concerning IGMP, MLD and snooping switches seem a have a hole
concerning roaming. A request for clarification to mcast-wifi@ietf.org
was left unanswered, unfortunately:
- https://mailarchive.ietf.org/arch/msg/mcast-wifi/Ghn2cGy1oN2ZwG1qaQO9SE13g6g

However, simply resending reports seems to be the straight forward way
to me to fix the issue mentioned above.
---
 net/ipv6/addrconf.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

Comments

David Miller Feb. 6, 2017, 4:43 p.m. UTC | #1
From: Linus Lüssing <linus.luessing@c0d3.blue>
Date: Fri,  3 Feb 2017 08:11:03 +0100

> When for instance a mobile Linux device roams from one access point to
> another with both APs sharing the same broadcast domain and a
> multicast snooping switch in between:
> 
> 1)    (c) <~~~> (AP1) <--[SSW]--> (AP2)
> 
> 2)              (AP1) <--[SSW]--> (AP2) <~~~> (c)
> 
> Then currently IPv6 multicast packets will get lost for (c) until an
> MLD Querier sends its next query message. The packet loss occurs
> because upon roaming the Linux host so far stayed silent regarding
> MLD and the snooping switch will therefore be unaware of the
> multicast topology change for a while.
> 
> This patch fixes this by always resending MLD reports when an interface
> change happens, for instance from NO-CARRIER to CARRIER state.
> 
> Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue>

Looks good to me, applied, thanks.
diff mbox

Patch

diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index f60e88e..81f7b4e 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -3386,9 +3386,15 @@  static int addrconf_notify(struct notifier_block *this, unsigned long event,
 			}
 
 			if (idev) {
-				if (idev->if_flags & IF_READY)
-					/* device is already configured. */
+				if (idev->if_flags & IF_READY) {
+					/* device is already configured -
+					 * but resend MLD reports, we might
+					 * have roamed and need to update
+					 * multicast snooping switches
+					 */
+					ipv6_mc_up(idev);
 					break;
+				}
 				idev->if_flags |= IF_READY;
 			}