Why IPv6 "Multicast Listener Report" packets are sent?

[giannici]

We have a large network with many Netonix switches.

Trying to limit the number of broadcast/multicast packets in the network, we found that the Netonixs send A LOT of IPv6 "Multicast Listener Report" (more then 25% of the broadcast/multicast traffic is done of these).

Is there any reason for this behavior if the switch don't even have an IPv6 address set (IPv6 section is disabled)???

Is there a way to stop them?

Thanks.

[Stephen]

That is odd. Multicast Listener Discovery is of course built into ICMPv6 so if you have IPv6 disabled completely it is a bit odd that you're seeing those messages.

Please post output from the following command:

Code: Select all

cmd
ip addr


[giannici]

OK, here is the output:

Code: Select all

Cammarata# cmd

BusyBox v1.19.4 (2018-06-01 19:38:44 EDT) built-in shell (ash)
Enter 'help' for a list of built-in commands.
root@Cammarata:/www# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue
 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
 inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
 inet6 ::1/128 scope host
 valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000
 link/ether ec:13:b2:61:3f:e0 brd ff:ff:ff:ff:ff:ff
 inet 10.10.252.127/16 brd 10.10.255.255 scope global eth0
 inet6 fe80::ee13:b2ff:fe61:3fe0/64 scope link
 valid_lft forever preferred_lft forever
root@Cammarata:/www#


This one is a WS-12-250-AC with 1.5.5, but the same occurs with different models and firmwares.

I confirm that the "IPv6" section of config is DISABLED.

[giannici]

About broadcasts, I noticed a lot of Gratuitous ARPs too. When are they sent?

[Stephen]

Do you mind sending me a pcap file that has the Multicast Listener Report packets captured? I want to compare it to mine to see if there are any difference's.

Yes the gratuitous arp's are VERY MUCH SO on purpose.
It is a bug fix mentioned here:
"- Boot up now performs a gratuitous ARP broadcast, which prevents the scenario where upon a warm or cold boot (aka firmware upgrade, etc) where loss of connectivity
to the switch's management site or ssh access occurs even if sometimes otherwise the switch is operating. "

[giannici]

Do you mind sending me a pcap file that has the Multicast Listener Report packets captured? I want to compare it to mine to see if there are any difference's.

Yes the gratuitous arp's are VERY MUCH SO on purpose.
It is a bug fix mentioned here:
"- Boot up now performs a gratuitous ARP broadcast, which prevents the scenario where upon a warm or cold boot (aka firmware upgrade, etc) where loss of connectivity
to the switch's management site or ssh access occurs even if sometimes otherwise the switch is operating. "

There are A LOT of gratuitous arps, they are NOT sent only on boot!

I'll send you a pcap later...

[Stephen]

Thank you giannici for sending the pcap file. I'll start examining it. During experiments we found that change's in the network can cause a loss of connectivity to a switch if the gratuitous arp's are not being sent at an interval. I can of course slow them down a but any decrease mean's an increase in waiting for access back to the switch in such a case as a topology change.
I will take a look at the pcap file and see if I can trace down why MTD is active when IPv6 is disabled.

[giannici]

Thank you giannici for sending the pcap file. I'll start examining it. During experiments we found that change's in the network can cause a loss of connectivity to a switch if the gratuitous arp's are not being sent at an interval. I can of course slow them down a but any decrease mean's an increase in waiting for access back to the switch in such a case as a topology change.
I will take a look at the pcap file and see if I can trace down why MTD is active when IPv6 is disabled.

Maybe an option could be added in the config for those that need to enable/disable these periodic gratuitous ARPs (except for those on boot that can be always useful and have almost no weight).

[Stephen]

Not a bad thought at all.

[giannici]

Hi.
Any news on this?
When will a new firmware be released?