[c-nsp] C3k: IPv6 multicast listener reports causes high CPU

[Phil Mayers]

On 03/03/2014 19:38, Peter Rathlev wrote:

> layer-2, is using the "desktop default" SDM template and has no IPv6
> features (like MLD snooping) configured.

Are you absolutely sure about this? Isn't MLD snooping on by default?

> They are multicast listener reports, and there are far far too many of
> them from each PC. That is a "problem" with the PC which other people
> are (supposed to be) looking at.

I wonder if it's the same thing we (and others) saw:

http://lists.cluenet.de/pipermail/ipv6-ops/2014-February/009835.html

> What I don't understand is why this ends up in a CPU queue on a 3560.
> Shouldn't the "Router Alert" option only be picked up by devices doing
> L3 forwarding? We see several switches with these symptoms but also
> several without. Even switches which are almost identical in
> configuration, hardware and IOS behave differently.

Big surprise there; Cisco being inconsistent!

> So apart from just blocking all IPv6 (which wouldn't be hard, but feels
> like going the wrong way) what can I do? :-)

I can't remember if c3k can do multicast storm limiting without bad side 
effects; that is how we solved our problem. In particular if it is the 
same thing (AMT/vPro) we found that even a modest rate-limit closed the 
flood storms down completely, as they seem to be caused by "reflection".