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

Peter Rathlev peter at rathlev.dk
Mon Mar 3 15:37:52 EST 2014


On Mon, 2014-03-03 at 19:50 +0000, Phil Mayers wrote:
> I wonder if it's the same thing we (and others) saw:
> 
> http://lists.cluenet.de/pipermail/ipv6-ops/2014-February/009835.html

It sounds very much like it, now that you say it. The part about two of
these goading each other turned out true; I just now tried temporarily
shutting one of the interfaces (connected at 100-FD, not answering ARP,
probably sleeping) and the "storm" stopped altogether.

The NIC OUIs of the two participants just now were "Dell Inc" and "Dell
PCBA Test". I'm unsure if they're Intel but they might very well be.
I'll ask our PC people to take a look at them

> > 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".

We'll try this the next time the problem shows up. All the places we've
seen the problem so far are places where we can rate-limit multicast a
lot without complaints. But we better also contact Cisco and ask for an
explanation of why this affects the control-plane. :-)

Thank you so much for the insight!

-- 
Peter






More information about the cisco-nsp mailing list