[nsp] Catalyst errdisable

[sthaug at nethelp.no]

> Our IOS version is the same as your problem code...  We'll get a
> good, err newer, version from Cisco.  I think the MALLOCFAIL
> messages are not related to the LOOP_BACK_DETECTED, it just
> coincides with turning up a new switchport.

Yes and no. The MALLOCFAIL messages are often signs of spanning tree
problems, and the same spanning tree problems can *lead to* keepalive
packets coming back to the switch, causing LOOP_BACK_DETECTED.

> I did find some additional info...
> http://groups.google.com/groups?ie=UTF-8&oe=UTF-8&as_umsgid=%3Cbmn1cm.1m5j.1@verdi.nethelp.no%3E&lr=&hl=no

I was just going to post that link :-)

Additional info: Instead of twiddling the errdisable know, you can
also turn off keepalive on the interface - this will prevent the
keepalive packets from being sent. Unfortunately it could also mean
slower detection of an interface going down.

Steinar Haug, Nethelp consulting, sthaug at nethelp.no