[c-nsp] loop guard still useful?

[Lee]

On 1/18/16, Michele Bergonzoni <bergonz at labs.it> wrote:

>> So it seems like loop guard isn't needed if rstp is enabled.
>
> I have no operational experience with loop guard, but from the description
> it seems to me that in order to trigger it the interface must become
> unidirectional *after* link up.

Right

> Thus, if your Joe Average while
> troubleshooting does a shut/no shut, he actually gets the loop.

I'm not sure about shut/no shut but a reboot after the link goes
unidirectional -- yes, you get a loop.

> So it will protect you on the other unidirectionality side, but not in all
> possible sequences of events.
>
> If you are operating an all-cisco net you might take a look at bridge
> assurance. I have no operational experience with it as well (apart from
> disabling it in the nexus), but looks much more like a bidirectional
> keepalive at the STP layer. It is proprietary and violates the standard as I
> understand it.

Sounds like loop guard except there's now edge, normal and network
port types with network ports going into blocking/inconsistent state
if they don't see BPDUs.   Loop guard puts a port into
blocking/inconsistent state if it _stops_ seeing BPDUs on a port.

>> No, I don't like UDLD at all - too many bad experiences with it
>
> In fact after what Saku said I would consider trusting the layer 1, but I
> usually work in a multivendor environment, YMMV.

Right - it does sound like rstp might be good enuf.

Regards,
Lee