Peter Memishian wrote:

> > In the past, interface naming has been like this:
> > IPv4 and IPv6 address names always match the device name and instance with
> > which they are associated and always match themselves.  bge0 for ipv4 will
> > be bge0 for ipv6 and is bge0 underneath.
>
>By convention yes, but not enforced by the kernel.
>

Whatever; the point is that's how it works and it largely makes sense.

> > What are the new rules for associating Ipv4 and Ipv6 interfaces names
> > with the devices and their instances?  Does vanity naming allow for an
> > interface to be given the same name in v4 and v6 but be associated with
> > different devices/instances?  Is there any reason that it should allow
> > net0 for Ipv4 to be a different interface to net0 for Ipv6?
>
>There's no reason to allow it, but it would require additional complexity
>in IP to prevent it.  How much code should we add to IP to prevent a
>demented privileged application from tying the system in knots?
>

Hmm, then there would seem to be an architectural deficency in
vanity naming.  I'm sorry I didn't catch this before commitment.

btw, it isn't the privileged applications that you're protecting,
it is the users themselves - it looks like the choice is to protect
them when they run ifconfig rather than dladm.  I hope that doesn't
lead to too much confusion...because while the dladm command has
succeeded but the ifconfig one failed, there would still appear to
be room for confusion, vis a vis:

# ifconfig vni0 inet6 plumb
# dladm rename-link ce0 vni0
# snoop -d vni0

What happens now?

I *really* think you need to solve the rename-link problem.

Darren


Reply via email to