I'm studying the CCNA R&S at the moment and I'm a bit confused about why E-Tree services require all leaf nodes to belong in the same subnet, despite the leaves only being able to send layer 2 frames to the root.

[snip] an E-Tree can present some challenges for routing protocols, because it is the only design in which some of the sites in the same EVC cannot send frames directly to each other.

What's the benefit/rationale for setting it up like this? Wouldn't it be simpler to keep each leaf node in a seperate subnet/vlan?

  • "Wouldn't it be simpler to keep each leaf node in a seperate subnet/vlan?" That will waste two addresses (network and broadcast) for IPv4 networks for each customer vs. for all the customers on the same network. Remember that there is a shortage of IPv4 addresses, and .they are not making any more IPv4 addresses.
    – Ron Maupin
    Jul 17 '19 at 0:00

One reason might be scaling. Maybe you do not want a separate VLAN / Subnet per site. If you have a lot of small sites this may add up. Basically its Private VLANs implemented in MPLS.

That being said we never had a customer request this or any other reason to implement E-tree.


E-Tree is essentially an L2 service, so the end nodes share the same segment and usually the same IP subnet. If you don't want that you need to choose another type of service.

Before you ask "why do I use XY" you should consider your requirements and choose the appropriate service type. E-Tree may be the right choice if you want to connect branch locations with headquarters without them being able to talk to each other.

There may be better solutions than that. Bridging over (bandwidth-limited and latency-ridden) WAN is not necessarily a very good idea...

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.