EVPN-ETREE Overview
The EVPN-ETREE service is a VPN service where each attachment circuit is designated as either root or leaf. The EVPN E-Tree feature implements E-Tree service as defined by the Metro Ethernet Forum (MEF) in draft-sajassi-l2vpn-evpn-etree-03. The E-Tree service is a rooted-multipoint service that is supported only with EVPN over MPLS in the core. The EVPN E-Tree feature provides a way to categorize the interfaces as either “root” or “leaf” in a routing instance. In an EVPN E-Tree service, each Customer Edge devices attached the service is either a root or a leaf. The EVPN E-Tree service adheres to the following forwarding rules:
-
A leaf can send or receive traffic only from a root.
-
A root can send traffic to another root or any of the leaves.
-
A leaf or root can be connected to provider edge (PE) devices in singlehoming mode or multihoming mode.
The EVPN ETREE service has all the benefits of EVPN such as active-active multihoming and load balancing loop detection for E-Tree.
In an EVPN ETREE service, the forwarding rule depends on the traffic source and destination for known unicast traffic or unknown unicast, broadcast, and multicast (BUM) traffic. Table 1 shows the forwarding rules within the ETREE service.
We don't support IGMP snooping, MLD snooping, or PIM snooping multicast optimizations with EVPN-ETREE.
Type of Traffic |
Allowed/Not-Allowed |
Filtering Location |
Known Unicast Traffic from Root to Root |
Allowed |
|
Known Unicast Traffic from Root to Leaf |
Allowed |
|
BUM Traffic from Root to Root |
Allowed |
|
BUM Traffic from Root to Leaf |
Allowed |
|
Known Unicast Traffic from Leaf to Leaf |
Not Allowed |
At the ingress Packet Forwarding Engine |
Known Unicast Traffic from Leaf to Root |
Allowed |
|
BUM Traffic from Leaf to Leaf |
Not Allowed |
At the Egress Packet Forwarding Engine |
BUM Traffic from Leaf to Root |
Allowed |
If you do not configure a role for an interface, it will be assigned the role of “root” by default. All leaf interfaces are assigned a new mesh group with no local switching set to TRUE. This enables the ingress filtering for unicast traffic and all the leaf-to-leaf traffic will get dropped at ingress leaf interface. For BUM traffic, the filtering will happen at the egress Provider Edge based on the root/leaf label being carried in the packet.
NSR and Unified ISSU Support for EVPN-ETREE
Nonstop active routing (NSR) and graceful Routing Engine switchover (GRES) minimize traffic loss when there is a Routing Engine switchover. When a Routing Engine fails, NSR and GRES enable a routing platform with redundant Routing Engines to switch over from a primary Routing Engine to a backup Routing Engine and continue forwarding packets. Unified in-service software upgrade (ISSU) allows you to upgrade your Junos OS software on your MX Series router with no disruption on the control plane and with minimal disruption of traffic. Both GRES and NSR must be enabled to use unified ISSU.
To enable GRES, include the graceful-switchover
statement at the
[edit chassis redundancy]
hierarchy level.
Junos OS mirrors essential data when NSR is enabled. For EVPN ETREE, the local EVPN ETREE leaf label that is advertised to other PE as part of the ETREE extended community will be mirrored on the standby Routing Engine. For information on other mirrored data and NSR data flow, see NSR and Unified ISSU Support for EVPN.
To enable NSR, include the nonstop-routing
statement at the
[edit routing-options]
hierarchy level and the commit
synchronize
statement at the [edit system]
hierarchy
level.
EVPN-ETREE on ACX5448 Routers
Starting in Junos OS Release 19.4R2, ACX5448 routers support EVPN-ETREE feature. To
enable EVPN-ETREE on ACX5448 routers, include the evpn-mh-profile
configuration statement at the [edit system packet-forwarding-options
firewall-profile
] hierarchy level.
user@host# set system packet-forwarding-options firewall-profile ? Possible completions: default-profile Set the profile to support default services. evpn-mh-profile Set the profile to support evpn-mh
After changing the profile and committing it, you need to restart the chassis
management process by issuing the restart chassis-control
CLI
command to bring up the new profile.
A syslog warning appears to restart the PFE.