Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

 
 

Routing Protocols

  • Maximum reference bandwidth increased to 4 TB for IGP protocols (ACX7100-32C, ACX7100-48L, ACX7509, PTX10001-36MR, PTX10003, PTX10004, PTX10008, PTX10016, QFX5130-32CD, QFX5700, and QFX5220)—Starting with Junos OS Evolved Release 22.3R1, we've increased the maximum reference bandwidth for IS-IS and OSPF interior gateway protocol (IGP) protocols from 1 Tbps to 4 Tbps. The default bandwidth is 100 Mbps. You can increase the reference bandwidth to adjust the path metrics, which you use to determine the preferred path in case of multiple equal-cost routes to a destination.

    To configure the reference bandwidth, use the reference-bandwidth reference-bandwidth statement at the [edit protocols isis] hierarchy level or the [edit protocols (ospf | ospf3)] hierarchy level.

    [See reference-bandwidth (Protocols IS-IS) and reference-bandwidth (Protocols OSPF).]

  • Support for 256-way ECMP (PTX10001-36MR, PTX10004, PTX10008, and PTX10016)—Starting in Junos OS Evolved Release 22.3R1, you can configure a maximum of 256 equal-cost multipath (ECMP) next hops for external BGP (EBGP) peers. This feature increases the number of direct BGP peer connections, which improves latency and optimizes data flow. However, we support 128 ECMP next hops for MPLS routes.

    [See Examples: Configuring BGP Multipath. ]

  • Fast lookup of origin and neighbor ASs (ACX7100-32C, ACX7100-48L, ACX7509, PTX10001-36MR, PTX10004, PTX10008, PTX10016, QFX5130-32CD, QFX5700, and QFX5220)—Starting in Junos OS Evolved Release 22.3R1, you can use the new asregex-optimize configuration statement at the [edit policy-options defaults] hierarchy level to perform a fast lookup of origin and neighbor autonomous systems (ASs). The asregex-optimize configuration statement is not enabled by default.

    [See Improve the Performance of AS Path Lookup in BGP Policy.]

  • Origin validation communities conversion to keywords (PTX10001-36MR, PTX10008, PTX10016, QFX5130-32CD, and QFX5220)—Starting in Junos OS Release Evolved 22.3R1, you can choose to accept or reject the origin validation extended communities received from an external BGP (EBGP) peer. The default behavior of Origin Validation State Extended Community (OVS EC) is changed to rejected if received from an EBGP peer. You can configure OVS EC to accept the community when needed. We also support the configuration of distinguished communities with keywords (valid, invalid, and unknown) at all the three layers of the BGP configuration hierarchy: global, group, and per-neighbor. If you enable the OVS EC at a hierarchy level, it’s enabled for the lower levels as well. However, you can choose to disable it explicitly at a lower layer if required in any instance.

    To accept origin validation communities from an EBGP peer, use origin-validation accept at the [edit protocols bgp ebgp-community-cleanup], edit protocols bgp group <group-name> ebgp-community-cleanup], or [edit protocols bgp group <group-name> neighbor <address> ebgp-community-cleanup] hierarchy level.

    To reject origin validation communities from an EBGP peer, use origin-validation reject at the [edit protocols bgp ebgp-community-cleanup], edit protocols bgp group <group-name> ebgp-community-cleanup], or [edit protocols bgp group <group-name> neighbor <address> ebgp-community-cleanup] hierarchy level.

    [See BGP Origin Validation.]

  • BGP LU Prefix SID redistribution between IGP domains (PTX10001-36MR, PTX10004, PTX10008, and PTX10016)— Starting in Junos OS Release 22.3R1, we support BGP Labeled Unicast (BGP-LU) prefix-sid redistribution between IGP domains. We also support installing an mpls.0 stitch route to a BGP-LU next-hop instance.

    To set prefix segment attributes, use the set prefix-segment statement at the [edit policy-options policy-statement <policy-name> term <tern-name> from] hierarchy level.

    You can now enable prefix-sid redistribution between BGP and IS-IS via policy configuration without specifying the index under prefix-segment. To do this, use the set prefix-segment redristribute option at the [edit policy-options policy-statement <policy-name> term <term-name> then] hierarchy level.

    [See Link-State Distribution Using BGP.]

  • Strengthening ARP/NDP Control for IPv4/IPv6 Networks (PTX10001-36MR, PTX10003, PTX10004, PTX10008, and PTX10016)— To protect routers from DoS attacks that exploit IPv6’s large address space, administrators can now configure limits on ARP/NDP neighbor entries using new CLI commands. These limits, applied globally and per interface, help prevent resource exhaustion by controlling the number of cached entries. The system issues warnings at 80% capacity and logs errors when it reaches maximum limits, ensuring network stability during potential attacks.

    [See nd6-max-cache.]