Navigation  Back up to About Overview 
ContentIndex
  
[+] Expand All
[-] Collapse All

No index entries found.

New Features in Junos OS Release 12.3 for M Series, MX Series, and T Series Routers

The following features have been added to Junos OS Release 12.3. Following the description is the title of the manual or manuals to consult for further information.

Hardware

  • SFP-GE80KCW1470-ET, SFP-GE80KCW1490-ET, SFP-GE80KCW1510-ET, SFP-GE80KCW1530-ET, SFP-GE80KCW1550-ET, SFP-GE80KCW1570-ET, SFP-GE80KCW1590-ET, and SFP-GE80KCW1610-ET (MX Series)—These transceivers provide a duplex LC connector and support operation and monitoring with links up to a distance of 80 km. Each transceiver is tuned to a different transmit wavelength for use in CWDM applications. These transceivers are supported on the following interface module. For more information about interface modules, see the Interface Module Reference for your router.
    • Gigabit Ethernet MIC with SFP (model number: MIC-3D-20GE-SFP) in all versions of MX-MPC1, MX-MPC2, and MX-MPC3 —Supported in Junos OS Release 12.3R5, 13.2R3, 13.3R1, and later.

    [See Gigabit Ethernet SFP CWDM Optical Interface Specification.]

  • CFP-GEN2-CGE-ER4 (MX Series, T1600, and T4000)—The CFP-GEN2-CGE-ER4 transceiver (part number: 740-049763) provides a duplex LC connector and supports the 100GBASE-ER4 optical interface specification and monitoring. The “GEN2” optics have been redesigned with newer versions of internal components for reduced power consumption. The following interface modules support the CFP-GEN2-CGE-ER4 transceiver. For more information about interface modules, see the Interface Module Reference for your router.

    MX Series routers:

    • 100-Gigabit Ethernet MIC with CFP (model number: MIC3-3D-1X100GE-CFP)—Supported in Junos OS Release 12.1R1 and later
    • 2x100GE + 8x10GE MPC4E (model number: MPC4E-3D-2CGE-8XGE)—Supported in Junos OS Release 12.3R2 and later

    T1600 and T4000 routers:

    • 100-Gigabit Ethernet PIC with CFP (model numbers: PD-1CE-CFP-FPC4 and PD-1CGE-CFP)—Supported in Junos OS Releases 12.3R5, 13.2R3, 13.3R1, and later

    [See 100-Gigabit Ethernet 100GBASE-R Optical Interface Specifications.]

  • CFP-GEN2-100GBASE-LR4 (T1600 and T4000)—The CFP-GEN2-100GBASE-LR4 transceiver (part number: 740-047682) provides a duplex LC connector and supports the 100GBASE-LR4 optical interface specification and monitoring. The “GEN2” optics have been redesigned with newer versions of internal components for reduced power consumption. The following interface modules support the CFP-GEN2-100GBASE-LR4 transceiver. For more information about interface modules, see the Interface Module Reference for your router.
    • 100-Gigabit Ethernet PIC with CFP (model numbers: PD-1CE-CFP-FPC4 and PD-1CGE-CFP)—Supported in Junos OS Releases 12.3R5, 13.2R3, 13.3R1, and later

    [See 100-Gigabit Ethernet 100GBASE-R Optical Interface Specifications.]

  • Support for 32 GB RE-S-1800 Routing Engine (MX Series)—Starting with Junos OS Release 12.3R4, the 32-GB RE-S-1800 Routing Engine (MX240, MX480, and MX960 model number RE-S-1800X4-32G-S; MX2010 and MX2020 model number REMX2K-1800-32G-S) is supported on MX Series routers. The 32 GB RE-S-1800 Routing Engine has a 1800 GHz processor with 32-GB of memory, and supports both 32-bit and 64-bit Junos OS builds. On the MX2010 and MX2020 routers, the Routing Engine supports only 64-bit Junos OS build. The new Routing Engine supports a 4-GB CompactFlash card. All CLI commands supported on the older Routing Engines are supported on the new Routing Engine.

Class of Service

  • Support for class-of-service features to ensure quality of service for real-time traffic that is sensitive to latency on a network (MX240, MX480, MX960 Routers with Application Services Modular Line Card)—The new Application Services Modular Line Card (AS MLC) supports the following CoS features on MX240, MX480, and MX960 routers:
    • Code-point aliases—A code-point alias is a meaningful name that can be associated with CoS values such as Differentiated Services code points (DSCPs), DSCP IPv6, IP precedence, IEEE 802.1, and MPLS experimental (EXP) bits that can then be used while configuring CoS components.
    • Classification—Packet classification associates the packet with a particular CoS servicing level. In Junos OS, classifiers associate incoming packets with a forwarding class and loss priority and, based on the associated forwarding class, assign packets to output queues.
      • Behavior Aggregate—A method of classification that operates on a packet as it enters the router.
      • Multifield Classification— A method of classification that can examine multiple fields in the packet.
      • Fixed Classification—A method of classification that refers to the association of a forwarding class with a packet regardless of its packet contents.

      [See Class of Service on Application Services Modular Line Card Overview.]

    • Scheduling—Schedulers are used to define the properties of output queues. On the AS modular carrier card (AS MCC), the following scheduling features are supported (physical interfaces only):
      • Buffer sizes
      • Delay buffer size
      • Drop profile map
      • Excess priority
      • Excess rate percentage
      • Output-traffic-control profile
      • Priority
      • Scheduler-map
      • Shaping rate
      • Transmit rate
      • WRED rules

      [Junos OS Class-of-Service Configuration Guide]

  • Setting the 802.1p field for host-generated traffic—On MPCs and Enhanced Queuing DPCs, you can now configure the IEEE 802.1p bits in the 802.1p field—also known as the Priority Code Point (PCP) field—in the Ethernet frame header for host outbound packets (control plane traffic). In earlier releases, this field is not configurable; instead it is set by CoS automatically for host outbound traffic.

    To configure a global default value for this field for all host outbound traffic, include the default value statement at the [edit class-of-service host-outbound-traffic ieee-802.1] hierarchy level. This configuration has no effect on data plane traffic; you configure rewrite rules for these packets as always.

    You cannot configure a default value for the 802.1p bits for host outbound traffic on a per-interface level. However, you can specify that the CoS 802.1p rewrite rules already configured on egress logical interfaces are applied to all host outbound packets on that interface. To do so, include the rewrite-rules statement at the [edit class-of-service host-outbound-traffic ieee-802.1] hierarchy level. This capability enables you to set only the outer tags or both the outer and the inner tags on dual-tagged VLAN packets. (On Enhanced Queuing DPCs, both inner and outer tags must be set.)

    This feature includes the following support:

    • Address families—IPv4 and IPv6
    • Interfaces—IP over VLAN demux, PPP over VLAN demux, and VLAN over Gigabit Ethernet
    • Packet types—ARP, ANCP, DHCP, ICMP, IGMP, and PPP
    • VLANs—Single and dual-tagged

    [Class of Service]

  • Software feature support on the MX2020 routers—Starting with Release 12.3, all MPCs and MICs supported on the MX Series routers in Junos OS Release 12.3 continue to be supported on the MX2020 routers. Also, the MX2020 routers support all software features that are supported by other MX Series routers in Junos OS Release 12.1.

    The following key Junos OS features are supported:

    • Basic Layer 2 features including Layer 2 Ethernet OAM and virtual private LAN service (VPLS)
    • Class-of-service (CoS)
    • Firewall filters and policers
    • Integrated Routing and Bridging (IRB)
    • Interoperability with existing DPCs and MPCs
    • Layer 2 protocols
    • Layer 2 VPNs, Layer 2 circuits, and Layer 3 VPNs
    • Layer 3 routing protocols and MPLS
    • Multicast forwarding
    • Port mirroring
    • Spanning Tree Protocols (STP)
    • Synchronous Ethernet and Precision Time Protocol (IEEE 1588)
    • Tunnel support

    [Class of Service, Ethernet Interfaces Configuration Guide, System Basics and Services Command Reference]

  • Ingress CoS on MIC and MPC interfaces (MX Series routers)—You can configure ingress CoS parameters, including hierarchical schedulers, on MX Series routers with MIC and MPC interfaces. In general, the supported configuration statements apply to per-unit schedulers or to hierarchical schedulers.

    To configure ingress CoS for per-unit schedulers, include the following statements at the [edit class-of-service interfaces interface-name] hierarchy level:

    input-scheduler-mapinput-shaping-rateinput-traffic-control-profileinput-traffic-control-profile-remaining

    To configure ingress CoS for hierarchical schedulers, include the interface-set interface-set-name statement at the [edit class-of-service interfaces] hierarchy level.

    Note: The interface-set statement supports only the following options:

    input-traffic-control-profileinput-traffic-control-profile-remaining

    To configure ingress CoS at the logical interface level, include the following statements at the [edit class-of-service interfaces interface interface-name unit logical-unit-number] hierarchy level:

    input-scheduler-mapinput-shaping-rateinput-traffic-control-profile

    [See Configuring Ingress Hierarchical CoS on MIC and MPC interfaces.]

  • Extends explicit burst size configuration support on IQ2 and IQ2E interfaces–The burst size for shapers can be configured explicitly in a traffic control profile for IQ2 and IQ2E interfaces. This feature is supported on M71, M10i, M40e, M120, M320, and all T Series routers.

    To enable this feature, include the burst-size statement at the following hierarchy levels:

    [edit class-of-service traffic-control-profiles shaping-rate][edit class-of-service traffic-control-profiles guaranteed-rate]

    Note: The guaranteed-rate burst size value cannot be greater than the shaping-rate burst size.

    [See Configuring Traffic Control Profiles for Shared Scheduling and Shaping.]

  • Classification and DSCP marking of distributed protocol handler traffic—The scope of traffic affected by the host-outbound-traffic statement is expanded. When it was introduced in Junos OS Release 8.4, the host-outbound-traffic statement at the [edit class-of-service] hierarchy level enabled you to specify the forwarding class assignment and DiffServ code point (DSCP) value for egress traffic sent from the Routing Engine. Affected traffic included control plane packets (such as OSPF hello and ICMP echo reply [ping] packets) and TCP-related packets (such as BGP and LDP control packets).

    In Junos OS 12.2R2, the same configuration applies to distributed protocol handler traffic in addition to Routing Engine traffic. Distributed protocol handler traffic refers to traffic from the router’s periodic packet management process (ppm) sessions, and it includes both IP (Layer 3) traffic such as BFD keepalive (KA) messages and non-IP (Layer 2) traffic such as LACP control traffic on aggregated Ethernet. DSCP changes do not apply to MPLS EXP bits or IEEE 802.1p bits. The specified queue must be correctly configured. The affected traffic includes distributed protocol handler traffic as well as Routing Engine traffic for egress interfaces hosted on MX Series routers with Trio-based or I-chip based Packet Forwarding Engines, and on M120, M320, and T Series routers.

    If you need the Routing Engine traffic and distributed protocol handler traffic to be classified in different forwarding classes or marked with different DSCP values, then you need to configure some additional steps. Apply a standard firewall filter to the loopback interface and configure the filter actions to set the forwarding class and DSCP value that override the host-outbound-traffic settings.

    For interfaces on MX80 routers, LACP control traffic is sent through the Routing Engine rather than through the Packet Forwarding Engine.

    Note: Any DSCP rewrite rules configured on a 10-Gigabit Ethernet LAN/WAN PIC with SFP+ overwrite the DSCP value rewritten as specified under the host-outbound-traffic statement.

    The following partial configuration example classifies egress traffic from the Routing Engine as well as distributed protocol handler traffic:

    [edit]
    class-of-service {host-outbound-traffic {forwarding-class my_fc_control-traffic_dph;dscp-code-point 001010;}forwarding-classes {queue 5 my_fc_control-traffic_dph;queue 6 my_fc_control_traffic_re;}}
    interfaces {lo0 {unit 0 {family inet {filter {output my_filter_reclassify_re;}}}}}
    firewall {filter my_filter_reclassify_re {term 1 {then {forwarding-class my_fc_control_traffic_re;dscp code-points 000011;accept;}}}}

    The statements in the example configuration cause the router to classify egress traffic from the Routing Engine and distributed protocol handler traffic as follows:

    • Distributed protocol handler traffic is classified to the my_fc_control-traffic_dph forwarding class, which is mapped to queue 5. Of those packets, Layer 3 packets are marked at egress with DSCP bits 001010 (10 decimal), which is compatible with ToS bits 00101000 (40 decimal).
    • Routing Engine traffic is classified to the my_fc_control-traffic_re forwarding class, which is mapped to queue 6. Of those packets, Layer 3 packets are marked at egress with DSCP bits 001100 (12 decimal), which is compatible with ToS bits 00110000 (48 decimal).

    If you do not apply the firewall filter to the loopback interface, Routing Engine-sourced traffic is classified and marked using the forwarding class and DSCP value specified in the host-outbound-traffic configuration statement.

    If you omit both the firewall filter and the host-outbound-traffic configuration shown in the previous configuration, then all network control traffic—including Routing Engine-sourced and distributed protocol handler traffic—uses output queue 3 (the default output queue for control traffic), and DSCP bits for Layer 3 packets are set to the default value 0 (Best Effort service).

  • Enhancements to scheduler configuration on FRF.16 physical interfaces—Starting with Release 12.3R2, Junos OS extends the class-of-service scheduler support on FRF.16 physical interfaces to the excess-rate, excess-priority, and drop-profile-map configurations. The excess-rate, excess-priority, and drop-profile-map statements are configured at the [edit class-of-service schedulers scheduler-name] hierarchy level.
    • Support for the drop-profile-map configuration enables you to configure random early detection (RED) on FRF.16 bundle physical interfaces.
    • Support for the excess-rate configuration enables you to specify the percentage of the excess bandwidth traffic to share.
    • Support for the excess-priority configuration enables you to specify the priority for excess bandwidth traffic on a scheduler.

    This feature is supported only on multiservices PICs installed on MX Series routers.

  • Accurate reporting of output counters for MLFR UNI NNI bundles—Starting with Release 12.3R2, Junos OS reports the actual output counters in the multilink frame relay (MLFR) UNI NNI bundle statistics section of the show interfaces lsq-interface statistics command output. From this release on, Junos OS also provides per-DLCI counters for logical interfaces. In earlier releases, there was a discrepancy between the actual output counters and the reported value because of errors in calculating the output counters at the logical interface level. That is, at the logical interface level, the output counter was calculated as the sum of frames egressing at the member links instead of roviding the output counter as the sum of per-DLCI output frames
  • Extended MPC support for per-unit schedulers—Enables you to configure per-unit schedulers on the non-queuing 16x10GE MPC and the MPC3E, meaning you can include the per-unit-scheduler statement at the [edit interfaces interface name] hierarchy level. When per-unit schedulers are enabled, you can define dedicated schedulers for logical interfaces by including the scheduler-map statement at the [edit class-of-service interfaces interface name unit logical unit number] hierarchy level. Alternatively, you can include the scheduler-map statement at the [edit class-of-service traffic-control-profiles traffic control profile name] hierarchy level and then include the output-traffic-control-profile statement at the [edit class-of-service interfaces interface name unit logical unit number] hierarchy level.

    Enabling per-unit schedulers on the 16x10GE MPC and the MPC3E adds additional output to the show interfaces interface name [detail | extensive] command. This additional output lists the maximum resources available and the number of configured resources for schedulers.

    [Applying Scheduler Maps and Shaping Rate to DLCIs and VLANs]

Forwarding and Sampling

  • Increased forwarding capabilities for MPCs and Multiservices DPCs through FIB localization (MX Series routers)—Forwarding information base (FIB) localization characterizes the Packet Forwarding Engines in a router into two types: FIB-Remote and FIB-Local. FIB-Local Packet Forwarding Engines install all of the routes from the default route tables into Packet Forwarding Engine forwarding hardware. FIB-Remote Packet Forwarding Engines create a default (0.0) route that references a next hop or a unilist of next hops to indicate the FIB-Local that can perform full IP table looks-ups for received packets. FIB-Remote Packet Forwarding Engines forward received packets to the set of FIB-Local Packet Forwarding Engines.

    The capacity of MPCs is much higher than that of Multiservices DPCs, so an MPC is designated as the local Packet Forwarding Engine, and a Multiservices DPC is designated as the remote Packet Forwarding Engine. The remote Packet Forwarding Engine forwards all network-bound traffic to the local Packet Forwarding Engine. If multiple MPCs are designated as local Packet Forwarding Engines, then the Multiservices DPC will load-balance the traffic using the unilist of next hops as the default route.

High Availability (HA) and Resiliency

  • Protocol Independent Multicast nonstop active routing support for IGMP-only interfaces—Starting with Release 12.3, Junos OS extends the Protocol Independent Multicast (PIM) nonstop active routing support to IGMP-only interfaces.

    In Junos OS releases earlier than 12.3, the PIM joins created on IGMP-only interfaces were not replicated on the backup Routing Engine and so the corresponding multicast routes were marked as pruned (meaning discarded) on the backup Routing Engine. Because of this limitation, after a switchover, the new master Routing Engine had to wait for the IGMP module to come up and start receiving reports to create PIM joins and to install multicast routes. This causes traffic loss until the multicast joins and routes are reinstated.

    However, in Junos OS Release 12.3 and later, the multicast joins on the IGMP-only interfaces are mapped to PIM states, and these states are replicated on the backup Routing Engine. If the corresponding PIM states are available on the backup, the multicast routes are marked as forwarding on the backup Routing Engine. This enables uninterrupted traffic flow after a switchover. This enhancement covers IGMPv2, IGMPv3, MLDv1, and MLDv2 reports and leaves.

    [High Availability]

  • RSVP

    Nonstop active routing support for RSVP includes:

    • Point-to-Multipoint LSPs
      • RSVP Point-to-Multipoint ingress, transit, and egress LSPs using existing non-chained next hop.
      • RSVP Point-to-Multipoint transit LSPs using composite next hops for Point-to-Multipoint label routes.
    • Point-to-Point LSPs
      • RSVP Point-to-Point ingress, transit, and egress LSPs using non-chained next hops.
      • RSVP Point-to-Point transit LSPs using chained composite next hops.
  • Configuration support to include GTP TEID field in hash key for load-balancing GTP-U traffic (MX Series routers with MPCs and MX80)—On an MX Series router with MPCs, when there are multiple equal-cost paths to the same destination for the active route, Junos OS uses a hash algorithm to choose one of the next-hop addresses from the forwarding table when making a forwarding decision. Whenever the set of next hops for a destination changes in any way, the next-hop address is rechosen using the hash algorithm. For GPRS tunneling protocol (GTP)-encapsulated traffic, the tunnel endpoint identifier (TEID) field changes for traffic traversing through peer routers. To implement load balancing for GTP-encapsulated traffic on the user plane (GTP-U), the TEID should be included in the hash key.

    In Junos OS Release 12.3R2, you can configure GTP hashing on MX Series routers with MPCs and on MX80, to include the TEID field in hash calculations for IPv4 and IPv6 packets. To configure GTP hashing and include the GTP TEID field in hash calculations, configure the gtp-tunnel-end-point-identifier statement at the [edit forwarding-options enhanched-hash-key family] hierarchy level. GTP hashing is supported for both IPv4 and IPv6 packets received for GTP-U traffic at the MPC. For bridging and MPLS packets, GTP hashing is supported for IPv4 and IPv6 packets that are carried as payload for GTP-encapsulated traffic.

    Note: For IPv4 packets, GTP hashing is supported only for the nonfragmented packets.

    [Overview of Per-Packet Load Balancing]

Interfaces and Chassis

  • Support for fabric management features (MX240, MX480, MX960 Routers with Application Services Modular Carrier Card)—The Application Services Module Line Card (AS MLC) is supported on MX240, MX480, and MX960 routers. The AS MLC consists of the following components:
    • Application Services Modular Carrier Card (AS MCC)
    • Application Services Modular Processing Card (AS MXC)
    • Application Services Modular Storage Card (AS MSC)

    The AS MCC plugs into the chassis and provides the fabric interface. On the fabric management side, the AS MLC provides redirection functionality using a demultiplexer. The following CLI operational mode commands display fabric-related information for the AS MCC:

    • show chassis fabric fpcs
    • show chassis fabric map
    • show chassis fabric plane
    • show chassis fabric plane-location
    • show chassis fabric reachability
    • show chassis fabric summary

    [Junos OS System Basics Configuration Guide, Junos OS System Basics and Services Command Reference]

    [See Fabric Plane Management on AS MLC Modular Carrier Card Overview.]

  • Support for chassis management (MX240, MX480, MX960 Routers with Application Services Modular Line Card)—The Application Services Modular Line Card (AS MLC) is a Modular Port Concentrator (MPC) that is designed to run services and applications on MX240, MX480, and MX960 routers.

    The following CLI operational mode commands support the chassis management operations of the modular carrier card on the AS MLC:

    • show chassis environment fpc
    • show chassis firmware
    • show chassis fpc
    • show chassis hardware
    • show chassis pic
    • show chassis temperature-thresholds
    • request chassis fpc
    • request chassis mic
    • request chassis mic fpc-slot mic-slot

    [Junos OS System Basics Configuration Guide, Junos OS System Basics and Services Command Reference]

  • 16-Port Channelized E1/T1 Circuit Emulation MIC (MX Series routers)—Starting with Junos OS Release 12.3, the 16-Port Channelized E1/T1 Circuit Emulation MIC (MIC-3D-16CHE1-T1-CE) is supported on MX80, MX240, MX480, and MX960 routers. [See 16-Port Channelized E1/T1 Circuit Emulation MIC Overview.]
  • Extends signaling support for SAToP/CESoPSN for E1/T1 interfaces ( MX Series routers)—Starting with Junos OS Release 12.3, the E1/T1 interfaces support signaling for Structure-Agnostic TDM over Packet (SAToP) and Circuit Emulation Services over Packet-Switched Network (CESoPSN) through Layer 2 VPN using BGP.

    [See Configuring SAToP on Channelized E1/T1 Circuit Emulation MIC and Configuring CESoPSN on Channelized E1/T1 Circuit Emulation MIC.]

  • Extends support for diagnostic, OAM, and timing features to 16-port Channelized E1/T1 Circuit Emulation MIC (MX Series routers)—Starting with Junos OS Release 12.3, the 16-port Channelized E1/T1 Circuit Emulation MIC (MIC-3D-16CHE1-T1-CE) supports the following features:
    • Diagnostic features:
      • Loopback: Support for E1/T1-level payload, local line, remote line, and NxDS0 payload loopbacks.
      • Bit error rate test (BERT): Support for the following BERT algorithms:
        • pseudo-2e11-o1520
        • pseudo-2e15-o152
        • pseudo-2e20-o150
    • Operation, Administration, and Maintenance (OAM) features:
      • Performance monitoring: Supports the following Layer 1 performance-monitoring statistics at the E1/T1 interface level for all kinds of encapsulations:
        • E1 interfaces
          • BPV—Bipolar violation
          • EXZ—Excessive zeros
          • SEF—Severely errored framing
          • BEE—Bit error event
          • LCV—Line code violation
          • PCV—Pulse code violation
          • LES—Line error seconds
          • ES—Errored seconds
          • SES—Severely errored seconds
          • SEFS—Severely errored framing seconds
          • BES—Bit error seconds
          • UAS—Unavailable seconds
          • FEBE—Far-end block error
          • CRC—Cyclic redundancy check errors
          • LOFS—Loss of frame seconds
          • LOSS—Loss of signal seconds
        • T1 interfaces
          • BPV—Bipolar violation
          • EXZ—Excessive zeros
          • SEF—Severely errored framing
          • BEE—Bit error event
          • LCV—Line code violation
          • PCV—Pulse code violation
          • LES—Line error seconds
          • ES—Errored seconds
          • SES—Severely errored seconds
          • SEFS—Severely errored framing seconds
          • BES—Bit error seconds
          • UAS—Unavailable seconds
          • LOFS—Loss of frame seconds
          • LOSS—Loss of signal seconds
          • CRC—Cyclic redundancy check errors
          • CRC Major—Cyclic redundancy check major alarm threshold exceeded
          • CRC Minor—Cyclic redundancy check minor alarm threshold exceeded
    • Timing features: Support for the following transmit clocking options on the E1/T1 interface:
      • Looped timing
      • System timing

    Note: In Junos OS Release 12.3, IMA Link alarms are not supported on the 16-port Channelized E1/T1 MIC.

    [See Configuring E1 Loopback Capability, Configuring E1 BERT Properties, and Interface Diagnostics.

  • Extends support for SAToP features to 16-port Channelized E1/T1 Circuit Emulation MIC (MX Series routers)—Starting with Junos OS Release 12.3, the 16-port Channelized E1/T1 Circuit Emulation MIC (MIC-3D-16CHE1-T1-CE) supports E1/T1 SAToP features.

    [See Configuring SAToP on Channelized E1/T1 Circuit Emulation MIC.]

  • CESoPSN encapsulation support extended to 16-Port Channelized E1/T1 Circuit Emulation MIC (MX Series routers)—Starting with Junos OS Release 12.3, support for CESoPSN encapsulation is extended to the 16-port Channelized E1/T1 Circuit Emulation MIC (MIC-3D-16CHE1-T1-CE).

    [See Configuring CESoPSN on Channelized E1/T1 Circuit Emulation MIC.]

  • SNMP and MIB support (MX2020 routers)—Starting with Junos OS Release 12.3, the enterprise-specific Chassis Definitions for Router Model MIB, jnx-chas-defines.mib, is updated to include information about the new MX2020 routers. The Chassis Definitions for Router Model MIB contains the object identifiers (OIDs) used by the Chassis MIB to identify platform and chassis components of each router.

    [See jnxBoxAnatomy, Chassis Definitions for Router Model MIB, and MIB Objects for the MX2020 3D Universal Edge Router.]

  • Junos OS support for FRU management of MX2020 routers—Starting with Release 12.3, Junos OS supports the new MX2020 routers. The MX2020 routers are the next generation of MX Series 3D Universal Edge Routers. The Junos OS chassis management software for the MX2020 routers provides enhanced environmental monitoring and field-replaceable unit (FRU) control. FRUs supported on the MX2020 routers include:
    • RE and CB—Routing Engine and Control Board including a Processor Mezzanine Board (PMB)
    • PDM—Power distribution module
    • PSM—Power supply module
    • Fan trays
    • SFB—Switch Fabric Board
    • Front panel display
    • Adapter cards
    • Line cards

    The MX2020 router supports up to two Control Boards (CBs) with the second CB being used as a redundant CB. The CB provides control and monitoring functions for the router. Adapter card and switch fabric board FRU management functionality is controlled by a dedicated processor housed on the Processor Mezzanine Board. The MX2020 router supports 20 adapter cards and 8 Switch Fabric Boards (SFBs).

    The MX2020 chassis has two cooling zones. Fans operating in one zone have no impact on cooling in another zone, enabling the chassis to run fans at different speeds in different zones. The chassis can coordinate FRU temperatures in each zone and the fan speeds of the fan trays in these zones.

    The power system on the MX2020 routers consists of three components: the power supply modules (PSMs), the power distribution module (PDM), and the power midplane. The MX2020 router chassis supplies N +N feed redundancy, N +1 power supply redundancy for line cards, and N +N power supply redundancy for the critical FRUs. The critical FRUs include two CBs, eight SFBs, and three fan trays (two fan trays in one zone and one fan tray in the other zone.) In cases where all PSMs are not present, or some PSMs fail or are removed during operation, service interruption is minimized by keeping the affected FPCs online without supplying redundant power to these FPCs. You can use the following configuration statement to monitor power management on the switch chassis:

    • fru-poweron-sequence—Include the fru-poweron-sequence statement at the [edit chassis] hierarchy level to configure the power-on sequence for the FPCs in the chassis.

    Table 1: Maximum FRUs Supported on the MX2020 Router

    FRUMaximum Number
    Routing Engines and CB

    2

    PDM

    4

    PSM

    18

    Fan trays

    4

    SFB

    8

    Front panel display

    1

    Adapter cards

    20

    Line cards

    20

    The following CLI operational mode commands support the various FRU and power management operations on MX2020 routers:

    Show commands:

    • show chassis adc
    • show chassis alarms
    • show chassis environment
    • show chassis environment adc adc-slot-number
    • show chassis environment cb cb-slot-number
    • show chassis environment fpc fpc-slot-number
    • show chassis environment fpm fpm-slot-number
    • show chassis environment monitored
    • show chassis environment psm psm-slot-number
    • show chassis environment routing-engine routing-engine-slot-number
    • show chassis environment sfb sfb-slot-number
    • show chassis craft-interface
    • show chassis ethernet-switch < errors | statistics >
    • show chassis fabric destinations
    • show chassis fabric fpcs
    • show chassis fabric plane
    • show chassis fabric plane-location
    • show chassis fabric summary
    • show chassis fan
    • show chassis firmware
    • show chassis fpc < detail | pic-status | fpc-slot-number >
    • show chassis hardware < clei-models | detail | extensive | models >
    • show chassis in-service-upgrade
    • show chassis mac-addresses
    • show chassis network-services
    • show chassis pic fpc-slot fpc-slot-number pic-slot pic-slot-number
    • show chassis power
    • show chassis power sequence
    • show chassis routing-engine < routing-engine-slot-number | bios >
    • show chassis sfb < slot sfb-slot-number >
    • show chassis spmb
    • show chassis temperature-thresholds
    • show chassis zones < detail >

    Request commands:

    • request chassis cb ( offline | online ) slot slot-number
    • request chassis fabric plane ( offline | online ) fabric-plane-number
    • request chassis fpc ( offline | online | restart ) slot fpc-slot-number
    • request chassis fpm resync
    • request chassis mic ( offline | online ) fpc-slot fpc-slot-number mic-slot mic-slot-number
    • request chassis routing-engine master ( acquire | release | switch ) < no-confirm >
    • request chassis sfb ( offline | online ) slot sfb-slot-number
    • request chassis spmb restart slot spmb-slot-number

    Restart command:

    • restart chassis-control < gracefully | immediately | soft >

    For details of all system management operational mode commands and the command options supported on the MX2020 router, see the System Basics and Services Command Reference.

    [See System Basics: Chassis-Level Features Configuration Guide.]

  • SAToP support extended to Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP ( MX Series routers)—Starting with Junos OS Release 12.3R1, support for Structure-Agnostic Time-Division Multiplexing over Packet (SAToP) is extended to MIC-3D-4COC3-1COC12-CE. You can configure 336 T1 channels on each COC12 interface on this MIC.

    [See Configuring SAToP on Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP and Configuring SAToP Encapsulation on T1/E1 Interfaces on Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP.]

  • CESoPSN support extended to Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP (MX Series routers)—Starting with Junos OS Release 12.3, support for Circuit Emulation Service over Packet-Switched Network (CESoPSN) is extended to MIC-3D-4COC3-1COC12-CE. You can configure 336 CT1 channels on each COC12 interface on this MIC.

    [See Configuring CESoPSN on Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP and Configuring CESoPSN Encapsulation on DS Interfaces on Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP.]

  • Support for ATM PWE3 on Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP (MX80 routers with a modular chassis, and MX240, MX480, and MX960 routers)—Starting with Junos OS Release 12.3, ATM Pseudowire Emulation Edge to Edge (PWE3) is supported on channelized T1/E1 interfaces of the Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP (MIC-3D-4COC3-1COC12-CE). The following PWE3 features are supported:
    • ATM pseudowire encapsulation. The pseudowire encapsulation can be either cell-relay or AAL5 transport mode. Both modes enable the transport of ATM cells across a packet-switched network (PSN).
    • Cell-relay VPI/VCI swapping. The Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP can overwrite the virtual path identifier (VPI) and virtual channel identifier (VCI) header values on egress and on both ingress and egress.

      Note: Cell-relay VPI swapping on both ingress and egress is not compatible with the ATM policing feature.

      To configure the Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP to modify both the VPI and VCI header values on both ingress and egress, you must specify the psn-vci statement at the following hierarchy level:

      [edit interface at-interface-name/pic/port unit logical-unit-number]

      To configure the Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP to modify only the VPI header values on both ingress and egress, you must specify the psn-vpi statement at the following hierarchy level:

      [edit interface at-interface-name/pic/port unit logical-unit-number]

      To configure the Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP to pass the VPI and VCI header values transparently, you must specify the no-vpivci-swapping statement at the following hierarchy level:

      [edit interface at-interface-name/pic/port unit logical-unit-number]

      If none of the aforementioned configuration statements are included, for virtual path pseudowires, VPI values are modified on egress, whereas for virtual channel pseudowires, both VPI and VCI header values are modified on egress.

    [See Configuring ATM Cell-Relay Pseudowire.]

  • Pseudowire ATM MIB support for Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP (MX80 routers with a modular chassis, and MX240, MX480, and MX960 routers)—Starting with Release 12.3, Junos OS extends Pseudowire ATM MIB support to the Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP (MIC-3D-4COC3-1COC12-CE).

    [See Interpreting the Enterprise-Specific Pseudowire ATM MIB.]

  • Multiple VRRP owners per physical port—Support for multiple owner addresses per physical interface, allowing users to reuse interface address identifiers (IFAs) as virtual IP addresses (VIPs).
  • Chassis daemon enhancements for the MFC application on the Routing Engine—The chassis daemon (chassisd) process runs on the Routing Engine to communicate directly with its peer processes running on the Packet Forwarding Engine. Starting with Junos OS Release 12.1, the chassisd process has been enhanced to enable the Media Flow Controller (MFC) application to run on a Dense Port Concentrator (DPC) with an x86 blade for high application throughput and a large amount of solid state storage on MX Series routers. The chassisd process detects the installation of the modular x86 blade for MFC services and monitors the physical status of hardware components and the field-replaceable units (FRUs) that enable MFC to be run on the x86 blade.

    [System Basics]

  • Support for aggregated SONET/SDH Interfaces (MX Series Routers)—Starting with Junos OS Release 12.3, you can configure aggregated SONET bundles with the member links of SONET/SDH OC3/STM1 (Multi-Rate) MICs with SFP, that is, MIC-3D-8OC3OC12-4OC48 and MIC-3D-4OC3OC12-1OC48.

    Junos OS enables link aggregation of SONET/SDH interfaces; this is similar to Ethernet link aggregation, but is not defined in a public standard. Junos OS balances traffic across the member links within an aggregated SONET/SDH bundle based on the Layer 3 information carried in the packet. This implementation uses the same load-balancing algorithm used for per-packet load balancing.

    The following features are supported on MIC-3D-8OC3OC12-4OC48 and MIC-3D-4OC3OC12-1OC48:

    • Encapsulation—Point-to-Point Protocol (PPP) and Cisco High-Level Data Link Control (Cisco HDLC)
    • Filters and policers—Single-rate policers, three-color marking policers, two-rate three-color marking policers, hierarchical policers, and percentage-based policers. By default, policer bandwidth and burst size applied on aggregated bundles are not matched to the user-configured bandwidth and burst size.
    • Mixed mode links
  • Support for synchronizing an MX240, MX480, or MX960 router chassis with an Enhanced MX SCB to an external BITS timing source—This feature uses the Building Integrated Timing Supply (BITS) external clock interface (ECI) on the Enhanced MX SCB. The BITS ECI can also be configured to display the selected chassis clock source (SETS) or a recovered line clock source (Synchronous Ethernet or Precision Time Protocol). You can configure the BITS ECI by using the synchronization statement at the [edit chassis] hierarchy level. You can view the BITS ECI information with the show chassis synchronization extensive command.
  • Aggregated interfaces support increased to 64 links (MX Series)—This feature adds support for specifying up to 64 links for aggregated devices. You set the number of links in the new maximum-links statement at the [chassis aggregated-devices] hierarchy level.
  • Junos OS support for new MX2010 routers—Starting with Release 12.3, Junos OS supports the new MX2010 routers. The MX2010 routers are an extension of the MX2020 routers and support all features supported by the MX2020 routers. Also, the MX2010 routers support all software features that are supported by other MX Series routers in Junos OS Release 12.1.

    The power system on the MX2010 routers consists of three components: the power supply modules (PSMs), the power distribution module (PDM), and the power midplane. The power feed (AC or DC) is connected to the PDM. The PDM delivers the power from the feeds to the power midplane. The power from the power midplane is provided to the PSMs. Output from the PSMs is sent back to the power midplane and then eventually to the field-replaceable units (FRUs). The MX2010 router chassis supplies N +N feed redundancy and N + 1 PSM redundancy for line cards. In case some PSMs fail or are removed during operation, service interruption is minimized by keeping as many affected FPCs online by supplying redundant power to these FPCs. Unlike the MX2020 router chassis, the MX2010 router chassis does not provide redundancy for the critical FRUs because there is only one power zone.

    Include the following existing configuration statement at the [edit chassis] hierarchy level to configure the power-on sequence for the FPCs in the chassis:

    [edit chassis]

    fru-poweron-sequence fru-poweron-sequence

    Junos OS also supports the following CLI operational mode commands for chassis management of MX2010 routers:

    Show commands

    Request commands

    Restart commands

    show chassis adc

    request chassis cb (offline | online) slot slot-number

    restart chassis-control < gracefully | immediately | soft >

    show chassis alarms

    request chassis fabric plane (offline | online) fabric-plane-number

    show chassis environment adc <adc-slot-number>

    request chassis fpc (offline | online | restart) slot fpc-slot-number

    show chassis environment cb <cb-clot-number>

    request chassis fpm resync

    show chassis environment fpc <fpc-slot-number>

    request chassis mic (offline | online) fpc-slot fpc-slot-number mic-slot mic-slot-number

    show chassis environment fpm

    request chassis routing-engine master (acquire | release | switch) <no-confirm>

    show chassis environment monitored

    request chassis sfb (offline | online) slot sfb-slot-number

    show chassis environment psm <psm-slot-number>

    request chassis spmb restart slot spmb-slot-number

    show chassis environment routing-engine <routing-engine-slot-number>

    show chassis environment sfb <sfb-slot-number>

    show chassis environment <adc | cb | fpc | fpm | monitored | psm | routing-engine | sfb>

    show chassis craft-interface

    show chassis ethernet-switch <(errors | statistics)>

    show chassis fabric destinations <fpc fpc-slot-number>

    show chassis fabric ( destinations | fpcs | plane | plane-location | summary )

    show chassis fan

    show chassis firmware

    show chassis fpc <slot> detail | <detail <slot>> | <pic-status <slot>> |<fpc-slot-number>

    show chassis hardware < (clei-models | detail | extensive | models)>

    show chassis in-service upgrade

    show chassis mac-addresses

    show chassis network-services

    show chassis pic fpc-slot fpc-slot-number pic-slot pic-slot-number

    show chassis power <sequence>

    show chassis routing-engine <slot-number | bios>

    show chassis sfb <slot slot-number>

    show chassis spmb

    show chassis temperature-thresholds

    show chassis zones <detail>

    For details of all system management operational mode commands and the command options supported on the MX2010 router, see the System Basics and Services Command Reference.

    [System Basics and Services Command Reference]

  • SNMP and MIB support for MX2010 routers—Starting with Junos OS Release 12.3, the enterprise-specific Chassis Definitions for Router Model MIB, jnx-chas-defines.mib, is updated to include information about the new MX2010 routers. The Chassis Definitions for Router Model MIB contains the object identifiers (OIDs) used by the Chassis MIB to identify platform and chassis components of each router.

    [See jnxBoxAnatomy, Chassis Definitions for Router Model MIB, and MIB Objects for the MX2010 3D Universal Edge Router.]

  • Improvements to interface transmit statistics reporting (MX Series devices)—On MX Series devices, the logical interface-level statistics show only the offered load, which is often different from the actual transmitted load. To address this limitation, Junos OS introduces a new configuration option in Releases 11.4 R3 and 12.3 R1 and later. The new configuration option, interface-transmit-statistics at the [edit interface interface-name] hierarchy level, enables you to configure Junos OS to accurately capture and report the transmitted load on interfaces.

    When the interface-transmit-statistics statement is included at the [edit interface interface-name] hierarchy level, the following operational mode commands report the actual transmitted load:

    • show interface interface-name <detail | extensive>
    • monitor interface interface-name
    • show snmp mib get objectID.ifIndex

    Note: This configuration is not supported on Enhanced IQ (IQE) and Enhanced IQ2 (IQ2E) PICs.

    The show interface interface-name command also shows whether the interface-transmit-statistics configuration is enabled or disabled on the interface.

    [See Improvements to Interface Transmit Statistics Reporting.]

  • Extends support for encapsulating TDM signals as pseudowires for E1/T1 Circuit Emulation MIC (MX Series routers)—Starting with Junos OS Release 12.3, the Channelized E1/T1 Circuit Emulation MIC (MIC-3D-16CHE1-T1-CE) supports encapsulating structured (NxDS0) time division multiplexed (TDM) signals as pseudowires over packet-switch networks (PSNs).

    [See Configuring SAToP Emulation on T1/E1 Interfaces on Circuit Emulation PICs.]

  • RE-JCS-1X2400-48G-S Routing Engine—The JCS-1200 Control System now supports the RE-JCS-1X2400-48G-S Routing Engine. The RE-JCS-1X2400-48G-S Routing Engine requires the enhanced management module (model number MM-E-JCS-S). The RE-JCS-1X2400-48G-S Routing Engine provides a 2.4-GHz dual core Xeon processor, 48 GB of memory, and two 128 GB hot-pluggable solid state drives. The RE-JCS-1X2400-48G-S Routing Engine supports the same functionality as the other routing engines supported on the JCS-1200.

    [See JCS1200 Control System Hardware Guide PDF Document.]

  • SFPP-10GE-ZR transceiver—The following PICs on the T640, T1600, and T4000 routers now support the SFPP-10GE-ZR transceiver. The SFPP-10GE-ZR transceiver supports the 10GBASE-Z optical interface standard. For more information, see “Cables and connectors” in the PIC guide.

    T640 Router:

    • 10-Gigabit Ethernet LAN/WAN PIC with SFP+ (Model number: PD-5-10XGE-SFPP)

    T1600 Router:

    • 10-Gigabit Ethernet LAN/WAN PIC with Oversubscription and SFP+ (Model number: PD-5-10XGE-SFPP)

    T4000 Router:

    • 10-Gigabit Ethernet LAN/WAN PIC with SFP+ (Model number: PF-12XGE-SFPP)
    • 10-Gigabit Ethernet LAN/WAN PIC with Oversubscription and SFP+ (Model numbers: PD-5-10XGE-SFPP for 10-Port Type 4 PIC and PF-24XGE-SFPP for 24-Port Type 5 PIC)

    [See 10-Gigabit Ethernet 10GBASE Optical Interface Specifications, T640 Core Router PIC Guide PDF Document, T1600 Core Router PIC Guide PDF Document, and T4000 Core Router PIC Guide PDF Document.]

  • CFP-100GBASE-ER4 and CFP-100GBASE-SR10 Transceivers—The following PICs on the T1600 and T4000 routers now support the CFP-100GBASE-ER4 and CFP-100GBASE-SR10 transceivers. The CFP-100GBASE-ER4 transceiver supports the 100GBASE-ER4 optical interface standard.  The CFP-100GBASE-SR10 transceiver supports the 100GBASE-SR10 optical interface standard.  For more information, see “Cables and connectors” in the PIC guide.
    • T1600 Router: 100-Gigabit Ethernet PIC with CFP (Model number: PD-1CE-CFP-FPC4)
    • T4000 Router: 100-Gigabit Ethernet PIC with CFP (Model numbers: PF-1CGE-CFP for Type 5 and PD-1CE-CFP-FPC4 for Type 4)

    [See  100-Gigabit Ethernet 100GBASE-R Optical Interface Specifications, T1600 Core Router PIC Guide PDF Document, and T4000 Core Router PIC Guide PDF Document.]

  • Accounting of system statistics for IPv4 and IPv6 traffic—On MX Series routers, you can enable accounting of system statistics for IPv4 and IPv6 traffic by including the extended-statistics statement at the [edit chassis] hierarchy level. By default, accounting of system statistics is disabled.

    [See extended-statistics.]

  • Fabric enhancements for MX2020 and MX2010 routers —MX2020 and MX2010 routers now support all existing fabric hardening enhancements.
  • Support for unified in-service software upgrade (TX Matrix Plus router)—Starting with Junos OS Release 12.3R2, unified in-service software upgrade (unified ISSU) is supported on a routing matrix based on a TX Matrix Plus router with the TXP-T1600 configuration.

    Unified ISSU is a process to upgrade the system software with minimal disruption of transit traffic and no disruption on the control plane. In this process, the new system software version must be later than the previous system software version. When unified ISSU completes, the new system software state is identical to that of the system software when the system upgrade is performed by powering off the system and then powering it back on.

  • Enhancement to ping ethernet command—Enables you to specify a multicast MAC address. For example:
    user@host> ping ethernet maintenance-domain md3 maintenance-association ma3 01:80:c2:00:00:33
  • Symmetric load balancing on MX Series routers with MPCs—Enables support for symmetrical load balancing over 802.3ad link aggregation groups (LAGs) on MX Series routers with MPCs.

    [See Configuring Symmetrical Load Balancing on an 802.3ad Link Aggregation Group on MX Series Routers.]

  • Computation of the Layer 2 overhead attribute in interface statistics (MX Series routers)—On MX Series routers, you can configure the physical interface and logical interface statistics to include the Layer 2 overhead size (header and trailer bytes) for both ingress and egress interfaces. Both the transit and total statistical information are computed and displayed for each logical interface. This functionality is supported on 1-Gigabit and 10-Gigabit Ethernet interfaces on Dense Port Concentrators (DPCs) and Modular Port Concentrators (MPCs).

    You can enable the Layer 2 overhead bytes for computation in the logical interface statistics by configuring the account-layer2-overhead (value | <ingress bytes | egress bytes>) statement at the [edit interface interface-name unit logical-unit-number] hierarchy level. If you configure this capability, all the Layer 2 header details (L2 header and cyclic redundancy check [CRC]) based on the Layer 2 encapsulation configured for an interface are calculated and displayed in the physical and logical interface statistics for ingress and egress interfaces in the output of the show interfaces interface-name commands. For physical and logical interfaces, the Input bytes and Output bytes fields under the Traffic statistics section in the output of the show interfaces interface-name <detail | extensive> command include the Layer 2 overhead of the packets. For physical and logical interfaces, the Input Rate and Output Rate fields under the Traffic statistics section in the output of the show interfaces interface-name <media | statistics> command include the Layer 2 overhead of the packets. For logical interfaces, the values for the newly added Egress accounting overhead and Ingress accounting overhead fields display the Layer 2 overhead size for transmitted and received packets respectively.

    The ifInOctets and the ifOutOctets MIB objects display statistics that include Layer 2 overhead bytes if you configured the setting to account for Layer 2 overhead at the logical interface level.

  • New label-switching router (LSR) FPC (model number T4000-FPC5-LSR)—The new LSR FPC in a T4000 core router provides LSR capability with the following scaling numbers:

    Feature

    LSR FPC Scale

    RIB capacity

    28 million

    FIB (IPv4 and IPv6 unicast)

    64,000

    MPLS label push

    48,000

    MPLS label FIB/MPLS swap table

    256,000

    IP multicast route capacity

    256,000

    Multicast forwarding table (S, G)

    128,000

    RSVP LSPs

    32,000 (ingress/egress)

    64,000 (transit)

    Layer 2 VPN ingress/egress with family CCC/TCC

    8000

    The LSR FPC operates in the following modes:

    • Packet transport mode—When the LSR FPC operates as an LSR only, the LSR FPC scaling numbers are supported.
    • Converged P/PE mode—In a mixed provider (P) and provider edge (PE) router deployment, the LSR FPC might receive routes and next hops that exceed the LSR scaling numbers that are supported. In that case, extended scaling numbers, such as for the T4000-FPC5-3D, are supported.
    • PE router mode—In PE router mode, running a Layer 3 VPN or peering services from the LSR FPC is not supported.

    [See the T4000 Core Router Hardware Guide PDF Document.]

  • Support for new fixed-configuration MPC on MX240, MX480, MX960, and MX2020 routers—MX2020, MX960, MX480, and MX240 routers support a new MPC, MPC4E. MPC4E provides scalability in bandwidth and services capabilities of the routers. MPC4E, like other MPCs, provides the connection between the customer's Ethernet interfaces and the routing fabric of the MX Series chassis. MPC4E is a fixed-configuration MPC and does not contain separate slots for Modular Interface Cards (MICs). MPC4E is available in two models: MPC4E-3D-32XGE-SFPP and MPC4E-3D-2CGE-8XGE.

    MPC4E, like MPC3E, requires the Enhanced MX Switch Control Board (SCBE) for fabric redundancy. MPC4E does not support legacy SCBs. MPC4E interoperates with existing MX Series line cards, including Dense Port Concentrators (DPCs) and Modular Port Concentrators (MPCs).

    MPC4E contains two Packet Forwarding Engines––PFE0 hosts PIC0 and PIC1 while PFE1 hosts PIC2 and PIC3.

    MPC4E supports:

    • Forwarding capability of up to 130 Gbps per Packet Forwarding Engine. On MX240, MX480, and MX960 routers with Enhanced MX Switch Control Boards (SCBEs), each Packet Forwarding Engine can forward up to 117 Gbps because of the Packet Forwarding Engine and fabric limitations. On M2020 routers, each Packet Forwarding Engine can forward up to 130 Gbps.
    • Both 10-Gigabit Ethernet interfaces and 100-Gigabit Ethernet interfaces.
    • Small form-factor pluggable (SFP) and C form-factor pluggable (CFP) transceivers for connectivity.
    • Up to 240 Gbps of full-duplex traffic.
    • Intelligent oversubscription services.
    • WAN-PHY mode on 10-Gigabit Ethernet Interfaces on a per-port basis.
    • Up to four full-duplex tunnel interfaces on each MPC4E.
    • Effective line rate of 200 Gbps for packets larger than 300 bytes.

    MPC4E supports feature parity with the Junos OS Release 12.3 software features:

    • Basic Layer 2 features and virtual private LAN service (VPLS) functionality, including Operation, Administration, and Maintenance (OAM)
    • Class-of-service (CoS) support
    • Firewall filters and policers
    • Interoperability with existing DPCs and MPCs
    • Internet Group Management Protocol (IGMP) snooping with bridging, integrated routing and bridging (IRB), and VPLS
    • Layer 3 routing protocols
    • J-Flow monitoring and services
    • MPLS
    • Multicast forwarding
    • Precision Time Protocol (IEEE 1588)
    • Tunnel Interfaces support

    The following features are not supported on the MPC4E:

    • Fine-grained queuing and input queuing
    • Intelligent hierarchical policers
    • Layer 2 trunk port
    • MPLS fast reroute (FRR) VPLS instance prioritization
    • Multilink services
    • Virtual Chassis support

    For more information about the supported and unsupported Junos OS software features for this MPC, see Protocols and Applications Supported by the MX240, MX480, MX960, and MX2000 MPC4E in the MX Series Line Card Guide.

  • Support for Ethernet synthetic loss measurement—You can trigger on-demand and proactive Operations, Administration, and Maintenance (OAM) for measurement of statistical counter values corresponding to ingress and egress synthetic frames. Frame loss is calculated using synthetic frames instead of data traffic. These counters maintain a count of transmitted and received synthetic frames and frame loss between a pair of maintenance association end points (MEPs).

    The Junos OS implementation of Ethernet synthetic loss measurement (ETH-SLM) is fully compliant with the ITU-T Recommendation Y.1731. Junos OS maintains various counters for ETH-SLM PDUs, which can be retrieved at any time for sessions that are initiated by a certain MEP. You can clear all the ETH-SLM statistics and PDU counters.

    The ETH-SLM feature provides the option to perform ETH-SLM for a given 802.1p priority; to set the size of the ETM-SLM protocol data unit (PDU); and to generate XML output.

    You can perform ETH-SLM in on-demand ETH-SLM mode (triggered through the CLI) or in proactive ETH-SLM mode (triggered by the iterator application). To trigger synthetic frame loss measurement (on-demand mode) and provide a run-time display of the measurement values, use the monitor ethernet synthetic-loss-measurement (remote-mac-address | mep mep-id) maintenance-domain md-name maintenance-association ma-name count frame-count wait time priority 802.1p value size xml operational mode command.

    To display the archived on-demand synthetic frame loss measurement values, use the show oam ethernet connectivity-fault-management synthetic-loss-statistics maintenance-domain md-name maintenance-association ma-name local-mep local-mep-id remote-mep remote-mep-id count entry-count operational mode command. To display the cumulative on-demand synthetic frame loss measurement values, use the show oam ethernet connectivity-fault-management interfaces detail operational mode command.

    To perform proactive ETH-SLM, you need to create an SLA iterator profile and associate the profile with a remote MEP. To create an SLA iterator profile for ETH-SLM, include the measurement-type slm statement at the [edit protocols oam ethernet connectivity-fault-management performance-monitoring sla-iterator-profiles profile-name] hierarchy level. To display proactive synthetic loss measurement values, use the show oam ethernet connectivity-fault-management sla-iterator-statistics maintenance-domain md-name maintenance-association ma-name local-mep local-mep-id remote-mep remote-ip-id sla-iterator identifier operational mode command.

    You can reset the SLM statistics by clearing the currently measured ETH-SLM statistical counters. To clear the existing on-demand Ethernet loss statistics measured for a specific maintenance domain and maintenance association local and remote MEP and restart the counter, use the clear oam ethernet connectivity-fault-management synthetic-loss-measurement maintenance-domain md-name maintenance-association ma-name local-mep local-mep remote-mep remote-mep operational mode command. To clear the existing proactive ETH-SLM counters for a specific maintenance domain, maintenance association, local MEP, remote MEP, and an SLA iterator, use the clear oam ethernet connectivity-fault-management sla-iterator-statistics maintenance-domain md-name maintenance-association ma-name local-mep local-mep-id remote-mep remote-mep-id sla-iterator identifier operational mode command.

    The following list consists of the connectivity fault management (CFM)-related operational mode commands that display ETH-SLM statistics:

    • The show oam ethernet connectivity-fault-management interfaces detail command is enhanced to display on-demand ETH-SLM statistics for MEPs in the specified CFM maintenance association within the specified CFM maintenance domain.
    • The show oam ethernet connectivity-fault-management mep-statistics command is enhanced to display on-demand ETH-SLM statistics and frame counts for MEPs in the specified CFM maintenance association within the specified CFM maintenance domain.
    • The show oam ethernet connectivity-fault-management mep-database command is enhanced to display on-demand ETH-SLM frame counters for MEPs in the specified CFM maintenance association within the specified CFM maintenance domain.
    • The show oam ethernet connectivity-fault-management sla-iterator-statistics command is enhanced to display service-level agreement (SLA) iterator statistics for ETH-SLM.

    [Release Notes]

  • Support for OSS mapping to represent a T4000 chassis as a T1600 or a T640 chassis (T4000 routers)—Starting with Junos OS Release 12.3R3, you can map a T4000 chassis to a T1600 chassis or a T640 chassis, so that the T4000 chassis is represented as a T1600 chassis or a T640 chassis, respectively, without changing the operations support systems (OSS) qualification. Therefore, you can avoid changes to the OSS when a T1600 chassis or a T640 chassis is upgraded to a T4000 chassis.

    You can configure the OSS mapping feature with the set oss-map model-name t640|t1600 configuration command at the [edit chassis] hierarchy level. This command changes the chassis field to the known chassis field in the output of the show chassis hardware and the show chassis oss-map operational mode commands. You can verify the change with the show snmp mib walk system and show snmp mib walk jnxBoxAnatomy operational commands as well.

    You can delete the OSS mapping feature by using the delete chassis oss-map model-name t640|t1600 configuration command.

  • Enhanced load balancing for MIC and MPC interfaces (MX Series) — Starting with Junos OS Release 12.3R4, the following load-balancing solutions are supported on an aggregated Ethernet bundle to correct genuine traffic imbalance among the member links:
    • Adaptive — Uses a real-time feedback and control mechanism to monitor and manage traffic imbalances.
    • Per-packet random spray — Randomly sprays the packets to the aggregate next hops to ensure that the next hops are equally loaded resulting in packet reordering.

    The aggregated Ethernet load-balancing solutions are mutually exclusive. To configure these solutions, include the adaptive or per-packet statement at the [edit interfaces aex aggregated-ether-options load-balance] hierarchy level.

  • SFPP-10G-CT50-ZR (MX Series)—The SPFF-10G-CT50-ZR tunable transceiver provides a duplex LC connector and supports the 10GBASE-Z optical interface specification and monitoring. The transceiver is not specified as part of the 10-Gigabit Ethernet standard and is instead built according to Juniper Networks specifications. Only WAN-PHY and LAN-PHY modes are supported. To configure the wavelength on the transceiver, use the wavelength statement at the [edit interfaces interface-name optics-options] hierarchy level. The following interface module supports the SPFF-10G-CT50-ZR transceiver:

    MX Series:

    • 16-port 10-Gigabit Ethernet MPC (model number: MPC-3D-16XGE-SFPP)—Supported in Junos OS Release 12.3R6, 13.2R3, 13.3R2, 14.1, and later.

    For more information about interface modules, see the “Cables and Connectors” section in the Interface Module Reference for your router.

    [See 10-Gigabit Ethernet 10GBASE Optical Interface Specifications and wavelength.]

  • SFPP-10G-ZR-OTN-XT (MX Series, T1600, and T4000)—The SFPP-10G-ZR-OTN-XT dual-rate extended temperature transceiver provides a duplex LC connector and supports the 10GBASE-Z optical interface specification and monitoring. The transceiver is not specified as part of the 10-Gigabit Ethernet standard and is instead built according to ITU-T and Juniper Networks specifications. In addition, the transceiver supports LAN-PHY and WAN-PHY modes and OTN rates and provides a NEBS-compliant 10-Gigabit Ethernet ZR transceiver for the MX Series interface modules listed below. The following interface modules support the SFPP-10G-ZR-OTN-XT transceiver:

    MX Series:

    • 10-Gigabit Ethernet MIC with SFP+ (model number: MIC3-3D-10XGE-SFPP)—Supported in Junos OS Release 12.3R5, 13.2R3, 13.3, and later
    • 16-port 10-Gigabit Ethernet (model number: MPC-3D-16XGE-SFPP)—Supported in Junos OS Release 12.3R5, 13.2R3, 13.3, and later
    • 32-port 10-Gigabit Ethernet MPC4E (model number: MPC4E-3D-32XGE-SFPP)—Supported in Junos OS Release 12.3R5, 13.2R3, 13.3, and later
    • 2-port 100-Gigabit Ethernet + 8-port 10-Gigabit Ethernet MPC4E (model number: MPC4E-3D-2CGE-8XGE)—Supported in Junos OS Release 12.3R5, 13.2R3, 13.3, and later

    T1600 and T4000 routers:

    • 10-Gigabit Ethernet LAN/WAN PIC with Oversubscription and SFP+ (model numbers: PD-5-10XGE-SFPP and PF-24XGE-SFPP)—Supported in Junos OS Release 12.3R5, 13.2R3, 13.3, and later
    • 10-Gigabit Ethernet LAN/WAN PIC with SFP+ (model number: PF-12XGE-SFPP)—Supported in Junos OS Release 12.3R5, 13.2R3, 13.3, and later

    For more information about interface modules, see the “Cables and Connectors” section in the Interface Module Reference for your router.

    [See 10-Gigabit Ethernet 10GBASE Optical Interface Specifications.]

Junos OS XML API and Scripting

  • Support for service template automation—Starting with Junos OS Release 12.3, you can use service template automation to provision services such as VPLS VLAN, Layer 2 and Layer 3 VPNs, and IPsec across similar platforms running Junos OS. Service template automation uses the service-builder.slax op script to transform a user-defined service template definition into a uniform API, which you can then use to configure and provision services on similar platforms running Junos OS. This permits you to create a service template on one device, generalize the parameters, and then quickly and uniformly provision that service on other devices. This decreases the time required to configure the same service on multiple devices, and reduces configuration errors associated with manually configuring each device.

    [See Service Template Automation.]

  • Support for configuring limits on concurrently running event policies and memory allocation for scripts—Junos OS Release 12.3 supports configuring limits on the maximum number of concurrently running event policies and the maximum amount of memory allocated for the data segment for scripts of a given type. By default, the maximum number of event policies that can run concurrently in the system is 15, and the maximum amount of memory allocated for the data segment portion of an executed script is half of the total available memory of the system, up to a maximum value of 128 MB.

    To set the maximum number of event policies that can run concurrently on a device, configure the max-policies policies statement at the [edit event-options] hierarchy level. You can configure a maximum of 0 through 20 policies. To set the maximum memory allocated to the data segment for scripts of a given type, configure the max-datasize size statement under the hierarchy appropriate for that script type, where size is the memory in bytes. To specify the memory in kilobytes, megabytes, or gigabytes, append k, m, or g, respectively, to the size. You can configure the memory in the range from 2,3068,672 bytes (22 MB) through 1,073,741,824 bytes (1 GB).

    [See Configuring Limits on Executed Event Policies and Memory Allocation for Scripts.]

Layer 2 Features

  • Support for Synchronous Ethernet and Precision Time Protocol on MX Series routers with 16-port Channelized E1/T1 Circuit Emulation MIC (MX Series Routers)—Starting with Junos OS Release 12.3, Synchronous Ethernet and Precision Time Protocol (PTP) are supported on MX Series routers with the 16-port Channelized E1/T1 Circuit Emulation MIC (MIC-3D-CH-16E1T1-CE). The clock derived by Synchronous Ethernet, PTP, or an internal oscillator is used to drive the T1/E1 interfaces on the 16-port Channelized E1/T1 Circuit Emulation MIC.
  • E-TREE with remote VSI support on NSN Carrier Ethernet Transport—The NSN Carrier Ethernet Transport solution supports Metro Ethernet Forum (MEF) Ethernet Tree (E-TREE) services using centralized virtual switch instances (VSIs). E-TREE is a rooted multipoint service, where end points are classified as Roots and Leaves. Root end points can communicate with both Root and Leaf end points, but Leaf end points can only communicate with the Root end points.

    The NSN CET solution employs E-TREE services using a centralized VSI model. This means that VSIs are only provisioned on certain selected PEs. End points are connected to these central VSIs using spoke pseudowires. The centralized VSI model uses a lower number of pseudowires and less bandwidth than the distributed VSI model.

  • Adds support to send and receive untagged RSTP BPDUs on Ethernet interfaces (MX Series platforms)–VLAN Spanning Tree Protocol (VSTP) can send and receive untagged Rapid Spanning Tree Protocol (RSTP) bridge protocol data units (BPDUs) on Gigabit Ethernet (ge), 10 Gigabit Ethernet (xe), and aggregated Ethernet (ae) interfaces.

    To configure this feature, include the access-trunk statement at the following hierarchy levels:

    [edit protocols vstp vlan vlan-identifier interface interface-name][edit routing-instances routing-instance-name instance-type (layer2-control | virtual-switch)][edit logical-systems logical-system-name protocols vstp][edit logical-systems logical-system-name routing-instances routing-instance-name protocols vstp]

    [See access-trunk.]

  • Extends support for multilink-based protocols on T4000 and TX Matrix Plus routers—Starting with Junos OS Release 12.3R3, multilink-based protocols are supported on the T4000 and TX Matrix Plus routers with Multiservices PICs.
    • Multilink Point-to-Point Protocol (MLPPP)—Supports Priority-based Flow Control (PFC) for data packets and Link Control Protocol (LCP) for control packets. Compressed Real-Time Transport Protocol (CRTP) and Multiclass MLPPP are supported for both data and control packets.
    • Multilink Frame Relay (MLFR) end-to-end (FRF.15)—Supports Ethernet Local Management Interface (LMI), Consortium LMI (C-LMI), and Link Integrity Protocol (LIP) for data and control packets.
    • Multilink Frame Relay (MFR) UNI NNI (FRF.16)—Supports Ethernet Local Management Interface (LMI), Consortium LMI (C-LMI), and Link Integrity Protocol (LIP) for data and control packets.
    • Link fragmentation and interleaving (LFI) non multilink MLPPP and MLFR packets.
    • Communications Assistance for Law Enforcement Act (CALEA)--Defines electronic surveillance guidelines for telecommunications companies.
    • Two-Way Active Measurement Protocol (TWAMP)-- Adds two-way or round-trip  measurement capabilities

    [Interfaces Command Reference]

  • Extends support of IPv6 statistics for MLPPP bundles on T4000 and TX Matrix Plus routers—Starting with Junos OS Release 12.3R3, the show interfaces lsq-fpc/pic/port command displays the packet and byte counters for IPv6 data for Multilink Point-to-Point Protocol (MLPPP) bundles on link services intelligent queuing (LSQ) interfaces.

    [Interfaces Command Reference]

  • Link Layer Discovery Protocol (LLDP) support (MX240, MX480, and MX960)—You can configure the LLDP protocol on MX Series routers with MPC3E and MPC4E. To configure and adjust default parameters, include the lldp statement at the [edit protocols] hierarchy level.

    LLDP is disabled by default. At the [edit protocols lldp] hierarchy level, use the enable statement to enable LLDP, and the interfaces statement to enable LLDP on all or some interfaces. Use the following statements at the [edit protocols lldp] hierarchy level to configure or adjust the default LLDP parameters:

    • advertisement-interval
    • hold-multiplier
    • lldp-configuration-notification-interval.
    • ptopo-configuration-maximum-hold-time
    • ptopo-configuration-trap-interval
    • transmit-delay
  • Configuration support for manual and automatic link switchover mechanism on multichassis link aggregation interface—You can configure a multichassis link aggregation (MC-LAG) interface in active-standby mode to automatically revert to a preferred node. In an MC-LAG topology with active-standby mode, a link switchover happens only if the active node goes down. With this configuration, you can trigger a link switchover to a preferred node even when the active node is available.

    To enable automatic link switchover for an multichassis link aggregation (mc-ae) interface, you must configure the switchover-mode revertive statement at the [edit interfaces aex aggregated-ether-options mc-ae] hierarchy level. You can also specify the revert time for the switchover by using the revert-time statement. To continue using the manual switchover mechanism, you must configure the switchover-mode non-revertive statement at the [edit interfaces aex aggregated-ether-options mc-ae] hierarchy level. For nonrevertive mode, you can configure manual switchover to the preferred node by using the switchover immediate and mc-ae-id statements at the [request interface mc-ae] hierarchy level.

    With this feature, you can use the show interfaces mc-ae revertive-info command to view the switchover configuration information.

  • Uniform Enhanced Layer 2 Software CLI configuration statements and operational commands—Enhanced Layer 2 Software (ELS) provides a uniform CLI for configuring and monitoring Layer 2 features on MX Series routers in LAN mode (MX-ELM). With ELS, for example, you can configure a VLAN and other Layer 2 features on an MX-ELM router by using the same configuration commands.

    [See the ELS CLI documentation for MX Series routers: Junos OS for EX9200 Switches, Release 12.3.]

  • When changing modes the user must delete any unsupported configurations.

    [See Configuring MX Enhanced LAN Mode.]

  • The web-based ELS Translator tool is available for registered customers to help them become familiar with the ELS CLI and to quickly translate existing MX Series router CLI configurations into ELS CLI configurations.

    [See ELS Translator.]

Layer 2 Tunneling Protocol

  • Support for filtering trace results by subscriber or domain for AAA, L2TP, and PPP (MX Series routers)—You can now filter trace results for AAA (authd), L2TP (jl2tpd), and PPP (jpppd) by subscribers or domains. Specify the filter user username option at the appropriate hierarchy level:
    • AAA—[edit system processes general-authentication-service traceoptions filter]
    • L2TP—[edit services l2tp traceoptions filter]
    • PPP—[edit protocols ppp-service traceoptions filter]

    For subscriber usernames that have the expected form of user@domain, you can filter on either the user or the domain. The filter supports the use of a wildcard (*) at the beginning or end of the user, the domain, or both. For example, the following are all acceptable uses of the wildcard: tom@example.com, tom*, *tom, *ample.com, tom@ex*, tom*@*example.com.

    You cannot filter results using a wildcard in the middle of the user or domain. For example, the following uses of the wildcard are not supported: tom*25@example.com, tom125@ex*.com.

    When you enable filtering by username, traces that have insufficient information to determine the username are automatically excluded.

MPLS

  • Link protection for MLDP—MLDP link protection enables fast reroute of traffic carried over LDP LSPs in case of a link failure. LDP point-to-multipoint LSPs can be used to send traffic from a single root or ingress node to a number of leaf nodes or egress nodes traversing one or more transit nodes. When one of the links of the point-to-multipoint tree fails, the subtrees may get detached until the IGP reconverges and MLDP initiates label mapping using the best path from the downstream to the new upstream router. To protect the traffic in the event of a link failure, you can configure an explicit tunnel so that traffic can be rerouted using the tunnel. Junos OS supports make-before-break (MBB) capabilities to ensure minimum packet loss when attempting to signal a new LSP path before tearing down the old LSP path. This feature also adds targeted LDP support for MLDP link protection.

    To configure MLDP link protection, use the make-before-break, and link-protection-timeout statements at the [edit protocols ldp] hierarchy level. To view MBB capabilities, use the show ldp session detail command. To verify that link protection is active, use the show ldp interface extensive command. To view adjacency type, use the show ldp neighbor extensive command. To view MBB interval, use the show ldp overview command.

    [See Example: Configuring LDP Link Protection.]

  • Ultimate-hop popping feature now available for LSPs configured on M Series, MX Series, and T Series platforms—An ultimate-hop popping LSP pops the MPLS label at the LSP egress. The default behavior for an LSP on a Juniper Networks device is to pop the MPLS label at the penultimate-hop router (the router before the egress router). Ultimate-hop popping is available on RSVP-signaled LSPs and static LSPs.

    The following network applications could require that you configure UHP LSPs:

    • MPLS-TP for performance monitoring and in-band OAM
    • Edge protection virtual circuits
    • UHP static LSPs

    To enable ultimate-hop popping on an LSP, include the ultimate-hop-popping statement at the [edit protocols mpls label-switched-path lsp-name] hierarchy level to enable ultimate-hop popping on a specific LSP or at the [edit protocols mpls] hierarchy level to enable ultimate-hop popping on all of the ingress LSPs configured on the router. When you enable ultimate-hop popping, RSVP attempts to resignal existing LSPs as ultimate-hop popping LSPs in a make-before-break fashion. If an egress router does not support ultimate-hop popping, the existing LSP is torn down. If you disable ultimate-hop popping, RSVP resignals existing LSPs as penultimate-hop popping LSPs in a make-before-break fashion.

    [See Configuring Ultimate-Hop Popping for LSPs.]

  • Enable local receivers on the ingress of a point-to-multipoint circuit cross-connect (CCC)—This feature enables you to switch the traffic entering a P2MP LSP to local interfaces. On the ingress PE router, CCC can be used to switch an incoming CCC interface to one or more outgoing CCC interfaces. To configure the output interface, include the output-interface statement at the [edit protocols connections p2mp-transmit-switch <p2mp-lsp-name-on-which-to-transmit>] hierarchy level. One or more output interfaces can be configured as local receivers on the ingress PE router using this statement. Use the show connections p2mp-transmit-switch (extensive | history | status), show route ccc <interface-name> (detail | extensive), and show route forwarding-table ccc <interface-name> (detail | extensive) commands to view details of the local receiving interfaces at ingress.

    [MPLS]

  • Support for Bidirectional Forwarding Detection protocol, LSP traceroute, and LSP ping on Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP (MX Series Routers)—Starting with Junos OS 12.3, support for Bidirectional Forwarding Detection (BFD) protocol, LSP traceroute, and LSP ping is extended to Channelized OC3/STM1 (Multi-Rate) Circuit Emulation MIC with SFP (MIC-3D-4COC3-1COC12-CE).

    The BFD protocol is a simple hello mechanism that detects failures in a network. You can configure Bidirectional Forwarding Detection (BFD) for LDP LSPs. You can also use the LSP ping commands to detect LSP data plane faults. You can trace the route followed by an LDP-signaled LSP.

    LDP LSP traceroute is based on RFC 4379, Detecting Multi-Protocol Label Switched (MPLS) Data Plane Failures. This feature allows you to periodically trace all paths in a Forwarding Equivalence Class (FEC). The FEC topology information is stored in a database accessible from the CLI.

  • Host fast reroute (HFRR)—Adds a precomputed protection path into the Packet Forwarding Engine, such that if a link between a provider edge device and a server farm becomes unusable for forwarding, the Packet Forwarding Engine can use another path without having to wait for the router or the protocols to provide updated forwarding information. HFRR is a technology that protects IP endpoints on multipoint interfaces, such as Ethernet. This technology is important in data centers where fast service restoration for server endpoints is critical. After an interface or a link goes down, HFRR enables the local repair time to be approximately 50 milliseconds. You can configure HFRR by adding the link-protection statement to the interface configuration in the routing instance. We recommend that you include this statement on all provider edge (PE) devices that are connected to server farms through multipoint interfaces.

    [See Example: Configuring Host Fast Reroute.]

  • Support of Path Computation Element Protocol for RSVP-TE—Starting with Junos OS Release 12.3, the MPLS RSVP-TE functionality is extended to provide a partial client-side implementation of the stateful Path Computation Element (PCE) architecture (draft-ietf-pce-stateful-pce). The PCE computes path for the traffic engineered LSPs (TE LSPs) of ingress routers that have been configured for external control. The ingress router that connects to a PCE is called a Path Computation Client (PCC). The PCC is configured with the Path Computation Client Protocol (PCEP) (defined in RFC 5440, but limited to the functionality supported on a stateful PCE only) to facilitate external path computing by a PCE.

    In this new functionality, the active stateful PCE sets parameters for the PCC's TE LSPs, such as bandwidth, path (ERO), and priority. The TE LSP parameters configured from the PCC's CLI are overridden by the PCE-provided parameters. The PCC re-signals the TE LSPs based on the path specified by the PCE. Since the PCE has a global view of the bandwidth demand in the network and performs external path computations after looking up the traffic engineering database, this feature provides a mechanism for offline control of TE LSPs in the MPLS RSVP TE enabled network.

    To enable external path computing by a PCE, include the lsp-external-controller statement on the PCC at the [edit mpls] and [edit mpls lsp lsp-name] hierarchy levels. To enable PCE to PCC communication, configure pcep on the PCC at the [edit protocols] hierarchy level.

    [See PCEP Configuration Guide.]

Multicast

  • Redundant virtual tunnel (VT) interfaces in Multiprotocol BGP (MBGP) multicast VPNs (MVPNs)—VT interfaces are needed for multicast traffic on routing devices that function as combined provider edge (PE) and provider core (P) routers to optimize bandwidth usage on core links. VT interfaces prevent traffic replication when a P router also acts as a PE router (an exit point for multicast traffic). You can configure up to eight VT interfaces in a routing instance, thus providing Tunnel PIC redundancy inside the same multicast VPN routing instance. When the active VT interface fails, the secondary one takes over, and you can continue managing multicast traffic with no duplication. To configure, include multiple VT interfaces in the routing instance and, optionally, apply the primary statement to one of the VT interfaces.

    [See Example: Configuring Redundant Virtual Tunnel Interfaces in MBGP MVPNs.]

  • Enhancements to RPD_MC_OIF_REJECT and RPD_MC_OIF_RE_ADMIT system log messages—When multicast call admission control (CAC) is enabled on an interface, the routing software cannot add a multicast flow to that interface if doing so exceeds the maximum configured bandwidth for that interface. Consequently, the interface is rejected for that flow due to insufficient bandwidth, and the router writes the RPD_MC_OIF_REJECT system log message to the log file at the info severity level. When bandwidth again becomes available on the interface, interfaces previously rejected for a flow are readmitted. In that case, the router writes the RPD_MC_OIF_RE_ADMIT system log message to the log file at the info severity level.

    Both the RPD_MC_OIF_REJECT system log message and the RPD_MC_OIF_RE_ADMIT system log message include the interface-name. In RPD_MC_OIF_REJECT messages, interface-name identifies the interface that was rejected for a multicast flow due to insufficient bandwidth. In RPD_MC_OIF_RE_ADMIT messages, interface-name identifies the interface that was re-admitted for a multicast flow due to newly available bandwidth on the interface.

    The RPD_MC_OIF_REJECT and RPD_MC_OIF_RE_ADMIT system log messages have been enhanced in this release to include the following information in addition to the interface-name:

    • group-address—IP address of the multicast group
    • source-address—Source IP address of the multicast flow
    • flow-rate—Bandwidth of the multicast flow, in bits per second (bps)
    • maximum-flow-rate—Maximum bandwidth that is the sum of all multicast flows on the interface, in bps
    • admitted-flow-rate—Admitted bandwidth that is the sum of all multicast flows on the interface, in bps

    When the maximum allowable bandwidth is exceeded on the logical interface (also known as the map-to interface) to which an outgoing interface (OIF) map directs (maps) multicast traffic, the RPD_MC_OIF_REJECT and RPD_MC_OIF_RE_ADMIT system log messages also display the oif-map-interface-name string. The oif-map-interface-name is an optional string that identifies one or more subscriber interfaces that requested the multicast traffic and are associated with the OIF map.

    The oif-map-interface-name string appears in the RPD_MC_OIF_REJECT and RPD_MC_OIF_RE_ADMIT system log messages when all of the following conditions are met:

    • The subscriber interface has CAC enabled, and is associated with an OIF map.
    • The map-to interface (also known as the multicast VLAN, or M-VLAN) has CAC enabled.
    • The subscriber interface that maps traffic to the M-VLAN receives an IGMP or MLD join message.
    • The M-VLAN is not already sending the group and source of the multicast flow.
    • Adding a multicast flow to the M-VLAN exceeds the maximum bandwidth configured on the M-VLAN interface.
    • The group and source is source-specific multicast (SSM), or multicast data traffic is flowing.

    Being able to view all of this information in a single system log message makes it easier for you to identify, troubleshoot, and resolve problems when using multicast protocols in your network. In earlier Junos OS releases, the RPD_MC_OIF_REJECT and RPD_MC_OIF_RE_ADMIT system log messages included only interface-name, but did not include group-address, source-address, or oif-map-interface-name.

    The following example shows an RPD_MC_OIF_REJECT system log message for an oversubscribed interface. Because the interface is not configured with an OIF map, the oif-map-interface-name string does not appear.

    Oct 26 08:09:51 wfpro-mx1-c r1:rpd[12955]: RPD_MC_OIF_REJECT: 225.1.0.2 193.0.1.2 (5000000 bps) rejected due to lack of bandwidth on ge-4/1/0.1 (maximum 12000000 bps, admitted 10000000 bps)

    This example includes the following information:

    • The group-address is 225.1.0.2
    • The source-address is 193.0.1.2
    • The flow-rate is 5000000 bps
    • The interface-name is ge-4/1/0.1
    • The maximum-flow-rate is 12000000 bps
    • The admitted-flow-rate is 10000000 bps

    The following example shows the same RPD_MC_OIF_REJECT system log message for an interface configured with an OIF map. All of the values are the same as in the preceding RPD_MC_OIF_REJECT example except for the addition of the oif-map-interface-name string, which is requested from ge-4/1/0.1 ge-4/1/0.2 ge-4/1/0.3.

    Oct 26 08:17:05 wfpro-mx1-c r1:rpd[15133]: RPD_MC_OIF_REJECT: 225.1.0.2 193.0.1.2 (5000000 bps) rejected due to lack of bandwidth on ge-4/1/0.4 (maximum 12000000 bps, admitted 10000000 bps) requested from ge-4/1/0.1 ge-4/1/0.2 ge-4/1/0.3

    The enhancements to the RPD_MC_OIF_REJECT and RPD_MC_OIF_RE_ADMIT system log messages make no changes to how bandwidth is managed on the router for multicast configurations.

    [Multicast Protocols Configuration Guide]

  • Static ARP with multicast MAC address for an IRB interface—Enables you to configure a static ARP entry with a multicast MAC address for an IRB interface which acts as the gateway to the network load balancing (NLB) servers. Earlier, the NLB servers dropped packets with a unicast IP address and a multicast MAC address. Junos OS 12.3 supports the configuration of a static ARP with a multicast MAC address.

    To configure a static ARP entry with a multicast MAC address for an IRB interface, configure the ARP entry at the [edit interfaces irb unit logical-unit-number family inet address address] hierarchy level.

    irb {unit logical-unit-number{family inet {address address{arp address multicast-mac mac-add;}}}}

Power Management

  • Power management support on T4000 routers with six-input DC power supply —Starting with Junos OS Release 12.3, the power management feature is enabled on a Juniper Networks T4000 Core Router. This feature enables you to limit the overall chassis output power consumption. That is, power management enables you to limit the router from powering on a Flexible PIC Concentrator (FPC) when sufficient output power is not available to power on the FPC.

    The power management feature is enabled only when six input feeds with 40 amperes (A) each or four input feeds with 60 A each are configured on the router. The power management feature is not enabled for any other input feed–current combination. When the power management feature is not enabled, Junos OS tries to power on all the FPCs connected to the router.

    Caution: If you do not configure the power management feature and the maximum power draw is exceeded by the router, FPCs’ states might change from Online to Offline or Present, some traffic might drop, or the interfaces might flap.

    After you connect the input feeds to the router, you must configure the number of input feeds connected to the router and the amount of current received at the input feeds. Use the feeds statement and the input current statement at the [edit chassis pem] hierarchy level to configure the number of input feeds and the amount of current received at the input feeds, respectively.

    Note: You can connect three 80 A DC power cables to the six-input DC power supply by using terminal jumpers. When you do this, ensure that you set the value of feeds statement to 6 and that of the input current statement to 40. If these configurations are not set, the power management feature is not enabled and, therefore, Junos OS tries to power on all the FPCs connected to the router.

    When the power management feature is enabled, FPCs connected to the router are powered on based on the power received by the router. If the router receives sufficient power to power on all the FPCs connected to the router, all the FPCs are powered on. If sufficient power is not available, Junos OS limits the number of FPCs brought online. That is, Junos OS uses the total available chassis output power as a factor to decide whether or not to power on an FPC connected to the router.

    [See T4000 Power Management Overview and T4000 Core Router Hardware Guide PDF Document.]

Routing Policy and Firewall Filters

  • Source checking for forwarding filter tables—On MX Series 3D Universal Edge Routers, you can apply a forwarding table filter by using the source-checking statement at the [edit forwarding-options family inet6] hierarchy level. This discards IPv6 packets when the source address type is unspecified, loopback, multicast, or link-local. RFC 4291, IP Version 6 Addressing Architecture, refers to four address types that require special treatment when they are used as source addresses. The four address types are: Unspecified, Loopback, Multicast, and Link-Local Unicast. The loopback and multicast addresses must never be used as a source address in IPv6 packets. The unspecified and link-local addresses can be used as source addresses but routers must never forward packets that have these addresses as source addresses. Typically, packets that contain unspecified or link-local addresses as source addresses are delivered to the local host. If the destination is not the local host, then the packet must not be forwarded. Configuring this statement filters or discards IPv6 packets of these four address types.

    [See Applying Filters to Forwarding Tables.]

  • Unidirectional GRE tunnels across IPv4 without tunnel interfaces—For Junos OS Release 12.3R2 and later, you can configure a tunnel that transports IPv4, IPv6, protocol-independent, or MPLS traffic across an IPv4 network without having to create tunnel interfaces on services PICs. This type of GRE tunnel is unidirectional and transports unicast or multicast transit traffic as clear text. Encapsulation, de-encapsulation, and forwarding of payloads is executed by Packet Forwarding Engine processes for logical Ethernet interfaces or aggregated Ethernet interfaces hosted on MICs and MPCs in MX Series routers. Two MX Series routers installed as PE routers provide network connectivity to two CE routers that lack a native routing path between them. This feature is also supported in logical systems.

    Specify tunnel characteristics by configuring the tunnel-end-point statement on the ingress PE router:

    firewall {tunnel-end-point tunnel-name {ipv4 {source-address source-host-address;destination-address destination-host-address;}gre [key number];}}

    To configure the ingress PE router to encapsulate passenger protocol packets, attach a passenger protocol family firewall filter at the input of a supported interface. The following terminating firewall filter action refers to the specified tunnel and initiates encapsulation of matched packets:

    encapsulate tunnel-name

    To configure the egress PE router to de-encapsulate GRE packets and forward the original passenger protocol packets, attach an IPv4 firewall filter at the input of all interfaces that are advertised addresses for the router. The following terminating firewall filter action initiates de-encapsulation of matched packets:

    decapsulate [routing-instance instance-name]

    By default, the Packet Forwarding Engine uses the default routing instance to forward payload packets to the destination network. If the payload is MPLS, the Packet Forwarding Engine performs route lookup on the MPLS path routing table using the route label in the MPLS header.

    If you specify the decapsulate action with an optional routing instance name, the Packet Forwarding Engine performs route lookup on the routing instance, and the instance must be configured.
    [Firewall Filters Configuration Guide]

Routing Protocols

  • Expanded support for advertising multiple paths to a destination in BGP—This feature now supports graceful restart and additional address families. Previously, graceful restart was not supported and only the IPv4 address family was supported with the BGP add-path feature. Now the following address families are supported:
    • IPv4 unicast (net unicast)
    • IPv6 unicast (inet6 unicast)
    • IPv4 labeled unicast (inet labeled-unicast)
    • IPv6 labeled unicast (inet6 labeled-unicast)

    To configure these address families, include the family <address-family> add-path statement at the [edit protocols bgp] hierarchy level.

    To configure graceful restart, include the graceful-restart statement at the [edit routing-options] hierarchy level.

    [See Example: Advertising Multiple BGP Paths to a Destination.]

  • Support for multihop BFD session—One desirable application of BFD is to detect connectivity to routing devices that span multiple network hops and follow unpredictable paths. This is known as a multihop session. Until Junos OS Release 12.3, multihop BFD was non-distributed and ran on the Routing Engine. Starting in Junos OS 12.3, multihop BFD is distributed, meaning that it runs on the Packet Forwarding Engine. This change provides multiple scalability improvements.

Security

  • DDoS protection flow detection (MX Series routers)—Flow detection is an enhancement to DDoS protection that supplements the DDoS policer hierarchies. When you enable flow detection by including the flow-detection statement at the [edit system ddos-protection global] hierarchy level, a limited amount of hardware resources are used to monitor the arrival rate of host-bound flows of control traffic. This behavior makes flow detection highly scalable compared to filter policers, which track all flows and therefore consume a considerable amount of resources.

    Flows that violate a DDoS protection policer are tracked as suspicious flows; they become culprit flows when they violate the policer bandwidth for the duration of a configurable detection period. Culprit flows are dropped, kept, or policed to below the allowed bandwidth level. Suspicious flow tracking stops if the violation stops before the detection period expires.

    Most flow detection attributes are configured at the packet level or flow aggregation level. Table 2 lists these statements, which you can include at the [edit system ddos-protection protocols protocol-group packet-type] hierarchy level. You can disable flow detection, configure the action taken for culprit flows, specify a bandwidth different than the policer bandwidth, configure flows to be monitored even when a policer is not in violation, disable automatic event reporting, or enable a timeout period that automatically removes flows as culprit flows after the timeout has expired.

    Table 2: Flow Detection Packet-Level Statements

    flow-detection-mode

    flow-level-detection

    no-flow-logging

    flow-detect-time

    flow-recover-time

    physical-interface

    flow-level-bandwidth

    flow-timeout-time

    subscriber

    flow-level-control

    logical-interface

    timeout-active-flows

    By default, flow detection automatically generates reports for events associated with the identification and tracking of culprit flows and bandwidth violations. You can include the flow-report-rate and violation-report-rate statements at the [edit system ddos-protection global] hierarchy level to configure the event reporting rate.

    Use the show ddos-protection protocols flow-detection command to display flow detection information for all protocol groups or for a particular protocol group. Use the show ddos-protection protocols culprit-flows command to display information about culprit flows for all packet types, including the number of culprit flows discovered, the protocol group and packet type, the interface on which the flow arrived, and the source address for the flow. The show ddos-protection statistics command now provides a global count of discovered and currently tracked culprit flows. You can use the clear ddos-protection protocols culprit-flows command to clear all culprit flows, or just those for a protocol group or individual packet type.

    [DDoS Configuration]

Subscriber Access Management

  • Support for PPP subscriber services over ATM networks (MX Series routers with MPCs and ATM MICs with SFP)—Enables you to create PPP-over-ATM (PPPoA) configurations on an MX Series router that has an ATM MIC with SFP (model number MIC-3D-80C3-20C12-ATM) and a supported MPC installed. PPPoA configurations support statically created PPP logical subscriber interfaces over static ATM underlying interfaces. (Dynamic creation of the PPP interfaces is not supported.) Most features supported for PPPoE configurations are also supported for PPPoA configurations on an MX Series router. You can dynamically apply subscriber services such as CoS and firewall filters to the static PPP logical subscriber interface by configuring the services in the dynamic profile that creates the PPP logical interface.

    PPPoA configurations on an MX Series router support two types of encapsulation on the ATM underlying interface:

    • To configure PPPoA encapsulation that uses LLC, you must configure the ATM underlying interface with PPP-over-AAL5 LLC encapsulation. To do so, include the encapsulation atm-ppp-llc statement at the [edit interfaces interface-name unit logical-unit-number] hierarchy level.
    • To configure PPPoA encapsulation that uses VC multiplexing, you must configure the ATM underlying interface with PPP-over-ATM AAL5 multiplex encapsulation. To do so, include the encapsulation atm-ppp-vc-mux statement at the [edit interfaces interface-name unit logical-unit-number] hierarchy level.

    PPPoA configurations enable the delivery of subscriber-based services, such as CoS and firewall filters, for PPP subscribers accessing the router over an ATM network. You use the same basic statements, commands, and procedures to create, verify, and manage PPPoA configurations as you use for PPPoA configurations on M Series routers and T Series routers.

    [Subscriber Access, Network Interfaces]

  • Support for adjusting shaping rate and overhead accounting attributes based on PPPoE access line parameters for agent circuit identifier interface sets (MX Series routers with MPCs/MICs)—Extends the functionality available in earlier Junos OS releases to enable you to configure the router to use the Actual-Data-Rate-Downstream [26-130] and Access-Loop-Encapsulation [26-144] DSL Forum vendor-specific attributes (VSAs) found in PPPoE Active Discovery Initiation (PADI) and PPPoE Active Discovery Request (PADR) control packets to adjust the shaping-rate and overhead-accounting class of service (CoS) attributes, respectively, for dynamic agent circuit identifier (ACI) interface sets. In earlier Junos OS releases, you used this feature to adjust the shaping-rate and overhead-accounting attributes only for dynamic subscriber interfaces not associated with ACI interface sets.

    The shaping-rate attribute is based on the value of the Actual-Data-Rate-Downstream VSA. The overhead-accounting attribute is based on the value of the Access-Loop-Encapsulation VSA, and specifies whether the access loop uses Ethernet (frame mode) or ATM (cell mode) encapsulation. In subscriber access networks where the router passes downstream ATM traffic to Ethernet interfaces, the different Layer 2 encapsulations between the router and the PPPoE Intermediate Agent on the digital subscriber line access multiplexer (DSLAM) make managing the bandwidth of downstream ATM traffic difficult. Using the Access-Loop-Encapsulation VSA to shape traffic based on frames or cells enables the router to adjust the shaping-rate and overhead-accounting attributes in order to apply the correct downstream rate for the subscriber.

    You can enable this feature in either the dynamic profile that defines the ACI interface set, or in the dynamic profile for the dynamic PPPoE (pp0) subscriber interface associated with the ACI interface set, as follows:

    • To configure the router to use the Actual-Data-Rate-Downstream VSA to adjust the shaping-rate CoS attribute, include the vendor-specific-tags actual-data-rate-downstream statement at the [edit dynamic-profiles profile-name class-of-service dynamic-class-of-service-options] hierarchy level.
    • To configure the router to use the Access-Loop-Encapsulation VSA to adjust the overhead-accounting CoS attribute, include the vendor-specific-tags access-loop-encapsulation statement at the [edit dynamic-profiles profile-name class-of-service dynamic-class-of-service-options] hierarchy level.

    When you enable this feature, the router adjusts the shaping-rate and overhead-accounting attributes when the dynamic ACI interface set is created and the router receives the PADI and PADR packets from the first subscriber interface member of the ACI interface set. The value of the Actual-Data-Rate-Downstream VSA in the PADI and PADR control packets overrides the shaping-rate value configured at the [edit dynamic-profiles profile-name class-of-service traffic-control-profiles] hierarchy level only if the Actual-Data-Rate-Downstream value is less than the shaping-rate value configured with the CLI. The value of the Access-Loop-Encapsulation VSA always overrides the overhead-accounting value configured at the [edit dynamic-profiles profile-name class-of-service traffic-control-profiles] hierarchy level.

    As part of this feature, the output of the following operational commands has been enhanced to display the adjustment value (frame mode or cell mode) for the overhead-accounting attribute:

    • show class-of-service interface
    • show class-of-service interface-set
    • show class-of-service traffic-control-profile

    [Subscriber Access]

  • DHCP relay agent selective traffic processing based on DHCP options (MX Series routers)—Subscriber management enables you to configure DHCP relay agent to provide subscriber support based on information in DHCP options. For DHCPv4 relay agent, you use DHCP option 60 and option 77 to identify the client traffic. For DHCPv6 relay agent, you use DHCPv6 option 15 and option 16.

    You can use the DHCP option information to specify the action DHCP relay agent takes on client traffic that meets the specified match criteria, such as forwarding traffic to a specific DHCP server, or dropping the traffic. You can also specify a default action, which DHCP relay agent uses when the option string in the client traffic does not satisfy any match criteria or when no other action is configured.

    To configure DHCP relay agent selective processing, you use the relay-option statement at the [edit forwarding-options dhcp-relay] or [edit forwarding-options dhcp-relay dhcpv6] hierarchy level. To display statistics for the number of forwarded packets, use the show dhcp relay statistics and show dhcpv6 relay statistics commands.

    [Subscriber Access]

  • Ensuring that RADIUS clears existing session state before performing authentication and accounting for new sessions (MX Series routers)—At subscriber session startup, the Junos OS authd process sends an Acct-On message to RADIUS servers. In some service provider environments, upon receipt of the Acct-On message, the RADIUS server cleans up the previous session state and removes accounting statistics. However, authentication or accounting for the new session can start before the RADIUS cleanup of the previous session—this can result in RADIUS deleting the new session’s authentication and accounting information (which might include billing information).

    To ensure that the new session’s authentication and accounting information is not deleted, you can optionally configure authd to wait for an Acct-On-Ack response message from RADIUS before sending the new authentication and accounting updates to the RADIUS server. When this feature is enabled, all authentication requests fail until the router receives the Acct-On-Ack response from at least one configured RADIUS server.

    To enable this feature, you configure the wait-for-acct-on-ack statement at the [edit access profile profile-name accounting] hierarchy level. To display the response status of the Acct-On messages (for example, Ack, Pending, None), use the show network-access aaa accounting command.

    [Subscriber Access]

  • Enhanced local configuration of DNS name server addresses (MX Series routers)—You can now configure the DNS name server addresses locally per routing instance or per access profile. The new configuration applies to both terminated and tunneled PPP subscribers (IPv4 and IPv6), DHCP subscribers (DHCPv4 and DHCPv6), and IP-over-Ethernet (VLAN) subscribers. In earlier releases, the local configuration for the DNS server address applied only to DHCP subscribers (configured as a DHCP attribute), and only at the more granular level of the address pool.

    As with the address-pool configuration, the new statements enable you to configure multiple DNS name server addresses per routing instance and access profile by issuing the statement for each address.

    Because you can both configure name server addresses at more than one level and configure more than one address within a level, a preference order for the configurations determines which address is returned to the client.

    • Within a configuration level, the preference order for the address matches the order in which the address is configured. For example, the first address configured within an access profile is preferred to the second address configured in that profile.
    • Among configuration levels, the preference order depends on the client type:

      • For DHCP subscribers, the preference in descending order is:

        RADIUS > DHCP address pool > access profile > global

      • For non-DHCP subscribers, the preference in descending order is:

        RADIUS > access profile > global

    • Accordingly, all subscriber types prefer a name server address configured in RADIUS to the address configured anywhere else. When a name server address is configured only in a DHCP address pool, then no address is available to non-DHCP subscribers. For all subscriber types, the global name server address is used only when no other name server addresses are configured.

    To configure a name server address in a routing instance, include the domain-server-name-inet or domain-name-server statement for IPv4 addresses, or the domain-name-server-inet6 statement for IPv6 addresses, at the [edit access] hierarchy level.

    To configure a name server address in an access profile, include any of the same statements at the [edit access profile] hierarchy level.

    Best Practice: In practice, choose either the domain-name-server statement or the domain-name-server-inet statement for IPv4 addresses. They both have the same effect and there is no need to use both statements.

    [Subscriber Access]

  • Gx-Plus support for service provisioning (MX Series routers)—Gx-Plus now supports service (policy rule) provisioning, service activation, threshold notifications, threshold updates, service termination, and recovery. Previously, Gx-Plus supported only notification, termination, and recovery. To request subscriber service provisioning from the Policy Control and Charging Rules Function (PCRF), include the provisioning-order gx-plus statement in the subscriber access profile.

    By default, Gx-Plus provisioning requests are made only for IPv4 subscribers. To enable requests to be made also for IPv6 subscribers, include the include-ipv6 statement at the [edit access gx-plus global] hierarchy level.

    The PCRF can request usage monitoring for the provisioned services for one or more of the following: number of bytes transmitted (CC-Output-Octets), number of bytes received (CC-Input-Octets), number of bytes transmitted and received (CC-Total-Octets), and elapsed time (CC-Time). If the specified threshold is reached, the router sends a usage report back to the PCRF. The PCRF can then return new threshold triggers and request that services be activated or deactivated.

    When a subscriber has been provisioned with Gx-Plus, only the PCRF can activate or deactivate services for that subscriber. Accordingly, AAA rejects any RADIUS CoA or CLI service activation or deactivation requests for these subscribers. You can override PCRF control on an individual session, which is useful for session and service troubleshooting. To do so, issue the new request network-access aaa subscriber set session-id command. You can then activate and deactivate services with the existing request network-access aaa subscriber add session-id and request network-access aaa subscriber delete session-id commands, respectively.

    [Subscriber Access]

  • Support for maintenance of CoS shaping rates for ANCP subscribers across ANCP restarts (MX Series routers)—When ANCP stops due to a process restart or graceful Routing Engine switchover (GRES), CoS now enforces the ANCP downstream shaping-rates until the CoS keepalive timer expires. When the timer expires, CoS reverts to its configured shaping-rate for the interfaces.

    You can configure the CoS keepalive timer by including the existing maximum-helper-restart-time seconds statement at the [edit protocols ancp] hierarchy level. It specifies how much time other daemons such as CoS will wait for ANCP to restart and is used to configure the CoS rate update keepalive timer.

    ANCP does not maintain TCP sessions from neighbors across the restart or graceful Routing Engine switchover (GRES). When it restarts, it must re-establish sessions with neighbors and subscriber sessions before the timer expires. For all the re-established sessions, ANCP updates CoS with the updated downstream shaping rates and provides DSL line attributes to the session database for AAA.

    If CoS stops or restarts while ANCP is up, ANCP retransmits all known subscriber downstream rates to CoS. Any existing adjusted shaping rates that have not been updated revert to the configured CoS shaping rates when the CoS restart timer expires.

    [Subscriber Access]

  • MAC address validation in enhanced network services modes—MAC address validation is now optimized for scaling when the router is configured for Enhanced IP Network Services mode or Enhanced Ethernet Network Services mode. When MAC address validation is enabled, the router compares the IP source and MAC source addresses against trusted addresses, and forwards or drops the packets according to the match and the validation mode. This feature is not available for IPv6.

    Note: When the router is configured for either of the enhanced network services modes, MAC address validation is supported only on MPCs. If the router has both DPCs and MPCs, or only DPCs, you cannot configure the chassis to be in enhanced mode.

    In contrast, when the router is configured for a normal (non-enhanced) network services mode, MAC address validation is supported on both DPCs and MPCs. The router can be populated completely with one or the other type of line card, or have a mix of both types. Normal network services mode is the default.

    To configure an enhanced network services mode, include the network-services service statement at the [edit chassis] hierarchy level, and then configure MAC address validation as usual.

    Note: In normal network services mode, you can use the show interfaces statistics interface-name command to display a per-interface count of the packets that failed validation and were dropped. In enhanced network services modes, this command does not count the dropped packets; you must contact Juniper Networks Customer Support for assistance in collecting this data.

    [Subscriber Access]

  • Fail filters for RPF checks in dynamic profiles—By default, unicast RPF checks prevent DHCP packets from being accepted on interfaces protected by the RPF check. When you enable an RPF check with a dynamic profile, you must configure a fail filter that identifies and passes DHCP packets.

    To configure a fail filter, include the fail-filter filter-name statement at the [edit dynamic-profiles profile-name interfaces interface-name unit logical-unit-number family family rpf-check] hierarchy level. To configure the terms of the fail filter, include the filter filter-name statement at the [edit firewall family family] hierarchy level. Include conditions in a filter term to identify DHCP packets, such as from destination-port dhcp and from destination-address 255.255.255.255/32. Define another filter term to drop all other packets that fail the RPF check. This feature is available for both IPv4 and IPv6 address families.

    To confirm that the fail filter is active, you can issue the show subscribers extensive command, which displays the name of active filters.

    [Subscriber Access]

  • Filtering traffic that is mirrored using DTCP-initiated subscriber secure policy—You can now filter mirrored traffic before it is sent to a mediation device. This feature allows service providers to reduce the volume of traffic sent to a mediation device. For some types of traffic, such as IPTV or video on demand, it is not necessary to mirror the entire content of the traffic because the content might already be known or controlled by the service provider.

    To configure, create a policy at the [edit services radius-flow-tap policy policy-name] hierarchy level. You can set up the policy to filter IPv4 or IPv6 traffic by source or destination address or port, protocol, or DSCP value. You then apply the policy by using the new DTCP attribute X-Drop-Policy. You can use the X-Drop-Policy attribute with the ADD DTCP command to begin filtering traffic when mirroring is triggered using the ADD DTCP command. To begin filtering traffic that is currently being mirrored, use the X-Drop-Policy attribute with the new ENABLE DTCP command. To stop filtering traffic that is currently being mirrored, use the X-Drop-Policy attribute with the new DISABLE DTCP command.

    [Subscriber Access Configuration Guide]

  • Enhancements to multicast subscriber flow distribution in an aggregated Ethernet bundle (MX Series routers)—Enables you to both target and separate the distribution of multicast subscriber traffic using enhanced IP chassis network services mode in an aggregated Ethernet bundle that is configured without link protection.

    This feature enhances already released scheduling and scaling improvements made for subscribers in an aggregated Ethernet bundle and includes support for the following:

    • IP demux subscriber interfaces on the EQ DPC and MPC/MIC modules and VLAN demux subscriber interfaces on MPC/MIC modules.

      Note: This feature is not supported for VLAN subscriber interfaces.

    • Multicast using the enhanced-ip mode setting at the [edit chassis network-services] hierarchy level.
    • Multicast traffic to egress in parallel with unicast traffic, sharing the CoS hierarchy and aggregated Ethernet flow distribution.
    • Targeted multicast flow distribution over inter-chassis redundancy (ICR) configurations where multicast traffic flows toward the subscriber primary interface even if that interface resides on a remote chassis within the virtual system.
    • The ability to separate unicast and multicast subscriber traffic on a per VLAN basis using OIF mapping.

    Targeted distribution enables you to target egress traffic for subscribers on a link. The system distributes subscriber interfaces equally among the links. For multicast traffic to egress in parallel with unicast traffic, share the CoS hierarchy and aggregated Ethernet flow distribution:

    Separated target distribution enables you to target multicast traffic to use a specific VLAN over the aggregated Ethernet interface instead of flowing over the same interface in parallel. To configure separated targeted distribution for a multicast link:

    When links are removed, affected flows are redistributed among the remaining active backup links. When links are added to the system, no automatic redistribution occurs. New subscriber and multicast flows are assigned to the links with the least number of subscribers (typically, the new links). You can configure the system to periodically rebalance the distribution of subscribers on the links by including the rebalance-periodic time hours:minutes interval hours statement at the [edit interfaces ae0 aggregated-ether-options] hierarchy level. To manually rebalance the subscribers on the interface, issue the request interface rebalance interface interface-name command.

    To display a summary of the targeted distribution on a logical interface, issue the show interface interface-name extensive command. To display the targeted distribution on a specific aggregated Ethernet bundle, issue the show interface targeting aex command.

    [Subscriber Access, Network Interfaces]

  • Layer-2 control packets—The forwarding path supports the following types of Layer-2 control packets (excluding Operation, Administration, and Maintenance (OAM) packets) in both directions, receiving and forwarding:
    • Ethernet control packets—ARP, IS-IS, 1588v2, Ethernet Synchronization Messaging Channel (ESMC).
  • Host path—The host path to and from the CPU is supported in the following ways:
    • Host-bound traffic, prioritized into multiple queues, to support various levels of traffic.
    • Hardware-based policing used to limit denial of service attacks.
    • Protocol and flow-based policing.
    • Code point-based classification and prioritization of packets from the host to the external world.
  • Counters and statistics—Most packet and byte-level statistics for various entities in the forwarding path available in Junos OS are supported. The following counters and statistics are supported:
    • Ingress and egress packet and byte counters for logical interfaces, Ethernet pseudowires, and MPLS transit label-switched paths.
    • Discard packets counter for system-wide global Packet Forwarding Engine statistics.
  • Statistics collection and reporting for Gigabit Ethernet interfaces—For Gigabit Ethernet interfaces, Packet Forwarding Engine statistics are disabled by default. To enable Gigabit Ethernet interface statistics, you must specifically configure them. To configure Gigabit Ethernet interface statistics, include the new statistics statement at the [edit interfaces interface-name unit logical-unit-number] hierarchy level. To display statistics, issue the show interfaces interface-name (brief-|-extensive) operational mode command.
  • Address Resolution Protocol (ARP) parameters—The maximum number of ARP entries is 7,000.
  • Support for configuring NAS-Port and NAS-Port-Type RADIUS attributes per physical interface, VLAN, or S-VLAN (MX Series routers with MPCs/MICs)—Enables you to configure the NAS-Port-Type (61) RADIUS IETF attribute, and an extended format for the NAS-Port (5) RADIUS IETF attribute, on a per-physical interface, per-static VLAN, or per-static stacked VLAN (S-VLAN) basis. The router passes the NAS-Port and NAS-Port-Type attributes to the RADIUS server during the authentication, authorization, and accounting (AAA) process.

    The NAS-Port-Type attribute specifies the type of physical port that the network access server (NAS) uses to authenticate the subscriber. The NAS-Port attribute specifies the physical port number of the NAS that is authenticating the user, and is formed by a combination of the physical port’s slot number, port number, adapter number, VLAN ID, and S-VLAN ID. The NAS-Port extended format configures the number of bits (bit width) for each field in the NAS-Port attribute: slot, adapter, port, VLAN, and S-VLAN.

    Configuring the NAS-Port-Type and the extended format for NAS-Port on a per-VLAN, per–S-VLAN, or per-physical interface basis is useful in the following network configurations:

    • 1:1 access model (per-VLAN basis)—In a 1:1 access model, dedicated customer VLANs (C-VLANs) provide a one-to-one correspondence between an individual subscriber and the VLAN encapsulation.
    • N:1 access model (per–S-VLAN basis)—In an N:1 access model, service VLANs are dedicated to a particular service, such as video, voice, or data, instead of to a particular subscriber. Because a service VLAN is typically shared by many subscribers within the same household or in different households, the N:1 access model provides a many-to-one correspondence between individual subscribers and the VLAN encapsulation.
    • 1:1 or N:1 access model (per-physical interface basis)—You can configure the NAS-Port-Type and NAS-Port format on a per-physical interface basis for both the 1:1 access model and the N:1 access model.

    To configure the NAS-Port-Type and the format for NAS-Port on a per-VLAN, or per–S-VLAN, or per-physical interface basis, you must create a NAS-Port options definition. The NAS-Port options definition includes the NAS-Port extended format, the NAS-Port-Type, and either the VLAN range of subscribers or the S-VLAN range of subscribers to which the definition applies.

    The basic tasks for configuring a NAS-Port options definition are as follows:

    • To create a named NAS-Port options definition, include the nas-port-options nas-port-options-name statement at the [edit interfaces interface-name radius-options] hierarchy level.
    • To configure the extended format for the NAS-Port, include the nas-port-extended-format statement and appropriate options at the [edit interfaces interface-name radius-options nas-port-options nas-port-options-name] hierarchy level. To include S-VLAN IDs, in addition to VLAN IDs, in the extended format, include the stacked statement at the [edit interfaces interface-name radius-options nas-port-options nas-port-options-name nas-port-extended-format] hierarchy level.
    • To configure the NAS-Port-Type, include the nas-port-type port-type statement at the [edit interfaces interface-name radius-options nas-port-options nas-port-options-name] hierarchy level.
    • To configure the VLAN range of subscribers to which the NAS-Port options definition applies, include the vlan-ranges statement at the [edit interfaces interface-name radius-options nas-port-options nas-port-options-name] hierarchy level. To specify all VLANs in the VLAN range, include the any statement at the [edit interfaces interface-name radius-options nas-port-options nas-port-options-name vlan-ranges] hierarchy level.
    • To configure the S-VLAN range of subscribers to which the NAS-Port options definition applies, include the stacked-vlan-ranges statement at the [edit interfaces interface-name radius-options nas-port-options nas-port-options-name] hierarchy level. To specify all VLAN IDs in the outer tag of the S-VLAN range, include the any statement at the [edit interfaces interface-name radius-options nas-port-options nas-port-options-name stacked-vlan-ranges] hierarchy level. You cannot configure the inner tag (S-VLAN ID) of the S-VLAN range; the inner tag is always specified as any to represent all S-VLAN IDs.

    Note: You can create a maximum of 16 NAS-Port options definitions per physical interface. Each definition can include a maximum of 32 VLAN ranges or 32 S-VLAN ranges, but cannot include a combination of VLAN ranges and S-VLAN ranges.

    [Subscriber Access]

  • Support for one dynamic profile for both single-stack and dual-stack subscribers—On PPP access networks, you can use one dynamic profile to support the following address combinations: IPv4 only, IPv6 only, and IPv4 and IPv6 dual stack.

    [Designing an IPv6 Architecture and Implementing IPv4 and IPv6 Dual Stack for Broadband Edge]

  • Support for DHCPv6 requests that include a request for both DHCPv6 IA_NA and DHCPv6 prefix delegation—For DHCPv6 subscribers on DHCP access networks, a client can solicit both an IA_NA address and a prefix for DHCP prefix delegation, and the session comes up even if either the address or the prefix is not allocated. In earlier releases, an error was returned if the BNG did not return both an address for DHCPv6 IA_NA and a prefix for DHCPv6 prefix delegation.

    [Designing an IPv6 Architecture and Implementing IPv4 and IPv6 Dual Stack for Broadband Edge]

  • Support for new Juniper Networks Diameter AVP (MX Series routers)—Junos OS supports a new Juniper Networks Diameter AVP, Juniper-State-ID (AVP code 2058). Juniper-State-ID specifies the value assigned to each synchronization cycle for the purpose of identifying which messages to discard. The Juniper-State-ID AVP can be included in Diameter messages and used by supported Diameter applications such as JSRC and PTSP.

    [Subscriber Access Configuration Guide]

  • Subscriber management and services feature and scaling parity (MX2010 and MX2020)—Starting in Junos OS Release 12.3R4, the MX2010 router and the MX2020 router support all subscriber management and services features that are supported by the MX240, MX480, and MX960 routers. In addition, the scaling and performance values for the MX2010 router and the MX2020 router match those of MX960 routers.

System Logging

New and deprecated system log tags—The following set of system log message is new in this release:

  • LLDP—This section describes messages with the LLDP prefix. They are generated by the link layer discovery protocol process (lldpd), which is used by EX Series switches to learn and distribute device information on network links. The information allows the switch to quickly identify a variety of devices, including IP telephones, resulting in a LAN that interoperates smoothly and efficiently.

The following system log messages are new in this release:

  • ASP_NAT_PORT_BLOCK_ACTIVE
  • ASP_PCP_NAT_MAP_CREATE
  • ASP_PCP_NAT_MAP_DELETE
  • ASP_PCP_TPC_ALLOC_ERR
  • ASP_PCP_TPC_NOT_FOUND
  • AUTHD_ACCT_ON_ACK_NOT_RECEIVED
  • CHASSISD_FPC_OPTICS_HOT_NOTICE
  • CHASSISD_MAC_ADDRESS_VIRB_ERROR
  • CHASSISD_RE_CONSOLE_ME_STORM
  • COSD_CLASS_NO_SUPPORT_IFD
  • COSD_CLASS_NO_SUPPORT_L3_IFL
  • COSD_MAX_FORWARDING_CLASSES_ABC
  • DDOS_SCFD_FLOW_AGGREGATED
  • DDOS_SCFD_FLOW_CLEARED
  • DDOS_SCFD_FLOW_DEAGGREGATED
  • DDOS_SCFD_FLOW_FOUND
  • DDOS_SCFD_FLOW_RETURN_NORMAL
  • DDOS_SCFD_FLOW_TIMEOUT
  • ESWD_VMEMBER_MAC_LIMIT_DROP
  • FC_PROXY_NP_PORT_RESTORE_FAILED
  • LIBJNX_PRIV_RAISE_FAILED
  • LLDP_NEIGHBOR_DOWN
  • LLDP_NEIGHBOR_UP
  • PPMD_MIRROR_ERROR
  • RPD_PARSE_BAD_COMMAND
  • RPD_PARSE_BAD_FILE
  • RPD_PIM_JP_INFINITE_HOLDTIME
  • UFDD_LINK_CHANGE
  • WEB_CERT_FILE_NOT_FOUND_RETRY
  • WEB_DUPLICATE_HTTPD

The following system log messages are no longer documented, either because they indicate internal software errors that are not caused by configuration problems or because they are no longer generated. If these messages appear in your log, contact your technical support representative for assistance:

  • FABOAMD_TASK_SOCK_ERR
  • JCS_EXT_LINK_STATE
  • JCS_RSD_LINK_STATE
  • JCS_SWITCH_COMMUNICATION_OK
  • LIBJNX_AUDIT_ERROR
  • LIBJNX_COMPRESS_EXEC_FAILED
  • LIBJNX_INVALID_CHASSIS_ID
  • LIBJNX_INVALID_RE_SLOT_ID
  • LIBJNX_REPLICATE_RCP_EXEC_FAILED

User Interface and Configuration

  • Support for HTTP reverse proxy and HTTP transparent proxy on Application Services Modular Line Card (MX240, MX480, MX960 routers)—The Application Services Modular Line Card with Media Flow Controller software installed enables configuring support for HTTP reverse proxy and HTTP transparent proxy caching.

    The Application Services Modular Line Card (AS MLC) has three components:

    • Application Services Modular Carrier Card (AS MCC)
    • Application Services Modular Processing Card with 64G (AS MXC)
    • Application Services Modular Storage Card with 6.4 TB capacity (AS MSC)

    The AS MLC for MX Series routers supports high throughput for applications developed with Juniper Networks Media Flow Controller software. A Media Flow Controller application functions as a web-caching proxy server that processes HTTP traffic. HTTP requests are routed to the Media Flow Controller either explicitly for a domain (reverse proxy) or by redirecting traffic based on a policy (transparent proxy).

    Media Flow Controller software can operate in HTTP reverse proxy mode, HTTP transparent proxy mode, or mixed mode.

    In HTTP reverse proxy configurations, the service provider provides services to a set of domains (content providers) that buy content caching capability from the service provider. Clients connect to content providers through virtual IP (VIP) addresses. Service providers in the reverse proxy scenario generally deploy the routers with AS MLC hardware to honor service requests (such as caching) from the domain users.

    HTTP reverse proxy supports the following features:

    • Retrieve and deliver content from content providers in response to client requests as if the content originated at the proxy
    • Prevent attacks from the Web when a firewall is included in the reverse proxy configuration
    • Load balance client requests among multiple servers
    • Lessen load on origin servers by caching both static and dynamic content

    In HTTP transparent proxy configurations, the service provider implements the AS MLC to improve its own caching capability and to reduce the load on its own network. Implementing caching on an MX Series router with an AS MLC improves the retrieval speeds for data and optimizes the back-end network utilization. Typically, HTTP transparent proxy retrieves content for clients from the Internet. The client identifies the target of the request, which is commonly a location on the Internet.

    HTTP transparent proxy does not enforce local policies: it does not add, delete, or modify information contained in the messages it forwards. HTTP transparent proxy is a cache for data. HTTP transparent proxy satisfies client requests directly because it retains the data that was previously requested by the same or by a different client. HTTP transparent proxy improves the efficiency and performance of network bandwidth within the content provider’s data center.

    In mixed mode, both reverse proxy and transparent proxy are configured on the same router.

    [Junos OS Ethernet Interfaces Configuration Guide]

  • Support for 10-port 10-Gigabit Ethernet MIC with SFPP on MPC3E (MX240, MX480, and MX960 routers)—Starting with Junos OS Release 12.3, the MPC3E supports the 10-port 10-Gigabit Ethernet MIC with SFPP (MIC3-3D-10XGE-SFPP). The 10-port 10-Gigabit Ethernet MIC with SFPP uses SFP+ optical transceiver modules for connectivity. The MIC supports up to ten 10-Gigabit Ethernet interfaces and occupies MIC slot 0 or 1 in the MPC3E.

    The MIC supports both LAN-PHY and WAN-PHY interface framing modes. You can configure the framing mode on a per-port basis. Use the existing command to switch between LAN-PHY and WAN-PHY modes:

    set interfaces interface-name framing (lan-phy | wan-phy)

    The 10-Gigabit Ethernet MIC with SFPP supports the same features as the other MICs supported on the MPC3E.

    [See MPC3E MIC Overview, MX Series 3D Universal Edge Router Line Card Guide, Ethernet Interfaces Configuration Guide, System Basics.]

  • Inline flow monitoring support (MX Series routers with MPC3E)—Junos OS Release 12.3 supports inline flow monitoring and sampling services on MX Series routers with MPC3E. To configure inline flow monitoring, include the inline-jflow statement at the [edit forwarding-options sampling instance instance-name family inet output] hierarchy level. Inline flow monitoring supports a specified sampling output format designated as IP_FIX and uses UDP as the transport protocol. Inline flow monitoring supports both IPv4 and IPv6 formats.

    [See Configuring Inline Sampling, and Protocols and Applications Supported by MX240, MX480, MX960 MPC3E.]

  • Enhancements to IPv4 and IPv6 inline-jflow flow IPFIX record templates—Junos OS Release 12.3 introduces the VLAN ID field in the inline-jflow flow IPFIX record templates for IPv4 and IPv6 traffic. The VLAN ID field is not valid for egress traffic, and returns a value of 0 for egress traffic. Note that the VLAN ID field is updated while creating a new flow record, and any change in VLAN ID after that might not be updated in the record.

    [Services Interfaces]

  • Support for IPv6 flow servers on interfaces hosted on MICs or MPCs—Starting with Release 12.3, Junos OS enables you to configure IPv6 flow servers for inline flow monitoring. When you configure an IPv6 address for the flow-server statement at the [edit forwarding-options sampling instance instance-name family (inet |inet6 |mpls) output] hierarchy level, you must also configure an IPv6 address for the inline-jflow source-address statement at the [edit forwarding-options sampling instance instance-name family (inet | inet6 | mpls) output] hierarchy level. You can configure different families that use IPv4 and IPv6 flow servers under the same sampling instance. However, you can configure only one flow server per family.

    [Services Interfaces]

  • Optical transceiver support for MIC3-3D-1X100GE-CFP on MPC3E (MX240, MX480, and MX960 routers)—Starting with Junos OS Release 12.3, the 100-Gigabit Ethernet MIC with CFP (MIC3-3D-1X100GE-CFP) on MPC3E supports the CFP-100GBase-ER4 optical transceiver.

    If the ambient temperature exceeds 40° C and the other MIC slot is not empty, the CFP-100GBase-ER4 optical transceiver is put on low power mode, which disables the transmitter and takes the optic modules on the MIC offline. This protects the optical transceiver and also prevents damage to adjacent components.

    When the optical transceiver is taken offline, you might see the following system log (syslog) message:

    PIC 1 optic modules in Port 0 8 have been disabled since ambient temperature is over threshold.

    Note: The CFP-100GBase-ER4 optical transceiver is NEBS (Network Equipment Building System) compliant only when plugged into the 100-Gigabit Ethernet MIC with CFP and when the other MIC slot is empty.

    To reactivate the optical transceiver, use the request chassis optics fpc-slot fpc-slot-number reactivate operational mode command.

    [System Basics Configuration Guide]

  • Optical transceiver support for MIC3-3D-10XGE-SFPP on MPC3E (MX240, MX480, and MX960 routers)—Starting with Junos OS Release 12.3, the 10-port 10-Gigabit Ethernet MIC with SFPP (MIC3-3D-10XGE-SFPP) on MPC3E supports the SFPP-10GE-ZR optical transceiver.

    If the ambient temperature exceeds 40° C, the transmitter on the SFPP-10GE-ZR optical transceiver is disabled, which takes the optic modules on the MIC offline. This protects the optical transceiver and also prevents damage to adjacent components.

    When the optical transceiver is taken offline, you might see the following system log (syslog) message:

    PIC 1 optic modules in Port 0 8 have been disabled since ambient temperature is over threshold.

    Note: The SFPP-10GE-ZR optical transceiver is not NEBS (Network Equipment Building System) compliant when plugged into the 10-port 10-Gigabit Ethernet MIC with SFPP. If other optical transceivers have been added, they can continue to operate.

    To reactivate the optical transceiver, use the request chassis optics fpc-slot fpc-slot-number reactivate operational mode command.

    [System Basics Configuration Guide]

VPLS

  • PIM snooping for VPLS—PIM snooping is introduced to restrict multicast traffic to interested devices in a VPLS. A new statement, pim-snooping, is introduced at the [edit routing-instances instance-name protocols] hierarchy level to configure PIM snooping on the PE device. PIM snooping configures a device to examine and operate only on PIM hello and join/prune packets.

    A PIM snooping device snoops PIM hello and join/prune packets on each interface to find interested multicast receivers and populates the multicast forwarding tree with this information. PIM snooping can also be configured on PE routers connected as pseudowires, which ensures that no new PIM packets are generated in the VPLS, with the exception of PIM messages sent through LDP on the pseudowire.

    PIM snooping improves IP multicast bandwidth in the VPLS core. Only devices that are members of a multicast group receive the multicast traffic meant for the group. This ensures network integrity and reliability, and multicast data transmission is secured.

    [See Example: Configuring PIM Snooping for VPLS.]

  • Improved VPLS MAC address learning on T4000 routers with Type 5 FPCs—Junos OS Release 12.3 enables improved virtual private LAN service (VPLS) MAC address learning on T4000 routers with Type 5 FPCs by supporting up to 262,143 MAC addresses per VPLS routing instance. In Junos OS releases before Release 12.3, T4000 routers with Type 5 FPCs support only 65,535 MAC addresses per VPLS routing instance.

    To enable the improved VPLS MAC address learning on T4000 routers with Type 5 FPCs:

    • Include the enhanced-mode statement at the [edit chassis network-services] hierarchy level and perform a system reboot. By default, the enhanced-mode statement is not configured.
    • Include the mac-table-size statement at the [edit routing-instances vpls protocols vpls ] hierarchy level.

    Note:

    • You can configure the enhanced-mode statement only on T4000 routers with Type 5 FPCs.
    • The enhanced-mode statement supports up to 262,143 MAC addresses per VPLS routing instance. However, the MAC address learning limit for each interface remains the same (that is, 65,535 MAC addresses).
    • You must reboot the system after configuring the enhanced-mode statement. Otherwise, the improved VPLS MAC address learning does not take effect.
    • When the T4000 router reboots after the enhanced-mode statement has been configured, all Type 4 FPCs go offline.

    [See Configuring Improved VPLS MAC Address Learning on T4000 Routers with Type 5 FPCs.]

  • VPLS Multihoming support extended to FEC 129–Enables you to connect a customer site to two or more PE routers to provide redundant connectivity. A redundant PE router can provide network service to the customer site as soon as a failure is detected. VPLS multihoming helps to maintain VPLS service and traffic forwarding to and from the multihomed site in the event of network failures. BGP-based VPLS autodiscovery (FEC 129) enables each VPLS PE router to discover the other PE routers that are in the same VPLS domain. VPLS autodiscovery also automatically detects when PE routers are added or removed from the VPLS domain. You do not need to manually configure the VPLS and maintain the configuration when a PE router is added or deleted. VPLS autodiscovery uses BGP to discover the VPLS members and to set up and tear down pseudowires in the VPLS. To configure, include the multi-homing statement at the [edit routing-instances instance-name] hierarchy level.

    [See Example: Configuring VPLS Multihoming (FEC 129).]

  • BGP path selection for Layer 2 VPNs and VPLS—By default, Juniper Networks routers use just the designated forwarder path selection algorithm to select the best path to reach each Layer 2 VPN or VPLS routing instance destination. However, you can now configure the routers in your network to use both the BGP path selection algorithm and the designated forwarder path selection algorithm. The Provider routers within the network can use the standard BGP path selection algorithm. Using the standard BGP path selection for Layer 2 VPN and VPLS routes allows a service provider to leverage the existing Layer 3 VPN network infrastructure to also support Layer 2 VPNs and VPLS. The BGP path selection algorithm also helps to ensure that the service provider’s network behaves predictably with regard to Layer 2 VPN and VPLS path selection. This is particularly important in networks employing route reflectors and multihoming.

    The PE routers continue to use the designated forwarder path selection algorithm to select the preferred path to reach each CE device. The VPLS designated forwarder algorithm uses the D-bit, preference, and PE router identifier to determine which path to use to reach each CE device in the Layer 2 VPN or VPLS routing instance.

    To enable the BGP path selection algorithm for Layer 2 VPN and VPLS routing instances, do the following:

    • Specify a unique route distinguisher on each PE router participating in a Layer 2 VPN or VPLS routing instance.
    • Configure the l2vpn-use-bgp-rules statement on all of the PE and Provider routers participating in Layer 2 VPN or VPLS routing instances. You can configure this statement at the [edit protocols bgp path-selection] hierarchy level to apply this behavior to all of the routing instances on the router or at the [edit routing-instances routing-instance-name protocols bgp path-selection] hierarchy level to apply this behavior to a specific routing instance.

    On all of the PE and Provider routers participating in Layer 2 VPN or VPLS routing instances, run Junos OS Release 12.3 or later. Attempting to enable this functionality on a network with a mix of routers that both do and do not support this feature can result in anomalous behavior.

    [See Enabling BGP Path Selection for Layer 2 VPNs and VPLS.]

VPNs

  • Provider edge link protection in Layer 3 VPNs—A precomputed protection path can be configured in a Layer 3 VPN such that if a link between a CE router and a PE router goes down, the protection path (also known as the backup path) between the CE router and an alternate PE router can be used. This is useful in an MPLS service provider network, where a customer can have dual-homed CE routers that are connected to the service provider through different PE routers. In this case, the protection path avoids disruption of service if a PE-CE link goes down.

    The protection path can be configured on a PE router in a Layer 3 VPN by configuring the protection statement at the [edit routing-instances instance-name protocols bgp family inet unicast] or [edit routing-instances instance-name protocols bgp family inet6 unicast] hierarchy level.

    The protection statement indicates that protection is required on prefixes received from a particular neighbor or family. After protection is enabled for a given family, group, or neighbor, protection entries are added for prefixes or next hops received from the respective peer.

    A protection path can be selected only if the best path has already been installed by BGP in the forwarding table. This is because a protection path cannot be used as the best path. There are two conditions under which the protection path will not work:

    • When configured for an internal BGP peer.
    • When configured with external and internal BGP multipath.

    [See Example: Configuring Provider Edge Link Protection in Layer 3 VPNs.]

  • Edge node failure protection for LDP-signaled pseudowires—This feature provides a fast protection mechanism against egress PE router failure when transport LSPs are RSVP-TE LSPs. This is achieved by using multihomed CEs, upstream assigned labels, context-specific label switching (egress-protection and context-identifier statements), and by extending RSVP facility backup fast reroute (FRR) to enable node protection at the penultimate hop router of the LSP. With node protection capability, the penultimate hop router can perform local repair upon an egress PE failure and redirect pseudowire traffic very quickly to a protector PE through a bypass LSP. You must configure a Layer 2 circuit and transport LSP to enable this feature. Use the show rsvp session and show mpls lsp commands to view bypass LSP and backup LSP information on the penultimate hop router and a protector PE router.

    [VPNs]

  • Support for configuring more than one million Layer 3 VPN labels—For Layer 3 VPNs configured on Juniper Networks routers, Junos OS normally allocates one inner VPN label for each customer edge (CE)-facing virtual routing and forwarding (VRF) interface of a provider edge (PE) router. However, other vendors allocate one VPN label for each route learned over the CE-facing interfaces of a PE router. This practice increases the number of VPN labels exponentially, which leads to slow system processing and slow convergence time.

    For Juniper Networks routers participating in a mixed vendor network with more than one million Layer 3 VPN labels, include the extended-space statement at the [edit routing-options forwarding-table chained-composite-next-hop ingress l3vpn] hierarchy level. The extended-space statement is disabled by default.

    We recommend that you configure the extended-space statement in mixed vendor networks containing more than one million BGP routes to support Layer 3 VPNs. However, because using this statement can also enhance the Layer 3 VPN performance of Juniper Networks routers in networks where only Juniper Networks routers are deployed, we recommend configuring the statement in these networks as well.

    [See Accepting BGP Updates with Unique Inner VPN Labels in Layer 3 VPNs.]

  • Layer 2 circuit switching protection—Provides traffic protection for the Layer 2 circuit paths configured between PE routers. In the event the path (working path) used by a Layer 2 circuit fails, traffic can be switched to an alternate path (protection path). Switching protection is supported for locally switched Layer 2 circuits and provides 1 to 1 protection for each Layer 2 circuit interface.

    Each working path can be configured to have a either a protection path routed directly to the neighboring PE router or indirectly using a pseudowire configured through an intermediate PE router. The protection path provides failure protection for the traffic flowing between the PE routers. Ethernet OAM monitors the status of these paths. When OAM detects a failure, it reroutes the traffic from the failed working path to the protection path. You can configure OAM to revert the traffic automatically to the working path when it is restored. You can also manually switch traffic between the working path, the protection path, and back.

    Layer 2 circuit switching protection is supported on MX Series routers only. Nonstop routing (NSR) and graceful Routing Engine switchover (GRES) are not supported.

    To enable Layer 2 circuit switching protection, include the connection-protection statement at the [edit protocols l2circuit local switching interface interface-name end-interface] hierarchy level. You also need to configure OAM for the working path and the protection path by configuring the maintenance-association statement and sub-statements at the [edit protocols oam ethernet connectivity-fault-management maintenance-domain maintenance-domain-name] hierarchy level.

    [See Example: Configuring Layer 2 Circuit Switching Protection]

  • History enhancements for Layer 2 circuit, Layer 2 VPN, and FEC 129-based pseudowires—Adds the instance-history option to the show vpls connections and show l2vpn connections commands. Also adds instance-level logs for the following events:
    • Catastrophic events
    • Route withdrawals
    • Pseudowire switchovers
    • Connect protect switchovers
    • Protect interface swaps
    • Interface flaps (interface down events)
    • Label block changes

    These logs are maintained until the instance is deleted from the configuration.

Related Documentation

Modified: 2016-06-09