Hardware

Class of service

• Support for class of service (CoS) configuration with these limitations:

  • 802.3X Ethernet PAUSE is not supported.
  • CoS flexible hierarchical scheduling, also known as enhanced transmission selection (ETS), is not supported.
  • Neither MPLS EXP ingress packet classification nor egress rewrite rules are supported.
  • Classifiers and rewrite rules are applied to logical interfaces instead of physical interfaces.

  [See CoS Support on QFX Series Switches, EX4600 Line of Switches, and QFabric Systems.]

• CoS support on EVPN-Virtual Extensivle LAN (VXLAN). Support is provided for defining classifiers and rewrite rules on leaf (initiation and terminations) and spine nodes for EXPN VXLANs. Support is also provided for defining schedulers, interpolated drop profiles, explicit congestion notification (ECN), and priority based flow control (PFC).

  [See CoS Support on EVPN VXLANs.]

DHCP

• Support for DHCPv4 and DHCPv6 stateless relay over Layer 3 (L3) interfaces.

  Support includes:

  • Option-82 for DHCPv4.
  • Option-18 and Option-37 for DHCPv6.

  • Virtual router aware DHCPv4/v6 stateless relay.

    Note:

    Support does not include IRB interfaces.

    [See DHCP Relay Agent.]

EVPN

• EVPN-VXLAN support for unicast traffic using media access control (MAC) virtual routing and forwarding (VRF) routing instances. The switch can be a leaf device or a spine device in an edge-routed bridging or centrally routed bridging overlay design. You can configure multiple EVPN instances (EVIs) of type mac-vrf. Each EVI can support a different EVPN service type (vlan-based, vlan-aware, or vlan-bundle). The switch creates and uses one virtual tunnel endpoint (VTEP) logical interface per remote provider edge (PE) device by default, regardless of the number of routing instances. This implementation improves VXLAN VTEP scaling with multiple routing instances. EVPN-VXLAN support on this switch includes:
  • External BGP (EBGP) and internal BGP (IBGP) overlays.
  • Single-homing and all-active multihoming.
  • Layer 2 and Layer 3 unicast for IPv4 and IPv6 with ARP suppression.
  • Proxy ARP and ARP suppression for Layer 2 and Layer 3 IRB traffic.
  • Proxy Neighbor Discovery Protocol (NDP) and NDP suppression for Layer 2 and Layer 3 IRB traffic.
  • Service provider style configuration with Q-in-Q support, supported for Layer 2 gateways only and only with MAC VRF vlan-bundle service type.
  • Storm control.
  • MAC mobility, MAC limiting, MAC move limiting, and duplicate MAC detection.
  • Core isolation.
  • Proxy re-advertisement of EVPN Type 2 MAC+IP routes by all provider edge devices on the same Ethernet segment identifier (ESI).
  • Virtual machine traffic optimization (VMTO).
  • OSPF and BGP routing protocols on IRB interfaces.

  [See EVPN User Guide.]

• Support for sFlow monitoring technology in EVPN-VXLAN fabrics.

  [See Overview of sFlow Technology.]

• Support for port mirroring in EVPN-VXLAN fabrics.

  [See How to Configure Remote Port Mirroring for EVPN-VXLAN Fabrics.]

• Support for EVPN Type-5 routes in EVPN-VXLAN fabrics.

  [See Understanding EVPN Pure Type-5 Routes.]

• Support for EVPN-VXLAN firewall filtering and policing.

  [See Firewall Filter Match Conditions and Actions (QFX and EX Series Switches).]

Hardware

• The QFX5700 switch supports up to 8 line cards and each line card supports 16 QSFP28 ports that operate at 100GbE speed or 4 QSFP56-DD ports that operate at 400GbE speed. The QFX5700 switch has AC or DC power supplies and front-to-back airflow.

  [See QFX5700 Hardware Guide.]

• Resiliency support for Routing and Control Board (RCB) includes CPU memory and dual in-line memory module. You can configure fault-handling actions such as logging the error, raising alarms, sending SNMP traps, and indicating error conditions by using LEDs.

  [See routing-engine (Chassis).]

• Resiliency support for the chassis, line card (JNP-FPC-4CD), and Forwarding Engine Board (FEB)—QFX5130-FEB—includes handling the faults related to the links between components; for example, between a line card (such as JNP-FPC16C) and the switch fabric.

  [See QFX Hardware Components.]

Interfaces and chassis

• Support for GRE tunneling.

  [See Generic Routing Encapsulation.]

• Support for qualifying optics on the 100GbE and the 400GbE FPC line cards includes software monitoring, sensor diagnostics, FPC interfaces node level failure or restoration, events, and error logging.

  [See QFX Hardware Components.]

• Support for two new FPCs introduced for QFX5700 switches:

  • JNP-FPC-16C: The line card contains a total of 16 QSFP28 ports that support 100Gbps and 40Gbps speeds. You can channelize the ports operating at:
    • 100Gbps to four 25Gbps channels.
    • 40Gbps to four 10Gbps channels.
  • JNP-FPC-4CD: The line card contains a total of four QSFP56-DD ports that support 400Gbps, 100Gbps, and 40Gbps speeds. You can channelize the ports operating at:
    • 400Gbps to four 100Gbps channels.
    • 100Gbps to four 25Gbps channels.
    • 40Gbps to four 10Gbps channels.

    [See Port Settings.]

• Support for one or two RCBs.

  [See QFX5700 Routing and Control Board.]

• Support for the FEB with field-replaceable unit (FRU) management that includes:
  • Health monitoring
  • Fault handling
  • Systems alarms
  • Notification by LEDs
  • Power budgeting
  • Cooling
  • Management

  [See request chassis feb.]

• Support for IPv4 and IPv6 unicast IP-over-IP (IPv4) filter based decapsulation.

  [See Overview of Next-Hop-Based Dynamic Tunneling Using IP-Over-IP Encapsulation.]

Juniper extension toolkit

• Support for Firewall Service and Interfaces Service JET APIs.

  [See Juniper Engineering Network website.]

L2 features

• Support for L2 control protocols: xSTP, LACP, and LLDP.

  [See Ethernet Switching User Guide.]

• Support for these L2 features:

  • Enhanced L2 Software (ELS)
  • 802.1D
  • 802.1Q VLAN tagging
  • 802.1Q VLAN trunking
  • 802.1p
  • Routed VLAN interface (RVI)
  • MAC address aging configuration
  • Static MAC address assignment for an interface
  • Disable MAC learning

  [See Ethernet Switching User Guide.]

RSVP-Traffic Engineering (TE) supports preempting secondary label-switched paths (LSPs) that are signaled but not active.

[See RSVP Overview.]

Multicast

• Support for these multicast forwarding features:

  • IPv4 and IPv6 multicast
  • IGMP
  • Multicast Listener Discovery (MLD) protocol
  • Protocol Independent Multicast source-specific multicast (PIM SSM)
  • Protocol Independent Multicast sparse mode (PIM SM)
    Note:

    Support does not extend to features beyond those listed above. In this release, IGMP snooping, MLD snooping, multicast virtual private network (MVPN) Multicast, PIM multicast-only fast reroute (MoFRR), PIM first hop router (FHR), rendezvous point (RP), and last hop router (LHR) are not supported. In addition, the IRB interface is not supported as either a source or as a receiver. Make-before-break (MBB) is not supported for existing L3 aggregated Ethernet (AE) or LAG receivers (member addition/deletions or up/downs).

  [See Multicast Overview.]

• Support for these IGMP snooping with IRB features:

  • IGMP V1, V2, and V3 with plain L2 snooping with IRB (only if device acts as an LHR)
  • Proxy mode
  • Enterprise-style CLI only
  • Any-source multicast (ASM) and source-specific multicast (SSM) modes

  Limitations include:

  • IGMP group-specific queries received on a multicast-router interface are forwarded to all other interfaces in the VLAN.
  • MBB on existing L2aggregated Ethernet or LAG interfaces, including for member additions, deletions, and link up or down events.
  • All unregistered IPv4 and IPv6 multicast packets are forwarded to the multicast router interfaces in the VLAN, even if an interface is configured as a multicast router interface only for IGMP snooping.
  • Neither MLD snooping, PIM snooping, snooping with VPLS, EVPN-VXLAN, nor EVPN-MPLS is supported.

  [See IGMP Snooping Overview and Integrated Routing and Bridging.]

Network management and monitoring

• Support for sFlow monitoring technology.

  [See Understanding How to Use sFlow Technology for Network Monitoring.]

• Support for port mirroring.

  [See Understanding Port Mirroring and Analyzers.]

Protection against distributed denial of service (DDoS) attacks

• Support for DDoS protection, which is enabled by default.

  [See Control Plane Distributed Denial-of-Service (DDoS) Protection Overview.]

Routing policy and firewall filters

• L3 support for firewall filters and policers.

  [See Firewall Filter Match Conditions and Actions (QFX and EX Series Switches).]

• Support for Unified Forwarding Table (UFT) and L3 sub-interface features—The QFX5700 line of switches supports these UFT and L3 sub-interface features:

  The UFT feature enables you to allocate forwarding table resources to optimize the memory available for different address types based on the needs of your network. The UFT stores both the L2 and L3 entries that enable you to set default sizes for different entries.

  These applications share UFT search banks:

  • L2 MAC addresses.
  • IPv4 and IPv6 host routes.
  • L2 and L3 multicast routes.
  • IPv4 and IPv6 longest prefix match (LPM) or prefix entries that are supported in UFT through algorithmic longest prefix match (ALPM).

  [See forwarding-options.]

• Support for redistribution of IPv4 routes with IPv6 next hop into BGP.

  [See Understanding Redistribution of IPv4 Routes with IPv6 Next Hop into BGP.]

Software installation and upgrade

• Support for zero-touch provisioning (ZTP) on the management and the WAN interfaces.

  [See Zero Touch Provisioning Overview.]

• Support for secure boot—The implementation is based on the UEFI 2.4 standard.

  [See Junos OS Software Installation and Upgrade Guide.]