Release 8.3 Features

The following features have been added to JUNOS Release 8.3. Following the description is the title of the manual or manuals to consult for further information. For a complete list of manuals, see Table 7, Table 8, and Table 10.

Hardware

• Multi-Rate PICs—New SONET/SDH PICs with small form-factor pluggable (SFP) transceivers allow you to provision and control bandwidths. These new PICs use the next-generation ASIC and support Type 2 FPC interfaces. Hot-pluggable SFPs are used as optical transponders. The following rate-selectable PICs are supported:
  • 1-port SONET/SDH OC48/STM16 Multi-Rate PIC with SFP
  • 4-port SONET/SDH OC12/STM4 Multi-Rate PIC with SFP
  • 4-port SONET/SDH OC3/STM1 Multi-Rate PIC with SFP

  These PICs support existing SONET/SDH configurations and also support bandwidth control. For example, you can configure each port of a 4-port OC12 PIC to be in OC3 or OC12 speed independently when this PIC is in 4xOC12 concatenated mode.

  To specify speed in concatenated mode (the default), include the speed (oc3 | oc12 | oc48) statement at the [edit interfaces so-fpc/pic/port] hierarchy level. To specify the speed in nonconcatenated mode, include the speed (oc3 | oc12) statement at the [edit interfaces so-fpc/pic/port] hierarchy level. To configure the PIC to operate in nonconcatenated mode, include the no-concatenate statement at the [edit chassis fpc slot-number pic pic-number] hierarchy level. To display interface information, use existing SONET/SDH interface operational mode commands. For digital diagnostics monitoring (DDM), issue the show interfaces diagnostics optics so-fpc/pic/port command. [Network Interfaces, System Basics, PIC Guides]

• Optics support for SFP transceivers on Multi-Rate PICs—You can display SFP transceiver optics information for the Multi-Rate SONET/SDH Type 2 PICs.

  Support for diagnostics can vary from one SFP transceiver to another. The SFP toolkit detects the level of diagnostic support provided by the SFP transceiver being queried and displays output accordingly.

  To view SFP diagnostics, issue the show interfaces diagnostics optics so-fpc/pic/port operational mode command. PIC types that do not support diagnostic monitoring will not display output. [Interfaces Command Reference]

• Support for 1-port SONET/SDH OC192 PIC with XFP (M120 routing platform)—The M120 routing platform now supports the 1-port OC192 PIC with XFP. [M120 PIC Guide]
• New Gigabit Ethernet uPIMs (J4350 and J6350 Services Routers)—Universal PIMs (uPIMs) are compatible with both high-speed slots and regular slots. The three uPIMs are 8-port and 16-port with RJ-45 connectors, and 6-port with SFP connector. The 16-port uPIM is two-high (uses two slots). The primary purpose of these new uPIMs is to provide port density. The uPIMs can be installed in either high-speed or regular slots on J4350 and J6350 Services Routers running JUNOS Release 8.3 and later. [J4350 and J6350 Getting Started Guide]

  Caution: You can install a maximum of three 8-port Gigabit Ethernet uPIMs in a single J4350 or J6350 chassis. If you install the maximum of three 8-port Gigabit Ethernet uPIMs, only one of the remaining slots can be occupied, and it cannot contain an ePIM, uPIM, or Avaya VoIP module.

• New MultiServices PIC (M120, M320, T320, and T640 routing platforms)—The new MultiServices 500 PIC is now available. It currently supports only stateful firewall, NAT, and flow monitoring services. [Services Interfaces, PIC Guides]

User Interface and Configuration

• Enhanced IPv6 statistics—Service providers can now use JUNOS software to track ingress and egress IPv6-specific packets and bytes traversing the router. When you enable IPv6 statistics, the Packet Forwarding Engine differentiates between IPv4 and IPv6 traffic, providing statistical accounting.

  You can enable IPv6 statistics from the Packet Forwarding Engine or from the CLI. To enable IPv6 statistics from the Packet Forwarding Engine, issue the set route-accounting inet6 (enable | disable) command. IPv6 statistics are disabled by default after a reboot, prior to interaction with the Routing Engine. To enable IPv6 statistics from the CLI, include the route-accounting statement at the [edit forwarding-options family inet6] hierarchy level.

  To display IPv6 statistics from the Packet Forwarding Engine, issue the show interfaces statistics device command to display statistics for the device and for all logical interfaces connected to the device. To display IPv6 statistics from the CLI, use the show interfaces interface-name statistics detail command. [Network Interfaces, Policy]

  Note: IQ2 PICs do not support IPv6 statistics. [PR/102075]

• J-Web CLI terminal access—Provides the ability to connect to the CLI from the J-Web user interface by means of a Java applet running on the Web client that provides SSH2 support to connect to the router. To enable this capability, select Diagnose > CLI Terminal. [J-Web User Guide]
• J-Web enhancements—Enables the Web administrator to configure the maximum number of simultaneous user sessions and a default session timeout for all users. To configure user sessions, include the session-limit and idle-timeout statements at the [edit system services web-management session] hierarchy level. You can also configure the maximum number of simultaneous child processes in response to user requests by including the active-child-process statement at the [edit system services web-management limits] hierarchy level. The show system users command adds tracking of who has logged in or out through J-Web.
• Accounting file storage location setting (J-series Services Routers only)—Enables you to set the default storage device for accounting log files to the DRAM disk file system instead of the compact flash. To configure this feature, include the nonpersistent statement at the [edit accounting-options file filename] hierarchy level. We recommend that you use the nonpersistent statement for all accounting files configured on your system. Using this setting minimizes the read/write traffic to the compact flash. [Network Management, J-series Administration]

  Caution: Because all files saved to DRAM are deleted when the system reboots, be sure to back up these files to ensure that they are not lost.

• Output from the show | compare command has been enhanced to more accurately reflect configuration changes. This includes more intelligent handling of order changes in lists. [CLI User Guide]

Interfaces and Chassis

• Link protection support on aggregated Ethernet interfaces—With this feature, VLAN shaping can be supported on aggregated Ethernet interfaces when link protection is enabled on the aggregated Ethernet interface. When VLAN shaping is configured on aggregated Ethernet interfaces with link protection enabled, the shaping is actually applied to the active child link. To configure link protection on aggregated Ethernet interfaces, include the link-protection statement in the configuration and designate a primary and backup link. Traffic passes only through the designated primary link. This includes transit traffic and locally generated traffic on the router. When the primary link fails, traffic is routed through the backup link. You also can reverse traffic from the designated backup link to the designated primary link.

  To enable link protection on aggregated Ethernet interfaces, include the link-protection statement at the [edit interfaces aex aggregated-ether-options] hierarchy level. To revert to sending traffic to the primary designated link when traffic is passing through the designated backup link, issue the revert command option, for example request interfaces revert aex. To configure a primary and a backup link, include the primary and backup statements at the [edit interfaces ge-fpc/pic/port gigether-options 802.3ad aex] or the [edit interfaces fe-fpc/pic/port fastether-options 802.3ad aex] hierarchy level. To disable link protection, delete the link-protection statement at the [edit interfaces aex aggregated-ether-options] hierarchy level. To display the active, primary, and backup link for an aggregated Ethernet interface, issue the show interfaces redundancy aex operational mode command. [Network Interfaces, Interfaces Command Reference]

• Support for untagged packets on ports configured for 802.1Q flexible VLAN tagging (1-, 4-, and 8-port Gigabit Ethernet IQ2 PICs, and 1-port 10-Gigabit Ethernet IQ2 PICs)—Previously, when you configured a port for flexible VLAN tagging, packets received on the port that matched the VLAN tagging parameters were accepted and untagged packets were dropped. Now, untagged packets are accepted on the same flexible VLAN tagged port. To accept untagged packets, include the native-vlan-id and flexible-vlan-tagging statements at the [edit interfaces interface-name] hierarchy level. You must configure the logical interface on which untagged packets are to be received with the same native VLAN ID as that configured on the physical interface. To configure the logical interface, include the vlan-id statement (matching the native-vlan-id value on the physical interface) at the [edit interfaces interface-name unit logical-unit-number] hierarchy level. [Network Interfaces]
• Layer 2.5 VPNs support ISO family and MPLS family over TCC (M320 and T-series routing platforms)—Interfaces supporting TCC (Ethernet, extended VLANs, PPP, HDLC, ATM, and Frame Relay) now support ISO traffic and MPLS traffic on Layer 2.5 VPNs. Previously, Layer 2.5 VPNs supported only inet traffic. For a protocol to be supported on a Layer 2.5 VPN, you must configure both ends of the VPN with the protocol configuration. IPv6 is not supported.

  To enable ISO or MPLS traffic over TCC, include the mpls or iso statement at the [edit interfaces interface-name unit logical-unit-number family tcc protocol] hierarchy level. To display which protocol is supported for an interface, issue the show interfaces interface-name extensive operational mode command. The protocol is displayed in the Flags field.

  To enable ISO over TCC in cases in which the Ethernet interface is on a customer-edge (CE) router, include the point-to-point statement at the [edit protocols isis interface interface-name] hierarchy level on the CE router. When you include this statement, the IS-IS protocol treats the Ethernet interface as point to point, even though the actual interface is a LAN interface.

  The M-series routing platforms continue to support only inet traffic for Layer 2.5 VPNs. [Network Interfaces]

  Note: The following PICs have been specifically verified to support this feature: 10-Gigabit Ethernet PIC with XENPAK Long Reach Optic Module Gigabit Ethernet IQ PIC, 2-port SFP 10-Gigabit Ethernet IQ2 PIC with XFP

• Extended support for external synchronization (M40e routing platform)—You can configure external synchronization to synchronize the internal Stratum 3 clock with an external source and then synchronize the chassis interface clock to that source. This feature can use external timing sources connected to the craft interface panel. External synchronization was previously implemented on the M320 routing platform only. [System Basics]
• PAP support for PPP authentication. (J-series, M-series, and T-series routing platforms)—Previously, PPP supported only CHAP authentication. PPP now supports the Password Authentication Protocol (PAP), a simple two-way handshake to establish identity. PAP is used after the link establishment phase (LCP up), during the authentication phase. JUNOS software can support PAP in one direction (egress or ingress), and CHAP in the other.

  To configure PAP support on the physical interface, include the pap statement with options at the [edit interfaces interface-name ppp-options] hierarchy level. To configure PAP support on the logical interface, include the pap statement with options at the [edit interfaces interface-name unit logical-unit-number ppp-options] hierarchy level. To view information about the PAP states, issue the show ppp summary and show ppp interfaces operational mode commands. [Network Interfaces, Interfaces Command Reference]

• Unnumbered Ethernet enhancements—The unnumbered Ethernet feature, which was introduced in JUNOS Release 8.2, has been enhanced. Unnumbered Ethernet enables you to configure IPv4 processing on an Ethernet interface without assigning an explicit IP address to the interface. Using unnumbered Ethernet interfaces conserves IPv4 address space by enabling many ports to share a single IP address, and provides a way to point a static route to an interface on broadcast media.

  The following enhancements have been made to the unnumbered Ethernet feature. These enhancements remove certain restrictions that were in effect for the feature in JUNOS Release 8.2.

  • You can now configure a loopback interface as the donor interface for an unnumbered Ethernet interface. (In the previous release, the donor interface could not be a loopback interface.)
  • Addresses configured on the point-to-point donor interface can now be /32 addresses. (In the previous release, you could not configure a /32 address on a point-to-point donor interface.)
  • Addresses configured on the donor interface need not be subnet addresses if the donor interface is a point-to-point interface. (In the previous release, addresses on the point-to-point donor interface had to be subnet addresses.)
  • Routes added on an unnumbered Ethernet interface need not be in the subnet of an address on the donor interface if the donor interface has a /32 address configured on it. (In the previous release, routes on the unnumbered Ethernet interface had to be in the subnet of an address on the donor interface.)
  • Running IGMP and PIM is now supported on unnumbered Ethernet interfaces that directly face the host and have no downstream PIM neighbors. IGMP and PIM are not supported on unnumbered Ethernet interfaces that act as upstream interfaces in a PIM topology.

  [Network Interfaces, Routing Protocols, Interfaces Command Reference]

• Clear MAC address information (Gigabit Ethernet IQ2 interfaces)—Enables you to issue the clear interfaces mac-database operational mode command to clear media access control (MAC) address information. [Interfaces Command Reference]
• Ethernet link aggregation support (M120 routing platform)—As with T-series routing platforms, the M120 routing platform now supports 128 groups per chassis and 16 links per group. To configure an aggregated link, include the 802.3ad aex statement at the [edit interfaces interface-name fastether-options] or [edit interfaces interface-name gigether-options] hierarchy level. [Network Interfaces]
• ISDN PRI support on channelized T1 and E1 PIMs (J-series Services Routers)—Channelized T1 and E1 interfaces now provide ISDN Primary Rate Interface (PRI) connectivity for dial-in and callback and for use as primary or backup network connections. You can configure up to 23 channelized T1 time slots and up to 30 channelized E1 time slots as an ISDN PRI group, with the 24th T1 time slot and the 16th E1 time slot operating as the D-channel to control the group of time slots as B-channels. T1 and E1 time slots unused by ISDN PRI can operate normally as DS0 interfaces. ISDN PRI B-channels run at 64 Kbps, but do not support the 56-Kbps line rate. ISDN PRI operates in a Services Router chassis with ISDN Basic Rate Interface PIMs and supports ATT5ESS, ETSI (NET5), National ISDN (NI2), DMS100, and NTT switch types. The following ISDN features are supported:
  • Bandwidth on demand using Multilink Point-to-Point Protocol (MLPPP)
  • Dial-on-demand routing (DDR) backup
  • Dial backup
  • Authentication (CHAP and PAP)
  • On-demand routing
  • Dialer profiles and maps
  • Dial-in
  • Dialer callback

  ISDN PRI options include a new bchannel-allocation option that enables you to allocate a free B-channel for dial-out calls. This setting directs the router to select the first free B-channel with the highest number (default) or lowest when dialing out. [J-series Basic Configuration Guide]

• CoS scheduler transmit-rate enhancement (J-series Services Routers only)—The minimum size of the class-of-service (CoS) scheduler transmit rate is increased to 1/10,000 of the interface speed. The smaller increments provide finer control of the transmit rate, which is especially useful for high-speed interfaces such as Fast Ethernet and Gigabit Ethernet interfaces on J4350 and J6350 routers. Previously, you could only assign 1/1,000th part of the interface speed; if the speed of the interface was 1000 Mbps, the minimum transmit rate would have been 1000 Mbps x 1/1000=1 Mbps. Now with the minimum transmit rate of 1/10,000 for the same interface, you can set the transmit rate of a queue to be 1000 x 1/10,000 = 100 Kbps. [J-series Advanced Configuration Guide]

Services Applications

• Support for twice NAT—Enables you to configure both source addresses and destination addresses for translation as packets traverse the NAT router. For example, you would use twice NAT when you are connecting two networks in which all or some addresses in one network overlap with addresses in another network (whether the network is private or public).

  Twice NAT is supported on all AS PICs and MultiServices PICs that support traditional NAT. It is also supported on J-series Services Routers.

  To configure this feature, include the destination-address or destination-address-range statement and the source-address or source-address range statement at the [edit services nat rule rule-name term term-name from] hierarchy level. An enhancement to the show services stateful-firewall flows command displays the two NAT actions on the same flow. AS PIC system log messages for a twice NAT rule now display information about both the source address and the destination address.

  Note: For twice NAT, you can apply an overload pool for source addresses and combined source and destination addresses (twice NAT). In the current release, you cannot configure application-layer gateways (ALGs) under stateful firewall rules or CoS rules when twice NAT is configured in the same service set. Do not include the application-protocol statement with the application statement for twice NAT configurations. By default, the twice NAT feature can affect IP, TCP, and UDP headers embedded in the payload of ICMP error messages. You can include the protocol tcp and protocol udp statements with the application statement for twice NAT configurations. When you configure one or more stateful firewall rules with an ALG in the same service set as a twice NAT configuration, the router ignores the ALG configuration. NAT applies only the IP address and TCP or UDP headers, but not the payload.

  [Services Interfaces, System Basics and Services Command Reference]

• LSQ no-termination-request enhancement—In interchassis and intrachassis redundant LSQ configurations that use MLPPP and SONET APS, the feature that enables you to inhibit termination requests when the primary LSQ PIC fails has been extended. You can now also inhibit termination requests when the link PICs fail. To configure, include the no-termination-request statement at the [edit interfaces interface-name ppp-options] hierarchy level for each of the link PICs. The supported PIC types include Channelized OC3, OC12, STM1, and STM4 IQ PICs. [Services Interfaces, Network Interfaces]
• RPM jitter measurement enhancement—Changes the way in which jitter is calculated. Previously, jitter calculations included delay on the other system and resulted in artificially high jitter numbers, which could imply that the network was experiencing more congestion than it actually was.

  On M-series and T-series platforms, AS PIC timestamping enables measurement of time delays for probe transit in egress (source to destination) and ingress (destination to source) directions. On J-series Services Routers, this feature enhances jitter calculation on RPM probes that provide the hardware timestamp capability. The jitter is calculated for two consecutive RPM probes, which helps the jitter value remain low and provides measurable values for egress jitter, ingress jitter, and interarrival jitter. The show services rpm probe-results and show services rpm history-results detail commands provide enhanced statistics for these measurements. [System Basics and Services Command Reference, J-series Administration Guide]

• New templates for exporting IPv4 and MPLS flow records—Support is added for version 9 flow record formats, as described in RFC 3954, Cisco Systems NetFlow Services Export Version 9. The version 9 format allows you to define a flow record template suitable for active monitoring of IPv4 traffic, MPLS traffic, or a combination of the two. Templates and the fields included in the template are transmitted to the collector periodically, and the collector need not be aware of the router configuration.

  To configure a version 9 template, include the template statement at the [edit services flow-monitoring version9] hierarchy level. You specify the traffic type by including either the ipv4-template, mpls-ipv4-template, or mpls-template statement within the template definition, and you can optionally include settings for flow-active-timeout, flow-inactive-timeout, option-refresh-rate, and template-refresh-rate. In addition, the templates for MPLS traffic allow you to specify up to three positions for the MPLS label information. To activate the new template definition, include the template statement at the [edit forwarding-options sampling output cflowd host version9] hierarchy level. If you define a template for MPLS traffic, you also must include the family mpls statement at the [edit forwarding-options sampling input] hierarchy level. For verification purposes, a new show services accounting aggregation template template-name operational command is added. [Services Interfaces, Policy Framework, Feature Guide, System Basics and Services Command Reference]

  Note: The new MPLS and MPLS-IPv4 flow aggregation templates are not supported on TX Matrix platforms in JUNOS Release 8.3R1. However, the IPv4 template is supported. [PR/100103] Packets originating from the Routing Engine (for example, outbound OSPF hello packets) cannot be sampled. This restriction applies to cflowd v5 also.

General Routing

• Support for unicast RPF statement extended to routing instances—You can now configure the operation of unicast reverse-path forwarding (RPF) check for a routing instance. To configure this feature, include the unicast-reverse-path (active-paths | feasible-paths) statement at the [edit routing-instances instance-name routing-options forwarding-table] hierarchy level. Unicast RPF check is a tool to reduce forwarding of IP packets that might be spoofing an address. You must enable unicast RPF check on an interface by including the rpf-check <fail-filter filter-name> statement at the [edit interfaces interface-name unit logical-unit-number family (inet | inet6)] hierarchy level. Previously, you could configure the operation of unicast RPF check only for a router or a logical router. [Routing Protocols, Network Interfaces]

Routing Protocols

• Support for IPSec authentication for OSPFv2—Enables you to configure IPSec authentication for an OSPFv2 interface, the remote endpoint of a sham link, or an OSPFv2 virtual link:
  • To configure IPSec authentication for an OSPFv2 interface, include the ipsec-sa statement at one or more of the following hierarchy levels: [edit logical-routers logical-router-name protocols ospf area area-id interface interface-name], [edit logical-routers logical-router-name routing-instances instance-name protocols ospf area area-id interface interface-name], [edit protocols ospf area area-id interface interface-name], or [edit routing-instances instance-name protocols ospf area area-id interface interface-name].
  • To configure IPSec authentication for the remote endpoint of a sham link, include the ipsec-sa statement at the following hierarchy levels: [edit logical-routers logical-router-name routing-instances instance-name protocols ospf area area-id sham-link-remote address] and [edit routing-instances instance-name protocols ospf area area-id sham-link-remote address].
  • To configure IPSec authentication for an OSPFv2 virtual link, include the ipsec-sa statement at one or more of the following hierarchy levels: [edit logical-routers logical-router-name protocols ospf area area-id virtual link], [edit logical-routers logical-router-name routing-instances instance-name protocols ospf area area-id virtual link], [edit protocols ospf area area-id virtual link], or [edit routing-instances instance-name protocols ospf area area-id virtual link].

  You configure IPSec separately. Previously, JUNOS supported IPSec authentication for OSPFv3 only. [Routing Protocols, Services Interfaces]

• Enhancement to clear ospf database operational mode command—The clear ospf database command includes new options that enable you to purge specific link-state advertisement (LSA) entries. Previously, the only options were to purge all the entries in the LSA database or all the LSAs associated either with a routing instance or a logical router. The new options are advertising-router, area, asbrsummary, external, link-local, lsa-id, netsummary, network, nssa, opaque-area, and router. [Protocols Command Reference]
• Support for BFD protocol extended to logical routers (M-series, MX-series, and T-series routing platforms)—You can now configure Bidirectional Failure Detection (BFD) in logical router configurations. To configure BFD timers, include the bfd-liveness-detection statement at the [edit logical-routers logical-router-name protocols] and the [edit logical-routers logical-router-name routing-instances instance-name protocols] hierarchy levels for any of the following protocols: RIP, BGP, OSPF, or IS-IS. To configure BFD timers for a static route, include the bfd-liveness-detection statement at the [edit logical-routers logical-router-name routing-options static route destination-prefix] and [edit logical-routers logical-router-name routing-instances instance-name routing-options static route destination-prefix] hierarchy levels. Previously, you could configure BFD only for routing protocols and static routes configured on a main router and for routing instances configured in a main router. [Routing Protocols]

MPLS Applications

• ECMP reliability—The reliability of an equal cost multipath (ECMP) set has been improved. Fewer packets are dropped in the event of an ECMP set failure. [MPLS Applications]
• Point-to-multipoint LSP object formats—The MX960 Ethernet Services Router now supports a number of new point-to-multipoint LSP object formats. These new objects are described in the Internet draft, draft-ietf-mpls-rsvp-te-p2mp-02.txt, “Extensions to RSVP-TE for Point to Multipoint TE LSPs.” [MPLS Applications]

Multicast

• IGMP and MLD enhancements—Enhancements to the IGMP and MLD protocols include the immediate-leave statement (for both IGMP and MLD) and the promiscuous-mode statement (for IGMP only). The immediate-leave statement enables you to remove a multicast group membership from the router as soon as the router receives a leave group message (for IGMP) or the last host leaves the group (for MLD) and without sending out a group membership query message. The promiscuous-mode statement (IGMP only) enables the router to accept IGMP reports from subnetworks that are not associated with its interfaces. To configure these protocol enhancements, include the appropriate statement at the [edit protocols (igmp | mld) interface interface-name] hierarchy level. The show igmp interfaces and show mld interfaces commands have been updated to display the state (On or Off) of the immediate leave option. [Multicast, Protocols Command Reference]
• IGMP and PIM support for unnumbered interfaces—Configuring unnumbered access (Ethernet) interfaces enables you to configure IPv4 processing on an access interface without assigning an explicit IP address to the interface. Support for this feature has been carried over for interfaces that also run IGMP and PIM. [Network Interfaces, Routing Protocols, Interfaces Command Reference, Feature Guide]

  Note: When you enable IGMP on an unnumbered Ethernet interface that uses a /32 loopback address as a donor address, you must configure IGMP promiscuous mode to accept the IGMP packets received on this interface.

• Bandwidth management—You can better manage multicast traffic by applying bandwidth admission control where an outgoing interface is added to the forwarding path only if there is still enough bandwidth. This is done by specifying the maximum bandwidth for each multicast interface and defining the bandwidth used by each multicast flow. The maximum-bandwidth statement is applied to a physical or logical interface at the [edit routing-options multicast] hierarchy level to define the available bandwidth in bits per second for the interface. The bandwidth statement at the [edit routing-options multicast flow-map map-name] hierarchy level enables you to apply bandwidth values to individual flows defined by a flow map. Either you can statically define the bandwidth values for any flows defined by the flow map or you can use the adaptive option to enable adaptive bandwidth autosense. This feature enables the router to calculate the bandwidth for each new flow based on periodically queried statistics. [Multicast]
• Support for mtrace functionality is now available on logical router configurations. [Multicast]

Routing Policy and Firewall Filters

• Extended DHCP relay agent (MX-960, M-series, and T-series platforms)—Enables you to configure extended DHCP relay options on the router and to enable the router to function as a DHCP relay agent. A DHCP relay agent forwards DHCP request and reply packets between a DHCP client on a subscriber’s computer and a DHCP server. You can use DHCP relay in carrier-edge applications such as video or IPTV to obtain configuration parameters, including an IP address, for your subscribers.

  To configure this feature, include the dhcp-relay statement at any of the following hierarchy levels:

  • [edit forwarding-options]
  • [edit logical routers logical-router-name forwarding-options]
  • [edit logical routers logical-router-name routing-instances instance-name forwarding-options]
  • [edit routing-instances instance-name forwarding-options]

  You can include the server-group statement at the [edit forwarding-options dhcp-relay] hierarchy level to configure a named group of DHCP server addresses for use by the extended DHCP relay agent on the router.

  To clear or display the DHCP relay agent configuration, issue the following new operational commands: clear dhcp relay binding, clear dhcp relay statistics, show dhcp relay binding, or show dhcp relay statistics. [Policy Framework, Protocols Command Reference]

  Note: The extended DHCP relay agent options configured with the dhcp-relay statement are incompatible with the DHCP/BOOTP relay agent options configured with the bootp statement. As a result, you cannot enable both the extended DHCP relay agent and the DHCP/BOOTP relay agent on the router at the same time.

• Access and access-internal routes—Enables support for two new types of routes: access routes and access-internal routes. These routes are used by the DHCP application on a video services router to represent either the end users or the networks behind the attached video services router.
  • An access route represents a network behind an attached video services router, and is set to a preference of 13.
  • An access-internal route is a /32 route that represents a directly attached end user, and is set to a preference of 12.

  To configure, import, and export access and access-internal routes in a routing policy, include the new access and access-internal match conditions at the [edit policy-options policy-statement policy-name term term-name from protocol] hierarchy level. To display configuration information for access and access-internal routes, issue the following operational commands: show route extensive, show route protocol access, and show route protocol access-internal. [Policy Framework, Protocols Command Reference]

VPNs

• Point-to-multipoint LSP support for VPLS—Enabling this feature for a VPLS routing instance enables the flooding of unknown unicast, broadcast, and multicast traffic using point-to-multipoint LSPs. For each VPLS routing instance, the provider edge (PE) router creates a dedicated point-to-multipoint LSP. Whenever VPLS discovers a new neighbor through BGP, a sub-LSP for this neighbor is added to the point-to-multipoint LSP for the routing instance. If there are n PE routers in the VPLS routing instance, the above procedure creates n point-to-multipoint LSPs in the network, in which each PE router is the root of the point-to-multipoint tree and includes the rest of the n–1 PE routers as leaf nodes.

  You can enable this feature incrementally on any PE router that is a part of a specific VPLS routing instance. The PE routers on which this feature is enabled use point-to-multipoint LSPs to flood traffic, whereas other PE routers in the same VPLS routing instance can use ingress replication to flood traffic. However, when you enable this feature on any PE router, you must ensure that all PE routers that are part of the same VPLS routing instance are upgraded to JUNOS Release 8.3 to support this feature.

  To enable this feature, configure either the static or label-switched-path-template option for the rsvp-te statement at the [edit routing-instance instance-name provider-tunnel] hierarchy level. You cannot include both of these options in the configuration at the same time. The static option creates a static point-to-multipoint LSP that includes all neighbors in the VPLS routing instance. The label-switched-path-template option provides a set of parameters for dynamically created point-to-multipoint LSPs. If you specify the default-template option for label-switched-path-template, the point-to-multipoint LSP is created with the default parameters. If you configure label-switched-path-template, you also need to include the template statement at the [edit protocols mpls label-switched-path lsp-name] hierarchy level. [VPNs]

Class of Service

• Average buffer WRED—By default, IQ PICs perform weighted random early detection (WRED) based on instantaneous buffer occupancy. You can now enable IQ PICs to perform WRED based on average buffer occupancy. To configure the weight value used in average buffer computations, include the red-buffer-occupancy statement at the [edit chassis fpc fpc-number pic slot-number] hierarchy level. The show interfaces extensive command output includes a new field (WRED Buffer Occupancy). [System Basics, Class of Service, Interfaces Command Reference]
• IQ2 interface forwarding class policer—Support is added for a rate limit applied at any of the eight egress queues of IQ2-based PICs. A transmit rate limit of zero used with the percentage option drops all packets in the queue.

  You implement the rate limit by applying a policer, and only one policer per queue is allowed. Policers are applied in this order: any queue-level policer, then interface-level, and finally a Layer 2 policer. To configure the rate limit, include the transmit-rate statement at the [edit class-of-service scheduler scheduler-name] hierarchy level. The show interfaces queue command output includes additional fields (RL-Dropped pkts and bytes). A new system log error message issues a warning when queue transmission rates have been exceeded. [Class of Service, Interfaces Command Reference]

• CoS for L2TP sessions—Enables session-aware CoS for IQ2 PICs. Dynamic (nonconfigured) CoS is enabled on L2TP LNS sessions or L2TP sessions with ATM VCs, as long as you configure L2TP to use an IQ2 PIC on the egress interface. By default, no CoS is applied to L2TP sessions. A new show class-of-service l2tp session command option displays related output. [Class of Service, Services Interfaces, System Basics and Services Command Reference]
• Egress GRE classification based on DSCP—Adds CoS support on generic routing encapsulation (GRE) and IP-IP tunnels. A BA classifier on an incoming physical interface can be applied to a GRE or IP-IP packet. This information is carried through the tunnel so that the IP packet on the customer-facing egress interface is queued based on the DiffServ code point (DSCP) setting in the header. No specific configuration is necessary. [Class of Service, Services Interfaces]

Network Management

• Support for sending traps over routing instances—You can now configure a router to send traps over any routing instance. Previously, traps were sent over the default routing instance (inet.0) only. To specify the routing instance for SNMPv1 and SNMPv2 trap targets, include the routing-instance instance-name statement at the [edit snmp trap-group group-name] hierarchy level. To specify the routing instance for SNMPv3 trap targets, include the routing-instance instance-name statement at the [edit snmp v3 target-address target-address-name] hierarchy level. [Network Management]
• Support for Internet draft draft-ietf-ospf-ospfv3-mib-11.txt, Management Information Base for OSPFv3 (Represented by mib-jnx-ospfv3mib.txt and implemented under the Juniper Enterprise jnxExperiment branch)—Read-only support is added for ospfv3NbrTable only. Object names are prefixed by jnx; for example: jnxOspfv3NbrTable, jnxOspfv3NbrAddressType, and jnxOspfv3NbrPriority. [Network Management]
• Expanded RMON alarm support—You can now adjust the sampling interval based on the alarm state and monitor CoS packet drop statistics for several interfaces simultaneously. To configure, include the falling-threshold-interval and request-type statements at the [edit snmp rmon alarm index] hierarchy level. In addition, the CoS MIB has been extended to include a new table (jnxCosInvQstatTable). [Network Management]

JUNOScope

• Script and configuration management—JUNOScope users can now edit archived configurations, save archived configurations to a local file system, and delete user-imported configurations. You can also manage JUNOS-based scripts such as commit scripts and event scripts. You can import these scripts into the JUNOScope CVS repository from the local file system and deploy them to a group of routers. In addition, you can view, edit, and delete these scripts using a text editor. To edit archived configurations, select Configuration > Repository > Edit. To delete archived configurations, select Configuration > Repository > Delete. To save archived configurations, select Configuration > Repository > Save. To manage JUNOS-based scripts, select Configuration > Repository > Scripts. [JUNOScope User Guide, Automation Guide]
• Inventory management—The JUNOScope software now provides Field Replaceable Unit (FRU) model numbers and Common Language Equipment Identifier (CLEI) code in Inventory Management system reports such as Hardware Inventory All and Hardware Inventory Search reports. This added information enables you to easily identify a hardware item and facilitates tracking and ordering of equipment and inventory management. To use this feature, from the JUNOScope main window, select Inventory Management > Reports > Inventory > Hardware Inventory All or select Inventory Management > Reports > Inventory > Hardware Inventory Search. [JUNOScope User Guide]
• TGM configuration—JUNOScope users can now configure the IP address of a Telephony Gateway Module (TGM) on J4350 and J6350 Services Routers and also archive and restore these configurations. You can view interface information for the TGM using the Looking Glass, and inventory information as part of hardware inventory reports. Configuration editing capability is not available for the TGM. [JUNOScope User Guide]

Automation and JUNOS XML API

• Merging of the event process (eventd) and system log process (syslogd)—Enables you to base your event policies on a larger set of system log messages. Each newly supported system log message has an event ID and a message attribute. Event IDs are based on the origin of the message, as follows:

  Event ID	Origin
  SYSTEM	For messages coming through UNIX domain sockets
  KERNEL	For messages coming from the kernel
  PIC	For messages coming from PICs
  PFE	For messages coming from the Packet Forwarding Engine
  LCC	For messages coming from an LCC
  SCC	For messages coming from an SCC

  To specify the newly supported message types in your event policy configuration, include the events event-id statement and the attributes-match statement with the event-id.message matches “message” attribute at the [edit event-options policy policy-name] hierarchy level. [Automation]

• Unique filenames for uploaded files—When an event policy uploads files for analysis, the files are timestamped to ensure the file names are unique, as follows: router-name_filename_YYYYMMDD_HHMMSS. If multiple files are uploaded within a 1-second period, each file is given a unique number as well, as follows: router-name_filename_YYYYMMDD_HHMMSS_number. [Automation]
• Upload files created by event scripts—Allows you to create an output file directly in an event script. The output file is automatically uploaded to the destination you specify at the [edit event-options policy policy-name then event-script filename destination] hierarchy level. To specify an output file in an event script, include the <event-script-output-filename>filename </event-script-output-filename> tag as the first child element of the <output> tag. If you also include the output-filename statement in the configuration, the configuration overrides the <event-script-output-filename> tag in the script. [Automation]
• Emit event script output in XML—Enables you to specify that the output of an event script be emitted in XML instead of text. To configure the format, include the output-format statement at the [edit event-options policy policy-name then event-script script-name] hierarchy level. [Automation]

• Table 1 lists the JUNOS Extensible Markup Language (XML) operational request tag elements that are new in JUNOS Release 8.3, along with the corresponding CLI command and response tag element for each one.

  Table 1: JUNOS XML Tag Elements and CLI Command Equivalents New in JUNOS 8.3

  Request Tag Element	CLI Command	Response Tag Element
  <clear-dhcp-relay-binding-information>	clear dhcp relay binding	None
  <clear-dhcp-relay-statistics-information>	clear dhcp relay statistics	None
  <get-bfd-session-information>	show bfd session	<bfd-session-information>
  <get-cos-l2tp-session-map-information>	show class-of-service l2tp-session	<cos-l2tp-session-information>
  <get-dhcp-relay-binding-information>	show dhcp relay binding	<dhcp-relay-binding-information>
  <get-dhcp-relay-statistics-information>	show dhcp relay statistics	<dhcp-relay-statistics-information>
  <get-multicast-interface-information>	show multicast interface	<multicast-interface-information>
  <get-service-accounting-aggregation- template-information>	show services accounting aggregation template	<service-accounting-aggregation- template-information>
  <request-background-task-close>	request system services web-management task close	None
  <request-background-task-signal>	request system services web-management task signal	None
  <request-background-task-start-junoscript>	request system services web-management task start junoscript	<background-task-information>

  [JUNOS XML Operational Reference]

System Log Tags

New System Log Tags

The following system log tags are new in JUNOS Release 8.3:

• ACCT_FORK_LIMIT_EXCEEDED
• CHASSISD_ANTICF_RE_CHECK_FAILED
• CHASSISD_ANTICF_RE_ROM_READ_FAIL
• CHASSISD_ANTICF_RE_SHA_READ_FAIL
• CHASSISD_IFDEV_GET_BY_NAME_FAIL
• CHASSISD_PIC_SPEED_INVALID
• COSD_EXACT_KEYWORD_NOT_SUPPORTED
• COSD_L2TP_COS_NOT_SUPPORTED
• COSD_MANY_HIGH_PRIORITY_QUEUES
• COSD_PER_SESSION_SCHED_NEEDED
• COSD_PIC_NOT_SESSION_QUEUED
• COSD_RATE_LIMIT_NOT_SUPPORTED
• COSD_TRI_COLOR_NOT_SUPPORTED
• EVENTD_POLICY_UPLOAD_FAILED
• EVENTD_XML_FILENAME_INVALID
• JDHCPD_ARGUMENT_FAILURE
• JDHCPD_BIND_FAILURE
• JDHCPD_DAEMON_FAILURE
• JDHCPD_EVLIB_CREATE_FAILURE
• JDHCPD_EVLIB_MAINLOOP_FAILURE
• JDHCPD_PERMISSION_FAILURE
• JDHCPD_PIDLOCK_FAILURE
• JDHCPD_PIDUPDATE_FAILURE
• JDHCPD_RECVMSG_FAILURE
• JDHCPD_RTSOCK_FAILURE
• JDHCPD_SENDMSG_FAILURE
• JDHCPD_SENDMSG_NOINT_FAILURE
• JDHCPD_SETSOCKOPT_FAILURE
• JDHCPD_SOCKET_FAILURE
• L2TPD_COS_PROFILE_ADD
• L2TPD_COS_PROFILE_DELETE
• L2TPD_FILTER_FILE_OPEN_FAILED
• L2TPD_TUNNEL_DEST_IF_LOOKUP_FAIL
• PPPD_PAP_GETHOSTNAME_FAILED
• PPPD_PAP_INVALID_IDENTIFIER
• PPPD_PAP_INVALID_OPCODE
• PPPD_PAP_LOCAL_PASSWORD_UNAVAIL
• PPPD_PAP_OPERATION_UNEXPECTED
• VRRPD_AUTH_INFO_INVALID
• VRRPD_GET_TRAP_HEADER_FAILED
• VRRPD_NEW_BACKUP
• VRRPD_NEW_MASTER

Removed System Log Tags

The following system log tags are no longer supported in JUNOS Release 8.3:

• AUTOD_MEMORY_ALLOCATION_FAILURE
• CHASSISD_OVER_TEMP_SHUTDOWN
• ECCD_DAEMONIZE_FAILED
• ECCD_DUPLICATE
• ECCD_LOOP_EXIT_FAILURE
• ECCD_NOT_ROOT
• ECCD_PCI_FILE_OPEN_FAILED
• ECCD_PCI_READ_FAILED
• ECCD_PCI_WRITE_FAILED
• ECCD_PID_FILE_LOCK
• ECCD_PID_FILE_UPDATE
• ECCD_TRACE_FILE_OPEN_FAILED
• ECCD_USAGE
• SNMPD_LOOPBACK_ADDR_ERROR

Changed System Log Tags

The logged text for the following system log messages has changed to the indicated form:

• CHASSISD_COMMAND_ACK_ERROR—Error occurred when fru-name fru-slot reported its online status: error-message (error code error-code)
• UI_COMMIT—User 'username' requested 'command' operation (comment: message)

System Log

• The structured-data statement at the [edit system syslog file filename] hierarchy level formats system log messages that are directed to a file in the format specified in section 6.3 of Internet draft http://www.ietf.org/internet-drafts/draft-ietf-syslog-protocol-19.txt, The Syslog Protocol.