JNCIE-ENT key terms

A Junos commit mode that activates a configuration change for a limited period, typically 10 minutes, and rolls it back automatically unless the operator issues a second commit to confirm it.

The Junos operation that restores a previously committed configuration, referenced by number such as rollback 1 for the prior commit, and requires a commit to activate.

A saved known-good configuration that can be activated with a single command to recover a device whose current configuration has failed or become unreachable.

Reusable named blocks of Junos configuration applied to interfaces or the whole device, letting common settings be defined once and inherited rather than repeated per stanza.

The Junos statement that attaches a configuration group to a hierarchy level so the group's statements are inherited there, with local statements taking precedence over inherited ones.

A Python library for automating Junos devices through NETCONF, offering tables, views, and operational RPC calls usable over console, SSH, or Telnet connections.

An XML-based network management protocol that carries configuration and operational data between a manager and a device, and the transport Junos uses for programmatic configuration and telemetry retrieval.

A data modeling language that defines the structure, types, and constraints of configuration and state data exposed by a network device, commonly consumed through NETCONF.

A push model in which a device periodically exports selected operational data, such as interface counters, to a collector instead of waiting for a polling request.

An unsolicited notification a device sends to a management station when a defined event occurs, such as an interface transition or a threshold crossing.

A packet-sampling technology that exports truncated headers and interface counters to a collector, giving traffic visibility without capturing every packet.

Rapid Spanning Tree Protocol, defined in IEEE 802.1w, which converges faster than classic STP by using proposal and agreement handshakes and explicit port roles.

Multiple Spanning Tree Protocol, defined in IEEE 802.1s, which maps VLANs to independent spanning-tree instances so different topologies can carry different VLAN groups.

VLAN Spanning Tree Protocol, a Junos feature that runs a separate spanning-tree instance per VLAN, interoperating with PVST+ and allowing per-VLAN load balancing.

An edge-port safeguard that disables an access interface if it receives a bridge protocol data unit, preventing an attached host or switch from altering the spanning-tree topology.

A Junos port setting that keeps a designated port in a blocking state if it stops receiving BPDUs, guarding against unidirectional-link loops that spanning tree alone would miss.

A provider bridging technique that pushes an outer service VLAN tag onto customer frames, preserving the original customer VLAN tag across the provider network.

A feature that rewrites a frame's VLAN identifier as it enters or leaves an interface, allowing two sites that use the same VLAN ID to be joined transparently.

A VLAN segmentation scheme in which ports are isolated, community, or promiscuous, restricting which hosts can exchange frames within a single primary VLAN.

An integrated routing and bridging logical interface that carries Layer 3 configuration for a VLAN, letting the switch route between that VLAN and other subnets.

Virtual Router Redundancy Protocol, which lets multiple routers share a virtual IP address and MAC so hosts retain a default gateway if the master fails.

Bidirectional Forwarding Detection, a lightweight hello protocol that detects path failures in sub-second time and can trigger routing or redundancy failover faster than routing protocol timers.

A link-state interior gateway protocol that floods link-state advertisements within an area and computes shortest paths using the Dijkstra algorithm.

A Junos value ranking the trustworthiness of routes from different sources, used to select among competing paths to the same destination before metric comparison.

The compiled table, built from the routing table, that the packet forwarding engine consults to decide each packet's next hop and outgoing interface.

Ethernet VPN, a BGP control-plane technology that distributes MAC and IP reachability among participating devices, enabling multipoint Ethernet services with efficient flooding control.

A tunneling encapsulation that carries Layer 2 frames inside UDP over an IP network, using a 24-bit VNI to identify up to millions of logical segments.

A VXLAN tunnel endpoint that encapsulates frames from a local segment into VXLAN and decapsulates traffic arriving from remote endpoints.

The EVPN route that advertises a MAC address, optionally with an IP address, allowing remote devices to learn end-host location and install forwarding entries.

The EVPN inclusive multicast route that signals which devices participate in a given EVPN instance, building the flood list for broadcast, unknown unicast, and multicast traffic.

A design in which two or more devices attach to the same customer segment and coordinate through EVPN routes, providing redundant active-active or active-standby access without spanning tree.

The elected device in an EVPN multihoming group responsible for forwarding broadcast, unknown unicast, and multicast traffic toward a segment, chosen per Ethernet segment.

The component that inspects incoming packets, using behavior aggregates or multifield filters, and assigns a forwarding class and loss priority before queuing.

A label grouping traffic with the same per-hop behavior, mapped to an output queue on an interface and used for scheduling and buffering decisions.

A CoS component that meters traffic against a bandwidth and burst allowance and applies actions such as discard or loss-priority marking to conforming or exceeding packets.

A CoS resource that defines transmit rate, buffer size, and priority for an output queue, controlling how bandwidth is shared among forwarding classes.

A set of fill-level thresholds and drop probabilities applied to a queue, enabling weighted random early discard before the queue is completely full.

A CoS rule that sets the CoS or DSCP bits of outgoing packets to match their assigned forwarding class and loss priority, so downstream devices see consistent marking.