Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?


interface (Protocols IS-IS)



Hierarchy Level

Release Information

Statement introduced before Junos OS Release 7.4.

Statement introduced in Junos OS Release 9.0 for EX Series switches.

Statement introduced in Junos OS Release 12.1 for the QFX Series.

Statement introduced in Junos OS Release 14.1X53-D20 for the OCX Series.

no-eligible-remote-backup option introduced in Junos OS Release 14.2 for the MX Series.

interface-group-holddown-delay option introduced in Junos OS Release 15.2 for the MX Series.

max-hello-size option introduced in Junos OS Release 17.1 for the MX Series.


Configure interface-specific IS-IS properties. To configure more than one interface, include the interface statement multiple times.

Enabling IS-IS on an interface (by including the interface statement at the [edit protocols isis] or the [edit routing-instances routing-instance-name protocols isis] hierarchy level), disabling it (by including the disable statement), and not actually having IS-IS run on an interface (by including the passive statement) are mutually exclusive states.


all | interface-nameSelect either automatic interfaces or specify the name of an existing interface.
  • all—Have Junos OS create IS-IS interfaces automatically. If you include this option, disable IS-IS on the management interface (fxp0).

  • interface-name—Name of an interface. Specify the full interface name, including the physical and logical address components.

checksumEnable checksums for packets on this interface. Junos OS supports IS-IS checksums as documented in RFC 3358, Optional Checksums in Intermediate System to Intermediate System (ISIS).

The checksum cannot be enabled with MD5 hello authentication on the same interface.

csnp-interval seconds | disableConfigure the interval between complete sequence number PDUs (CSNPs) on a LAN interface, or disable it. If the routing device is the designated router on a LAN, IS-IS sends CSN packets every 10 seconds.

The csnp-interval statement is effective only when configured on LAN interfaces.

secondsNumber of seconds between the sending of CSNPs. The range is 1 through 65,535 seconds.
disableDo not send CSNPs on this interface.
disableDisable IS-IS on the routing device, on an interface, or on a level. Enabling IS-IS on an interface (by including the interface statement at the [edit protocols isis] or the [edit routing-instances routing-instance-name protocols isis] hierarchy level), disabling it (by including the disable statement), and not actually having IS-IS run on an interface (by including the passive statement) are mutually exclusive states. IS-IS is enabled for Level 1 and Level 2 routers on all interfaces on which family iso is enabled.
flood-groupEnables IS-IS support for flood-group. This feature limits link-state packet data unit (PDU) flooding over IS-IS interfaces.
hello-padding adaptive | disable | loose | strictConfigure padding on hello packets to accommodate asymmetrical maximum transfer units (MTUs) from different hosts. This helps to prevent a premature adjacency Up state when one routing device’s MTU does not meet the requirements to establish the adjacency.

As an OSI Layer 2 protocol, IS-IS does not support data fragmentation. Therefore, maximum packet sizes must be established and supported between two routers. During adjacency establishment, the IS-IS protocol makes sure that the link supports a packet size of 1492 bytes by padding outgoing hello packets up to the maximum packet size of 1492 bytes. This is the default behavior of the Junos OS IS-IS implementation. However, Junos OS provides an option to disable hello padding that can override this behavior.

adaptiveOn point-to-point connections, the hello packets are padded from the initial detection of a new neighbor until the neighbor verifies the adjacency as Up in the adjacency state type, length, and value (TLV) tuple. If the neighbor does not support the adjacency state TLV, then padding continues. On LAN connections, padding starts from the initial detection of a new neighbor until there is at least one active adjacency on the interface. Adaptive padding has more overhead than loose padding and is able to detect MTU asymmetry from one side of the connection. This one-sided detection can result in generation of extra link-state PDUs that are flooded throughout the network. Specify the adaptive option to configure enough padding to establish an adjacency to neighbors.
disablePadding is disabled on all types of interfaces for all adjacency states. Specify the disable option to accommodate interfaces that support less than the default packet size of 1492 bytes.
looseThe hello packet is padded from the initial detection of a new neighbor until the adjacency transitions to the Up state. Loose padding might not be able to detect certain situations such as asymmetrical MTUs between the routing devices. Specify the loose option to configure enough padding to initialize an adjacency to neighbors.
strictPadding is done on all interface types and for all adjacency states, and is continuous. Strict padding has the most overhead. The advantage is that strict padding detects MTU issues on both sides of a link. Specify the strict option to configure padding to allow all adjacency states with neighbors.
interface-group-holddown-delay secondsNumber of seconds before the routing device replaces the bandwidth based metric. This value configures the time interval that the IS-IS takes before replacing the metric with the new metric value and before flooding the new metric to the labeled-switched paths (LSPs). When the bundle changes from a worse bandwidth based metric to a better metric the system waits for the configured time before switching to the new metric. The system also uses this time delay when a link status changes from down to up or when a member link changes status from degrade to non-degrade.

By default, the time delay for a routing device is 20 seconds.

ldp-synchronizationEnable synchronization by advertising the maximum cost metric until LDP is operational on the link. LDP distributes labels in non-traffic-engineered applications. Labels are distributed along the best path determined by IS-IS. If the synchronization between LDP and IS-IS is lost, the label-switched path (LSP) goes down. Therefore, IS-IS and LDP synchronization is beneficial. When LDP synchronization is configured and when LDP is not fully operational on a given link (a session is not established and labels are not exchanged), IS-IS advertises the link with the maximum cost metric. The link is not preferred but remains in the network topology.

LDP synchronization is supported only on point-to-point interfaces and LAN interfaces configured as point-to-point interfaces under IS-IS. LDP synchronization is not supported during graceful restart. To advertise the maximum cost metric until LDP is operational for LDP synchronization, include the ldp-synchronization statement.

disableDisable LDP synchronization for IS-IS.
hold-time seconds The time period to advertise the maximum cost metric for a link that is not fully operational. The range is 1 through 65,535 seconds. The default is infinity.

When an interface has been in the holddown state for more than 3 minutes, a system log message with a warning level is sent. This message appears in both the messages file and the trace file.

lsp-interval millisecondsConfigure the link-state PDU interval time, in milliseconds. By default, the routing device sends one link-state PDU packet out an interface every 100 milliseconds. The valid range is 0 through 1000 milliseconds.

To disable the transmission of all link-state PDUs, set the interval to 0. Link-state PDU throttling by use of the lsp-interval statement controls the flooding pace to neighboring routing devices in order to not overload them.

Also, consider that control traffic (such as link-state PDUs and related packets) might delay user traffic (information packets) because control traffic always has precedence in terms of scheduling on the routing device interface cards. Unfortunately, the control traffic transmission rate is not decreased on low-bandwidth interfaces, such as DS-0 or fractional T1 and E1 interface. Line control traffic stays the same. On a low-bandwidth circuit that is transmitting 30 full-MTU-sized packets, there is not much bandwidth left over for other types of packets.

max-hello-size sizeModify the maximum size of IS-IS hello packets. IS-IS sends hello packets out of all IS-IS enabled interfaces to discover neighbors and form adjacencies between the devices. Based on the actual MTU of the physical interface, you can configure upto 16000 bytes as the maximum size for IS-IS packets. The valid size range is 512 through 16000 bytes, while the default is 1492 bytes.
mesh-group value | blockedConfigure an interface to be part of a mesh group, which is a set of fully connected nodes. A mesh group is a set of routing devices that are fully connected. That is, they have a fully meshed topology. When link-state PDUs are being flooded throughout an area, each router within a mesh group receives only a single copy of a link-state PDU instead of receiving one copy from each neighbor, thus minimizing the overhead associated with the flooding of link-state PDUs.

To create a mesh group and designate that an interface be part of the group, assign a mesh-group number to all the routing device interfaces in the group. To prevent an interface in the mesh group from flooding link-state PDUs, configure blocking on that interface.

valueThe number that identifies the mesh group. The valid range is 1 through 4,294,967,295 (232 – 1; 32 bits are allocated to identify a mesh group).
blockedConfigure the interface so that it does not flood link-state PDUs.
no-adjacency-holddownDisable the hold-down timer for IS-IS adjacencies. A hold-down timer delays the advertising of adjacencies by waiting until a time period has elapsed before labeling adjacencies in the up state. You can disable this hold-down timer, which labels adjacencies up faster. However, disabling the hold-down timer creates more frequent link-state PDU updates and SPF computation.
no-eligible-backupExclude the specified interface as a backup interface for IS-IS interfaces on which link protection or node-link protection is enabled.
no-eligible-remote-backupDisable remote LFA backup calculation for the specified interface. If remote LFA is disabled, Junos OS does not consider the interface for calculating the remote LFA next hop.
no-ipv4-multicastExclude an interface from IPv4 multicast topologies.
no-ipv6-multicastExclude an interface from IPv6 multicast topologies.
no-ipv6-unicastExclude an interface from IPv6 unicast topologies. This enables you to exercise control over the paths that unicast data takes through a network.
no-unicast-topologyExclude an interface from the IPv4 unicast topologies.
node-link-protectionEnable node-link protection on the specified IS-IS interface. Junos OS creates an alternate loop-free path to the primary next hop for all destination routes that traverse a protected interface. This alternate path avoids the primary next-hop routing device altogether and establishes a path through a different routing device.
point-to-pointConfigure an IS-IS interface to behave like a point-to-point connection. You can use the point-to-point statement to configure a LAN interface to act like a point-to-point interface for IS-IS. You do not need an unnumbered LAN interface, and it has no effect if configured on an interface that is already point-to-point. The point-to-point statement affects only IS-IS protocol procedures on that interface. All other protocols continue to treat the interface as a LAN interface. Only two IS-IS routing devices can be connected to the LAN interface, and both must be configured as point-to-point.
Best Practice

On QFX10000 switches, we strongly recommend that you configure all IS-IS interfaces, including peer interfaces, as point-to-point interfaces. If you do not, you might experience session flaps, that is, IS-IS sessions that go down and then come back up, when IS-IS is configured in virtual routing and forwarding (VRF) instances. When you scale IS-IS in any scenario, you might also experience scaling issues if you do not configure IS-IS interfaces as point-to-point interfaces.

The remaining statements are explained separately. See CLI Explorer.

Required Privilege Level

routing—To view this statement in the configuration.

routing-control—To add this statement to the configuration.