Configuring Loopback Testing

Loopback testing allows you to verify the connectivity of a circuit. You can configure any of the following interfaces to execute a loopback test: Aggregated Ethernet, Fast Ethernet, Gigabit Ethernet, E1, E3, NxDS0, serial, SONET/SDH, T1, and T3.

The physical path of a network data circuit usually consists of segments interconnected by devices that repeat and regenerate the transmission signal. The transmit path on one device connects to the receive path on the next device. If a circuit fault occurs in the form of a line break or a signal corruption, you can isolate the problem by using a loopback test. Loopback tests allow you to isolate segments of the circuit and test them separately.

To do this, configure a line loopback on one of the routers. Instead of transmitting the signal toward the far-end device, the line loopback sends the signal back to the originating router. If the originating router receives back its own data link layer packets, you have verified that the problem is beyond the originating router. Next, configure a line loopback farther away from the local router. If this originating router does not receive its own data link layer packets, you can assume the problem is on one of the segments between the local router and the remote routing platform’s interface card. In this case, the next troubleshooting step is to configure a line loopback closer to the local router to find the source of the problem.

There are several types of loopback testing supported by the JUNOS software, as follows:

• DCE local—Loops packets back on the local DCE.
• DCE remote—Loops packets back on the remote DCE.
• Local—Useful for troubleshooting physical PIC errors. A local loopback loops packets, including both data and timing information, back on the local routing platform’s PIC. When you configure a local loopback, the interface transmits packets to the channel services unit (CSU) built into the interface. These packets are transmitted onto the circuit toward the far-end device. The PIC receives back its own transmission and ignores any data sent from the physical circuit and the CSU. To test a local loopback, issue the show interfaces interface-name command. If PPP keepalives transmitted on the interface are received by the PIC, the Device Flags field contains the output Loop-Detected.
• Payload—Useful for troubleshooting the physical circuit problems between the local router and the remote router. A payload loopback loops data only (without clocking information) on the remote routing platform’s PIC. With payload loopback, overhead is recalculated.
• Remote—Useful for troubleshooting the physical circuit problems between the local router and the remote router. A remote loopback loops packets, including both data and timing information, back on the remote routing platform’s interface card. A router at one end of the circuit initiates a remote loopback toward its remote partner. When you configure a remote loopback, the packets received from the physical circuit and CSU are received by the interface. Those packets are then retransmitted by the PIC back toward the CSU and the circuit. This loopback tests all the intermediate transmission segments.

Table 23 shows the loopback modes supported on the various interface types.

Table 23: Loopback Modes by Interface Type

To configure loopback testing, include the loopback statement:

You can include this statement at the following hierarchy levels:

• [edit interfaces interface-name aggregated-ether-options]
• [edit interfaces interface-name ds0-options]
• [edit interfaces interface-name e1-options]
• [edit interfaces interface-name e3-options]
• [edit interfaces interface-name fastether-options]
• [edit interfaces interface-name gigether-options]
• [edit interfaces interface-name serial-options]
• [edit interfaces interface-name sonet-options]
• [edit interfaces interface-name t1-options]
• [edit interfaces interface-name t3-options]

Copyright © 2008, Juniper Networks, Inc. All rights reserved. Trademark Notice.