Introduction To ISDN, Part II

Understanding Idle‑Timeout and Interesting Traffic on Cisco ISDN

When a Cisco router initiates an ISDN connection, the link behaves much like a traditional telephone call. Once the line is established, the phone system waits for a reason to hang up. If nothing tells it to break the connection, the call can stay open for days or weeks. For service providers and enterprise networks, that means a long‑distance bill that never ends. To keep costs under control, Cisco routers enforce a built‑in idle‑timeout that tears down an ISDN circuit after a period of inactivity.

The idle‑timeout mechanism relies on a concept called “interesting traffic.” Any packet that is not part of the ISDN signalling or low‑level protocol traffic counts as interesting. When a packet meets the interesting criteria, it resets the timeout counter. The default timeout is two minutes, meaning if the router does not see any interesting traffic for two minutes, it will shut the link down. This default value balances cost savings with the need for quick recovery in the event of a link failure.

While the two‑minute window seems short, many ISDN links are used as backup circuits. Picture a primary Frame Relay connection that drops. The backup ISDN line comes up automatically, but the cost of keeping it alive for an extended period without data can be significant. By letting the idle‑timeout cut the circuit after two minutes of silence, the backup link remains available for rapid failover while preventing runaway billing. Network administrators can tweak the timeout value in the ISDN interface configuration if longer or shorter intervals are required.

To view the current idle‑timeout setting on a Cisco router, use the show interface ISDN command. The output will list the Idle Timeout value and the amount of time left before the circuit will drop. You can also monitor traffic patterns with show ip interface brief and show interface counters to see how often interesting traffic flows across the ISDN link. If you notice frequent drops during business hours, adjusting the timeout or adding traffic‑generation mechanisms such as keep‑alive packets may be necessary.

When you modify the idle‑timeout, keep in mind that any protocol requiring periodic updates will automatically keep the circuit alive. Routing protocols like RIP v2 and EIGRP send frequent updates that satisfy the interesting‑traffic condition. If those updates are suppressed or filtered, the line will shut down, potentially breaking failover paths. Therefore, it is crucial to understand which traffic is classified as interesting on your specific router model and to ensure that critical keep‑alive or routing updates are exempt from any filtering that could reset the timeout.

In short, the idle‑timeout is a cost‑control feature that forces ISDN circuits to behave like paid phone calls. By configuring it correctly and monitoring traffic patterns, you can keep your backup links ready for use without letting the bill spiral out of control.

Optimizing Protocol Traffic and Myths on ISDN Links

Routing protocols that run over ISDN must be tuned to avoid unintentionally keeping the line alive. For RIP v2 and EIGRP, the most efficient strategy is to block the multicast addresses used for updates in the access‑list that defines interesting traffic. By filtering only the multicast update packets from the traffic that resets the idle‑timeout, the router still accepts the routing information but the idle timer doesn’t keep the link open. This selective filtering is preferable to blocking all traffic, which would render the routing protocol inoperative. The key is to apply the access‑list to the interface’s interesting‑traffic filter, not to the actual data path.

Open Shortest Path First (OSPF) behaves differently, thanks to Cisco’s demand‑circuit feature. The ip ospf demand‑circuit command lets an OSPF adjacency form over an ISDN link while suppressing the regular hello packets. The adjacency status stays at Full, as shown by show ip ospf adjacency, even though hello exchanges stop. Because the adjacency remains in place, the link can be dropped and restored without OSPF needing to re‑establish the neighbor relationship. When the circuit comes back up, traffic flows immediately, avoiding the usual OSPF convergence delay. This command is a staple for CCNA and CCNP labs, and mastering it can save time during troubleshooting.

Many network engineers mistakenly believe that Cisco Discovery Protocol (CDP) keeps an ISDN line alive. CDP is a proprietary protocol that exchanges neighbor information between directly connected Cisco devices. It is not designed to maintain ISDN circuits, and disabling it does not affect the idle‑timeout. In field and lab tests, CDP packets do not trigger the idle‑timeout timer. If you need to prevent a line from staying up, focus on filtering interesting traffic or adjusting the idle‑timeout rather than disabling CDP.

When configuring a backup ISDN link, start by ensuring that routing updates are appropriately filtered or allowed through the interesting‑traffic list. Next, apply ip ospf demand‑circuit if OSPF is your main routing protocol. Finally, monitor the interface with show interface ISDN to confirm that the idle‑timeout behaves as expected. By aligning protocol behavior with the idle‑timeout, you can keep the backup link ready for failover while preventing unnecessary charges.

For more detailed guidance on ISDN timing and OSPF demand‑circuit configuration, refer to Cisco’s official documentation at ISDN Timing Configuration and OSPF Demand Circuit. These resources offer configuration examples and best‑practice recommendations that align with the concepts discussed above.