352-001 · Question #591
You are redesigning a single-level IS-IS network with 500 routers, which have short-haul and long- haul links Most of the time the routing domain is stable, but periodically interfaces on long- haul l
The correct answer is D. Implement a delay between successive IS-IS LSP packet transmissions on routers with long-haul. IS-IS LSP pacing delays successive LSP transmissions on long-haul links to protect CPU during flapping bursts, while preserving fast convergence for normal single-failure events.
Question
You are redesigning a single-level IS-IS network with 500 routers, which have short-haul and long- haul links Most of the time the routing domain is stable, but periodically interfaces on long- haul links bounce for a short period of time , causing 10 to 20 flaps in a few minutes. The probable cause is local road construction. Although fast convergence important, the client has concerns about taxing CPU cycles on the older routing platforms. What change should you recommend that both protects the CPU of the older routers during the short periods of excessive flapping, yet does not have an impact on fast convergence for all interface failures?
Options
- AModify hello timers on routers with short-haul links
- BImplement LSP generation throttling on routers with long-haul links
- CModify the length of time than an LSP remains in the router database without being refreshed on
- DImplement a delay between successive IS-IS LSP packet transmissions on routers with long-haul
How the community answered
(30 responses)- A7% (2)
- B13% (4)
- C23% (7)
- D57% (17)
Why each option
IS-IS LSP pacing delays successive LSP transmissions on long-haul links to protect CPU during flapping bursts, while preserving fast convergence for normal single-failure events.
Hello timers govern adjacency formation and keepalive detection speed on short-haul links and have no effect on the rate of LSP flooding triggered by long-haul interface flapping.
LSP generation throttling delays how quickly the router creates a new LSP after a topology change, which would slow initial convergence for every interface failure - directly contradicting the fast convergence requirement.
The LSP refresh interval controls how long an LSP remains valid in the link-state database without being re-flooded, which is unrelated to the burst of CPU load caused by rapid short-duration flapping.
LSP pacing introduces a configurable inter-packet delay between successive LSP flood transmissions on a per-interface basis. Applying it only to long-haul link routers throttles the burst of LSPs produced by repeated flapping, protecting older CPUs from being overwhelmed, while the first LSP for any new failure is still sent immediately - so fast convergence is not degraded.
Concept tested: IS-IS LSP pacing to limit CPU load during link flapping
Source: https://www.cisco.com/c/en/us/td/docs/ios-xml/ios/iproute_isis/configuration/xe-16/iri-xe-16-book/iri-lsp-gen.html
Topics
Community Discussion
No community discussion yet for this question.