For instance, if your spanning tree network holds a critical position, or you simply cannot step away from nProbe or Wireshark, you need a reliable solution to fix problems that previously took hours to resolve. IEEE 802.1D supports loop-free connections, and Wireshark can clean up arbitrary packets with just a few simple operations. Yet, even with so many mature network management technologies and tools, certain network routing issues still require debugging via a secure shell. Route analysis is a perfect example.
Convenient Network Tools
We may not mount light blue network cables on the wall like fire extinguishers, but every administrator knows exactly where to find one in an instant. Solving routing problems requires repeatedly checking status tables through a command-line interface, and you always end up pacing back and forth between racks. You might even need to connect consoles to several devices just to trace a route. However, these methods may still fall short of addressing the real challenge of route debugging. Routing often seems to develop problems, and the most effective way to find the root cause is to capture error information in real time.
Even if you have well-configured syslog alerts, many routing changes occur only once every few days, making them difficult to catch. Moreover, users are not the best source for reporting routing errors: when a page loads slowly or a chat app has noticeable lag, they merely grumble to those around them (although some VoIP clients will deliberately give a low subjective rating when jitter or packet loss occurs on the transmission path). So the question arises: “How can you regularly query the routing table and receive an alert the moment a network routing issue appears?”
The Impact of Route Resolution
The actual routing of a network involves many factors that can鈥攖hough not always鈥攊nfluence one another in various ways. Routing Information Protocol, Open Shortest Path First, Border Gateway Protocol, Enhanced Interior Gateway Routing Protocol, and static configurations all affect the final path. In essence, they collectively form a kind of virtual configuration. However, the ultimate “physical” route can only be verified in one place: the routing table. But who has the time to stare at the routing tables of all devices all day (and there’s no reason to do so)?
First, whenever a route changes, it forces routers to recalculate the network topology, flooding the network with update packets. This is not good. Second, route change detection usually occurs after the problem appears, forcing you to spend time examining individual routing leaf nodes to discover what happened hours earlier. Sometimes, routing changes can slow down traffic transmission, potentially causing routes to roll back to a previously underperforming path. And sometimes, these changes can sever a link or disrupt critical user services.
Fortunately, many network performance monitoring solutions provide easy access to the routing tables of monitored devices. Because they can sense routing changes, you do not need to connect a console to receive alerts, generate reports, or view the routing table directly. Combine them with network topology discovery tools, and you can monitor the complete traffic route from end to end, inspecting routing