Monitoring Cisco 2600, 2800 Series

When monitoring the routers for performance, its hard to pin down any one resource that could lead to performance degradation. It could be any or a combination of the following reasons that lead to poor performance:

Performance of the CPU:Core router functions like, routing protocol processing, process packet switching etc, are handled in memory and share the CPU. Thus, if CPU utilization is very high, it is possible that a routing update is not handled or a process-switching packet is dropped. The CPU performance degrading on routers is a frequent issue caused due to interrupts or slow processes on the router. The router stops sending routing updates to other routers, the performance is slow and the services on the router fail to respond, impacting the business. Proactively watching the CPU health is a must. Read more on troubleshooting high cpu utilization issues.

Performance of RAM: Memory leaks or fragmentation are common causes leading to poor memory performance. Identifying memory allocation failures or leaks early on helps in optimzing the memory performance. Monitoring basic memory metrics like memory utilization, free memory, used memory etc will serve as a good indicator of the RAM health.Click here to read further on troubleshooting memory problems on Cisco routers

Buffer Monitoring: The Route Processor divides its processor memory into 6 pools with each pool comprising of many memory blocks of equal size. These memory blocks are called buffers. The buffers are used for handling traffic landing on the router or when packets are being switched. When a requested buffer is not available, a 'buffer failure' is detected and the buffer pool manager process tries creating new buffer to avert further failures. Insufficient memory limits creating of new buffers. So buffers are crucial resources to be monitored on a router, especially the number of failures and buffer-create failures due ti inadequate memory. A consistent failure of buffers requires tuning.

Bandwidth Monitoring: The percentage of network bandwidth used by the traffic coming into and going out of the network needs monitoring. If there is a consistent increase in the bandwidth utilized by the Rx Traffic (incoming) or the Tx Traffic(out-going), errors and discards occur leading to data loss. The network bottlenecks can be eliminated by analyzing the traffic indepth using Netflow.