Netcordia's NetMRI

Version 1.3 includes port scanning (which can be disabled if that sort of thing makes you nervous), support for Cisco ISO base wireless access points, retrieval of Cisco configs, and SNMP data collection for Juniper and Enterasys devices. NetMRI analyzes network health and assigns a relative quantification of 1 through 10. In my case, I ended up at a 7.2 health index, which Netcordia says is good.

It took me about five minutes to set up the brick-sized NetMRI Campus 200 model, and five minutes more to define a couple of local subnets for it to monitor. I had to wait until the next day to see results. The box is not interactive; rather, it checks the network on an ongoing basis and issues daily reports, which are found on the easy-to-use administrative Web interface.

The reports are logically divided into four sections. The results section lists network performance and history. The network-information section includes route, subnet, OS, VLAN and port summaries. The performance section shows a quick glimpse into CPU, memory, and device and interface availability. The historical section shows device, route, subnet, VLAN, HSRP, reboot, and configuration changes and stability for the past 30 days.

The second morning of my tests, I had a list of concerns to begin addressing on, prioritized by severity, including weak security SNMP strings, open HTTP interfaces, mismatched full- and half-duplex switch connections and a version of IOS that had an SNMP memory leak. When I drilled down into the SNMP memory leak, I was given two references--Google and Cisco TAC--from which to pinpoint the problem and the fix.

NetMRI includes typical network-management tools that warn of unusually high or low traffic utilization on an interface. A less common but more useful tool warns of any big change in traffic utilization. To accomplish this, NetMRI automatically compares utilization for seven days and displays interfaces that have a high percentage of change. This isn't necessarily an indication of trouble, but it gives you a heads-up.When a network is having intermittent problems, they're usually difficult to diagnose. NetMRI flags interfaces with two or more operational changes in a day or that are restarted a number of times, and also alerts you to OSPF intra-area routing tables that change more than 75 times in a day. Such occurrences are hard to track and important to be informed about.

Pricing is fair and returns immediate value. If you're looking for some network expertise, this is the management tool that's got it.

Bruce Boardman, executive editor of Network Computing, tests and writes about network management and systems. Write to him at [email protected].