The Challenge

Paul Glenn, Senior Systems Analyst at University of Kentucky is charged with maintaining 1000 desktops in 17 labs, 6 Netware servers, 2 Cisco routers and 50-plus Cisco switches in one of the University's campus LANs. The servers that his group maintains hold all the images for the lab machines, staff data and around 45,000 students' Home directories.

Ensuring an uninterrupted learning and working environment is Mr. Glenn's chief mission. One of his greatest concerns before deploying NetCrunch was the LAN's external connectivity, specifically port settings at switches and routers. "All of our switches in our 17 labs are supposed to be set to 100/full. However, sometimes a port would not be set correctly - this was a very big problem when we created ghost images of our labs. Like in so many other places, getting the WAN area to look at switches and routers when they may be suspect was quite a task." As a result, he and his team decided to look for a solution that would allow them to diagnose problem switches without having to turn to the busy WAN department for help.

Another problem for Glenn was the lack of a system that would automatically create a multi-dimensional graphical representation of the network topology and provide early alerts of failures in device, server or desktop performance and connectivity. Because he knew that in the past UK's WAN team had difficulties deploying a large systems management solution, he was looking for a "plug-and-play" type of tool that offers immediate help in key LAN monitoring duties.

The AdRem Solution

The implementation of NetCrunch on the administrator's desktop and the following network discovery procedure went smoothly. Paul Glenn recalls, "NetCrunch was a piece of cake to setup with no need for consulting. It did require some planning only because the program is so versatile with the way you can view the network maps."

Actually, maps of the network automatically created by NetCrunch turned out to offer many useful options in terms of the program's customizability. "We pretty much settled on creating a map for each lab, one for all of our switches, one for our servers, and one for staff machines," explains Glenn. With different maps grouping network resources according to user-defined functional criteria, the LAN team in the University of Kentucky can look at the network from many angles and even configure monitoring, alerting and reporting policies for each individual map.

With NetCrunch, Paul Glenn and his colleagues managed to quickly sort out the port settings problems. "With NetCrunch we can look and immediately see incorrect ports. I had been told for months everything was set just the way I had asked. When we went to image our machines, we noticed several labs that still had problems. Our tech team was on the phone with the WAN people quite often just getting them to look at each individual ports - many ports where found with incorrect settings. Now with NetCrunch, I can look at those ports myself. It's also much quicker for me to do this since I don't have to Telnet into the switch, and then look through the port listings. Sometimes it would take the WAN people 30 minutes just to let me know a port setting, now it only takes me seconds to see for myself."

Paul Glenn is not the sole beneficiary of NetCrunch though. "I had it at my desk for several months when my manager saw what it could do. At that point, she told me to order her an administrator's license and has been using it ever since to monitor our DART network interface controllers."

Before NetCrunch, the University's LAN team had continuous problems with DART controllers losing connectivity. However, now they can configure lab machines and devices to send SNMP traps to NetCrunch when any DART controller goes down. NetCrunch in turn converts these traps into alerts and notifies the proper administrator via email, SMS, pager or ICQ. As a result, WAN problems can be tracked with minimum human attention and can be instantly reported to the WAN department.

Additionally, NetCrunch is occasionally used by WAN professionals at the UK who, according to Paul Glenn, "were so impressed by this tool that they are looking at it for the entire network. The program they've spent thousands of dollars on takes forever to setup and doesn't give them everything NetCrunch gives me in a timely manner."

Leveraging NetCrunch's mapping, monitoring and alerting capabilities in their LAN environment, University of Kentucky administrators gained a better understanding and control of the connectivity and performance issues. "When I had problems before, I had to hunt down the WAN people, get them to find time to work me in and then try to fix the problem. Now we can rely on ourselves in our LAN for most of the time. I absolutely recommend NetCrunch to administrators in budget-sensitive public institutions like ours. The information and detail that can be garnered from this [tool] is unbelievable," Glenn concludes.