North Carolina State Employees’ Credit Union (NCSECU) is a financial institution that serves roughly 1.7 million members across 255 branches and more than 1,100 ATMs. NCSECU uses Citrix virtual desktop technologies to deliver applications to more than 4,800 concurrent end-users. The core applications include Office 2010, Internet Explorer and mortgage and imaging software. Al Boyce, Systems Manager for NCSECU, shares how his company uses ControlUp, and a few more tidbits.
One of our primary challenges involves monitoring hundreds of servers. We needed ControlUp to do this more efficiently. Now we can quickly detect and locate a performance problem across all these servers and fix the issue before it affects our users. In our environment, it’s a problem if the average logon duration is longer than 20 seconds, as this usually signifies a couple of servers with network issues. It’s also a problem if any of our 350 XenApp servers have zero users on them. ControlUp also allows us to perform maintenance tasks in bulk on multiple servers, such as executing scripts easily on multiple computers and changing registry keys. This saves us time and lowers our risks.
One of the best things about ControlUp is that it helps us understand our environment and get baselines of normal behavior so we know when we’re starting to have problems and can solve them quickly and efficiently. We love the console hypervisor view, which may indicate the symptoms of a battery failure in our HP blade servers, alert us to possible network and load balancing issues and reduce outage times by catching things early. Once I had the monitor up and running and suddenly every server in one of our buildings went red. As it happened, there was a switch failure on the network and we were immediately able to pinpoint which one using ControlUp.
Currently my favorite ControlUp feature is the event log aggregation with filtering. I’m monitoring more than 400 virtual servers and when I let the event logs roll in – thousands of events – I quickly get an idea of which ones I can ignore, like Printer creation and miscellaneous .Net errors. Then I enable filters until I only see 15 or 20 event log entries per hour. I might see a server with Office applications crashing, so I know to take that server out of load and let it reboot. If a print spooler crashes and can’t be restarted, I can prevent users from logging into it until I get it fixed. This is great, because I can be proactive on things before anyone contacts me about them.
In the future, I would like to see built-in PowerShell scripts, because I’m not a scripting guru. I’d like a dropdown that lets you change the load evaluators on the selected XenApp servers all at once. That would be very useful when you want to take servers out of load because you need to reboot their underlying hypervisors.
ControlUp provides a thorough understanding of your environment and lets you monitor everything in real time. You can take actions on multiple servers simultaneously. The software aggregates event logs with filtering so you can proactively monitor events that matter to your business. Best of all, ControlUp catches problems you don’t know are happening. You can then head off much larger issues by addressing them early.