I ran network and datacenter operations for the University of California system, the National Geographic Society, Fastclick (now part of ValueClick), ExpertCity (now Citrix Online), and I also consulted for a variety of organizations.
Monitoring was always a pain point.
There was too much reliance on manual processes to add things to monitoring; too much expectation that the end user of the equipment would know what to monitor.
The sysadmin, who already has way too much to do, doesn’t have time to figure out what matters on his new device or software. And he often won’t have the operational experience with it to know, as it’s new to him.
The other issue was that people, even good sysadmins, are imperfect.
If they make changes in the heat of a technical issue, such as creating a new storage volume, or virtual IP on a load balancer – they will intend to add this into monitoring later – and often forget. Then six months later, there’ll be another issue that should have been caught by monitoring, but wasn’t as the new object was never added to monitoring.
When I was consulting, many of the problems I was diagnosing and solving required good monitoring and trending – and just getting the data was consuming a fair bit of time.
I realized that if I could address these common needs in a way that let people get best of breed monitoring without having to be monitoring experts, they’d save a lot of time, and improve infrastructure availability.
A few months of market validation later, and LogicMonitor was born.
LogicMonitor addresses the needs I had when I was in charge of complex infrastructure. We actually do go through the MIBs, Mbeans, database data, etc and figure out what should be monitored, for all sorts of hardware and software.
We also automate the discovery and configuration of the monitoring – when you add a NetApp to LogicMonitor, because we have knowledge of what should be monitored (volumes, CPUs, snapshots, snapmirrors, etc), we can go and automatically discover those objects – and we periodically check for changes, so if you add or change volumes, or snapvaults, or whatever – the monitoring knows, automatically.
LogicMonitor deals with the case where there may be 100 web servers, some production, some staging, some QA, so different credentials and alert policies should be applied to each group.
It gives you expert monitoring, without requiring you to be a monitoring expert.
I hope you have the opportunity to give it a try and see if it helps you be both more proactive, and more productive.