This lens focuses on agentless server monitoring, which users can implement to get real-time updates on the availability of their servers. I've included a list of popular applications and some of the pros & cons of each.
I've included both commercial and free/open source software, as I believe you should seek out the best solution for your business, rather than focus on just one type. Additionally, note that this is a work in progress: I believe I have the product descriptions down correctly, but if I've made a mistake, please let me know! I'll be adding to this list as often as I can (I try to give each app a chance by installing and playing with it for a time, though sometimes this just isn't feasible). Eventually, I'd like to turn these into mini-reviews, but as always, time is a problem.
Enjoy the list!
MRTG
This project is an open-source (GPL) app that really integrates well with SNMP. If you're organization is using SNMP currently (or has a lot of devices that already support SNMP output, as you probably do), then this tool may be useful. It collects information from your SNMP-enabled devices and can generate graphs and alerts based on threshold. One major drawback is the lack of trending and other analysis features. IPMonitor
IPMonitor is a Windows application that regularly polls servers to determine their availability. In addition, it can do tests to check the quality of service your servers are operating at (i.e. is a web page responding slowly) and respond by issuing alerts and even executing scripts. BigBrother
BigBrother is a very simple monitoring solution: a web interface tells you whether your servers are up or down ("green is good and red is bad") from a single overview web page. In addition, a commercial version is available with numerous additional features (including redundancy and better integration with dashboards and other management consoles). Heroix: Longitude
Another agent-less monitoring system, Heroix Longitude boasts it can be setup in just 15 minutes, and monitors a number of different metrics. Unfortunately, I have yet to use this product, so I'd love to hear your opinions. One of the most interesting features of Longitude is the nice SLA monitoring reports it can generate, though I'm a little concerned as to the limited monitoring metrics it collects (for instance, it focuses on application-level monitoring, but I'm not sure if it can be adapted for use with different applications than the ones it supports out of the box). AlertSite
A service that probably belongs in a third category - website monitoring - AlertSite is designed to regularly poll your web site for availability (and can be combined with an additional tool to combine server performance metrics). AlertSite can test a website from many different places around the world and alert you to any performance degradation. I found the prices to be a bit steep, but having an additional third-party customer-facing performance monitor can be very useful to many companies. Coradiant: TrueSight
An interesting product provided by Coradiant is their TrueSight monitoring appliance. A computer appliance is deployed onto your network that sniffs packets from the network and gathers statistics from this information. I've found this approach to be very interesting, as it introduces zero additional overhead onto your network or servers, and can gather quite a bit of useful information. As customers access your servers, TruSight records the response time from your servers and the message (i.e. is there a "404" message from a web server?) and can alert you to these issues. Deploying the appliance takes some thought, as you need to ensure it's placed on a network segment between your servers and the outside world. IPSwitch: WhatsUp
WhatsUp is a Windows monitoring solution with multiple versions (aimed at different customers' needs, such as WhatsUp and WhatsUp Professional). One of the interesting features of WhatsUp is the discovery mechanism, whereby WhatsUp can find devices on your network that have SNMP enabled, or via ICMP active discovery, and create a graph of your network. It collects information by regularly polling these devices, though can also include performance data if available via WMI (Windows Management Instrumentation). The Bottom Line
Agentless server monitoring software is often the easiest to implement and provides a generally good overall idea of your servers' health. The biggest drawback to these solutions, though, is the lack of insight they provide for diagnostic/troubleshooting purposes, nor can they usually spot trends that might impact performance (i.e. a rapidly filling hard disk) that agent-based systems would normally catch.
Before you choose a monitoring application, be sure to determine if your needs actually call for an agent-based system, or if an agentless solution might be better suited. Be sure to check product trials of software before you commit to any one application. Good luck in your search. Be sure to let me know if you found this lens useful!
Before you choose a monitoring application, be sure to determine if your needs actually call for an agent-based system, or if an agentless solution might be better suited. Be sure to check product trials of software before you commit to any one application. Good luck in your search. Be sure to let me know if you found this lens useful!
