NSM7 is the latest version of GFI’s general server monitoring tool, which keeps an eye on the basic parameters of your network servers (ie. whether each is up and contactable), the key applications running on them (eg. an SMTP or web server) and key operating parameters (CPU usage, disk space, and the like). It runs on Windows 2000/XP and WS2003, and it can monitor both Windows and Linux servers.
When you run the installer you’re asked a few questions before the files are extracted and copied over. First you’re prompted to provide the address and, if appropriate, authentication details for your mail server in order that NSM can send out alerts. Then you’re asked whether you want to use Access or SQL Server for the underlying database - if you choose the latter, which you’ll probably do for all but the smallest installation, you’ll need to have SQL Server or MDSE installed somewhere. Once you’ve made your choices, the installation process only takes a couple of minutes.
When it first launches, NSM throws you into the wizard that walks you through adding a new server to be monitored. You select the OS (choices includes Windows 2000/XP, WS2003 and various Linuxes), the server applications to monitor (Windows print server or an IIS Web/SMTP/FTP server), then any third-party applications (the list includes a pile from the GFI family plus Symantec, Kaspersky and Norman AV), and then the address(es) of the machine(s) to be monitored.
Incidentally, the wizard covers just the common types of check; the complete list of checks that you can add from the main config window is much more extensive, and covers general services (DNS, POP, etc), applications (Exchange, SQL Server, etc) and OS-specific stuff such as checking Linux machines for free disk space, or Windows machines for CPU usage levels. You can also write your own scripts for monitoring applications that aren’t included already, and configure NSM to kick them off just like any of its in-built checks.
The machines you want to check, and the checks they perform, can be organised as a hierarchy of folders; this is sensible since most set-ups will have a number of different machines performing a number of different functions.
Once you’ve configured the stuff you want to monitor, you can use the various other tools in the NSM bundle to keep an eye on things. So there’s a monitoring tool, which is basically an event logger that tells you what checks have been performed when, and whether anything went wrong - this is handy for discovering that, say, the credentials you gave for a particular server are incorrect. Then there’s the Reporter, which is a wizard-based reporting tool that walks you through what type of report you want (summary or detail), the date range, the part of the hierarchy of checks and machines you want to report on, and whether you want the output in XML, HTML or as a CSV file. The Status Monitor is a quick “dashboard” onto how NSM is behaving and whether it has any problems, and the Troubleshooter is a tool for reporting bugs in NSM to GFI; it asks you for a problem description, then bundles this up along with an analysis of the machine you’re running on into a Zip file which you can then email to GFI support.
Back in the configuration tool (which is actually the main program you use for doing most things in NSM – it’s perhaps a misleading name) you can define users and groups and arrange for them to be automatically alerted to problems. Alerting preferences are hierarchical, so when you define them for a parent folder they’ll be inherited by all the child objects of that entity unless you specifically configure something different into one or more children.
As well as alerting people, you can tell the system either to run an external script (perhaps a script that does a stop/start on an offending service, for instance) or to reboot the monitored system in question. There’s also a set of network tools (a DNS lookup tool, the obligatory Ping/Traceroute, an SMTP walker, and so on) which, though they don’t add a great deal of value, might come in handy from time to time.
NSM is an inexpensive monitoring tool that’s simple to get up and running and which makes a decent fist of monitoring network services and Windows/Linux proprietary parameters.
NSM is a very attractive monitoring tool that does most of what the average SME needs. If you want something with a bigger feature list and better in-built support for additional applications, look at Up.Time