Staying Alive

Network administrators hate getting taken by surprise when their users find a problem like a server that's locked up.

Users love catching the admins with their pants down, so to speak, and admins feel like a hole has been poked in their bulletproof armor.

Enter ServersAlive from Woodstone, a flexible and powerful Windows program that monitors all of your servers around the clock and sends alerts at any signs of a problem. You set it up on any machine in the office and it keeps tabs on the other machines.

ServersAlive's dozens of checks help you keep tabs on every part of your server room. The free version limits you to 10 monitored services, while the $79 version will monitor an unlimited number of machines & services. Checks range from simple TCP/IP pings to diskspace checks, Windows processes, DNS checks, and external checks that users can design.

The HTTP check is particularly powerful: ServersAlive will monitor a given web page to make sure it includes a specific word or phrase, or the opposite - to make sure it does not include a specific word or phrase. For companies that worry about being hacked, the checksum check is perfect: if the web page is changed in any way, ServersAlive will catch it.

The database checks for Microsoft SQL Server and Oracle are another strong point. ServersAlive can test a database server to ensure that a given database is in operation. When an entire company relies on a database, this is the fastest way to know of impending danger: the database check will fail before the applications that rely on it start to fail.

Any given machine in your server room is complex: on one machine, multiple things rely on each other, and ServersAlive handles those dependencies with a simple drag and drop interface. For example, for a file server, the admin might choose to monitor it with a ping (to make sure the machine is up) and a disk space check (to make sure there's enough empty space). Just drag the disk space check on top of the ping check, and it's now dependent. When the machine is rebooted, only one alert will go out - obviously, users don't want to get diskspace alerts when a machine is down altogether.

Checks can be scheduled down to the days and times: in the file server example, checks would only be performed from 7 AM to 6 PM for regular business hours. Nobody wants to get an unnecessary warning, and if a server goes down when no one is supposed to be using it anyway, then there's no rush to fix it at midnight. Plus, if there's a regular scheduled maintenance time (such as all servers are rebooted at 6 PM on Fridays), then ServersAlive can be set to skip checks for that half-hour range.

When checks fail, ServersAlive's flexible alerting options give admins plenty of choices. Alerts can be sent by email, AOL or MSN instant messaging, alpha or numeric pager, WinPopup, audio, and more. There's a plethora of setup options here: with email alerting, for example, ServersAlive can automatically fail over to an alternate mail server if the regular one is down.

ServersAlive can handle multiple teams of people, each with their own alerting methods (mail, AOL or MSN instant messaging, pagers, etc.) and each with their own schedule and holidays. The network administrator might want to be alerted via instant messaging and email during business hours, but get alphanumeric pages on their cell phone during lunch hours, evenings and weekends. The built-in scheduler shows the week in easy-to-understand color-coded segments.

Most configuration changes are made on the machine running ServersAlive, but smaller changes can be made over Telnet or SSH, as ServersAlive has built-in servers for both.

Reporting capabilities start with a built-in HTML server, so you can see which machines are down via a web browser. ServersAlive can generate template-driven web pages into your existing site as well, so that it blends seamlessly with any site's existing look & feel. Simply create a page with the custom ServersAlive tags, and after each check cycle, ServersAlive will generate an HTML file based on that template.

The newest version takes reporting to another level: it adds ODBC logging for admins who want their uptime and outage information written into a database for further analysis. Even small shops can take advantage of this new, easy-to-access historical data by writing reports in their favorite programming language to analyze their uptime, see trends on which servers are becoming the most troublesome, and even enforce Service Level Agreements.

In summary, ServersAlive is a mature, flexible monitoring program with features missing from management packages that cost hundreds more. Windows network administrators with this tool in their arsenal will catch problems before the users do, and that's the difference between a reactive admin and a proactive admin.