Problem solve Get help with specific problems with your technologies, process and projects.

4 ways -- besides MOM -- to monitor Exchange's health

While using MOM to monitor your system's performance is optimum, Exchange has several built-in monitoring mechanisms that will alert you to potential issues with your server.

As all Exchange 2003 admins know, Microsoft included a management pack for Microsoft Operations Manager (MOM) when it released Exchange Server 2003.

In general, MOM lets you monitor countless aspects of your system's performance. If MOM detects a potential problem, it can either alert you or take corrective action on its own (or both).

As you can imagine, MOM is a great addition to Exchange and can be tremendously helpful in keeping Exchange running smoothly. The only problem is that MOM tends to be a little on the pricey side: a MOM 2005 license appropriate for an Exchange Server costs $729 per server. While Exchange Server does include a MOM management pack, it doesn't include MOM itself or a MOM license.

I don't think there is any substitute for MOM. No other product exists that will monitor an Exchange Server to the extent that MOM will. Having said that, though, MOM is often impractical for smaller organizations due to its cost and complexity.

Even though I do own a couple of MOM licenses, I do not use MOM in my own organization (except in a lab environment for MOM-related writing assignments). My production Exchange Server only has a few mailboxes and I'm just really not interested in granular monitoring on a server that only a few people ever access. Besides, all of that monitoring eats up system resources. That doesn't mean that I don't want to know what's going on with my server.

If you don't have the money for MOM or if using MOM in your organization is overkill, there are simpler ways of monitoring your Exchange Server's performance. You could run the Windows Performance Monitor as an alternative, but there is an even easier solution than that. Exchange Server has some of its own monitoring mechanisms built in. These mechanisms can be configured to alert you to potential issues with your server.

CPU monitoring

One aspect of your server's performance that you can monitor is its CPU use. It is normal for an Exchange Server's CPU utilization to spike to 100% on a regular basis. However, these spikes should be brief. If a server's CPU is more than 80% to 85% utilization for an extended period of time, it means that either the server is seriously underpowered or that some process has run amuck and is depleting the system of resources. Either way, it is a problem that you need to know about.

To configure Exchange Server to monitor CPU use, open the Server Manager and then navigate to Administrative Groups | your administrative group | Servers | your server. Right click on the server that you want to monitor and select the Properties command from the resulting shortcut menu. When you do, you will see the server's properties sheet.

At this point, select the properties sheet's Monitoring tab and click the Add button. When you do, you will see a list of items that you can monitor. Select CPU Utilization from the list and click OK. You must now fill in the duration, warning state and critical state thresholds. I recommend setting the duration to anywhere from 5 to 10 minutes. I would then recommend setting the warning state to somewhere between 85% and 90%. You should set the critical state to about 95%.

Virtual Memory

Another aspect to your system's performance that you can monitor is its virtual memory usage. To do so, click the Add button on the Monitoring tab of the server's properties sheet and choose the virtual memory option.

Configuring virtual memory monitoring is very similar to configuring CPU utilization monitoring. Once again, you will set a duration, a warning value and a critical state value. I recommend setting the duration to five minutes, the warning threshold to 15% and the critical threshold to 10%.

Disk Space

Of all of the things that you can monitor, perhaps none is as important as monitoring your server's free disk space. As you are no doubt aware, Exchange is a real disk hog. If you turn your back on your server, it can run out of disk space in the blink of an eye.

The process for configuring free disk space monitoring is pretty much identical to the procedure used for monitoring CPU and virtual memory utilization. The only difference is that you must set up free disk space monitoring on a per volume basis. The threshold values that you should use will vary widely depending on what the volume is being used for.

If a volume is being used solely for the Windows operating system and for the Exchange Server system files, then I would recommend setting a warning at around 768 MB free and setting a critical state value to 256 MB free. This probably sounds excessive, but keep in mind that Windows and Exchange are both constantly being updated and some of those updates can be large. Without sufficient updates you may not be able to apply the bigger updates.

For volumes containing Exchange database files, I would set a warning at 3 GB of free disk space and a critical state at 1 GB. If your Exchange Server doesn't have a large information store or isn't heavily used though, you may wish to adjust these values.

For a volume containing transaction logs, you should never have less than 10 MB of free disk space. Personally, I don't like to cut it that close, so I would recommend setting a warning at 512 MB free and setting the critical state to 100 MB free.

SMTP Queue Growth

You can also monitor the growth of the SMTP queues. If your SMTP queues start backing up, it often indicates that there is a major problem. Either the message transport has failed, the Internet connection has failed, or someone is using your server to send spam. In any event, you need to know that there is a problem.

SMTP queue growth monitoring works a little bit different than the other types of monitoring. You must still input a warning and a critical state threshold value, but these values are entered in minutes. The idea is that you must tell Exchange how many minutes the queue should be constantly growing for before you generate a warning or a critical state message. You can use any values that you want, but I recommend setting the warning threshold at 10 minutes and the critical state threshold at 15 minutes.

As you can see, you don't need to shell out big bucks for a copy of MOM in order to be able to monitor your server. Instead, you can use the monitoring mechanisms that are built into Exchange Server. Although I have shown you the most important mechanisms to monitor, there are a few others available, including Windows services and X.400 queue growth.

Brien M. Posey, MCSE, is a Microsoft Most Valuable Professional for his work with Windows 2000 Server and IIS. Brien has served as the CIO for a nationwide chain of hospitals and was once in charge of IT security for Fort Knox. As a freelance technical writer he has written for Microsoft, CNET, ZDNet, TechTarget, MSD2D, Relevant Technologies and other technology companies. You can visit Brien's personal Web site at

Do you have a useful Exchange tip to share? Submit it to our monthly tip contest and you could win a prize and a spot in our Hall of Fame.

Dig Deeper on Exchange Server setup and troubleshooting

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.