Messaging systems are so critical to organisations of all sizes that it is important to develop a plan for ongoing systems management, monitoring, and troubleshooting. The type of management plan needed depends on many factors, including size of the company, number of users, location of remote offices, type of messaging software used, network infrastructure and so on.

This need can be determined by both the administration and help desk structure that is in place as well as the business criticality of the messaging software. For example, if e-mail is mission critical, then a heavy amount of monitoring should be in place. If e-mail is not critical, or the e-mail infrastructure is small, less monitoring will suffice. The architecture of the messaging system at this company may or may not reflect your own, but the server roles detailed below will probably be in your architecture at one place or another.

The following provides a detailed view and explanation of the systems management requirements for Microsoft Exchange version 5.0 deployed at a large company with a large number of users, post-offices, etc. It is intended to serve as a starting point of what needs to be considered to monitor.
For smaller companies or those with only a few sites, some of the monitoring requirements may be dropped. These management requirements were developed because of both the business criticality of e-mail services as well as some problems that were encountered when it was too late to prevent them. With some monitors in place, problems and, more importantly, outages can be prevented and operations can be notified at a stage when they have enough time to proactively react to it with time to fix it. The systems management at the following levels of e-mail operations is covered:

• Tier 0-Enterprise send-mail relay (used for firm-wide external mail distribution)
• Tier I-IMS servers (servers that provide Internet Mail Services)
• Tier II-Bridgehead servers (servers that provide MTA services for site to site communication)
• Tier III-Postoffice (servers that contain the mailboxes of users).

Note: Systems administrators and helpdesk personnel need to ensure that the individual requirements for a specific environment meet the situation described here.

With these requirements implemented at each of the tiers, Exchange downtime has been minimized, but most importantly, problem resolution time decreased significantly. By monitoring critical Exchange pieces, the time needed to determine and resolve the problem was reduced because the cause of the problem was pinpointed as soon as troubles occurred. This has ensured that the business critical application of e-mail remains up at all times and system or resource problems are discovered with enough time to react and fix the problem. The goal was to discover e-mail problems BEFORE the users did and prevent calls to the help desk.