Summary


This chapter touched on many aspects of monitoring and alerting. You learned that monitoring can come from many sources. ICMP ping checks, OS service statuses, port query responses, and detailed application interrogation can all supply an administrator with information about the status of a system. Monitoring systems can leverage existing standards like SNMP (Simple Network Management Protocol), WMI, or NetBIOS queries or they might require dedicated agents to be installed on the systems to be monitored .

Monitoring information can be useful in both a short term and long term sense. Short term monitoring can determine whether a service stops, whether a piece of hardware changes status, or whether a system stops responding to service requests . Performance counters can be queried for point-in-time information to see if they are within expected ranges. This information can also be useful from a long- term tracking standpoint. By tracking when services go from up to down an administrator can begin to accurately track system uptime. Performance counters can be used from a capacity-planning standpoint to predict when resources will become insufficient. Administrators can plot system resource consumption against the number of users supported to gain valuable insight into the actual capacity of the systems. By having this information, upgrades to the system can be accurately measured in terms of the increase in capacity.

This chapter has shown the various types of monitoring available on the market and has made suggestions about how to determine what parameters should be monitored on different types of servers.

This chapter has introduced the reader to the concepts of scripting. Scripting is an excellent way to extend the capabilities of a monitoring and alerting system by creating complex responses to simple inputs. Rather than just raise a flag when an error occurs, a script can attempt to fix the situation and alert multiple resources about the situation.

You also saw how dedicated monitoring and alerting applications like MOM take the concept to new heights. By centralizing monitoring results, MOM also centralizes the information and expertise. By applying its own knowledge of software applications, MOM is often better able to diagnose problems than the administrator is. MOM intelligently links the administrator to knowledge specific to the situation encountered . By offering knowledge packs for all Microsoft applications, MOM has a significant advantage over other monitoring systems.



Microsoft Windows Server 2003 Insider Solutions
Microsoft Windows Server 2003 Insider Solutions
ISBN: 0672326094
EAN: 2147483647
Year: 2003
Pages: 325

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net