Discovering Problems

   

How do you discover directory problems in the first place? A problem might come to the attention of the directory administrator in any of the following ways:

  • The monitoring system (if you have one) may automatically detect a failure or degradation of the directory and notify you about it.

  • The maintenance or operations staff may notice problems with the directory as they go about their routine directory maintenance functions.

  • Administrators of dependent services such as e-mail servers may notice and report problems.

  • End users may notice a problem and report it to your Help Desk. The problem might be described by end users as a failure in a dependent system, such as an Internet Message Access Protocol (IMAP) server, or as a problem with a desktop application that relies on the directory.

  • Your performance monitoring system, if you have one, may alert you to degraded performance on one or more of your directory servers.

A failure may be detected and reported via all these methods nearly simultaneously . For example, if a directory server becomes unavailable, you might be notified by your network management system (NMS) software at the same time your Help Desk receives numerous calls from end users. Those end users may report problems with any of the dependent applications, such as address books, e-mail servers, and authenticated Web server access. Your operations staff might also be unable to run a regular data update procedure.

Receiving multiple reports of system failure from a variety of sources is typical of distributed systems in which the individual portions of the system are interdependent. Part of the problem resolution process, therefore, involves correlating the various reports and identifying the underlying causes. As soon as you know what the real problem is, you can then inform users of all the affected services.

Ideally you should strive to eliminate the possibility that your users will discover problems before you do. You can accomplish this through a well-designed, proactive monitoring system. Careful planning is required, but the payoffs are significant. More information on proactive monitoring can be found in Chapter 19, Monitoring.

When there is a problem, you should clearly communicate the following information to your users and frontline staff:

  • The fact that there is a known problem

  • How long you expect it to take to fix the problem

  • Any workarounds or alternative services that your users can employ in the meantime

Plan well in advance how you will notify your users about directory problems. Common methods include publishing outage information to Web pages, posting outage information to Usenet newsgroups, and providing status information in a recorded telephone message. Whatever method you use, make sure that it does not depend on the directory itself.

Providing good information to your end users and frontline support staff serves two important purposes. First, it allows your end users to remain productive because they know which applications are affected by the failure and how long the applications will be unavailable. Second, your frontline support staff can be more effective in helping your end users understand the implications of the outage. Your end users will also have a better overall impression of your directory service and dependent applications.

   


Understanding and Deploying LDAP Directory Services
Understanding and Deploying LDAP Directory Services (2nd Edition)
ISBN: 0672323168
EAN: 2147483647
Year: 2002
Pages: 242

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