Every day the system is compared to a list of
stability reports. It counts the number of software installs and uninstalls that have been
performed in the last 24 hours. It looks for application, hardware, Windows, and miscellaneous
failures that may have occurred as well. Using an algorithm, this information is
then translated into a stability index ranging from 1 to 10, where 10 is the most stable.
This index is displayed on a System Stability Chart so you can trend your server??™s reliability
over time (Figure 7-28).
If any recent changes or failures resulted in a lower index, you can find out more
information by expanding the relevant category in the System Stability Report section
under the chart. Each of these categories contains important information you can use for
troubleshooting your server. For example, for a Windows failure, it will indicate the failure
type (boot failure or OS crash), OS version, service pack level, failure details including
stop and reason codes, and of course the date and time when the failure occurred.
Figure 7-26. Selecting the launch schedule
247 Chapter 7: Resource Management and Performance Monitoring
You should note the following about Reliability Monitor, including how it comes up
with its stability index:
?–? Recent failures are weighted more heavily that past failures.
?– The system automatically excludes any days in which the server is shut off or
is in a sleep state.
Pages:
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292