Newbie -
I agree with fnunezsa - you're going to have to get a little dirty on this one. But this might work ...
Assuming it's used, try analyzing events in the Log Database, and subtract the time between events from the total scheduled available time for the month.
Example:
Assume that there was a host failure that impacted the TServer on 09/15 for an hour. You should see Alarm or Standard log records showing
- Loss of connection to the host
- The Backup TServer (if it exists) entering Primary mode
- Loss of Connection to the TServer
When it restarts, you should see alarm or Standard log records showing
- TServer failure alarm cleared
- Primary TSErver entering Primary mode
You then subtract the one hour from the scheduled available time (For September, assume 24 hours * 30 days = 720 hours), so Uptime = (730 - 1) * 100%/730 = (729 * 100%)/730 = 99.86%
Example 2:
If there were NO TServer failures during the month, then there would be NO events during that time period , and uptime would be 100%.
You'll need to determine for yourself what constitutes a "failure" and what does not - for example, does a swtchover from Primary to Backup constitute a failure? It will also need to recognize that there can be more than one event that constitutes a failure or recovery - such as "CTI Link Failure", "Application is not able to provide service", etc.
Good luck.
