0 Replies Latest reply: Nov 23, 2012 12:50 PM by fhlbcin RSS

CPU/Memory/Disk Load Report

fhlbcin

Here is the goal.  The report will show the number of minutes each time a node was over a cpu, memory or disk threshold.  There can be three seperate reports.

 

From a different area on the Thwack site, I found an example report that requires that an alert be created to identify the cpuLoad when it goes over, and then reset when it falls under, the desired threshold.  The trigger and reset action must write to the NPM event log.

 

The basic alert works fine and the reports works, but my problem is getting this into a readable and nice report for management.   The biggest item I am trying to fix is to total the minutes for each node.  It's fine to give the number of minutes for each time a node was in the "alert" stage, but it is more important for me to see the total time overall (sum).  From that, I could get the percent time that the node was in the "alert" stage for a month.  Hopefully this makes sense.  Our DBA was able to get the totals directly from SQL, but I can't get it to work from the Orion report writer.  The one other problem is that at the end of the month, if a node is still in an "alert" stage, the number of minutes is blank.

 

Could someone help me with this?  I have attached the basic orion report.