0

We have an Amazon EC2 instance which is firing off a High CPU alert once a day. It happens at exactly the same time every single day.

In looking at the cpu utilization history I see one spike that occurs at either 6:09, 6:14, 6:19 or 6:24 each day - only 1, but over the past 2 weeks that spike has rotated through those moments.

A second spike occurs at 8:59; this second one causes the alarm and is ALWAYS at 8:59.

How can I figure out what's going on?

The server is a web server running a .net application. There are no services installed other than what comes with Windows. There are no jobs firing from other parts of the system that hit this machine.

Improve this question
1
  • Is it the free tier micro instance? in this case the CPU is share between more VMs and you can get CPU performance issues.
    – POLLOX
    May 20, 2014 at 9:19

2 Answers 2

Reset to default
2
+100

Use perfmon (Windows Performance Monitor) to log All Instances of the counter Process\% Processor Time. Set it to log during the time frames you are experiencing alerts, and then review the data to determine the process spiking during those timeframes.

Possible dupe of: https://stackoverflow.com/questions/69332/tracking-cpu-and-memory-usage-per-process

Improve this answer
2
  • Another important parameters in the CPU performance are: Processor\% Interrupt Time System\Processor Queue Length
    – POLLOX
    May 20, 2014 at 9:17
  • I'm accepting this as it will likely lead to an explanation for someone else. However, I wanted to verify the server this alarm corresponded to. After going through all of our instances (4) none of them are associated with this alarm. An older instance that we had terminated used to be tied to it so I'm guessing it's just bad data.
    – NotMe
    May 20, 2014 at 13:57
0

Try Sysyinternal Process Monitor to inspect real time activity.

Improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .