1

Not sure what is happening, my instance goes 100% CPU usage two exact times during the day one at 11:33 UTC and another at 3:38 otherwise rest of the day usage is under 10%.

I didn't find: 1. Any crontab entry (scheduled task) 2. Any spike in usage of my app. In fact, app is used total 3-4 times throughout the day.

The problem is a few times the instance goes unresponsive either after 11:33 or 3:38 and only restart fixes the problem.

I am running word-press and tomcat server (with a very light weight app). The app has run fine on a micro instance. But, for better performance I chose to go for m1.small.

I could go for even bigger instance, but before I would like to know what exactly causes this 100% at 11:33 and 3:38.

See attached screenshot of monitoring details.

enter image description here

This might be related to CPU Steal as mentioned here https://serverfault.com/a/560330/204141 But, I don't know for sure.

  • 1
    yes CPU Steal might me the issue,but to know exactly what is going on,you need to check with top "command". And please provide logs(tomcat,mysql) or check the logs for any unusual traffic. – TBI Infotech Jul 19 '14 at 7:16
  • @Watt What didyou do? I am having the same exact issue! I moved my DB to RDS as well. Did that fix it for you? – Neo Nov 1 '14 at 21:39
  • It turned out that my apache+php were eating all the memory and after fix interval when memory is full, there is a spike in CPU which triggers killing/restarting processes. Had to spend few hundreds dollors on AWS and their Dev service, only to be misguided that spike in I/O is causing this issue. I/O was spiking due to the memory issue (OS wast trying to write to swap). Long story short, I moved out of AWS to bit.ly/13AU6vp and now saving in funds and trouble. Also I moved to ngix for better memory footprint. – Watt Nov 3 '14 at 22:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.