Click the Start a free trial link to start a 15-day SaaS trial of our product and join our community as a trial user. If you are an existing customer do not start a free trial.
AppDynamics customers and established members should click the sign in button to authenticate.
We have a situation where every now an then the CPU is increased considerably, and never recovers, until the consumption gets to 100%.
To give you an example, if the first day of month the CPU is OK, one week later all in a sudden the CPU gets a peak and increases to 40%, and stays like that for some days. Then two weeks later the CPU suffers another peak and increases from 40% to 75%... and so on until by the end of the month the CPU gets to 100% and then we are forced to reboot.
I have noticed that the problematic process is the WAS portal where the web application is hosted, and we have an App agent enabled there.
In order to find the root cause, I have created a bunch of Business Transactions, so I could see which ones are CPU expensive. Unfortunately this didnt help, and couldn't get any connection between these CPU expensive BTs, and the CPU peaks.
Do you know any more effective method to troubleshoot these CPU issues?
I have found the following article:
But this is not exactly our case because the CPU peaks don't occur just when the agent is enabled.
Solved! Go to Solution.
Your post suggests that the CPU consumption does not seem to be correlated to load on the JVM, is that correct?
Have you looked at the memory statistics? When CPU runs away like this, excessive GC activity is often the cause.
If that does not show any promise, you can use AppDynamics to collect some thread-dumps, which should give you an indication as to what is going on.
That is a very big topic! The documentation makes suggestions, and talks about the tooling we provide that may help here: https://docs.appdynamics.com/display/PRO45/Monitor+JVMs#MonitorJVMs-MemoryManagement
I hope this helps you get to root cause.