c1/c2 compiler thread eats a lot of cpu

from : https://community.atlassian.com/t5/Jira-questions/c1-c2-compiler-thread-eats-a-lot-of-cpu/qaq-p/51641

Aleksey Shirokih Jul 26, 2015

hi.

My JIRA instalation eats alot of cpu time during "not so high load" at night for example.

thread dump shows that most of the time that threads are on

"C1 CompilerThread3" #8 daemon prio=9 os_prio=0 tid=0x00007f534400a800 nid=0x56d3 runnable [0x0000000000000000]

   java.lang.Thread.State: RUNNABLE

 

"C2 CompilerThread2" #7 daemon prio=9 os_prio=0 tid=0x00007f5344008800 nid=0x56d2 waiting on condition [0x0000000000000000]

   java.lang.Thread.State: RUNNABLE

 

"C2 CompilerThread1" #6 daemon prio=9 os_prio=0 tid=0x00007f5344006800 nid=0x56d1 waiting on condition [0x0000000000000000]

   java.lang.Thread.State: RUNNABLE

 

"C2 CompilerThread0" #5 daemon prio=9 os_prio=0 tid=0x0000000001eb3000 nid=0x56d0 waiting on condition [0x0000000000000000]

   java.lang.Thread.State: RUNNABLE

 

any ideas how to debug it?

--------------------------------------------------------------

答案：

Matheus Fernandes ATLASSIAN TEAM Jul 27, 2015 

Hi, Aleksey, 

May I know if can see the following on the JIRA logs: 

Java HotSpot(TM) 64-Bit Server VM warning: CodeCache is full. Compiler has been disabled. I believe this may be related to the CompilerThreads using so much CPU. This behavior is fully explained on this blog post:

Matheus Fernandes ATLASSIAN TEAM Jul 27, 2015 Hi, Aleksey, May I know if can see the following on the JIRA logs: Java HotSpot(TM) 64-Bit Server VM warning: CodeCache is full. Compiler has been disabled. I believe this may be related to the CompilerThreads using so much CPU. This behavior is fully explained on this blog post:

We don’t have any statistics as to the impact of this on long running performance; the impression that I have is that CPU utilisation & request service time would increase over long running VMs for code that could have been optimised. 

For a system that is over-committed on CPU resources the application throughput could decrease noticeably. 

To try to mitigate that, could you try adding the following parameters to the JIRA startup and restart JIRA: 

-XX:ReservedCodeCacheSize=384m -XX:+UseCodeCacheFlushing 

Reference: https://confluence.atlassian.com/x/6xS_Eg I hope this helps!