Recently, on a project, the backend mule es CPU was found to be high. After Jstat-gcutil JVM process number 1000 10 analysis, found FGC times, more.
After the GC observation, there are a large number of new generation objects, directly into the Laosheng generation.
Modify mule/conf/wrapper.conf. Add the following options.
Wrapper.java.additional.5=-xx:+printgcdetails
Wrapper.java.additional.6=-xx:+printgcdatestamps
Wrapper.java.additional.7=-verbose:gc
Wrapper.java.additional.8=-xloggc:gc.log
wrapper.java.additional.9=-xx:parallelgcthreads=10
Wrapper.java.additional.10=-xx:+useconcmarksweepgc
Wrapper.java.additional.11=-xx:+useparnewgc
Wrapper.java.additional.12=-xx:newratio=3
wrapper.java.additional.13=-xss256k
wrapper.java.additional.14=-xx:survivorration=2
Wrapper.java.additional.15=-xx:targetsurvivorratio=8
The Newratio property increases the Cenozoic size.
Linux environment Mule JVM short life cycle Object performance tuning