async-profiler: Fresh segfault

Stack: [0x00007fb3fb887000,0x00007fb3fb988000],  sp=0x00007fb3fb9857f8,  free space=1017k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
j  org.apache.spark.SomeRDD$$anon$1.<init>(Lorg/apache/spark/SomeRDD;Lorg/apache/spark/TaskContext;Lorg/apache/spark/Partition;)V+431
j  org.apache.spark.SomeRDD.compute(Lorg/apache/spark/Partition;Lorg/apache/spark/TaskContext;)Lscala/collection/Iterator;+27
j  org.apache.spark.rdd.RDD.computeOrReadCheckpoint(Lorg/apache/spark/Partition;Lorg/apache/spark/TaskContext;)Lscala/collection/Iterator;+26
j  org.apache.spark.rdd.RDD.iterator(Lorg/apache/spark/Partition;Lorg/apache/spark/TaskContext;)Lscala/collection/Iterator;+42

@krzysztofslusarski gave me advice about wall mode

safemode=64,event=wall,thread,interval=1000000,jfr

caused segfault

About this issue

Original URL
State: closed
Created 2 years ago
Comments: 20 (5 by maintainers)

Most upvoted comments

@toktarev In case of stability - in my previous work I used Async-profiler 1.8.x in continuous mode. The profiler was always enabled on wall mode. The results were dump every minute. At the end of the last year we had 40 production JVMs profiled continuously, and even more test instances. I believe in one year of such a profiling we had 1 crash - https://github.com/jvm-profiling-tools/async-profiler/issues/461. My previous company is in the middle of switching to 2.x line.

I’m not as good as @apangin in analyzing hs_err files, but I cannot see anything related to profiler in a problematic thread.

I understand what you have written, that there is no error without async-profiler, and there is with async-profiler, but it doesn’t mean that async-profiler is guilty here.

One example from my life, there is a Spring component called Spring Session JDBC. It works great, has no problem at all. But if you add CAS Client to your application then you have a memory leak in Spring Session JDBC component. This is the same situation, you add CAS Client and your application has a problem, but it is a fault of Spring Session JDBC which doesn’t emit some events by design.

krzysztofslusarski on Feb 2, 2022