dd-trace-java: dd-trace-java v1.11.0 crashes the JVM
Our server automatically downloaded the latest trace agent. v1.11.0 After a few minutes, our servers started to crash and reboot in loop. After some investigation, it seems that the JVM was crashing. So maybe it’s more a JVM issue, but it seems to be your recent changes that introduced this behaviour.
#
# A fatal error has been detected by the Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x00007f171904b992, pid=23811, tid=24534
#
# JRE version: OpenJDK Runtime Environment Corretto-17.0.6.10.1 (17.0.6+10) (build 17.0.6+10-LTS)
# Java VM: OpenJDK 64-Bit Server VM Corretto-17.0.6.10.1 (17.0.6+10-LTS, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, g1 gc, linux-amd64)
# Problematic frame:
# C [libjavaProfiler16018171302964888844.so+0x7992] Buffer::putVar64(unsigned long long)+0x102
#
# No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# JFR recording file will be written. Location: /usr/share/tomcat/hs_err_pid23811.jfr
#
# If you would like to submit a bug report, please visit:
# https://github.com/corretto/corretto-17/issues/
#
--------------- S U M M A R Y ------------
--------------- T H R E A D ---------------
Current thread (0x00007f1755ab1900): JavaThread "dd-trace-processor" daemon [_thread_in_Java, id=24534, stack(0x00007f16fc328000,0x00007f16fc429000)]
Stack: [0x00007f16fc328000,0x00007f16fc429000], sp=0x00007f16fc4266d8, free space=1017k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
C [libjavaProfiler16018171302964888844.so+0x7992] Buffer::putVar64(unsigned long long)+0x102
C [libjavaProfiler16018171302964888844.so+0x1b956] Profiler::recordSample(void*, unsigned long long, int, int, Event*)+0x256
C [libjavaProfiler16018171302964888844.so+0x1c2c6] PerfEvents::signalHandler(int, siginfo_t*, void*)+0x116
siginfo: si_signo: 11 (SIGSEGV), si_code: 2 (SEGV_ACCERR), si_addr: 0x00007f17081c3673
About this issue
- Original URL
- State: closed
- Created a year ago
- Reactions: 17
- Comments: 18 (7 by maintainers)
Fixed by #4981
Seeing the same issue with Temurin base image and Java 11.0.18.
We managed to reproduce the issue and verified a fix for this, released in 1.11.2. If the crash persists after upgrading to 1.11.2, please report back with the backtrace from the hs_err file, the JDK version and the base docker base image being used (or the linux and libc versions otherwise).
We acknowledge that this crash is still possible, hence reopening the issue, and are working on getting 1.11.2 out with full mitigation.
Just in case people haven’t seen the workaround Jaroslav posted on slack:
1.11.2 has resolved this for me. Thank you.
1.11.1 didn’t solve the issue for us
We are validating the fix now and will do a patch release once we are sure the root cause is fixed. Will keep you posted.
Thank you, but this does not seem to be working yet. we are on Java 8. it still crashes with the latest agent.
SIGSEGV (0xb) at pc=0x00007fc0f07a28b2, pid=45, tid=0x00007fc078b5f700
+1 to above we ended up pinning to the latest stable version we observed
1.10.1instead of
We are seeing similar with Java 11 + CentOS. Identical application instances with version 1.10.0~c545cdc5a3 do not experience this issue.