Temurin-build: SIGSEGV (0xb) in C [libc++abi.dylib+0x30b4] for osX 10.15 Catalina

Created on 9 Oct 2019 · 15Comments · Source: adoptium/temurin-build

Platform:
OS:uname:Darwin 19.0.0 Darwin Kernel Version 19.0.0: Wed Sep 25 20:18:50 PDT 2019; root:xnu-6153.11.26~2/RELEASE_X86_64 x86_64
Architecture:
x86_64

hs_err_pid30602.log

After upgrade to os X 10.15 Catalina, previously working program failed with SIGSEGV.
How to reproduce:
create a basic maven project with 2 dependencies:

    <dependency>
      <groupId>org.deeplearning4j</groupId>
      <artifactId>deeplearning4j-core</artifactId>
      <version>1.0.0-beta4</version>
    </dependency>
    <dependency>
      <groupId>org.nd4j</groupId>
      <artifactId>nd4j-native-platform</artifactId>
      <version>1.0.0-beta4</version>
    </dependency>

and one class:

public class Main{

  public static void main( String[] args ){
    INDArray s = Nd4j.zeros(1, 1);
    System.out.println( s.isMatrix() );
  }

}

.
Under Os X 10.15 Catalina, expected result:

false

Process finished with exit code 0

Actual result:

#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fff666980b4, pid=31244, tid=10243
#
# JRE version: OpenJDK Runtime Environment (11.0.4+11) (build 11.0.4+11)
# Java VM: OpenJDK 64-Bit Server VM (11.0.4+11, mixed mode, tiered, compressed oops, g1 gc, bsd-amd64)
# Problematic frame:
# C  [libc++abi.dylib+0x30b4]  __cxxabiv1::__si_class_type_info::has_unambiguous_public_base(__cxxabiv1::__dynamic_cast_info*, void*, int) const+0x4
#
# No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /Users/ivanpetrychenko/src/testAdotOpenJdk/hs_err_pid31244.log
#
# If you would like to submit a bug report, please visit:
#   https://github.com/AdoptOpenJDK/openjdk-build/issues
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#

Process finished with exit code 134 (interrupted by signal 6: SIGABRT)

bug

Source

petrychenko

👍3

Most helpful comment

might be solved as mentioned here: https://github.com/eclipse/deeplearning4j/issues/8156

ticofab on 18 Nov 2019

👍2

All 15 comments

you have two libstdc++ loaded into memory, are you sure it's ok to do so ?
0x00007fff6860a000 /usr/lib/libstdc++.6.dylib
0x00000001351d8000 /Users/ivanpetrychenko/.javacpp/cache/mkl-dnn-0.18.1-1.5-macosx-x86_64.jar/org/bytedeco/mkldnn/macosx-x86_64/libstdc++.6.dylib

perunosx on 9 Oct 2019

I'm not sure if it is normal or not, I'm not the author of this library.
But as this code works for Oracle JDK, and for AdoptOpenJDK before update to Catalina, it seems it is acceptable.
The same bug is reported to deeplearning4j: https://github.com/eclipse/deeplearning4j/issues/8156 .

petrychenko on 10 Oct 2019

I am having exactly the same issue since upgrading to Catalina yesterday, adding my specs in case it can be useful. Haven't tried with Oracle JDK.

08:59:16.436 INFO  o.n.linalg.factory.Nd4jBackend - Loaded [CpuBackend] backend
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fff6e8160b4, pid=8675, tid=35587
#
# JRE version: OpenJDK Runtime Environment (11.0.4+11) (build 11.0.4+11)
# Java VM: OpenJDK 64-Bit Server VM (11.0.4+11, mixed mode, tiered, compressed oops, g1 gc, bsd-amd64)
# Problematic frame:
# C  [libc++abi.dylib+0x30b4]  __cxxabiv1::__si_class_type_info::has_unambiguous_public_base(__cxxabiv1::__dynamic_cast_info*, void*, int) const+0x4
#
# No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /Users/fabiotiriticco/Development/programming/ticofab-rep/experiments/deeplearning/planes/hs_err_pid8675.log
#
# If you would like to submit a bug report, please visit:
#   https://github.com/AdoptOpenJDK/openjdk-build/issues
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#

hs_err_pid8675.log

MBP 15 in ch 2019, processor 2,3 GHz 8-Core Intel Core i9

ticofab on 11 Oct 2019

Same error for me, using Oracle JDK (also happened with OpenJDK); macOS 10.15 on a 2018 MBP 15", 2.6 GHz 6-core i7:

#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fff6b0d40b4, pid=56262, tid=28679
#
# JRE version: Java(TM) SE Runtime Environment (10.0.2+13) (build 10.0.2+13)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (10.0.2+13, mixed mode, tiered, compressed oops, g1 gc, bsd-amd64)
# Problematic frame:
# C  [libc++abi.dylib+0x30b4]  __cxxabiv1::__si_class_type_info::has_unambiguous_public_base(__cxxabiv1::__dynamic_cast_info*, void*, int) const+0x4
#
# No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /Users/tylucaskelley/Code/topos/GeoGulp/bin/hs_err_pid56262.log
#
# If you would like to submit a bug report, please visit:
#   http://bugreport.java.com/bugreport/crash.jsp
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.

hs_err_pid56262.log

tylucaskelley on 14 Oct 2019

Same error for me for macOS 10.15 on a 2015 MBP 13" i5

Adding log:
hs_err_pid1656.log

drbitzer on 15 Oct 2019

Still an issue: macOS 10.15 (19A602) Darwin 19.0.0

14:33:15.855 [main] INFO  o.n.l.f.Nd4jBackend - Loaded [CpuBackend] backend
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007fff686400b4, pid=14351, tid=0x0000000000000e03
#
# JRE version: Java(TM) SE Runtime Environment (8.0_181-b13) (build 1.8.0_181-b13)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (25.181-b13 mixed mode bsd-amd64 compressed oops)
# Problematic frame:
# C  [libc++abi.dylib+0x30b4]  _ZNK10__cxxabiv120__si_class_type_info27has_unambiguous_public_baseEPNS_19__dynamic_cast_infoEPvi+0x4
#
# Failed to write core dump. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#

hs_err_pid14351.log

kyndi-barryevans on 17 Oct 2019

Same error here. I tried

Oracle JDK 8
AdoptOpenJDK 8
AdoptOpenJDK 11
Worked fine before upgrading to Mac OS 10.15. So I don't think it's an AdoptOpenJDK problem...

ff-will on 22 Oct 2019

Same here

fjauer on 23 Oct 2019

Can folks try the latest builds that came out this week?

https://adoptopenjdk.net/releases.html?variant=openjdk8&jvmVariant=hotspot#x64_mac
https://adoptopenjdk.net/releases.html?variant=openjdk11&jvmVariant=hotspot#x64_mac

karianna on 23 Oct 2019

@karianna Just tried it with OpenJDK11: Still the same. Not surprised though as this really seems to be a problem in the deeplearning4j library.

ff-will on 23 Oct 2019

Still the same, and still works under Orcale JDK
393 [main] INFO o.n.l.f.Nd4jBackend - Loaded [CpuBackend] backend

A fatal error has been detected by the Java Runtime Environment:

SIGSEGV (0xb) at pc=0x00007fff686c10b4, pid=95582, tid=9987

JRE version: OpenJDK Runtime Environment (11.0.5+10) (build 11.0.5+10)
Java VM: OpenJDK 64-Bit Server VM (11.0.5+10, mixed mode, tiered, compressed oops, g1 gc, bsd-amd64)
Problematic frame:
C [libc++abi.dylib+0x30b4] __cxxabiv1::__si_class_type_info::has_unambiguous_public_base(__cxxabiv1::__dynamic_cast_info, void, int) const+0x4

petrychenko on 24 Oct 2019

Just tried the quickstart with the latest version of AdoptOperJDK 11:
hs_err_pid15516.log

PMacho on 26 Oct 2019

Same here.
I tried AdoptOpenJDK 8, 11, 13
Catalina 10.15

mimkorn on 29 Oct 2019

might be solved as mentioned here: https://github.com/eclipse/deeplearning4j/issues/8156

ticofab on 18 Nov 2019

👍2

I would guess that the suggested solution is to upgrade to a newer version
of DL4J.
Maybe the bug still occurs in newer versions? I haven't worked with this in
a while so I can't be of much help I'm afraid.
Good luck!

On Sun, Aug 23, 2020, 00:46 rgampa notifications@github.com wrote:

@ticofab https://github.com/ticofab Can you point me to the solution in
thread eclipse/deeplearning4j#8156
https://github.com/eclipse/deeplearning4j/issues/8156.

After upgrading to MacOs Catalina tomcat (8.5.43 and 8.5.57) is crashing
with below error,

#

A fatal error has been detected by the Java Runtime Environment:

#

SIGSEGV (0xb) at pc=0x00007fff67945aa8, pid=3209, tid=0x0000000000009903

#

JRE version: Java(TM) SE Runtime Environment (8.0_221-b11) (build 1.8.0_221-b11)

Java VM: Java HotSpot(TM) 64-Bit Server VM (25.221-b11 mixed mode bsd-amd64 compressed oops)

Problematic frame:

C [libc++abi.dylib+0x10aa8] _ZNK10__cxxabiv120__si_class_type_info27has_unambiguous_public_baseEPNS_19__dynamic_cast_infoEPvi+0x4

#

Failed to write core dump. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again

#

If you would like to submit a bug report, please visit:

http://bugreport.java.com/bugreport/crash.jsp

The crash happened outside the Java Virtual Machine in native code.

See problematic frame for where to report the bug.

#

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
https://github.com/AdoptOpenJDK/openjdk-build/issues/1315#issuecomment-678704557,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/AANVXFK2JCZSKD5RIUEQQYTSCBDDBANCNFSM4I6Y4USA
.