7

I am doing reboot tests on Sles12sp2 using STAF v3.4.24 and after some time I get this error:

# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007f105e11e712, pid=6577, tid=0x00007f1027efe700
#
# JRE version: Java(TM) SE Runtime Environment (8.0_121-b13) (build 1.8.0_121-b13)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (25.121-b13 mixed mode linux-amd64 compressed oops)
# Problematic frame:
# V  [libjvm.so+0x63c712][thread 139707363079936 also had an error]
  IndexSet::alloc_block_containing(unsigned int)+0x42
#
# Core dump written. Default location: /root/core or core.6577
#
# An error report file with more information is saved as:
# /root/hs_err_pid6577.log

[error occurred during error reporting, id 0xb]
# If you would like to submit a bug report, please visit:
#   http://bugreport.java.com/bugreport/crash.jsp
#

I have searched on Google and most of the links were about C library. Java version is 8u121. No core dump was found at the location proposed.

Mario Boss
  • 1,784
  • 3
  • 20
  • 43
Adriana
  • 71
  • 1
  • 3
  • Some updates of the problem: also I tried to increase Java memory: -XXMaxPermGen=2048m but this parameter is ignored in Java 8 and this issue is still reproducible. I have tried also with -Xmx=1024m but the same result. – Adriana Apr 12 '17 at 14:40
  • 1
    Could you please paste the complete hs_error file – Fairoz Apr 12 '17 at 15:51
  • I've uploaded the file. Please check the link: https://drive.google.com/file/d/0BzW1jWdmyXcTM0NTNnRQZGtuRnM/view?usp=sharing – Adriana Apr 13 '17 at 07:07
  • @Fairoz did you take a look on the file uploaded on Google Drive? – Adriana Apr 18 '17 at 07:20
  • Sorry i will check that now... – Fairoz Apr 18 '17 at 07:38

1 Answers1

2

From the logs you got crash in C2 compiler thread, during compilation of "com.sun.org.apache.xerces.internal.impl.xs.XMLSchemaValidator::handleStartElement" it could be sereous compilation bug. Without proper reporducer it will be difficult to identify the root cause in compiler and fix as wee.

As a work around you should exclude this method from compilation using -XX:CompileCommand=exclude, com.sun.org.apache.xerces.internal.impl.xs.XMLSchemaValidator::handleStartElement check the proper usage of the -XX:CompileCommand=exclude command

Fairoz
  • 1,616
  • 13
  • 16
  • What do you mean by " Without proper reporducer it will be difficult to identify the root cause in compiler and fix as wee." ?? What should I do to find the root cause of this crash? – Adriana Apr 18 '17 at 13:18
  • Hi @andriana i work in Oracle in Java team. Issue you found is Java bug to fix that it will be difficult for us as the crash occuring during compilation of "com.sun.org.apache.xerces.internal.impl.xs.XMLSchemaValidator::handleStartElement" method which is not jdk source. You no need to find the root cause, if it is easily reproduce kindly share the steps to reproduce we will look for fix at our end. – Fairoz Apr 18 '17 at 14:23
  • 1
    What does the XX:CompileCommand=exclude exactly do? Can this option affect the generation of the XML files or the XML files in general? If yes, how? If no, why? – Adriana Apr 19 '17 at 13:35
  • 1
    Please read this - https://uberdev.wordpress.com/2010/07/21/xxcompilecommandexclude-jvm-option/ – Fairoz Apr 19 '17 at 13:43
  • 1
    I have done a reboot test with 7 hours with the JVM option that you gave and I got another crash. Please check https://drive.google.com/file/d/0BzW1jWdmyXcTRFdQWkdfcUpaZFU/view?usp=sharing – Adriana Apr 20 '17 at 07:09
  • And please also check the JVM log file https://drive.google.com/file/d/0BzW1jWdmyXcTM085cnhWM01MVzQ/view?usp=sharing. – Adriana Apr 20 '17 at 07:14
  • Will it be possible for you to share the test case, that i can reproduce at my end? As you got another C2 crash at org.jdom2.Element::getChild. – Fairoz Apr 20 '17 at 08:36
  • I have made a reboot script and I am trying to reproduce the issue without any specific application from my company. Also would you be so kind to check the core dump found? Link for the core dump: https://drive.google.com/file/d/0BzW1jWdmyXcTR1Y1VFpVMTFfSnc/view?usp=sharing – Adriana Apr 26 '17 at 09:16
  • Core dump is from the same crash you have posted https://drive.google.com/file/d/0BzW1jWdmyXcTM0NTNnRQZGtuRnM/view ? – Fairoz Apr 26 '17 at 10:46
  • can you please analyze the core dump? Meanwhile i am trying to reproduce the issue and give you the steps to follow. – Adriana Apr 26 '17 at 11:50
  • I managed to reproduce the issue by following the below steps: - install STAF 3.4.24 from https://sourceforge.net/projects/staf/files/staf/V3.4.24/ , choose STAF3424-setup-linux-amd64-NoJVM.bin - follow steps by step instalaltion - install staf in /usr/local/staf - install java 8u111 – Adriana May 04 '17 at 08:19
  • - create folder /home/go2test - copy scripts startstaf.sh and xstaf.sh in folder created above - put in /etc/xdg/autostart the files reboottests.desktop and staf.desktop - put the script reboot_test.sh in any folder you want e.g in /root - run reboot_test.sh The script will do 300 reboot cycles. In our case the issue has been reproduced after 83 cycles. – Adriana May 04 '17 at 08:19
  • Thanks Adriana i will try to reproduce once i am done with my current assignment. – Fairoz May 04 '17 at 08:47
  • But I am not sure that this will help you, anyway I have uploaded all the files needed in Google Drive: https://drive.google.com/file/d/0BzW1jWdmyXcTWGtPM21OcFZKVW8/view?usp=sharing – Adriana May 04 '17 at 09:18
  • Hi Andriana, looks like it is taking more time than i expected, could you file an issue with Java i will look into it. http://bugreport.java.com/submit_start.do – Fairoz May 09 '17 at 04:30
  • 1
    Hi Andriana, i have created issue on your behalf - https://bugs.openjdk.java.net/browse/JDK-8180406 – Fairoz May 16 '17 at 07:26
  • Ok. Thanks a lot. Please be aware that you will need also the files from Google Drive to reproduce the issue. – Adriana May 17 '17 at 11:18
  • Hi Andriana, we are unable to reproduce this issue at our end. As it has lot of dependencies and require multiple reboots. Will it be possible for you to reduce to smaller test case and share? – Fairoz May 24 '17 at 15:02
  • Hi Fairoz, could you please tell me what seems to be the problem? As I have reduced the test case to the minimum – Adriana May 26 '17 at 12:51
  • Hello @Fairoz, could you please answer my question above? – Adriana Jun 07 '17 at 09:22
  • Hi @Adriana, sorry i didn't saw your question, I am working on a remote machine where reboots are difficult, hence i won't be able to run the application completly, if you could make it simpler test case that will be greater help. In JDK9 there are are comple of issues fixed, it will be good you can check on that also. – Fairoz Jun 07 '17 at 12:48