FS#57677 - [jdk8-openjdk] 8.u162-1 causes segmentation faults

Attached to Project: Arch Linux
Opened by Stanislav T (stas-t) - Wednesday, 28 February 2018, 12:33 GMT
Last edited by Eli Schwartz (eschwartz) - Thursday, 01 March 2018, 19:29 GMT
Task Type Bug Report
Category Packages: Extra
Status Closed
Assigned To Guillaume ALAUX (galaux)
Architecture All
Severity High
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 8
Private No

Details

Description:
After upgrading jre8-openjdk-headless, jre8-openjdk and jdk8-openjdk from 8.u144-1 to 8.u162-1 maven run against big project causes segmentation fault on every build.

Additional info:
* package version(s)
linux 4.15.5-1
jre8-openjdk-headless 8.u162-1
jre8-openjdk 8.u162-1
jdk8-openjdk 8.u162-1
openjdk8-doc 8.u162-1
openjdk8-src 8.u162-1
maven 3.5.2-1

Steps to reproduce:

git clone https://github.com/dbeaver/dbeaver.git
cd dbeaver
env MAVEN_OPTS="-Xmx2G -XX:MaxPermSize=512m" mvn clean package

After a while the build interrupts and you see the message like this:

#
# A fatal error has been detected by the Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x00007ff2571163b8, pid=22837, tid=0x00007ff254b3d700
#
# JRE version: OpenJDK Runtime Environment (8.0_162-b12) (build 1.8.0_162-b12)
# Java VM: OpenJDK 64-Bit Server VM (25.162-b12 mixed mode linux-amd64 compressed oops)
# Problematic frame:
# V [libjvm.so+0x6b73b8]
#
# Core dump written. Default location: /home/stas/packages/dbeaver/core or core.22837
#
# An error report file with more information is saved as:
# /home/stas/packages/dbeaver/hs_err_pid22837.log
#
# If you would like to submit a bug report, please visit:
# http://bugreport.java.com/bugreport/crash.jsp
#
Aborted (core dumped)
This task depends upon

Closed by  Eli Schwartz (eschwartz)
Thursday, 01 March 2018, 19:29 GMT
Reason for closing:  Fixed
Additional comments about closing:  jdk8-openjdk 8.u162-2
Comment by Stanislav T (stas-t) - Wednesday, 28 February 2018, 12:36 GMT
Attached fatal error log.
Comment by kkl2401 (kkl2401) - Wednesday, 28 February 2018, 14:52 GMT
I've also been getting crashes with this version. Not during Maven invocation (I probably hadn't tried that) but from within IntelliJ IDEA during compilation. I can't be absolutely sure it's the same issue but it seems likely. I've immediately downgraded back to u144 which solved it.
Comment by nyyu (nyyu) - Wednesday, 28 February 2018, 17:08 GMT
The option "-fno-delete-null-pointer-check" is missing in CFLAGS
Comment by Mohammad AlSaleh (MoSal) - Wednesday, 28 February 2018, 23:23 GMT
I discovered 8.u162 crashes a private server software here the day it was pushed to [extra]. There is clearly something wrong with that release.

I didn't report it earlier because I searched the internet and didn't find anyone complaining (we are really early adopters). Unfortunately, downgrading is not an option security-wise.
Comment by Wai Wang (waicool20) - Thursday, 01 March 2018, 02:34 GMT
Also getting these SIGSEGV in all the java applications I'm using after upgrading to 8u162. Same as kkl2401, Intellij IDEA is very unstable and crashes intermittently during compilation or even normal usage. Same goes for other jvm apps. Downgraded back to 8u144 and have no more problems.
Comment by Denys Vitali (denvit) - Thursday, 01 March 2018, 16:04 GMT
Can confirm. I've had the exact same problem w/ openjdk 1.8.0_162:

```
Linux denvit-ws1 4.16.0-1-mainline #1 SMP PREEMPT Tue Feb 27 09:43:29 CET 2018 x86_64 GNU/Linux
```

```
openjdk version "1.8.0_162"
OpenJDK Runtime Environment (build 1.8.0_162-b12)
OpenJDK 64-Bit Server VM (build 25.162-b12, mixed mode)
```

```
jdk8-openjdk 8.u162-1
jre8-openjdk 8.u162-1
jre8-openjdk-headless 8.u162-1
```

Code that can reproduce the bug:
https://git.ded1.denv.it/dvitali/supsi-i2b-programmazione-concorrente-parallela/commit/20a9b4bd2dfca941079ae4d4749b78e8bd59257d

Specifically, line 19 causes the problem:
https://git.ded1.denv.it/dvitali/supsi-i2b-programmazione-concorrente-parallela/src/20a9b4bd2dfca941079ae4d4749b78e8bd59257d/src/it/denv/supsi/progconc/serie/serie2/es3/WhitdrawalProcess.java#L19
Comment by Guillaume ALAUX (galaux) - Thursday, 01 March 2018, 17:18 GMT
I've just pushed version 8.u162-2 with `-fno-delete-null-pointer-checks` as nyyu advised. Thanks for your report.
Could you please test and report here?
As it can take time to get to your mirrors, I've also made the files along with signatures available here: http://pkgbuild.com/~guillaume/misc/openjdk_8.u162/
Comment by Luca Weiss (z3ntu) - Thursday, 01 March 2018, 17:54 GMT
-2 seems to have fixed my problems with the atlassian plugin sdk (which uses maven behind the scenes).
Comment by Denys Vitali (denvit) - Thursday, 01 March 2018, 18:12 GMT
I can confirm that @galaux package fixed my issues. Well done!
Comment by Mohammad AlSaleh (MoSal) - Thursday, 01 March 2018, 19:18 GMT
Fixed here too. Thanks.

Loading...