IGV segfaulting on startup (Fedora 27)

[James Abbott]

Hello,

I'm experiencing segfaults on launching IGV on fedora 27 under OpenJDK 8.0_121-b15. I first saw this with 2.3.98 installed from bioconda, and have now downloaded the 2.4.4 binary distribution where I am seeing the same problem.

[jamesa@wssb-jamesa IGV_2.4.4]$ ./igv.sh 

WARN [2017-12-11 14:03:00,770]  [Globals.java:138] [AWT-EventQueue-0]  Development mode is enabled

INFO [2017-12-11 14:03:00,793]  [DirectoryManager.java:171] [AWT-EventQueue-0]  IGV Directory: /home/jamesa/igv

INFO [2017-12-11 14:03:00,793] [DirectoryManager.java:171]  IGV Directory: /home/jamesa/igv

INFO [2017-12-11 14:03:00,795]  [Main.java:154] [AWT-EventQueue-0]  Startup  IGV Version 2.4.4 11/09/2017 01:52 PM

INFO [2017-12-11 14:03:00,795]  [Main.java:155] [AWT-EventQueue-0]  Java 1.8.0_121

INFO [2017-12-11 14:03:00,795]  [DirectoryManager.java:76] [AWT-EventQueue-0]  Fetching user directory... 

INFO [2017-12-11 14:03:00,813]  [Main.java:156] [AWT-EventQueue-0]  Default User Directory: /home/jamesa

INFO [2017-12-11 14:03:00,813]  [Main.java:157] [AWT-EventQueue-0]  OS: Linux

Dec 11, 2017 2:03:00 PM java.util.prefs.FileSystemPreferences$6 run

WARNING: Prefs file removed in background /home/jamesa/.java/.userPrefs/prefs.xml

#

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

#

#  SIGSEGV (0xb) at pc=0x000000000000a256, pid=27381, tid=0x00007f3dd6dd5700

#

# JRE version: OpenJDK Runtime Environment (8.0_121-b15) (build 1.8.0_121-b15)

# Java VM: OpenJDK 64-Bit Server VM (25.121-b15 mixed mode linux-amd64 compressed oops)

# Problematic frame:

# C  0x000000000000a256

#

# Core dump written. Default location: /home/jamesa/Downloads/IGV_2.4.4/core or core.27381

#

# An error report file with more information is saved as:

# /home/jamesa/Downloads/IGV_2.4.4/hs_err_pid27381.log

#

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

#   http://www.azulsystems.com/support/

#

Aborted (core dumped)

I've attached the hs_err_pid log, and have the core available if this may help in debugging. The 'Problematic frame' reported is the same for both igv 2.3.98 and 2.4.4.

Both versions work fine on the same machines under the non-open Java(TM) SE Runtime Environment (build 1.8.0_144-b01).

Best Regards

James

[James Robinson]

Hi,  seg faults in Java are of course JVM bugs,  there isn't much we can do about it.    It looks like the specified problem is in some compiled code called from  "sun.net.spi.DefaultProxySelector.init".     Have you filed a bug report with OpenJDK?

I did add a flag to possibly work around this.   Its available in the latest snapshot version (http://software.broadinstitute.org/software/igv/download_snapshot).   To use it make sure IGV is not running, then edit your prefs.properties file in the "igv" folder under your home directory and add the following line

PROXY.DISABLE_CHECK=true

This will disable checks for a system defined proxy which seems to be triggering this OpenJDK bug.   If you use a proxy you will have to configure it manually in the IGV preferences window.  

Stack: [0x00007f3dd6cd5000,0x00007f3dd6dd6000],  sp=0x00007f3dd6dd2778,  free space=1013k

Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)

C  0x000000000000a256

j  sun.net.spi.DefaultProxySelector.init()Z+0

j  sun.net.spi.DefaultProxySelector.<clinit>()V+172

--

---
You received this message because you are subscribed to the Google Groups "igv-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to igv-help+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/igv-help/53614890-f933-4ed8-a1af-eab8896132ac%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.