Java Hotspot Crash Error

[digerata]

I'm in the process of setting up a tomcat appliance. When I start tomcat, it fails with a JVM crash. The error reported in tomcat logs:

#
# An unexpected error has been detected by Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x00002aaaaaab4112, pid=2577, tid=1076017472
#
# Java VM: Java HotSpot(TM) 64-Bit Server VM (10.0-b23 mixed mode linux-amd64)
# Problematic frame:
# C [ld-linux-x86-64.so.2+0x9112]
#
# An error report file with more information is saved as:
# /tmp/hs_err_pid2577.log
#
# If you would like to submit a bug report, please visit:
# http://java.sun.com/webapps/bugreport/crash.jsp
# The crash happened outside the Java Virtual Machine in native code.
# See problematic frame for where to report the bug.
#

The versions I'm using:

Java SE Development Kit 6u7 (jdk-6u7-linux-x64.bin)
Tomcat 6.0.16
WEB64 Appliance

Basically, I took a WEB64 appliance, renamed and branched it to TOMCAT. I resized the /usr and / volumes to 2GB each. I then started up the app and downloaded java and then tomcat. I extracted both of those to /usr/local/tomcat and java. Set my JAVA_HOME path. Created a service script: /etc/init.d/tomcat using a standard template. Ran service tomcat start and hit the problem.

I created a one line java class that prints "working." and can run that. However, that certainly isn't a full test of the JVM, but it does suggest that the JVM is installed and can perform basic operations.

Any ideas?

Thanks,

-Mike

[PeterNic]

Mike,

I would suggest:

1. Add a swap volume to the appliance (don't forget to make sure it is mounted in the appliance)

2. Take the 64-bit VPS, do a full yum update (except the kernel, of course), and try again. If it works, most likely some of the libraries need to be upgraded

3. Try a different version of the JVM if available.

Let me know how it goes... (we have a few people who use 64-bit JVM, no similar reports so far).

Regards,
-- Peter

[digerata]

I tried the yum update suggestion first. That worked! Thanks for the help!

On aside, the applogic shell brings Firefox crashing to its knees when running commands that output a ton of text. (E.g., yum update.) I gave it 20 minutes at 100% and yum update got to 9 of 254 items to update before not responding. Is this normal? If not, where do i submit a bug report?

Thanks again!

[PeterNic]

I tried the yum update suggestion first. That worked! Thanks for the help!

I am glad that worked out!


On aside, the applogic shell brings Firefox crashing to its knees when running commands that output a ton of text. (E.g., yum update.) I gave it 20 minutes at 100% and yum update got to 9 of 254 items to update before not responding. Is this normal? If not, where do i submit a bug report?


You just did. Can you please provide additional details (either here or via PM) -- host OS, Firefox version, is the problem reproducible (and how), how did FF crash... We know that FF 2.0 leaked memory (at least when used with our web shell), so sometimes if you just close all FF windows and open a fresh copy things work better. In any case, we'd like to know the details so we can reproduce/diagnose/fix. Note that long output does take a while -- but it does finish (tail /var/log/messages or big apache access logs does take several minutes).

Workaround (which you probably did): use ssh instead of the web shell in these cases.

Thanks,
-- Peter

[digerata]

PM'd you the details...

[PeterNic]

digerata,

Thanks. Some of the problems appear to be mac-specific (because we're not seeing them on non-macs) but this is valuable feedback; I have forwarded the issues to our engineering -- thanks for your help!

-- Peter