Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 03 Apr 2015 17:48:50 -0400
From:      Jesse White <jesse@opennms.org>
To:        OpenNMS Code Development and Bugs <opennms-devel@lists.sourceforge.net>, java@freebsd.org
Subject:   Re: [opennms-devel] OpenNMS 1.14+ on FreeBSD 8, 9, 10, 11
Message-ID:  <551F0AC2.1040100@opennms.org>
In-Reply-To: <551C2448.1010704@atlantisservices.net>
References:  <5516FBCD.4050009@atlantisservices.net>	<55197228.4050808@opennms.org>	<5519C7DC.9010801@atlantisservices.net>	<5519DF51.90909@opennms.org>	<5519FD02.60109@atlantisservices.net>	<551AD507.1050703@opennms.org> <551C2448.1010704@atlantisservices.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On 04/01/2015 01:00 PM, Paul Pathiakis wrote:
> Just curious as to what it looks like the problem may be.

I've been playing around with OpenNMS 15.0.2 on FreeBSD 10.1 and the 
latest port of OpenJDK 8 (1.8.0_40).

The SIGSEGV is easy to reproduce and the problematic frame always points 
to networking related code. I've seen it happen in eventd's TCP listener 
threads as well as snmp4j's UDP transport threads running the in the 
context of provisiond.

The issue also occurs with all of the JNI and JNA bits disabled, so 
those can be ruled out.

I tried to reproduce the error by running provisiond's code in a 
standalone .jar, making it easier for the port maintainers to reproduce 
the problem, but I haven't had any success in triggering the crash this way.

At this point everything points to a JVM bug. I think we'll need some 
help from the port maintainers to solve this one.

-Jesse




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?551F0AC2.1040100>