public inbox for [email protected]  

help / color / mirror / Atom feed

Subject: [Pljava-dev] PLJava SIGSEGV on Linux
Date: Thu, 3 Sep 2009 16:23:23 +0100
Message-ID: <519DE98D62C54D4A896E22A026CF1322502584@cdlon-ex.CoalitionDev.local> (raw)

Hi,

I'm new to PLJava and am experiencing a problem. I've done the usual
google searches and haven't really found anything to help with my
problem - so I'm posting my problem here in the hope that someone's seen
it before and can point me in the right direction.


I'm using the following systems/software:

Postgres: v8.3.3 on x86_64-pc-linux-gnu

PLJava: 1.4.0 (pljava-x86_64-unknown-linux-gnu-pg8.3-1.4.0.tar.gz
downloaded from
http://pgfoundry.org/frs/?group_id=1000038&release_id=1024)

Java: Sun Java HotSpot(TM) 64-Bit Server VM (14.2-b01) for linux-amd64
JRE (1.6.0_16-b01)

Running on a HP server running Debian (etch) Linux 2.6.18-5-amd64 #1 SMP
Sat Dec 22 20:43:59 UTC 2007 x86_64 GNU/Linux

JBoss: jboss-4.2.3.GA


Our PLJava application is straightforward enough. On startup, we create
a database connection and invoke some sql code that invokes a java
function that registers a pljava/postgres transaction listener. Then
during normal usage, a database transaction is started, data in various
tables is modified - and we invoke another java function multiple times
to keep "track" of which items in the database have been modified by
this transaction.

Finally, when the transaction commits, the previously registered
transaction listener is automatically executed. This listener publishes
the list of changed database items onto a JBoss JMS topic.

This appears to work OK - but we regularly see the postgres processes
restarting due to a SIGSEGV fault in the jvm. Under light loads (1
transaction an hour) it can be several days before we see the restart.
We can provoke the restart by firing lots of concurrent transactions at
the database server (most of which modify data and hence end up
publishing the changes) - this causes a restart within 15 minutes
usually.


The problem nearly always occurs in JNI_callStaticVoidMethodV.

I've attached the java crash-log.


Does anyone have any pointers on what could be causing this?

Thanks,
--Jatinder
-------------- next part --------------
A non-text attachment was scrubbed...
Name: hs_err_pid21360.log
Type: application/octet-stream
Size: 27209 bytes
Desc: hs_err_pid21360.log
URL: <http://lists.pgfoundry.org/pipermail/pljava-dev/attachments/20090903/cf05ec70/attachment.obj;

reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: [email protected]
  Subject: Re: [Pljava-dev] PLJava SIGSEGV on Linux
  In-Reply-To: <519DE98D62C54D4A896E22A026CF1322502584@cdlon-ex.CoalitionDev.local>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox