Mailing-List: contact pylucene-dev-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: pylucene-dev@lucene.apache.org
Received-SPF: neutral (nike.apache.org: local policy)
Message-ID: <cf535d8201dd35468eb703f590eea627.squirrel@sm.webmail.pair.com>
In-Reply-To: <E5082F2F-23F6-4CDE-958B-283947319546@apache.org>
References: <4D265465.7030401@ontrenet.com>
    <alpine.OSX.2.01.1101061842370.34034@yuzu.local>
    <4D26FAC8.9050602@ontrenet.com>
    <alpine.OSX.2.01.1101071508220.34651@yuzu.local>
    <4D27A372.5090502@ontrenet.com>
    <E5082F2F-23F6-4CDE-958B-283947319546@apache.org>
Date: Fri, 7 Jan 2011 19:03:56 -0500
Subject: Re: JVM errors
From: darren@ontrenet.com
To: pylucene-dev@lucene.apache.org
User-Agent: SquirrelMail/1.4.20
MIME-Version: 1.0
Content-Type: text/plain;charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Importance: Normal

Not really because each thread is a worker that stays alive and iterates
over work items. The fatality seems to happen when they all are using the
index (lucene is supposed to be multi-read threadsafe).

It does it without the detach, but when the detach is in there it fatals
_on_ the detach. wierd.

>
> On Jan 7, 2011, at 15:36, Darren Govoni <darren@ontrenet.com> wrote:
>
>> That's a good thought.
>>
>> I just wish I knew how to interpret the JVM fatal error. I rarely
>> ever see them working with straight java,
>> so I'm also wondering if something else is at play. Its pretty
>> predicatable when it happens.
>> Like at either a vm.detachCurrentThread
>
> Oh. I have no confidence in detachCurrentThread(). Can you pool and
> reuse threads instead ?
>
> Andi..
>
>> or perhaps when an object attached to the current thread
>> goes out of scope, but in a stressed environment.
>>
>> On 01/07/2011 06:09 PM, Andi Vajda wrote:
>>>
>>> On Fri, 7 Jan 2011, Darren Govoni wrote:
>>>
>>>> I'll try that and report.
>>>>
>>>> It seems to happen when many threads are attached at once and the
>>>> CPU throttles.
>>>
>>> Maybe some GC settings could also help work around this ?
>>>
>>> Andi..
>>>
>>>>
>>>> On 01/06/2011 09:42 PM, Andi Vajda wrote:
>>>>>
>>>>> On Thu, 6 Jan 2011, Darren Govoni wrote:
>>>>>
>>>>>> Hi,
>>>>>> I am getting these JVM fatal errors:
>>>>>>
>>>>>> #
>>>>>> # A fatal error has been detected by the Java Runtime Environment:
>>>>>> #
>>>>>> #  SIGSEGV (0xb) at pc=0x00007f11dc3c093a, pid=6268,
>>>>>> tid=139711024641792
>>>>>> #
>>>>>> # JRE version: 6.0_21-b06
>>>>>> # Java VM: Java HotSpot(TM) 64-Bit Server VM (17.0-b16 mixed
>>>>>> mode linux-amd64 )
>>>>>> # Problematic frame:
>>>>>> # C  [libjcc.so+0xa93a]
>>>>>> _ZNK6JCCEnv21getObjectArrayElementEP13_jobjectArrayi+0x2a
>>>>>> #
>>>>>> # An error report file with more information is saved as:
>>>>>> # /home/darren/Gridwave/product/trunk/symphony/server/software/
>>>>>> hs_err_pid6268.log
>>>>>> #
>>>>>> # If you would like to submit a bug report, please visit:
>>>>>> #   http://java.sun.com/webapps/bugreport/crash.jsp
>>>>>> #
>>>>>>
>>>>>> They seem to happen sporadically in a multithreaded situation
>>>>>> where many threads (attached, etc) are accessing the same index.
>>>>>>
>>>>>> Is there anything I can do on my end to correct this?
>>>>>
>>>>> I've never seen this. Maybe getting a newer JVM ?
>>>>>
>>>>> Andi..
>>>>
>