[27236] in Athena Bugs
Re: StarBiochem Java applet fails on Optiplex 745s in W20-575
daemon@ATHENA.MIT.EDU (William Cattey)
Mon Sep 15 17:38:01 2008
In-Reply-To: <50E141BF-C306-44CF-A280-9D19CD9B068D@mit.edu>
Mime-Version: 1.0 (Apple Message framework v753.1)
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Message-Id: <71F676AE-76A4-425F-B557-6E6E5010329A@mit.edu>
Content-Transfer-Encoding: 7bit
From: William Cattey <wdc@mit.edu>
Date: Mon, 15 Sep 2008 17:37:37 -0400
To: Charles Shubert <cshubert@mit.edu>
X-Spam-Flag: NO
X-Spam-Score: 0.00
Cc: Shelley S Gu <sgu@mit.edu>, James Cain <jrcain@mit.edu>, star@mit.edu,
bugs@mit.edu, Geoffrey Thomas <geofft@mit.edu>
Errors-To: bugs-bounces@mit.edu
This problem had the same root cause as the OpenOffice failure:
Unauthorized software installed on some cluster
machines, particularly in W20 had screwed up the OpenGL library.
Cluster Services is cleaning stuff up,
but if you find yourself on an affected machine in the meantime, you can
run, as root, /etc/athena/verify followed by
/etc/athena/install-proprietary-x-drivers.
-Bill
----
Important: IS&T IT staff will *NEVER* ask you for your password, nor
will MIT send you email requesting your password information. Please
continue to ignore any email messages that claim to require you to
provide such information.
----
William Cattey
Linux Platform Coordinator
MIT Information Services & Technology
N42-040M, 617-253-0140, wdc@mit.edu
http://web.mit.edu/wdc/www/
On Sep 11, 2008, at 7:05 PM, Charles Shubert wrote:
> Hi Geoffrey,
>
> Thanks for reporting this bug. I'd guess that your analysis is
> likely to be correct.
>
> The segfault looks like it is pretty deep in the code. It's been a
> while since I've seen this type of bug as the code in StarBiochem
> is written in Java and this looks to be a memory management flaw.
> I'd guess the Optiplex driver has a bug. Googling "Optiplex
> segfault" gets hits for both ubuntu and Red Hat which suggests
> Optiplex is where the problem lies.
>
> It would be helpful if you could give me a set of steps that would
> allow me to reproduce this problem in the W20-575 cluster.
>
> There may be an updated driver for Optiplex. This bug is a little
> out of the StarBiochem league, so I'll try to find out who owns the
> W20-575 cluster and ask them to take a look at both the installed
> drivers and the version of Java that is running on this machine.
>
> I'm assuming that since you can run StarBiochem on other machines
> this is not something that stands between you and turning in homework.
>
> This is a good catch!
>
> Thanks,
>
> --Chuck
>
> On Sep 11, 2008, at 6:26 PM, Geoffrey Thomas wrote:
>
>> Hi,
>>
>> On the new widescreen Optiplex 745s in the W20-575 cluster, the
>> StarBiochem Java Web Start applet from http://web.mit.edu/star/
>> biochem/ crashes with the following error:
>> #
>> # An unexpected error has been detected by HotSpot Virtual Machine:
>> #
>> # SIGSEGV (0xb) at pc=0x00000000, pid=31441, tid=1827867552
>> #
>> # Java VM: Java HotSpot(TM) Server VM (1.5.0_16-b02 mixed mode)
>> # Problematic frame:
>> # C 0x00000000
>> #
>> # An error report file with more information is saved as
>> hs_err_pid31441.log
>> #
>> # If you would like to submit a bug report, please visit:
>> # http://java.sun.com/webapps/bugreport/crash.jsp
>> #
>>
>> I've attached the file it mentioned. The error happens on the
>> fairly clean sipbtest account, and it can't be reproduced on other
>> machines including the HPs in the cluster and an older Optiplex.
>>
>> Since this is a null pointer segfault that only happens on the
>> widescreen Optiplex 745s, I'm wondering if it's related to the
>> OpenOffice.org null pointer segfault that only happens on the
>> widescreen Optiplex 745s that I reported earlier this week (http://
>> diswww.mit.edu/menelaus/bugs/27225).
>> It certainly seems likely that StarBiochem is using DRI for
>> rendering 3D molecules.
>>
>> --
>> Geoffrey Thomas
>> geofft@mit.edu<hs_err_pid31441.log>
>
