Subject: Re: S@h on MacIntel?
From: Chris Schram
Date: 30/07/2006, 06:25
Newsgroups: alt.sci.seti

On Sat, 29 Jul 2006 13:03:07 +0200, Lars Bausch wrote:

Hello Cris,

Chris Schram wrote:
I am running BOINCManager 5.4.9 on a new Intel iMac running Mac OS X
10.4.7, and have observed that the SETI@home application within BOINC
crashes quite frequently. I am also running Einstein@home on the same
machine and it never ever crashes.

When SETI@home crashes, BOINCManager always gets it running again, and
my workunits do get completed.

I haven't a clue how to interpret all these crash logs I've gotten, but
they all look quite similar. Does anyone know who handles the Mac
programming at S@h? I'd like to see if we can get this issue resolved.
Can you please post some of the <stderr> outputs from the aborted WUs. You
can see the stderr log via the SETI Web Sitte / Your Account / Computers
on this account / select Results (from the affected system) /
  Result ID.

It would appear that, with one exception, my crashed workunits are able to
restart and complete without producing anything useful in stderr. Here is
the one exception:

<http://setiathome.berkeley.edu/result.php?resultid=357584163>

I am currently detached from S@h on the Intel iMac, but I was getting
crash logs on every workunit. If someone out there wants to see what one
of those looks like, I'm sure I could reattach and generate one in no time
flat. ;-)


Any other MacIntel users out there getting these frequent crashes?

Are there any other BOINC projects beside the ones I mentioned that
support the MacIntel Screensaver PrefPane?
SIMAP has a Mac Intel application, too. Some other projects (Predictor,
Rosetta, SZTAKI, WCG) has binaries for Mac PPC which should run in the
rosetta emulation.

I was attached to Predictor for a long while, unaware that it was MacIntel
hostile. It kept me up-to-date on the Predictor activity on my dinosaur
PPC Mac, but was unable to actually do any work. I prefer not to use
Rosetta unless I'm backed up against the wall at gunpoint. Thanks for the
heads-up about SIMAP. I'll take a look and see if it appeals to me.

-- schram@webenet.net is a filtered spam magnet. Email replies may get lost. Try <http://public.xdi.org/=chris.schram> instead.
-- Posted via a free Usenet account from http://www.teranews.com