All WU error after few sec of runtime
All WU error after few sec of runtime
http://www.renderfarm.fi/results.php?hostid=674
example boinc log:
<core_client_version>6.10.43</core_client_version>
<![CDATA[
<message>
- exit code -3 (0xfffffffd)
</message>
<stderr_txt>
boinc_init_diagnostics() completed
boinc_init_options() completed
Basename: projects/www.renderfarm.fi/blender_4.74_windows_intelx86.exe --> blender_4.74_windows_intelx86.exe
Mapping logical files to physical destinations:
in => ../../projects/www.renderfarm.fi/file202021
out.zip => ../../projects/www.renderfarm.fi/ses0000000191frm0000000411prt00009_6_0
blender_4.74_windows_intelx86.exe.zip => ../../projects/www.renderfarm.fi/blender_4.74_windows_intelx86.exe.zip
./windows_zip.exe => ../../projects/www.renderfarm.fi/windows_zip.exe
./windows_unzip.exe => ../../projects/www.renderfarm.fi/windows_unzip.exe
Decompressing executable archive...
Running ../../projects/www.renderfarm.fi/windows_unzip.exe -o ../../projects/www.renderfarm.fi/blender_4.74_windows_intelx86.exe.zip
runcmd(../../projects/www.renderfarm.fi/windows_unzip.exe -o ../../projects/www.renderfarm.fi/blender_4.74_windows_intelx86.exe.zip) exitted with exit code 0
Creating worker...
Worker constructing...
Worker constructed.
$Id: glue.cpp,v 1.2 2008/08/27 08:56:13 jbk Exp $
$Id: BOINCHandler.cpp,v 1.1 2008/08/16 12:56:40 jbk Exp $
$Id: Controller.cpp,v 1.2 2008/08/27 08:56:09 jbk Exp $
$Id: ProgressMonitor.cpp,v 1.1 2008/08/16 12:56:45 jbk Exp $
Executing blender.exe -noaudio -y -b ../../projects/www.renderfarm.fi/file202021 -F PNG -t 1 -f 411 0.4375 0.0 0.5 1.0
Created pipes
Child created.
Worker thread started
Worker thread monitor almost up.
Worker thread monitor up.
|Shortname = 'blender.exe'
Application reports 'Booted'
|'import site' failed; use -v for traceback
|Compiled with Python version 2.5.
|Warning: could not determine argv[0] path
|Checking for installed Python... No installed Python found.
|Only built-in modules are available. Some scripts may not run.
|Continuing happily.
|cyclic OBCurveCircle.005
|cyclic OBCurveCircle.004
|cyclic OBCurveCircle.004
|cyclic OBCurveCircle.005
|cyclic OBCurveCircle.004
|cyclic OBCurveCircle.005
|cyclic OBCurveCircle.005
|cyclic OBCurveCircle.004
|ERROR: Border render and Buffer-save not supported yet
|
|Blender quit
Application reports 'Exit'
../../projects/www.renderfarm.fi/windows_zip.exe -9 ../../projects/www.renderfarm.fi/ses0000000191frm0000000411prt00009_6_0 out being called
Running ../../projects/www.renderfarm.fi/windows_zip.exe -9 ../../projects/www.renderfarm.fi/ses0000000191frm0000000411prt00009_6_0 out
runcmd(../../projects/www.renderfarm.fi/windows_zip.exe -9 ../../projects/www.renderfarm.fi/ses0000000191frm0000000411prt00009_6_0 out) exitted with exit code 12
---------------------------
Exception caught: Non-zero exit status in Portability::runcmd()
Status: -3
---------------------------
</stderr_txt>
]]>
Sorry about these problems. We are doing some stress tests on big scenes like this and trying to remove issues as they appear. This time the reason was buffer save being enabled in the .blend.
Same here, all tasks end in Computation error after about 13 seconds. Not nice after a 113MB d/l.
Tried today run this application on 2 old computers. I think, on Intel Pentium 1.8 it is working.
But on AMD Duron 1.3 all tasks ended with Computation error after several seconds.
Why?
It seems that alot of computers (including my Macbook Pro) are failing to compute WU's from session 667. We'll have to work on this tomorrow and try to find out what's wrong.
I'm the uploader of session 667 and i wonder about the maximum amount of ram it can use...
I set it to 512 MB now i noticed it's set to 2048 MB is it possible that those errors have something to do with it?
Yesterday it worked on my Pentium, who has 767 MB RAM with no problems, but today I got message:
15/09/2010 16:37:25 Open Rendering Environment Message from server: Blender needs 2048.00 MB RAM but only 767.49 MB is available for use.
Sad news, I think :)
Some sessions require more RAM than others to run. This does not mean that no work will be dealt to the computers with small amounts of RAM, but merely that they will unfortunately get less work than others because of the amount of RAM required for rendering the more elaborate scenes.
As to the other issue, we've unfortunately been unable to determine the problem's source so far. We will get to it eventually and I promise that everybody will get compensated for this in their credits. I just want to find the actual issue before starting to feed medicine to the system, if you get the point. :)
Hey aurimas,
An idea that we would investigate is that probably those work units were returned after the validation of at least three of them, making the part valid. We need to find out what really happens but this could be one possibility.
We would investigate more ASAP
Regards
I got the exact same problem