Output file absent
Hello,
this project is great, but I have a lot of errors. It worked some time, but now every work unit stops after 7 to 9 seconds.
This is a error from the account :
<core_client_version>6.10.58</core_client_version> <![CDATA[ <message> process exited with code 247 (0xf7, -9) </message> <stderr_txt> ---------------------------- GLUE ---------------------------- Detected input arg: argv[0] is ../../projects/www.renderfarm.fi/blender_4.83_x86_64-pc-linux-gnu Detected input arg: argv[1] is blender Detected input arg: argv[2] is -noaudio -y -b Detected input arg: argv[3] is -F PNG -t 1 -f 1772 0.0 0.0 1.0 1.0 $Id: BOINCHandler.cpp,v 1.1 2008/08/16 12:56:40 jbk Exp $ boinc_init_diagnostics() completed boinc_init_options() completed Basename: ../../projects/www.renderfarm.fi/blender_4.83_x86_64-pc-linux-gnu --> blender_4.83_x86_64-pc-linux-gnu Mapping logical files to physical destinations: in => ../../projects/www.renderfarm.fi/file531509 out.zip => ../../projects/www.renderfarm.fi/ses0000001468frm0000001772prt00001_5_0 blender_4.83_x86_64-pc-linux-gnu.zip => ../../projects/www.renderfarm.fi/blender_4.83_x86_64-pc-linux-gnu.zip ./linux_zip => ../../projects/www.renderfarm.fi/linux_zip ./linux_unzip => ../../projects/www.renderfarm.fi/linux_unzip Decompressing executable archive... Running ../../projects/www.renderfarm.fi/linux_unzip -o ../../projects/www.renderfarm.fi/blender_4.83_x86_64-pc-linux-gnu.zip runcmd(../../projects/www.renderfarm.fi/linux_unzip -o ../../projects/www.renderfarm.fi/blender_4.83_x86_64-pc-linux-gnu.zip) exitted with exit code 0 CHMODing executable... 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 -noaudio -y -b ../../projects/www.renderfarm.fi/file531509 -F PNG -t 1 -f 1772 0.0 0.0 1.0 1.0 Child created. Worker thread started Worker thread monitor almost up. Worker thread monitor up. |[-->] Info: |child <8517> to stdout |0 - ./blender |1 - -noaudio |2 - -y |3 - -b Application reports 'Booted' |found bundled python: /var/lib/boinc/slots/2/2.55/python |[-->] Info |Unhandled segmentation violation --------------------------- Exception caught: Worker application apparently died prematurely Status: -9 --------------------------- </stderr_txt> ]]>
Don't know what this mean.
I use ArchLinux, 4 cores, 8GB of RAM, 8 projects.
But even if renderfarm is alone, only 1 process, alone to execute, it makes error.
Even restarting (reset) the project don't help.
Any help would be appreciated since it will help you.
Offline
Thanks for reply,
sorry I posted twice and I'm not familiar with this forum,
After installing libjpeg6,
I donwloaded and executed as you said and got this error :
./blender: error while loading shared libraries: libgettextlib-0.17.so: cannot open shared object file: No such file or directory
Nothing more, blender didn't start.
Hope that helps.
P.S. : This are the packages I have installed.
extra/libjpeg 8.3.0-1 [installed]
Library of JPEG support functions
multilib/lib32-libjpeg 8.0.2-3 [installed]
Library of JPEG support functions (32-bit)
custom/lib32-libjpeg6 6b-2 (lib32) [installed]
An older version of a library of JPEG support functions.
custom/libjpeg6 6b-9 [installed]
Library of JPEG support functions
But for libgettextlib, I got nothing and it's not on the repos of my distro. Also, gettext is installed
well, on my system gettext is on version 0.18 so i had to copy libgettextlib-0.17.so manualy to make it work.
also do a ldd /where_the_file_is/blender_executable_name and it will tell u all the dependecies it uses and the ones that are missing, so u know what it must get installed.
well, on my system gettext is on version 0.18 so i had to copy libgettextlib-0.17.so manualy to make it work.
also do a ldd /where_the_file_is/blender_executable_name and it will tell u all the dependecies it uses and the ones that are missing, so u know what it must get installed.
Thanks for that hints,
I found that gettext is at version 0.18, I have libgettextlib-0.18.1.so . Also I installed libtiff4 wich was missed.
And python-pyffmpeg to get libavutil.
BUT : libgettextlib is 0.18 and should be 0.17 like you said and i have libavutil.so.50 instead of libavutil.so.49.
How do you copy a older version? From where do you get it?
I don't see others libraries missing.
first try to link the newer version as the older one to see if it works ( ln -s /path_to_you;re_libs/libgettextlib-0.18.so /path..../libget.....0.17.so)
if it doesn;t work, search with google the versions u need, and install them.
Symlinking semmed to work with ldd.
But when I started blender, I got :
./blender: /usr/lib/libavutil.so.49: version `LIBAVUTIL_49' not found (required by ./blender)
./blender: /usr/lib/libjpeg.so.62: no version information available (required by /usr/lib/libtiff.so.4)
So it seems to work with libgettextlib, but not libavutil, and another "appeared".
Probably the best is that I wait an update of blender comes out using the newer version.
I have had 3 tasks error out today on a 64bit Ubuntu host with this:
Exception caught: Worker application apparently died prematurely Status: -9
All tasks 'died' while awaiting to resume computation after being superseded to allow another projects tasks to complete. Is there a known problem with tasks failing after being suspended? Boinc manager stating output file absent. (leave apps in memory is on)
|
|228745756
|Fra:376 Mem:199.97M (18.18M, peak 287.99M) | Scene.001, Part 26-34
---------------------------
Exception caught: Worker application apparently died prematurely
Status: -9
---------------------------
That is one of the hardest nuts to figure out. Simply it means that for a reason or another, the render job crashed. This can be related to the computer running out of memory or Blender becoming unresponsive (although the error would be different if Blender didn't respond). Suspending could have had something to do with this, although it should work. Your computer seems to be returning results fine for most of the time though.
Yeah,it's a strange one as they 'died' before they had a chance to resume (ie. they weren't running at the time of crash). All tasks complete successfully if run without stop. Can't see it being memory related as I'm using less than 1of 4gb. They crashed after different time intervals too so could it be that something fails to stop and then crashes when it reaches it's supposed completion? I say this as the crashes occurred after 38 minutes for the task at Part 23-34, 28 mins at Part 26-34 and the last after 14 mins at Part 31-34. Can't be a coincidence could it?