Dear Johannes Problem solved.
I discovered the fenics.stable profile for the bash in my home directory! Now i can test this. On Wed, Mar 11, 2015 at 4:35 PM, moritz braun <[email protected]> wrote: > Dear Johannes > > Thanks a lot! > > My build seems to have been successful, since the .hastdist/tmp directory > is now empty > > However I have no clue how to use the FEniCS build that it is sitting in > the bld directory. > There seems to be a step missing that I did not see, since I left the > office before it was finished!! > > Tomorrow back at the office I will collect the names of the rpms that are > required to build fenics > and send them to you to include in the "HOW TO BUILD FENICS on > DISTRIBUTION XYZ "list > > > Thanks a lot for your hard work. > > > regards > > > Moritz > > On Wed, Mar 11, 2015 at 12:21 PM, Johannes Ring <[email protected]> wrote: > >> On Wed, Mar 11, 2015 at 11:14 AM, moritz braun <[email protected]> >> wrote: >> > Thanks again! >> > >> > However, why could he not find pthread?? >> >> I think pthread was found. See this in build.log: >> >> 2015/03/10 10:48:28 - INFO: [package:run_job] -- Looking for include >> file pthread.h >> 2015/03/10 10:48:28 - INFO: [package:run_job] -- Looking for include >> file pthread.h - found >> >> and >> >> 2015/03/10 10:48:29 - INFO: [package:run_job] -- Looking for >> pthread_create >> 2015/03/10 10:48:29 - INFO: [package:run_job] -- Looking for >> pthread_create - not found >> 2015/03/10 10:48:29 - INFO: [package:run_job] -- Looking for >> pthread_create in pthreads >> 2015/03/10 10:48:29 - INFO: [package:run_job] -- Looking for >> pthread_create in pthreads - not found >> 2015/03/10 10:48:29 - INFO: [package:run_job] -- Looking for >> pthread_create in pthread >> 2015/03/10 10:48:29 - INFO: [package:run_job] -- Looking for >> pthread_create in pthread - found >> 2015/03/10 10:48:29 - INFO: [package:run_job] -- Found Threads: TRUE >> >> Johannes >> >> > regards >> > >> > Moritz >> > >> > On Wed, Mar 11, 2015 at 11:59 AM, Johannes Ring <[email protected]> >> wrote: >> >> >> >> On Tue, Mar 10, 2015 at 4:27 PM, moritz braun <[email protected]> >> >> wrote: >> >> > Dear Johannes >> >> > >> >> > Now that I have arrived at home and had dinner here is >> >> > the build.log attached! >> >> >> >> This is the error: >> >> >> >> 2015/03/10 10:57:19 - INFO: [package:run_job] >> >> >> >> >> /home/braunm/.hashdist/tmp/vtk-7y5tduwxoiyh/Rendering/vtkXRenderWindowInteractor.h:37:74: >> >> fatal error: X11/StringDefs.h: No such file or directory >> >> 2015/03/10 10:57:19 - INFO: [package:run_job] #include >> >> <X11/StringDefs.h> // Needed for X types in the public interface >> >> 2015/03/10 10:57:19 - INFO: [package:run_job] >> >> ^ >> >> 2015/03/10 10:57:19 - INFO: [package:run_job] compilation terminated. >> >> >> >> The X11 headers must be in place in order to build VTK on Linux [1]. >> >> On Ubuntu this file is available in the libxt-dev package but I'm not >> >> sure which package you must install on Redhat. >> >> >> >> > By the way, it would be awesome, if you could add a way to indicate >> via >> >> > an environment variable to fenics-install.sh that vtk should not be >> >> > build , >> >> > because on servers >> >> > for number crunching it is not so essential. >> >> >> >> You can use one of the profiles from the fenics-developer-tools >> >> repository [2] and remove vtk from it. Then run "fenics-install.sh >> >> <yourprofile.yaml>" to build without VTK. >> >> >> >> [1] https://github.com/hashdist/hashstack/issues/634 >> >> [2] >> >> >> https://bitbucket.org/fenics-project/fenics-developer-tools/src/93b0d179b3e7f97eb224a52f8af5cf6ee50e71cc/install/profiles/?at=master >> >> >> >> Johannes >> >> >> >> > I remember from googling this >> >> > afternoon >> >> > that with dorsal this was possible. >> >> > >> >> > regards >> >> > >> >> > Moritz >> >> > >> >> > On Tue, Mar 10, 2015 at 3:19 PM, Johannes Ring <[email protected]> >> >> > wrote: >> >> >> >> >> >> [Please keep fenics-support in Cc] >> >> >> >> >> >> On Tue, Mar 10, 2015 at 1:17 PM, moritz braun < >> [email protected]> >> >> >> wrote: >> >> >> > Hi Johannes >> >> >> > >> >> >> > >> >> >> > I have attached the CMakeError.log from the directory >> >> >> > .hashdist/tmp/vtk-7y5tduwxoiyh/_build/CMakeFiles >> >> >> > >> >> >> > grepping on the file yields the following undefined references! >> >> >> > >> >> >> > CheckSymbolExists.c:(.text+0x1b): undefined reference to >> >> >> > `pthread_create' >> >> >> > CheckFunctionExists.c:(.text+0x15): undefined reference to >> >> >> > `setsysinfo' >> >> >> > CheckFunctionExists.c:(.text+0x15): undefined reference to >> >> >> > `sigsetjmp' >> >> >> > CheckFunctionExists.c:(.text+0x15): undefined reference to >> `_isinf' >> >> >> > >> >> >> > >> >> >> > Maybe the file you want is a different one! >> >> >> >> >> >> Yes, I want the build.log from hashdist. When the script fails you >> >> >> should see something like this: >> >> >> >> >> >> [vtk] Building vtk/dxyszop5gfft, follow log with: >> >> >> [vtk] tail -f >> /home/johannr/.hashdist/tmp/vtk-dxyszop5gfft/build.log >> >> >> [vtk|ERROR] Command '[u'/bin/bash', '_hashdist/build.sh']' returned >> >> >> non-zero exit status 1 >> >> >> [vtk|ERROR] command failed (code=1); raising >> >> >> Command exited with non-zero status 127 >> >> >> >> >> >> The file I'm looking for will be available in >> >> >> ~/.hashdist/tmp/vtk-<hash>/build.log. >> >> >> >> >> >> Johannes >> >> >> >> >> >> > I cannot be only one having this problem. >> >> >> > >> >> >> > vtk is a pain in a neck to get right. >> >> >> > >> >> >> > Thanks a lot! >> >> >> > >> >> >> > regards >> >> >> > >> >> >> > Moritz >> >> >> > >> >> >> > >> >> >> > On Tue, Mar 10, 2015 at 1:46 PM, Johannes Ring <[email protected] >> > >> >> >> > wrote: >> >> >> >> >> >> >> >> Hi Moritz, >> >> >> >> >> >> >> >> Can you post the build.log from vtk somewhere or attach it to >> this >> >> >> >> thread? >> >> >> >> >> >> >> >> Johannes >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > -- >> >> >> > Prof M Braun Tel.:27-12-4298006/8027 >> >> >> > Physics Department Fax.: 27-12-4293643 >> >> >> > University of South Africa (UNISA) >> >> >> > [email protected] >> >> >> > P.O. Box 392 >> >> >> > 0003 >> >> >> > UNISA >> >> >> > South Africa >> >> >> > http://moritz-braun.blogspot.com >> >> > >> >> > >> >> > >> >> > >> >> > -- >> >> > Prof M Braun Tel.:27-12-4298006/8027 >> >> > Physics Department Fax.: 27-12-4293643 >> >> > University of South Africa (UNISA) >> >> > [email protected] >> >> > P.O. Box 392 >> >> > 0003 >> >> > UNISA >> >> > South Africa >> >> > http://moritz-braun.blogspot.com >> > >> > >> > >> > >> > -- >> > Prof M Braun Tel.:27-12-4298006/8027 >> > Physics Department Fax.: 27-12-4293643 >> > University of South Africa (UNISA) >> > [email protected] >> > P.O. Box 392 >> > 0003 >> > UNISA >> > South Africa >> > http://moritz-braun.blogspot.com >> > > > > -- > Prof M Braun Tel.:27-12-4298006/8027 > Physics Department Fax.: 27-12-4293643 > University of South Africa (UNISA) > [email protected] > P.O. Box 392 > 0003 > UNISA > South Africa > http://moritz-braun.blogspot.com > -- Prof M Braun Tel.:27-12-4298006/8027 Physics Department Fax.: 27-12-4293643 University of South Africa (UNISA) [email protected] P.O. Box 392 0003 UNISA South Africa http://moritz-braun.blogspot.com
_______________________________________________ fenics-support mailing list [email protected] http://fenicsproject.org/mailman/listinfo/fenics-support
