On 02/23/2015 06:59 PM, Terje Kvernes wrote:
[ERROR] urllib failed to download (reason: [Errno -3] Temporary failure in name 
resolution): http://www.mpfr.org/mpfr-3.1.2/mpfr-3.1.2.tar.bz2
[CRITICAL] You may wish to check your Internet connection or the remote server
-------------------
This suggests that the website providing mpfr is having issues.  www.mpfr.org 
works fine here now, and this should only have been a temporary issue, however 
it would seem that http://mpfr.loria.fr/mpfr-3.1.2/mpfr-3.1.2.tar.gz works, 
while http://www.mpfr.org/mpfr-3.1.2/mpfr-3.1.2.tar.gz still fails.   The 
easiest solution would probably be to download 
http://mpfr.loria.fr/mpfr-3.1.2/mpfr-3.1.2.tar.gz manually and place it in the 
source_caches directory manually, something like this on one line:

wget  http://www.mpfr.org/mpfr-3.1.2/mpfr-3.1.2.tar.bz2 -O  $( printf "%s/%s"  
$( grep source_caches: -A1  ~/.hashdist/config.yaml | tail -1 | cut -f2 -d: ) 
phdt6yfpaefdbjocpkkvuhjnag5asw3s )

The checksum is gathered from hashstack/pkgs/mpfr.yaml and you could also 
modify the source from there if you wish.

Hi Terje

Ok, I found out what the problem is/was... I use a _*/etc/hosts*_-file to avoid too many advertisements/banners and although there was no entry for mpfr.org I thought it was really really strange that it worked on my windows-pc on the same network...

So I copied the old (almost empty) hosts-file into _*/etc/hosts*_ and then the problem was solved... But I'm not completely through this - now I'm back to a stage I think I've been in earlier:

---------
...
[=========================] 100.0% (0.2MB of 0.2MB) 0.069MB/s ETA 0s
[python-dateutil] Building python-dateutil/fpz7vv5p3nci, follow log with:
[python-dateutil] tail -f /home/laptop/.hashdist/tmp/python-dateutil-fpz7vv5p3nci/build.log Downloading https://pypi.python.org/packages/source/S/Sphinx/Sphinx-1.2b3.tar.gz...
[=========================] 100.0% (3.0MB of 3.0MB) 2.489MB/s ETA 0s
[sphinx] Building sphinx/fi6kdryuehzr, follow log with:
[sphinx]   tail -f /home/laptop/.hashdist/tmp/sphinx-fi6kdryuehzr/build.log
Downloading https://bitbucket.org/fenics-project/dolfin/downloads/dolfin-1.5.0.tar.gz...
[=========================] 100.0% (9.2MB of 9.2MB) 0.275MB/s ETA 0s  10s
[dolfin] Building dolfin/qosrpuzdfk6s, follow log with:
[dolfin]   tail -f /home/laptop/.hashdist/tmp/dolfin-qosrpuzdfk6s/build.log
[dolfin|_*ERROR*_] Command '[u'/bin/bash', '_hashdist/build.sh']' returned non-zero exit status 1
[dolfin|_*ERROR*_] command failed (code=1); raising
[laptop@worklaptop FEniCS]$ gcc --version
gcc (GCC) 4.8.2 20140120 (Red Hat 4.8.2-15)
Copyright (C) 2013 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
---------

Further investigation revealed this:
----
$ tail -f /home/laptop/.hashdist/tmp/dolfin-qosrpuzdfk6s/build.log
2015/02/24 13:29:15 - INFO: [package:run_job] -- Detecting CXX compile features - done 2015/02/24 13:29:15 - INFO: [package:run_job] CMake Error at CMakeLists.txt:25 (message): 2015/02/24 13:29:15 - INFO: [package:run_job] _*GCC version must be at least 4.6 (for sufficient C++11 support. You have*__*
*__*2015/02/24 13:29:15 - INFO: [package:run_job] version 4.4.7*_
2015/02/24 13:29:15 - INFO: [package:run_job]
2015/02/24 13:29:15 - INFO: [package:run_job]
2015/02/24 13:29:15 - INFO: [package:run_job] -- Configuring incomplete, errors occurred!
----

I've seen that before - described earlier in this thread - I thought I had taken care of this problem, but apparantly not. Let's have a look:

----
$ which gcc
/usr/local/bin/gcc

$ gcc --version
gcc (GCC) 4.8.2 20140120 (Red Hat 4.8.2-15)
Copyright (C) 2013 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
----

So it seems like it has something to do with cmake - which is bound to the "system default gcc":

$ locate gcc | grep -i bin
/opt/rh/devtoolset-2/root/usr/bin/gcc
/opt/rh/devtoolset-2/root/usr/bin/gcc-ar
/opt/rh/devtoolset-2/root/usr/bin/gcc-nm
/opt/rh/devtoolset-2/root/usr/bin/gcc-ranlib
/opt/rh/devtoolset-2/root/usr/bin/x86_64-redhat-linux-gcc
/usr/bin/gcc
/usr/bin/x86_64-redhat-linux-gcc
/usr/local/bin/gcc
/usr/local/bin/gcc-ar
/usr/local/bin/gcc-nm
/usr/local/bin/gcc-ranlib
/usr/local/bin/x86_64-redhat-linux-gcc


[laptop@worklaptop FEniCS]$ _*/usr/bin/gcc --version*_
gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-11)
Copyright (C) 2010 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.


/_*Conclusion - need help with cmake - should not use "system-installed gcc"...:*_/ I thought cmake would automatically use the first gcc version in the path environment variable... I'm not a cmake-expert at all, so I don't know how to tell cmake to NOT use /usr/bin/gcc but instead use the /usr/local/bin/gcc - I hope one of you could/would please advice me here...

I also hope this is the last step, before the whole thing can be compiled successfully... But I don't know... Please let me know what you think or what you could possibly recommend I try out.

I would be grateful for help into successfully compiling this...


_______________________________________________
fenics-support mailing list
[email protected]
http://fenicsproject.org/mailman/listinfo/fenics-support

Reply via email to