hi, this is a known issue with Anaconda python in some Linux distro.
The issue here is that Anaconda provides its own `ld` program that's not
compatitble with your system.
can you please try to rename
"/opt/amber19/amber18/miniconda/compiler_compat/ld" to
"/opt/amber19/amber18/miniconda/compiler_compat/ld-old" to force
to use system `ld` program? Then "make install" again.
Related:
-
https://github.com/pytorch/pytorch/issues/16683#issuecomment-459982988
- google search
<
https://www.google.com/search?ei=wRzwXImBGYWp_QaN1YvgBQ&q=miniconda+unable+to+initialize+decompress+status+for+section+.debug_info&oq=miniconda+unable+to+initialize+decompress+status+for+section+.debug_info&gs_l=psy-ab.3...84340.85138..85554...0.0..0.276.1030.1j1j3......0....1..gws-wiz._mHh8RYRQnE>
Hai
On Thu, May 30, 2019 at 10:00 AM Hashemi, Mohtadin <
mohtadin.hashemi.unmc.edu> wrote:
> I have a very similar issue during configure of AmberTools19, but my issue
> during make is with parmed. I did not troubleshoot for very long since it
> was in a production environment, but a similar failure also happens on my
> personal computer. My system is Arch Linux (fully updated as of Tuesday)
> using GCC 8.3.0. Amber18 and AmberTools18 compile and run normally on the
> same system; I did compilation from scratch earlier this morning to see if
> something was wrong with the host system.
>
> These are the steps for configure and make:
> ./configure -noX11 gnu
> source amber.sh
> make
>
> Attached are the stdout during configure, stderr during configure, and
> output during make.
>
> The relevant section of the make output:
>
> make[2]: Entering directory '/opt/amber19/amber18/AmberTools/src'
> (cd parmed && PYTHONPATH=/opt/amber19/amber18/lib/python2.7/site-packages
> /opt/amber19/amber18/bin/amber.python setup.py install -f
> --prefix=/opt/amber19/amber18 > build.log)
> warning: no files found matching 'versioneer.py'
> cc1plus: warning: command line option '-Wstrict-prototypes' is valid for
> C/ObjC but not for C++
> cc1plus: warning: command line option '-Wstrict-prototypes' is valid for
> C/ObjC but not for C++
> src/readparm.cpp: In function 'ExitStatus readparm(const string&,
> std::vector<std::__cxx11::basic_string<char> >&, ParmDataMap&,
> ParmStringMap&, ParmStringMap&, ParmFormatMap&, std::__cxx11::string&)':
> src/readparm.cpp:143:28: warning: 'char* strncpy(char*, const char*,
> size_t)' specified bound 8 equals destination size [-Wstringop-truncation]
> strncpy(d.c, word.c_str(), MAX_HOLLERITH_SIZE);
> ~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> /opt/amber19/amber18/miniconda/compiler_compat/ld:
> build/temp.linux-x86_64-2.7/src/_rdparm.o: unable to initialize decompress
> status for section .debug_info
> /opt/amber19/amber18/miniconda/compiler_compat/ld:
> build/temp.linux-x86_64-2.7/src/_rdparm.o: unable to initialize decompress
> status for section .debug_info
> /opt/amber19/amber18/miniconda/compiler_compat/ld:
> build/temp.linux-x86_64-2.7/src/_rdparm.o: unable to initialize decompress
> status for section .debug_info
> /opt/amber19/amber18/miniconda/compiler_compat/ld:
> build/temp.linux-x86_64-2.7/src/_rdparm.o: unable to initialize decompress
> status for section .debug_info
> build/temp.linux-x86_64-2.7/src/_rdparm.o: file not recognized: file
> format not recognized
> collect2: error: ld returned 1 exit status
> error: command 'g++' failed with exit status 1
> make[2]: *** [Makefile:525: parmed] Error 1
> make[2]: Leaving directory '/opt/amber19/amber18/AmberTools/src'
> make[1]: *** [Makefile:56: serial] Error 2
> make[1]: Leaving directory '/opt/amber19/amber18/AmberTools/src'
> make: *** [Makefile:7: install] Error 2
>
>
> Using the system Python (both 2.7 and 3.7) during configure/compilation I
> am able to successfully compile a working AmberTools19/Amber18 suite.
> However, it is not a viable solution because the production environment
> does not allow change to the python installation.
>
> Best,
> Mohtadin
>
>
> -----Original Message-----
> From: Vlad Cojocaru <vlad.cojocaru.mpi-muenster.mpg.de>
> Date: Tue, 28 May 2019 14:37:07 +0200
>
> Sorry for not being completely clear ...
>
> Basically I just did a simple "./configure gnu" from $AMBERHOME and said
> "Yes" when configure was asking me whether to download and install
> miniconda ....
>
> See attached the output of "$AMBERHOME/AmberTools/src/configure_python "
>
> It does look to me like some python packages get compiled at this stage
> and the error pops up during these compilations (e.g. line 1299 in the
> attached file) ...
>
> This may or may not be different between Tools 18 and 19. I forgot I
> performed an OS update before compiling Tools 19 (did not try to recompile
> 18 after this update) ... So, the error might have something to do with
> that OS update ... And of course, with software that is not under your
> control ...
>
> But maybe this thread helps to track this down or at least to point people
> with the solution with using the system's python just in case anybody gets
> similar issues ...
>
> Vlad
>
>
> On 5/28/19 1:38 PM, David A Case wrote:
> > On Tue, May 28, 2019, Vlad Cojocaru wrote:
> >> The first time I got the error was during the first configure step
> >> (serial compilation). The error appears when building package
> >> subprocess32 during the compilation of miniconda (I run configure
> >> with minicoda compilation). Parts of the error are below.
> > Can you say exactly what arguments you gave to the configure script?
> > Did you say "yes" to downloading miniconda? As far as I know, there is
> > nothing that is supposed to be compiled at that step, and I've never
> > seen an error like this before. It doesn't sound like anything that
> > changed between AmberTools18 and AmberTools19.
> >
> > As you found out, you can use the --with-python flag: it can point to
> > your system python, or to the minconda python that was downloaded in
> > your AmberTools18 install. That way, you will have all the needed
> > modules.
> >
> > [Aside: I'm not sure whether everything in Amber actually works with
> > Python 3.7, althought the things you are most likely to use are.]
> >
> > If you wish, can you post the entire output of running
> > AMBERHOME/AmberTools/src/configure_python (this is what the configure
> > script is running)? This seems like a miniconda problem--still our
> > problem, since we ask users to rely on that--but one that involves
> > third party codes that we don't control.
> >
> > ...thanks....dac
> >
> >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
>
> --
> Vlad Cojocaru, PD (Habil.), Ph.D.
> -----------------------------------------------
> Project Group Leader
> Department of Cell and Developmental Biology Max Planck Institute for
> Molecular Biomedicine Röntgenstrasse 20, 48149 Münster, Germany
> -----------------------------------------------
> Tel: +49-251-70365-324; Fax: +49-251-70365-399
> Email: vlad.cojocaru[at]mpi-muenster.mpg.de
> http://www.mpi-muenster.mpg.de/43241/cojocaru
>
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
> text/x-log attachment: configure_python.log
>
> The information in this e-mail may be privileged and confidential,
> intended only for the use of the addressee(s) above. Any unauthorized use
> or disclosure of this information is prohibited. If you have received this
> e-mail by mistake, please delete it and immediately contact the sender.
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu May 30 2019 - 11:30:02 PDT
