[Mesa-users] ./binary: /lib64/libm.so.6: version `GLIBC_2.29' not found (required by ./binary)

Fri Sep 6 07:11:05 UTC 2024

Hi Rich,

I am sending the output you requested:

(base) [wcss] amiszuda at r31c03b04 ~/.../V1216_Sco/cluster_template > which gfortran
/lustre/pd01/hpc-walczakp-1692967221/amiszuda/Programs/mesasdk/bin/gfortran
(base) [wcss] amiszuda at r31c03b04 ~/.../V1216_Sco/cluster_template > gfortran -v
Using built-in specs.
COLLECT_GCC=/lustre/pd01/hpc-walczakp-1692967221/amiszuda/Programs/mesasdk/bin/gfortran.exec
COLLECT_LTO_WRAPPER=/lustre/pd01/hpc-walczakp-1692967221/amiszuda/Programs/mesasdk/bin/../libexec/gcc/x86_64-pc-linux-gnu/13.1.0/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: /home/user/sdk2-tmp/build/gcc/configure CC= CXX= --build=x86_64-pc-linux-gnu --host=x86_64-pc-linux-gnu --target=x86_64-pc-linux-gnu --prefix=/home/user/sdk2-tmp/mesasdk --with-gmp=/home/user/sdk2-tmp/mesasdk --with-mpfr=/home/user/sdk2-tmp/mesasdk --with-mpc=/home/user/sdk2-tmp/mesasdk --enable-languages=c,c++,fortran --disable-multilib --disable-nls --disable-libsanitizer --enable-clocale=generic
Thread model: posix
Supported LTO compression algorithms: zlib
gcc version 13.1.0 (GCC) 

It should be fixed now… I got a message from the admin that said there has been some incidental overwriting of the paths, so probably that caused use of the wrong compiler during my previous ./mk

MESA is running again, many thanks Rich and Warrick for your help!

Cheers,
Amadeusz

> Wiadomość napisana przez RICHARD H D TOWNSEND <townsend at astro.wisc.edu> w dniu 05.09.2024, o godz. 21:52:
> 
> Hi Amadeusz —
> 
> From what I can tell from the binary file you sent me, none of the library search paths that the MESA SDK usually inserts into executables, are present in the binary file.
> 
> My suspicion is that, somehow, a different (non-MESA SDK) compiler got used when you ran the ./mk command to generate the binary. To see whether that’s the case, can you run the following commands inside your work directory:
> 
> ./clean
> ./mk
> which gfortran
> gfortran -v
> 
> (and post the output)
> 
> Many thanks,
> 
> Rich
> 
>> On Sep 2, 2024, at 11:49 PM, Amadeusz Miszuda <miszuda.amadeusz.astronomy at gmail.com> wrote:
>> 
>> Hi Rich,
>> 
>> The LD_LIBRARY_PATH is not set, it produces no output.
>> 
>> Cheers,
>> Amadeusz
>> 
>>> Wiadomość napisana przez RICHARD H D TOWNSEND <townsend at astro.wisc.edu> w dniu 03.09.2024, o godz. 01:37:
>>> 
>>> Thanks!
>>> 
>>> From what I can tell, the path to where ./binary looks for libraries is being incorrectly overridden. This could be because the LD_LIBRARY_PATH environment variable is  set (when it shouldn’t be). Can you post the output from
>>> 
>>> echo $LD_LIBRARY_PATH
>>> 
>>> (there might not be any output, FYI)
>>> 
>>> cheers,
>>> 
>>> Rich
>>> 
>>>> On Sep 2, 2024, at 2:16 PM, Amadeusz Miszuda <miszuda.amadeusz.astronomy at gmail.com> wrote:
>>>> 
>>>> linux-vdso.so.1 (0x00007fff5e7af000)
>>>> libcrmath.so => not found
>>>> libcrlibm.so.0 => not found
>>>> liblapack95.so => not found
>>>> libtmg.so => not found
>>>> liblapack.so => not found
>>>> libblas.so => not found
>>>> libhdf5_fortran.so.310 => not found
>>>> libhdf5.so.310 => not found
>>>> libz.so.1 => /usr/local/easybuild/software/zlib/1.2.12-GCCcore-11.3.0/lib/libz.so.1 (0x00001528d9e7c000)
>>>> libpgplot.so => not found
>>>> libgfortran.so.5 => /usr/local/easybuild/software/GCCcore/11.3.0/lib64/libgfortran.so.5 (0x00001528d99c7000)
>>>> libm.so.6 => /lib64/libm.so.6 (0x00001528d9645000)
>>>> libmvec.so.1 => /lib64/libmvec.so.1 (0x00001528d941a000)
>>>> libgomp.so.1 => /usr/local/easybuild/software/GCCcore/11.3.0/lib64/libgomp.so.1 (0x00001528d9e34000)
>>>> libgcc_s.so.1 => /usr/local/easybuild/software/GCCcore/11.3.0/lib64/libgcc_s.so.1 (0x00001528d9e1a000)
>>>> libquadmath.so.0 => /usr/local/easybuild/software/GCCcore/11.3.0/lib64/libquadmath.so.0 (0x00001528d9dd1000)
>>>> libpthread.so.0 => /lib64/libpthread.so.0 (0x00001528d91fa000)
>>>> libc.so.6 => /lib64/libc.so.6 (0x00001528d8e24000)
>>>> /lib64/ld-linux-x86-64.so.2 (0x00001528d9c76000)
>>>> libdl.so.2 => /lib64/libdl.so.2 (0x00001528d8c20000)
>>>> 
>>>>> Wiadomość napisana przez RICHARD H D TOWNSEND <townsend at astro.wisc.edu> w dniu 02.09.2024, o godz. 21:05:
>>>>> 
>>>>> OK, looks like glibc is sufficiently recent. What does the command
>>>>> 
>>>>> ldd ./binary
>>>>> 
>>>>> produce?
>>>>> 
>>>>>> On Sep 2, 2024, at 2:00 PM, Amadeusz Miszuda <miszuda.amadeusz.astronomy at gmail.com> wrote:
>>>>>> 
>>>>>> Hi Warrick, hi Rich,
>>>>>> 
>>>>>>> Can you post the output from running the command:
>>>>>>> 
>>>>>>> /lib64/libc.so.6
>>>>>> 
>>>>>> GNU C Library (GNU libc) stable release version 2.28.
>>>>>> Copyright (C) 2018 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.
>>>>>> Compiled by GNU CC version 8.5.0 20210514 (Red Hat 8.5.0-22).
>>>>>> libc ABIs: UNIQUE IFUNC ABSOLUTE
>>>>>> For bug reporting instructions, please see:
>>>>>> <https://urldefense.com/v3/__http://www.gnu.org/software/libc/bugs.html__;!!Mak6IKo!JSzNM71QrGKLQcJJEnUHLw6TQ2orNpHWKGxcreA8crwF9wCKl9eQwZ8oiKN4NtXlTh_0NHSn6d6MBZWdEDetA0ondYenbGE$ >
>>>>>> 
>>>>>>> Just to rule out a few more simple things (which I doubt are the issue), can I check that you've rebuilt your work folder too, i.e. run `./clean` and `./mk` before `./rn` again?
>>>>>> 
>>>>>> I have now, apparently I forgot to do that beforehand (sorry!). Now it complains about:
>>>>>> ./binary: error while loading shared libraries: libcrmath.so: cannot open shared object file: No such file or directory
>>>>>> 
>>>>>> Cheers,
>>>>>> Amadeusz
>>>>>> 
>>>>>>> Wiadomość napisana przez Warrick Ball <W.H.Ball at bham.ac.uk> w dniu 02.09.2024, o godz. 20:46:
>>>>>>> 
>>>>>>> Hi Amadeusz,
>>>>>>> 
>>>>>>> Just to rule out a few more simple things (which I doubt are the issue), can I check that you've rebuilt your work folder too, i.e. run `./clean` and `./mk` before `./rn` again?
>>>>>>> 
>>>>>>> Then, following this advice about glibc compatibility on the MESA SDK website:
>>>>>>> 
>>>>>>> http://user.astro.wisc.edu/~townsend/static.php?ref=mesasdk#Compatibility
>>>>>>> 
>>>>>>> "run the command /lib/libc.so.6 (or, possibly, /lib64/libc.so.6) and examine the first line of the output for the version number".
>>>>>>> 
>>>>>>> It needs to be >= 2.5, though I'd be surprised if this wasn't the case.
>>>>>>> 
>>>>>>> Regards,
>>>>>>> Warrick
>>>>>>> 
>>>>>>> ___________
>>>>>>> 
>>>>>>> Warrick Ball
>>>>>>> Senior Research Software Engineer
>>>>>>> University of Birmingham
>>>>>>> W.H.Ball at bham.ac.uk
>>>>>>> 
>>>>>> 
>>>>>>> Wiadomość napisana przez RICHARD H D TOWNSEND <townsend at astro.wisc.edu> w dniu 02.09.2024, o godz. 20:42:
>>>>>>> 
>>>>>>> Hi Amadeusz —
>>>>>>> 
>>>>>>> Can you post the output from running the command:
>>>>>>> 
>>>>>>> /lib64/libc.so.6
>>>>>>> 
>>>>>>> This will allow us to see what version of the GNU C library your cluster is using.
>>>>>>> 
>>>>>>> cheers,
>>>>>>> 
>>>>>>> Rich
>>>>>> 
>>>>>>> On Mon, 2 Sep 2024, Amadeusz Miszuda wrote:
>>>>>>> 
>>>>>>>> CAUTION: This email originated from outside the organisation. Do not click links or open attachments unless you recognise the sender and know the content is safe.
>>>>>>>> Hi Warrick,
>>>>>>>> I have, but just to be double sure I reinstalled both SDK and MESA once again. The new build.log is attached below.
>>>>>>>> Cheers,
>>>>>>>> Amadeusz
>>>>>>>> --------------------------------------------------------------
>>>>>>>> Amadeusz Miszuda, PhD
>>>>>>>> Postdoctoral researcher
>>>>>>>> Nicolaus Copernicus Astronomical Center 
>>>>>>>> Polish Academy of Sciences,
>>>>>>>> Warsaw, Poland
>>>>>>>> 
>>>>>>>>  Wiadomość napisana przez Warrick Ball <W.H.Ball at bham.ac.uk> w dniu 02.09.2024, o godz. 14:03:
>>>>>>>> Hi Amadeusz,
>>>>>>>> I probably won't be able to fully resolve this but just to check, have you reinstalled MESA since the cluster maintenance happened?  I.e., have you more recently loaded the SDK and run `./touch` then `./install` in your MESA directory?
>>>>>>>> Regards,
>>>>>>>> Warrick
>>>>>>>> ___________
>>>>>>>> Warrick Ball
>>>>>>>> Senior Research Software Engineer
>>>>>>>> University of Birmingham
>>>>>>>> W.H.Ball at bham.ac.uk
>>>>>>>> On Mon, 2 Sep 2024, Amadeusz Miszuda via Mesa-users wrote:
>>>>>>>> 
>>>>>>>>  CAUTION: This email originated from outside the organisation. Do not click links or open attachments unless you recognise the sender and know the content is safe.
>>>>>>>>  Hi all!
>>>>>>>>  Most recent maintenance on the cluster I am using for MESA computations results in some errors when executing the ./rn, to be more specific:
>>>>>>>>  ./binary: /lib64/libm.so.6: version `GLIBC_2.29' not found (required by ./binary)
>>>>>>>>  ./binary: /lib64/libc.so.6: version `GLIBC_2.33' not found (required by ./binary)
>>>>>>>>  ./binary: /lib64/libc.so.6: version `GLIBC_2.34' not found (required by ./binary)
>>>>>>>>  I install, both, MESA r23.05.1 and MESASDK 23.7.3 without issues.
>>>>>>>>  You can see my build.log file attached below.
>>>>>>>>  Thanks for any help!
>>>>>>>>  Amadeusz Miszuda
>>>>>>>>  --------------------------------------------------------------
>>>>>>>>  Amadeusz Miszuda, PhD
>>>>>>>>  Postdoctoral researcher
>>>>>>>>  Nicolaus Copernicus Astronomical Center 
>>>>>>>>  Polish Academy of Sciences,
>>>>>>>>  Warsaw, Poland
>>>>> 
>>>>> 
>>>>> --
>>>>> Rich Townsend • Professor of Astronomy
>>>>> Astronomy Department • University of Wisconsin-Madison
>>>>> Phone: 608-262-1752 • E-mail: rhtownsend at wisc.edu
>>> 
>>> 
>>> --
>>> Rich Townsend • Professor of Astronomy
>>> Astronomy Department • University of Wisconsin-Madison
>>> Phone: 608-262-1752 • E-mail: rhtownsend at wisc.edu
>> 
>> 
> 
> --
> Rich Townsend • Professor of Astronomy
> Astronomy Department • University of Wisconsin-Madison
> Phone: 608-262-1752 • E-mail: rhtownsend at wisc.edu
> 
> 
> 
> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20240906/d8cc17d1/attachment.htm>