Skip to content

Instantly share code, notes, and snippets.

Embed
What would you like to do?
Fixes for GLIBC errors when installing tensorflow or pytorch on older Red Hat or CentOS cluster environments

Goal

Install working tensorflow or pytorch via standard conda environment workflow.

Basic Setup : Install pytorch in a fresh conda environment

The recommended conda-based install process works smoothly:

$ # Create a fresh environment
$ conda create --name py37_torch python=3.7 --yes

$ # Activate new environment
$ source activate py37_torch

$ # Install tensorflow
$ conda install tensorflow --yes

$ # Install pytorch 
$ conda install pytorch-cpu torchvision-cpu -c pytorch --yes

Roadblock

The gotcha is that when we try to then use the package we just installed, we get an GLIBC error like this:

$ python -c "import torch"
ImportError: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by .../site-packages/torch/lib/libshm.so)

Badness! Clearly, the current computing system doesn't have a recent-enough GLIBC. However, if this is a cluster computing system, you often don't have root access and can't easily upgrade the GLIBC.

Step 1: Install recent copies of glibc and libc++ in userspace

Credit: StackOverflow answer by Theo T.

Step 1a: (NEW FOR PYTHON 3.7) Download and unpack some pre-compiled GLIBC shared libraries

This is for Python 3.7 (works for 3.6 too!) (See an older list for Python 2.7 at bottom of this doc).

$ # Make a folder within the environment to hold useful things
$ mkdir -p /path/to/conda/envs/py37_torch1.0/custom_libs/
$ cd /path/to/conda/envs/py37_torch1.0/custom_libs/

$ # Get libc files (URL verified by MCH on 2019/08/21)
$ wget http://mirrors.kernel.org/ubuntu/pool/main/g/glibc/libc6_2.23-0ubuntu10_amd64.deb
$ wget http://mirrors.kernel.org/ubuntu/pool/main/g/glibc/libc6-dev_2.23-0ubuntu10_amd64.deb

$ # Unpack files into current directory (will create usr/ and lib/ and lib64/ folders)
$ ar p libc6_2.23-0ubuntu10_amd64.deb data.tar.xz | tar xvJ
$ ar p libc6-dev_2.23-0ubuntu10_amd64.deb data.tar.xz | tar xvJ

What have we accomplished? You should have some new folders in your current directory, labeld usr/ and lib/ and lib64/.

We can verify that before, we had an OLD libc, and now we have a shiny new one!

Check the OLD location of libc.so.6

$ strings /lib/libc.so.6 | grep GLIBC_2. | tail -n3
GLIBC_2.10
GLIBC_2.11
GLIBC_2.12

NEW version of libc.so.6 in working directory

$ strings lib/x86_64-linux-gnu/libc.so.6 | grep GLIBC_2 | tail -n3
GLIBC_2.18
GLIBC_2.22
GLIBC_2.23

Step 1b: Download and unpack some pre-compiled LIBSTDC++ shared libraries

# Get libstdc++ (URL verified by MCH on 2019/02/18)
wget ftp://195.220.108.108/linux/mageia/distrib/4/x86_64/media/core/updates/libstdc++6-4.8.2-3.2.mga4.x86_64.rpm

# Alternative URL:
# wget http://ftp.riken.jp/Linux/scientific/6.0/x86_64/os/Packages/libstdc++-4.4.4-13.el6.x86_64.rpm

# Unpack into current directory (will add content to lib/ and lib64/ folders)
rpm2cpio libstdc++6-4.8.2-3.2.mga4.x86_64.rpm | cpio -idmv

Step 2: Use patchelf to make your python install use these userspace libraries instead of the system defaults

Credit: Stackoverview answer by Evalds Urtans

Step 2a: Install patchelf into current conda env

# Be sure correct environment is active
$ source activate py37_torch

# Install patchelf
(py37_torch) $ conda install patchelf -c conda-forge --yes

Step 2b: Use attached script to alter the conda env's python executable to use the custom GLIBC libraries

(py37_torch) $ bash rewrite_python_exe_glibc_with_patchelf.sh

-- DEPRECATED --

Step 1a: (OLD FOR PYTHON 2.7) Download and unpack some pre-compiled GLIBC shared libraries

$ # Make a folder within the environment to hold useful things
$ mkdir -p /path/to/conda/envs/py27_torch1.0/custom_libs/
$ cd /path/to/conda/envs/py27_torch1.0/custom_libs/

$ # Get libc files (URL verified by MCH on 2019/02/18)
$ wget https://launchpadlibrarian.net/137699828/libc6_2.17-0ubuntu5_amd64.deb
$ wget https://launchpadlibrarian.net/137699829/libc6-dev_2.17-0ubuntu5_amd64.deb

$ # Unpack files into current directory (will create usr/ and lib/ and lib64/ folders)
$ ar p libc6_2.17-0ubuntu5_amd64.deb data.tar.gz | tar zx
$ ar p libc6-dev_2.17-0ubuntu5_amd64.deb data.tar.gz | tar zx
#!/usr/env bash
# TODO edit this line to specify location of new glibc
export GLIBC_PATH=/cluster/tufts/hugheslab/miniconda2/envs/ape/custom_libs/
export GLIBC_LD_PATH=$GLIBC_PATH/lib/x86_64-linux-gnu/ld-2.23.so
if [[ ! -f $GLIBC_LD_PATH ]]; then
echo "ERROR: Provided GLIBC_LD_PATH not valid"
exit
fi
echo "OVERWRITING PYTHON EXECUTABLE:"
python_exe=`which python`
echo $python_exe
IS_CONDA_ENV=`python -c "print('$python_exe'.count('/envs/') > 0)"`
echo "IS_CONDA_ENV: $IS_CONDA_ENV"
if [[ $IS_CONDA_ENV -ne 'True' ]]; then
echo "ERROR: Current python executable not in conda env. Will not alter to avoid problems."
exit
fi
CONDA_ENV_LIB=`python -c "print('$python_exe'.replace('/bin/python', '/lib'))"`
echo "CREATING BACKUP PYTHON"
python_tmp_exe=`python -c "print('$python_exe'.replace('python', 'python_backup'))"`
cp $python_exe $python_tmp_exe
echo "$python_tmp_exe"
rpath=$GLIBC_PATH/lib/x86_64-linux-gnu:$CONDA_ENV_LIB:/usr/lib64:/lib64:/lib
echo "CALLING PATCHELF on 'python' binary"
patchelf --set-interpreter $GLIBC_LD_PATH --set-rpath $rpath $python_exe
echo "DONE! patchelf complete"
@geblanco

This comment has been minimized.

Copy link

@geblanco geblanco commented Apr 3, 2020

This was extremely useful, made us able to get up and running torch with an old centOS 6. However, we found various issues:

  • Your instructions are mixed with StackOverflow answer by Theo T. in that you download libc_2.23, but do the ar p command for lib_2.17, it's just a matter of changing the command to reflect the correct libs (happens with libcstd++ too). The same goes for ld_2.17 in the rewrite script.

When installing 2.23, for some reason, patching the python binary caused pip & network related stuff to break, yielding an environment unable to install more packages, but with the already installed ones running. We found the following solutions:

  • Reinstalling python and pip (conda install -c anaconda pip python==3.7), installing desired packages (pip install ...), and patching elf again, ./rewrite_python... step.
  • Installing libc_2.17 instead of libc_2.23, as in StackOverflow answer by Theo T., that, for some reason, doesn't break network, resulting in a functional, but patched environment.

Anyway, thank you very much for this gist, it has saved my day!

@michaelchughes

This comment has been minimized.

Copy link
Owner Author

@michaelchughes michaelchughes commented May 23, 2020

This has been updated to fix the version inconsistencies mentioned above.

Yes I agree this might break some pip stuff as is, so you can revert to the backup python as needed, then repatch.

@machineyearning

This comment has been minimized.

Copy link

@machineyearning machineyearning commented Jun 10, 2020

Hi, I'm getting this error when running the bash script:

CALLING PATCHELF on 'python' binary
patchelf: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.20' not found (required by patchelf)
patchelf: /usr/lib64/libstdc++.so.6: version `CXXABI_1.3.9' not found (required by patchelf)
patchelf: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.21' not found (required by patchelf)
DONE! patchelf complete

The output I get from strings lib/x86_64-linux-gnu/libc.so.6 | grep GLIBC_2 | tail -n3 looks correct:

GLIBC_2.18
GLIBC_2.22
GLIBC_2.23

Any idea how to fix this?

@ManifoldFR

This comment has been minimized.

Copy link

@ManifoldFR ManifoldFR commented Oct 30, 2020

I think the links for libc6 packages need to be updated to http://security.ubuntu.com/ubuntu/pool/main/g/glibc/libc6_2.23-0ubuntu11.2_amd64.deb

@XStargate

This comment has been minimized.

Copy link

@XStargate XStargate commented Dec 19, 2020

Hi, I'm getting this error when running the bash script:

CALLING PATCHELF on 'python' binary
patchelf: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.20' not found (required by patchelf)
patchelf: /usr/lib64/libstdc++.so.6: version `CXXABI_1.3.9' not found (required by patchelf)
patchelf: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.21' not found (required by patchelf)
DONE! patchelf complete

The output I get from strings lib/x86_64-linux-gnu/libc.so.6 | grep GLIBC_2 | tail -n3 looks correct:

GLIBC_2.18
GLIBC_2.22
GLIBC_2.23

Any idea how to fix this?

I also met the same problem. Did you fix it?

@Colhodm

This comment has been minimized.

Copy link

@Colhodm Colhodm commented Jan 14, 2021

Bump on the above, I'm having the same issue. Was anybody able to resolve this?

@hzcheney

This comment has been minimized.

Copy link

@hzcheney hzcheney commented Apr 9, 2021

Thanks a lot ! It really helps me !!!
But the link file mentioned above doesn't exist, so I googled it and find some alternative files that may be useful to you guys!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment