向已有conda环境安装fastspar时遇ResolvePackageNotFound错误求助

问题场景 #

尝试向qiime_1_9_0 conda环境安装fastspar，执行命令：

conda install -c bioconda fastspar

触发ResolvePackageNotFound错误：

Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.

ResolvePackageNotFound: 
  - defaults/linux-64::natsort==3.5.0=py27_0
  - defaults/linux-64::pyparsing==2.0.3=py27_0
  - defaults/linux-64::qt==4.8.7=3
  - defaults/linux-64::mkl==11.3.3=0
  - defaults/linux-64::mysql-python==1.2.5=py27_0
  - defaults/linux-64::freetype==2.5.5=1
  - defaults/linux-64::fontconfig==2.11.1=6
  - defaults/linux-64::numpy==1.10.4=py27_2
  - defaults/linux-64::cairo==1.12.18=6
  - defaults/linux-64::py2cairo==1.10.0=py27_2
  - defaults/linux-64::libgfortran==3.0.0=1
  - defaults/linux-64::libgcc-ng==7.2.0=hdf63c60_3
  - defaults/linux-64::scipy==0.17.1=np110py27_1
  - defaults/linux-64::libpng==1.6.17=0
  - defaults/linux-64::libstdcxx-ng==7.2.0=hdf63c60_3
  - defaults/linux-64::matplotlib==1.4.3=np110py27_2

执行conda list | grep free确认，这些包已安装但来自废弃的free通道：

conda list |grep free

输出：

cairo                     1.12.18                       6    https://repo.anaconda.com/pkgs/free
fontconfig                2.11.1                        6    https://repo.anaconda.com/pkgs/free
freetype                  2.5.5                         1    https://repo.anaconda.com/pkgs/free
gdata                     2.0.18                   py27_0    https://repo.anaconda.com/pkgs/free
libgfortran               3.0.0                         1    https://repo.continuum.io/pkgs/free
libpng                    1.6.17                        0    https://repo.anaconda.com/pkgs/free
matplotlib                1.4.3               np110py27_2    https://repo.anaconda.com/pkgs/free
mkl                       11.3.3                        0    https://repo.anaconda.com/pkgs/free
mysql-python              1.2.5                    py27_0    https://repo.anaconda.com/pkgs/free
natsort                   3.5.0                    py27_0    https://repo.anaconda.com/pkgs/free
numpy                     1.10.4                   py27_2    https://repo.anaconda.com/pkgs/free
py2cairo                  1.10.0                   py27_2    https://repo.anaconda.com/pkgs/free
pyparsing                 2.0.3                    py27_0    https://repo.anaconda.com/pkgs/free
qt                        4.8.7                         3    https://repo.anaconda.com/pkgs/free
scipy                     0.17.1              np110py27_1    https://repo.anaconda.com/pkgs/free

已尝试操作 #

添加free通道并调整配置：

conda config --add channels free
conda config --set restore_free_channel true
conda config --set channel_priority flexible

查看通道配置：

conda config --show channels

输出：

channels:
  - conda-forge
  - bioconda
  - defaults
  - free
  - biobakery

此时解决了free通道包的问题，但出现新的错误：

conda install -c bioconda fastspar

错误输出：

Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.

ResolvePackageNotFound: 
- defaults/linux-64::libgcc-ng==7.2.0=hdf63c60_3
- defaults/linux-64::libstdcxx-ng==7.2.0=hdf63c60_3

执行conda list确认这两个包已存在但无通道标识：

conda list|grep "libgcc-ng\|libstdcxx-ng"

输出：

libgcc-ng                7.2.0                hdf63c60_3  
libstdcxx-ng             7.2.0                hdf63c60_3

环境信息 #

conda info输出：

active environment : qiime_1_9_0
    active env location : /storage-daredevil/CONDA_ENVS/qiime_1_9_0
            shell level : 1
       user config file : $HOME/.condarc
 populated config files : $HOME/.condarc
                          /storage-daredevil/CONDA_ENVS/qiime_1_9_0/.condarc
          conda version : 4.11.0
    conda-build version : not installed
         python version : 3.9.7.final.0
       virtual packages : __linux=3.10.0=0
                          __glibc=2.17=0
                          __unix=0=0
                          __archspec=1=x86_64
       base environment : /opt/miniconda  (read only)
      conda av data dir : /opt/miniconda/etc/conda
  conda av metadata url : None
           channel URLs : https://conda.anaconda.org/conda-forge/linux-64
                          https://conda.anaconda.org/conda-forge/noarch
                          https://conda.anaconda.org/bioconda/linux-64
                          https://conda.anaconda.org/bioconda/noarch
                          https://repo.anaconda.com/pkgs/main/linux-64
                          https://repo.anaconda.com/pkgs/main/noarch
                          https://repo.anaconda.com/pkgs/free/linux-64
                          https://repo.anaconda.com/pkgs/free/noarch
                          https://repo.anaconda.com/pkgs/r/linux-64
                          https://repo.anaconda.com/pkgs/r/noarch
                          https://conda.anaconda.org/free/linux-64
                          https://conda.anaconda.org/free/noarch
                          https://conda.anaconda.org/biobakery/linux-64
                          https://conda.anaconda.org/biobakery/noarch
          package cache : /opt/miniconda/pkgs
                          $HOME/.conda/pkgs
       envs directories : /storage-daredevil/CONDA_ENVS
                          $HOME/.conda/envs
                          /opt/miniconda/envs
               platform : linux-64
             user-agent : conda/4.11.0 requests/2.27.1 CPython/3.9.7 Linux/3.10.0-1160.59.1.el7.x86_64 centos/7.9.2009 glibc/2.17
                UID:GID : 1060:20
             netrc file : $HOME/.netrc
           offline mode : False

解决方法 #

方法1：跳过依赖检查安装 #

如果确认fastspar运行不依赖环境中已存在的旧版本包，可直接跳过依赖校验：

conda install -c bioconda fastspar --no-deps

安装完成后手动验证fastspar是否能正常执行。

方法2：忽略冲突包 #

告诉conda忽略这两个无法解析的系统级包：

conda install -c bioconda fastspar --ignore-package libgcc-ng libstdcxx-ng

方法3：重建干净环境（推荐） #

由于当前环境状态混乱，最稳妥的方式是重新构建兼容环境：

1. 导出当前环境包列表：

conda env export --name qiime_1_9_0 > qiime_env.yml

2. 编辑qiime_env.yml，移除libgcc-ng、libstdcxx-ng以及来自free通道的旧包，替换为conda-forge或defaults通道的兼容版本。
3. 创建新环境：

conda env create -f qiime_env.yml -n qiime_1_9_0_new

4. 在新环境中安装fastspar：

conda activate qiime_1_9_0_new
conda install -c bioconda fastspar

方法4：手动标记包通道（风险较高） #

通过修改环境的conda元数据文件，手动标记包的来源通道：

1. 进入环境的conda-meta目录：

cd /storage-daredevil/CONDA_ENVS/qiime_1_9_0/conda-meta

2. 找到libgcc-ng-7.2.0-hdf63c60_3.json和libstdcxx-ng-7.2.0-hdf63c60_3.json文件，编辑其中的channel字段为defaults：

"channel": "defaults",

3. 保存文件后重新尝试安装。

内容的提问来源于stack exchange，提问作者Marco Severgnini