linux64 和 glibc 的 Conda UnsatisfiableError

Question

我在所有 miniconda 环境中都遇到错误。我刚刚新建了一个环境，仍然出现同样的错误：

\n

\nconda create --name py36 python=3.6 spyder jupyter\n\n

\n

^ 工作正常。然后：

\n

\nconda install -c conda-forge/label/rc_iris iris \n\n

\n

给出以下错误：

\n

Solving environment: failed with initial frozen solve. Retrying with flexible solve.\n\nSolving environment: failed with repodata from current_repodata.json, will retry with next repodata source.\n\nCollecting package metadata (repodata.json): done\n\nSolving environment: failed with initial frozen solve. Retrying with flexible solve.\n\nSolving environment: -\n\nFound conflicts! Looking for incompatible packages.\n\nThis can take several minutes. \xc2\xa0Press CTRL-C to abort.\n\nfailed\n\n\nUnsatisfiableError: The following specifications were found to be incompatible with each other:\n\n\nOutput in format: Requested package -> Available versionsThe following specifications were found to be incompatible with your system:\n\n\n\xc2\xa0 - feature:/linux-64::__glibc==2.17=0\n\n\xc2\xa0 - feature:|@/linux-64::__glibc==2.17=0\n\n\nYour installed version is: 2.17\n

\n

但由于某种原因，如果您创建一个全新的环境，这会起作用：

\n

conda create --name py36 -c conda-forge iris python=3.6 spyder jupyter\n

\n

有谁知道发生了什么事吗？当我尝试安装其他东西时，如何确保不会出现此 glibc 错误？

\n

Answer 1

这与 GLIBC 无关，而是 Conda 不擅长报告冲突。查看感兴趣的包清楚地表明它有一个python>=3.8要求。由于您将 Conda 锁定为python=3.6，因此它无法求解，并尝试解释原因。

如果您确实有兴趣安装该候选版本，则需要在至少具有 Python v3.8 的环境中执行此操作。