Synology DS220+迁移用户后scikit-learn 0.24.1运行报错求助

在Synology DS220+上将用户名从“admin”迁移至“username”后，原通过SSH每日运行的Python程序出现异常。已通过以下步骤在虚拟环境中安装scikit-learn 0.24.1：

sudo python3 -m pip install virtualenv
python3 -m virtualenv venv
source venv/bin/activate
pip install scikit-learn==0.24.1

执行pip show scikit-learn确认安装信息：

pip show scikit-learn
>Name: scikit-learn
Version: 0.24.1
Summary: A set of python modules for machine learning and data mining
Home-page: http://scikit-learn.org
Author:
Author-email:
License: new BSD
Location: /volume1/homes/username/venv/lib/python3.8/site-packages
Requires: joblib, numpy, scipy, threadpoolctl
Required-by:

但运行指定脚本时出现如下错误：

username@synology:~$ venv/bin/python3 /volume1/homes/username/Drive/stock/portfolio/price_updater.py
Traceback (most recent call last):
  File "/volume1/homes/username/venv/lib/python3.8/site-packages/sklearn/__check_build/__init__.py", line 44, in <module>
    from ._check_build import check_build  # noqa
ImportError: dlopen: cannot load any more object with static TLS

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/volume1/homes/username/Drive/stock/portfolio/price_updater.py", line 6, in <module>
    from new.portbot import PortfolioBot
  File "/volume1/homes/username/Drive/stock/new/portbot.py", line 25, in <module>
    from NewStock import Stock
  File "/volume1/homes/username/Drive/stock/NewStock.py", line 17, in <module>
    from new.buyingprice import BuyingPrice
  File "/volume1/homes/username/Drive/stock/new/buyingprice.py", line 13, in <module>
    from new.pricetarget import PriceTarget
  File "/volume1/homes/username/Drive/stock/new/pricetarget.py", line 19, in <module>
    from sklearn.linear_model import LinearRegression
  File "/volume1/homes/username/venv/lib/python3.8/site-packages/sklearn/__init__.py", line 81, in <module>
    from . import __check_build  # noqa: F401
  File "/volume1/homes/username/venv/lib/python3.8/site-packages/sklearn/__check_build/__init__.py", line 46, in <module>
    raise_build_error(e)
  File "/volume1/homes/username/venv/lib/python3.8/site-packages/sklearn/__check_build/__init__.py", line 31, in raise_build_error
    raise ImportError("""%s
ImportError: dlopen: cannot load any more object with static TLS
___________________________________________________________________________
Contents of /volume1/homes/username/venv/lib/python3.8/site-packages/sklearn/__check_build:
_check_build.cpython-38-x86_64-linux-gnu.sosetup.py                  __init__.py
__pycache__
___________________________________________________________________________
It seems that scikit-learn has not been built correctly.

If you have installed scikit-learn from source, please do not forget
to build the package before using it: run `python setup.py install` or
`make` in the source directory.

If you have used an installer, please check that it is suited for your
Python version, your operating system and your platform.

奇怪的是，Python交互环境中可正常导入sklearn：

username@Synology:~$ venv/bin/python3
Python 3.8.6 (default, Jan  4 2021, 16:11:49)
[GCC 4.9.3 20150311 (prerelease)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import sklearn
>>>

1. 调整脚本导入顺序 #

将sklearn相关的导入语句移至脚本最顶部，避免其他库先占用静态TLS槽位。例如，在price_updater.py的第一行添加：

from sklearn.linear_model import LinearRegression

2. 重新编译scikit-learn并禁用静态TLS #

激活虚拟环境后，使用以下命令重新安装scikit-learn，编译时禁用静态TLS：

CFLAGS="-fno-static-tls" pip install --force-reinstall --no-binary :all: scikit-learn==0.24.1

3. 升级scikit-learn版本 #

较新版本的scikit-learn已修复部分静态TLS问题，若脚本兼容，可直接升级：

pip install --upgrade scikit-learn

4. 调整系统环境变量 #

运行脚本前设置环境变量，预加载pthread库：

export LD_PRELOAD=/lib64/libpthread.so.0
venv/bin/python3 /volume1/homes/username/Drive/stock/portfolio/price_updater.py

也可将该环境变量添加到虚拟环境的venv/bin/activate脚本末尾，每次激活时自动生效。

内容的提问来源于stack exchange，提问作者mrkgoh