gpt4 book ai didi

linux - Elastic Beanstalk 安装 scipy

转载 作者:塔克拉玛干 更新时间:2023-11-03 00:40:38 25 4
gpt4 key购买 nike

我正在尝试在 Amazon Elastic Beanstalk 上运行的基本 Django 应用程序上安装 scipy,但我无法让它工作。

以下是重现我的问题的步骤:

来自 the guide 的常规内容:

# Create a new virtual environment
mkvirtualenv -p python2.7 django_eb

# Install Django on it
pip install django==1.9.12
# pip freeze should now show Django==1.9.12 and some other things

# Start a new Django project
# This creates a directory that has everything you need for Django
django-admin startproject django_eb
cd django_eb

# Optionally check that the site works
python manage.py runserver
Ctrl-C

# Store the pip requirements so that the remote host can install them
pip freeze > requirements.txt

# Tell the remote server where our wsgi file is
mkdir .ebextensions
cat <<EOT >> .ebextensions/django.config
option_settings:
aws:elasticbeanstalk:container:python:
WSGIPath: django_eb/wsgi.py
EOT

# Allow any host for our project
# If this is unset, you'll get a 404 on the deployed site
set ALLOWED_HOSTS = ['*'] in settings.py

# Create an EB project
# Will need AWS EB CLI for this
eb init -p python2.7 django_eb
# Choose some region
eb init
# choose Y so we can SSH and check logs

# Create a deployment environment
eb create django-eb-env
# This step takes around 5 minutes
# If it fails and you need to restart run 'eb deploy'

# Open the website in your OS's default browser
eb open
# If you get DisallowedHost at / error
# double check that ALLOWED_HOSTS = ['*']

安装 scipy:

# Now we'll install scipy and watch how it doesn't work remotely
pip install scipy==0.19.0
pip freeze > requirements.txt
eb deploy
# Should take forever and then finally print
# 'ERROR: Failed to deploy application.'

eb ssh
cat /var/log/eb-activity.log
# Should print
# numpy.distutils.system_info.NotFoundError: no lapack/blas resources found

# After reading this blog post
# https://medium.com/@DaveJMcKeown/deploying-scipy-into-aws-elastic-beanstalk-2e5e481155de
# I added this to .ebextensions/django.config:
packages:
yum:
make: []
gcc-c++: []
gcc-gfortran: []
python27-devel: [] # I used python27-devel instead of python-devel
atlas-sse3-devel: []
lapack-devel: []
libpng-devel: []
freetype-devel: []
zlib-devel: []
container_commands:
AddGlobalWSGIGroupAccess:
command: "if ! grep -q 'WSGIApplicationGroup %{GLOBAL}' ../wsgi.conf ; then echo 'WSGIApplicationGroup %{GLOBAL}' >> ../wsgi.conf; fi;"

# Unfortunately, this leads to an our of memory error
# Running dmesg now looks like the following:
# http://stackoverflow.com/a/35011967/2770572

我在这里不知所措。看起来我也许可以让它在具有更多 RAM 的 EC2 实例上运行,但我不能真的那样做,因为它会让我脱离免费层。有没有办法运行 make 命令,使其不会占用太多内存或其他解决方案?

最佳答案

您可以尝试将此添加到您的.ebextensions/django.config:

commands:
01_no_cache:
command: "/opt/python/run/venv/bin/pip install --no-cache-dir -r /opt/python/current/app/requirements.txt"

如果您检查 /var/log/eb-activity.log(或者如果您安装了 EB CLI,则只需键入 eb logs),您可以看到错误这里:

MemoryError
2017-08-10 11:40:25,557 ERROR Error installing dependencies: Command '/opt/python/run/venv/bin/pip install -r /opt/python/ondeck/app/requirements.txt' returned non-zero exit status 2

据我所知,如果将特定命令包含在 django.config 中,则可以为 EB 提供这些命令。如所写here ,您可以禁用缓存。

这对我有用。

关于linux - Elastic Beanstalk 安装 scipy,我们在Stack Overflow上找到一个类似的问题: https://stackoverflow.com/questions/43164961/

25 4 0
Copyright 2021 - 2024 cfsdn All Rights Reserved 蜀ICP备2022000587号
广告合作:1813099741@qq.com 6ren.com