We Keep Coding

Bash script gives psycopg2 error: symbol not found in flat namespace (_PQbackendPID)?

I am using Apple M1. Here is the bash script I wrote.
#!/bin/sh
# Open Activity Tracker website
python -m webbrowser http://127.0.0.1:8000/
# Activate virtual environment
. dj-env/bin/activate
# Navigate to Activity Tracker project folder
cd dj_activity_tracker
# Run Django local development server
python manage.py runserver
When you run manually are you doing that in the same virtualenv? Yes.
Are you doing the installs into the virtualenv? Yes.
What version of psycopg? Within my virtual environment, I run "pip show psycopg2-binary". It outputs "Version: 2.9.5".
When I run these commands on my terminal manually without the script, it works. However, when I use this script, I get the error:
File "/Users/doge/Desktop/dj-env/lib/python3.11/site-packages/django/db/backends/postgresql/base.py", line 28, in \<module\>
raise ImproperlyConfigured("Error loading psycopg2 module: %s" % e)
django.core.exceptions.ImproperlyConfigured: Error loading psycopg2 module: dlopen(/Users/doge/Desktop/dj-env/lib/python3.11/site-packages/psycopg2/\_psycopg.cpython-311-darwin.so, 0x0002): symbol not found in flat namespace (\_PQbackendPID)
Things I have tried:
Uninstall/reinstall psycopg2
brew install postgresql
I am stumped on how this is working manually but not when I run the script. What could be the issue? I appreciate your support.

Can't install gcloud on Amazon Linux : invalid syntax

I'm trying to install gcloud on my EC2 server running Amazon Linux 4.14.47-56.37 64bits, in interactive mode running the following command :
curl https://sdk.cloud.google.com | bash
The files download correctly, but the install then fails with the following Traceback :
File "/home/ec2-user/google-cloud-sdk/bin/bootstrapping/install.py", line 12, in <module>
import bootstrapping
File "/home/ec2-user/google-cloud-sdk/bin/bootstrapping/bootstrapping.py", line 32, in <module>
import setup # pylint:disable=g-import-not-at-top
File "/home/ec2-user/google-cloud-sdk/bin/bootstrapping/setup.py", line 55, in <module>
from googlecloudsdk.core import properties
File "/home/ec2-user/google-cloud-sdk/lib/googlecloudsdk/core/properties.py", line 291
self.__sections = {section.name: section for section in sections}
^
SyntaxError: invalid syntax
Any idea why this for is causing issues?
I am running python 2.7 (2.7.14) as recommended by Google.

On top of python 2.7 installed on the "python" command, I also had python 2.6 installed on the "python2" command.
Uninstalling python 2.6 solved the issue, Google Cloud install went through without issue on the next try.

You locally update the install.sh file to use python2.7 instead of python2. This worked for me

In case someone meets this problem again, and if you do not want to uninstall python 2.6 or to modify install.sh, here is a procedure that worked for me on an ec2 instance :
open a new terminal and type :
curl https://sdk.cloud.google.com | bash
The install should not work as CLOUDSDK_PYTHON=python2 by default, which leads to Python 2.6. To correct this, enter this command :
export CLOUDSDK_PYTHON=python
this links CLOUDSDK_PYTHON to Python 3.6
then
bash google-cloud-sdk/install.sh
and finally
./google-cloud-sdk/bin/gcloud init
Once this is done, you can use any gcloud or gsutil command by doing this :
./google-cloud-sdk/bin/gcloud <your gcloud command>
./google-cloud-sdk/bin/gsutil <your gsutil command>
For example, in order to copy a folder from a gs bucket to your ec2 instance :
./google-cloud-sdk/bin/gsutil -m cp -r gs://<bucket_name>/path/to/folder /path/on/your/ec2/instance
the -m option allows fast transfer (I could transfer with an average speed of 81 MiB/s with this option)
Hope this helps !

Setup yum repo:
# Update YUM with Cloud SDK repo information:
sudo tee -a /etc/yum.repos.d/google-cloud-sdk.repo << EOM
[google-cloud-sdk]
name=Google Cloud SDK
baseurl=https://packages.cloud.google.com/yum/repos/cloud-sdk-el7-x86_64
enabled=1
gpgcheck=1
repo_gpgcheck=1
gpgkey=https://packages.cloud.google.com/yum/doc/yum-key.gpg
https://packages.cloud.google.com/yum/doc/rpm-package-key.gpg
EOM
# The indentation for the 2nd line of gpgkey is important.
Install yum downloader and download the rpm, then install with "--nodeps":
yum install yum-utils
yumdownloader google-cloud-sdk-231.0.0-1.el7.noarch
mv 23873bd6e8459ba6e70e96eb8f03f6ac03cd707ce3c80baa8264c714e030c915-google-cloud-sdk-231.0.0-1.el7.noarch.rpm /usr/local/src/google-cloud-sdk-231.0.0-1.el7.noarch
rpm -ivh --nodeps /usr/local/src/google-cloud-sdk-231.0.0-1.el7.noarch

This solved issue:
export CLOUDSDK_PYTHON=python

ImportError libssl.so.6: cannot open shared object file error even after having the file in the path

I am getting following error even after having following packages installed. Recently os has been upgraded and python has been updated from 2.5 to 2.7 and before that application was using 2.7.11 which is installed on local folder.
After up gradation I am unable to run my scirpts and getting following ssl error.
File "/itStorage/tools/swutils/pylibs/shu/common.py", line 5, in <module>
import ssl
File "/itStorage/tools/swutils/deps/python_2.7.11/lib/python2.7/ssl.py", line 97, in
import _ssl # if we can't import it, let the error propagate
ImportError: libssl.so.6: cannot open shared object file: No such file or directory
After searching google it may be due to openssl packages or libssl path. we have every thing in place even after that we having issue.
[aafedc#afeae ~]$ sudo rpm -qa *openssl*
openssl-1.0.2k-8.el7.x86_64
openssl-devel-1.0.2k-8.el7.x86_64
openssl-libs-1.0.2k-8.el7.i686
openssl-libs-1.0.2k-8.el7.x86_64
openssl098e-0.9.8e-29.el7_2.3.i686
[abcd#abcd ~]$ sudo ls -l /usr/lib/libssl.so.6*
lrwxrwxrwx. 1 root root 16 May 25 03:34 /usr/lib/libssl.so.6 -> libssl.so.0.9.8e
any help is greatly appropriated!!..
Thanks in advance.

You could try to run the script using strace, then you should see what fails.. I'm not sure but I think it could be a missing dependency of libssl as well or permissions problem.

How to resolve cannot import name pages (django/wagtail)

Recently I have been trying to get wagtail to work with my existing Django application. I was experiencing an error: -
ImportError: No module named wagtail
unable to load app 0 (mountpoint='') (callable not found or import error)
After much troubleshooting I managed to fix this, by copying the folder wagtail from:
/usr/local/lib/python2.7/dist-packages/
into here
/opt/django/src/
Having resolved this error, I received another about a different module, and another... each time I copied the folder from /usr/local/lib/python2.7/dist-packages/ into /opt/django/src/ and it eventually resolved the issues I was having and uWSGI started.
Now when I access the homepage of my app, I receive this error
ImportError at /
cannot import name pages
Request Method: GET
Request URL: http://example.com
Django Version: 1.9
Exception Type: ImportError
Exception Value:
cannot import name pages
Exception Location: ./wagtail/wagtailadmin/urls/__init__.py in <module>, line 4
Python Executable: /usr/local/bin/uwsgi
Python Version: 2.7.3
Python Path:
['.',
'',
'/opt/django/src',
'/root/.python',
'/opt/django/env/lib/python2.7',
'/opt/django/env/lib/python2.7/plat-linux2',
'/opt/django/env/lib/python2.7/lib-tk',
'/opt/django/env/lib/python2.7/lib-old',
'/opt/django/env/lib/python2.7/lib-dynload',
'/usr/lib/python2.7',
'/usr/lib/python2.7/plat-linux2',
'/usr/lib/python2.7/lib-tk',
'/opt/django/env/local/lib/python2.7/site-packages',
'/opt/django/env/lib/python2.7/site-packages']
I have checked the file referenced (./wagtail/wagtailadmin/urls/__init__.py) and it looks like this:
from django.conf.urls import url, include
from django.views.decorators.cache import cache_control
from wagtail.wagtailadmin.urls import pages as wagtailadmin_pages_urls
from wagtail.wagtailadmin.urls import collections as wagtailadmin_collections_urls
from wagtail.wagtailadmin.urls import password_reset as wagtailadmin_password_reset_urls
from wagtail.wagtailadmin.views import account, chooser, home, pages, tags, userbar
from wagtail.wagtailadmin.api import urls as api_urls
from wagtail.wagtailcore import hooks
from wagtail.utils.urlpatterns import decorate_urlpatterns
from wagtail.wagtailadmin.decorators import require_admin_access
The offending line is the first wagtail 'from' ...
from wagtail.wagtailadmin.urls import pages as wagtailadmin_pages_urls
How can I resolve this error?

I've ran a few tests on wagtail setup. I made each test in a fresh Ubuntu 14 install (each test in a new workspace on Cloud9 IDE).
Test 1: Straight python 2.7
sudo pip install wagtail
ERRO: error: command 'x86_64-linux-gnu-gcc' failed with exit status 1
Comment: Lots of x86_64-linux-gnu-gcc: error: build/temp.linux-x86_64-2.7/libImaging/codec_fd.o: No such file or directory errors were output.
Didn't work.
Test 2 - Python 2.7 with virtualenv
sudo pip install virtualenv
virtualenv env
source env/bin/activate
Now from their github https://github.com/wagtail/wagtail
pip install wagtail
out: Successfully installed Django-1.10.7 Pillow-4.1.1 Unidecode-0.4.20 Willow-0.4 beautifulsoup4-4.5.3 django-modelcluster-3.1 django-taggit-0.22.1 django-treebeard-4.1.0 djangorestframework-3.6.2 html5lib-0.999999999 olefile-0.44 pytz-2017.2 requests-2.13.0 wagtail-1.9.1 webencodings-0.5.1
wagtail start mysite
cd mysite
python manage.py migrate
python manage.py createsuperuser
python manage.py runserver
I had to edit the settings/dev.py and include ALLOWED_HOSTS = ['*'] because I'm on cloud 9. Also needed the $IP:$PORT after runserver. WORKED!
Test3 - Python 3.4 with virtualenv
sudo pip3 install virtualenv
virtualenv -p python3 env
source env/bin/activate
pip install wagtail
Successfully installed Django-1.10.7 Pillow-4.1.1 Unidecode-0.4.20 Willow-0.4 beautifulsoup4-4.5.3 django-modelcluster-3.1 django-taggit-0.22.1 django-treebeard-4.1.0 djangorestframework-3.6.2 html5lib-0.999999999 olefile-0.44 pytz-2017.2 requests-2.13.0 wagtail-1.9.1 webencodings-0.5.1
Rest of the command are the same. And it WORKED!
Test 4 - Python 3.4 no virtualenv
sudo pip3 install wagtail
out: Successfully installed wagtail django-taggit requests Unidecode Django djangorestframework Pillow beautifulsoup4 Willow django-modelcluster django-treebeard olefile pytz
wagtail start mysite
python3 manage.py migrate
python3 manage.py createsuperuser
python3 manage.py runserver
EDIT: Made a mistake. Should've run with python3, not python in this case. Did it again and worked!
Conclusion: Looks that you need a virtualenv for this to work on Ubuntu 14 and python2. With python3 it works with and without virtualenv. I could not reproduce your error, but the idea is the same: it does not work because it looks for stuff in the wrong place.
Comments: There are two things you mention that I could not test:
I have been trying to get wagtail to work with my existing Django
application
With this big amount of heavy dependencies, (Pillow, Django rest, beatifullsoup and others I've never heard of) it would be strange if it worked out of the box with a previous setup. I mean, Django is a dependency for wagtail. So I would try the other way around. Start fresh with wagtail and port your code there (with python3 for the sake of the community ;) ). Also this manual copy of files is like programmers worst nightmare (at least to me). This looks like a version problem. Try to post the versions of everything involved, and the exact steps to reproduce the issue. I mean, did you pip install wagtail in the operating system that django was running?
I was having and uWSGI started.
So you have your app deployed already? This complicates things. If versions and virtualenv do not solve your problem, community needs setup details of deploy. Also you haven't mentioned your operational system.
I hope this helps! Good Luck!

I ran into the
cannot import name pages
error at:
from wagtail.wagtailadmin.urls import pages as wagtailadmin_pages_urls
while trying to deploy a wagtail app to Google App Engine and could not duplicate the problem on my local dev server.
A seemingly unrelated error that I saw only intermittently in GAE involved an import error in the PIL library. I was able to resolve BOTH errors by deleting PIL and pillow related folders from my vendored packages in my /lib directory. I had to import PIL through the app.yaml file instead due to the platform specific binary, but running pip install -t lib/ -r requirements-vendor.txt caused PIL to be installed into the lib folder as well (as a dependency of Wagtail)

Fail to build atom/electron on Windows 7

Try to build atom/electron on Windows 7, but failed.
OS Environment:
Windows 7
VS 2013 Update 5
Python 2.7
Node.js 4.2.3
Git 2.6.4
atom/electron - Latest pull (2015.12.17)
It failed when running the bootstrap script "$ python bootstrap.py -v", the screenshot is here:
Trying to find a way through, any help will be appreciated.
-- dean

Found a way...:
Situation failed one: "can't open file"
According to the official documentation: Build Instructions (Windows)
$ python script\bootstrap.py -v
the result will be:
D:\Program Files\Python27\python.exe: can't open file 'scriptbootstrap.py': [Errno 2] No such file or directory
Situation failed two: "WindowsError: [Error 3]...."
Pay attention to this:
$ cd script
$ python bootstrap.py -v
I cd into the script folder, and run the bootstrap script, and failed.
Situation OK: Use "/" instead of "\"
See the screenshot below:
Explanations are welcome...
--dean