I am using an anaconda's python 2.7 distribution as my default python distribution. I receive a
Fatal Python error: PyThreadState_Get: no current thread
This error seems to be a result of a python distribution conflict; I am using anconda's python distribution but Leap motion's sdk anticipates the default mac distribution.
I followed Leap motion's sdk instructions on changing the default python distribution as follows,
install_name_tool -change /Library/Frameworks/Python.framework/Versions/2.7/Python
//anaconda/bin/python LeapPython.so
But now I receive the following error, which I have not been able to make much of,
Traceback (most recent call last):
File "Sample.py", line 9, in <module>
import Leap, sys, thread, time
File "/Users/jkarimi91/Documents/Projects/hackingedu/LeapDeveloperKit_2.3.1+31549_mac/LeapSDK/lib/Leap.py", line 28, in <module>
LeapPython = swig_import_helper()
File "/Users/jkarimi91/Documents/Projects/hackingedu/LeapDeveloperKit_2.3.1+31549_mac/LeapSDK/lib/Leap.py", line 24, in swig_import_helper
_mod = imp.load_module('LeapPython', fp, pathname, description)
ImportError: dlopen(/Users/jkarimi91/Documents/Projects/hackingedu/LeapDeveloperKit_2.3.1+31549_mac/LeapSDK/lib/LeapPython.so, 2): Library not loaded: //anaconda/bin/python
Referenced from: /Users/jkarimi91/Documents/Projects/hackingedu/LeapDeveloperKit_2.3.1+31549_mac/LeapSDK/lib/LeapPython.so
Reason: no suitable image found. Did find:
//anaconda/bin/python: can't map unslidable segment __TEXT to 0x100000000 with size 0x1000
I think the problem is that you are trying to load the Python executable, not the shared library. Since Anaconda doesn't seem to give you a framework, try replacing /anaconda/bin/python in your command with /anaconda/pkgs/python-2.7.10.0 or if that doesn't work, specify the full path to the Python dylib /anaconda/pkgs/python-2.7.10.0/lib/libpython2.7.dylib.
Related
Possible duplicate of this question.
I have a gpu account to whom I connect through putty (ssh login). I have created a virtualenv there and I am installing tenorflow through pip for gpu. Everything works fine, when I run command
$ pip list
following list is being shown:
backports.weakref (1.0rc1)
bleach (1.5.0)
funcsigs (1.0.2)
html5lib (0.9999999)
Markdown (2.6.8)
mock (2.0.0)
numpy (1.13.1)
olefile (0.44)
pbr (3.1.1)
Pillow (4.2.1)
pip (9.0.1)
protobuf (3.3.0)
setuptools (36.0.1)
six (1.10.0)
tensorflow-gpu (1.2.1)
Werkzeug (0.12.2)
wheel (0.29.0)
But when I run:
$ python
>>> import tensorflow
It shows the following error:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/nauman/junaid/final/test/lib/python2.7/site-
packages/tensorflow/__init__.py", line 24, in <module>
from tensorflow.python import *
File "/home/nauman/junaid/final/test/lib/python2.7/site-
packages/tensorflow/python/__init__.py", line 49, in <module>
from tensorflow.python import pywrap_tensorflow
File "/home/nauman/junaid/final/test/lib/python2.7/site-
packages/tensorflow/python/pywrap_tensorflow.py", line 52, in <module>
raise ImportError(msg)
ImportError: Traceback (most recent call last):
File "/home/nauman/junaid/final/test/lib/python2.7/site-
packages/tensorflow/python/pywrap_tensorflow.py", line 41, in <module>
from tensorflow.python.pywrap_tensorflow_internal import *
File "/home/nauman/junaid/final/test/lib/python2.7/site-
packages/tensorflow/python/pywrap_tensorflow_internal.py", line 28, in
<module>
_pywrap_tensorflow_internal = swig_import_helper()
File "/home/nauman/junaid/final/test/lib/python2.7/site-
packages/tensorflow/python/pywrap_tensorflow_internal.py", line 24, in
swig_import_helper
_mod = imp.load_module('_pywrap_tensorflow_internal', fp, pathname,
description)
ImportError: libcusolver.so.8.0: cannot open shared object file: No such
file or directory
Failed to load the native TensorFlow runtime.
See https://www.tensorflow.org/install/install_sources#common_installation_problems
for some common reasons and solutions. Include the entire stack trace
above this error message when asking for help.
I have also set my environment variables like this:
export CUDA_HOME=/opt/cuda
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:$CUDA_HOME/lib64:$CUDA_HOME/extras/CUPTI/lib64"
Moreover, I found that libcusolver.so.8.0 is not there, while libcusolver.so.7.5
is there in cuda/lib64/.
Somehow tensorflow is finding wrong file or I ain't know nothing.
Any help would be appreciated as I am new to all this stuff.
Python version: 2.7
OS: Linux
This may be connected to the incorrect linking of your libraries.
Simply run sudo ldconfig /usr/local/cuda/lib64. It solved for me.
If you need to know more: ldconfig man page.
I solved the issue. Actually I have cuda 7.5 installed and I was installing latest tensorflow version which probably support cuda 8.0.
So I downgraded.
pip install --upgrade \ https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0-cp27-none-linux_x86_64.whl
If you installed tensorflow-gpu by using pip with prebuilt .whl, the cuda version were fixed. As far as I know, starting 0.11.0rc1, all the prebuilt packages are now built for cuda 8. So there are two ways to solve the problem:
install cuda 8 for the prebuilt packages tensorflow-gpu >= 0.11.orc1
keep cuda 7.5 stay, then build tensorflow-gpu from source code
If you are using cuda-9.0 try sudo apt install nvidia-cuda-dev (if you are using an Ubuntu distribution)
When I upgrade Python from 2.7.9 to 2.7.10 with anaconda, I get the following error in my apache2 log and my websites don't work any more.
mod_wsgi (pid=608): Target WSGI script '/www/wsgi_scripts/django_myproject.wsgi' cannot be loaded as Python module.
Exception occurred processing WSGI script '/www/wsgi_scripts/django_myproject.wsgi'.
Traceback (most recent call last):
File "/www/wsgi_scripts/django_myproject.wsgi", line 8, in <module>
import django.core.handlers.wsgi
File "/Users/xnx/anaconda/lib/python2.7/site-packages/django/core/handlers/wsgi.py", line 6, in <module>
from io import BytesIO
File "/Users/xnx/anaconda/lib/python2.7/io.py", line 51, in <module>
import _io
ImportError: dlopen(/Users/xnx/anaconda/lib/python2.7/lib-dynload/_io.so, 2): Symbol not found: __PyErr_ReplaceException
Referenced from: /Users/xnx/anaconda/lib/python2.7/lib-dynload/_io.so
Expected in: dynamic lookup
I expect this is some kind of mismatch in package library versions, but does anyone know which one?
Double check your PYTHONPATH if it's correct (you don't override it anywhere in your startup files).
Otherwise you may consider downgrading your Python from 2.7.10 to 2.7.9.
For example (as suggested in here):
conda install python=2.7.9
or when using Homebrew try:
brew switch python 2.7.9
as it could be some particular problem with 2.7.10.
Alternatively use Python 3.x instead.
My suspicion is that this is a problem with Anaconda as I've just hit the same problem. I've reported an issue here: https://github.com/conda/conda/issues/1367
update:
The conda guys have now released a fixed virtualenv package that no longer has this problem.
I cannot successfully run the first command in the Scrapy tutorial (http://doc.scrapy.org/en/latest/intro/tutorial.html). The code copy pasted below is the result (with the error at the end).
Python 2.7 is installed, and I followed the installation instructions for scrapy (http://doc.scrapy.org/en/latest/intro/install.html). I am running Python 2.7.6 32 bit on Windows 7 (64 bit).
Other aspects of installation:
Twisted-13.2.0.dist-info
zope.interface-4.1.1-py2.7
Scrapy-0.22.2-py2.7
lxml-3.3.3-py2.7
cssselect-0.9.1-py2.7
cryptography-0.3.dist-info
pyOpenSSL-0.14-py2.7
pywin32_system32
And here's the error message:
C:\Python27\Scripts>scrapy startproject tutorial
Traceback (most recent call last):
File "C:\Python27\lib\runpy.py", line 162, in _run_module_as_main
"__main__", fname, loader, pkg_name)
File "C:\Python27\lib\runpy.py", line 72, in _run_code
exec code in run_globals
File "C:\Python27\lib\site-packages\scrapy\cmdline.py", line 9, in <module>
from scrapy.crawler import CrawlerProcess
File "C:\Python27\lib\site-packages\scrapy\crawler.py", line 3, in <module>
from twisted.internet import reactor, defer
File "twisted\internet\reactor.py", line 38, in <module>
from twisted.internet import default
File "twisted\internet\default.py", line 56, in <module>
install = _getInstallFunction(platform)
File "twisted\internet\default.py", line 50, in _getInstallFunction
from twisted.internet.selectreactor import install
File "twisted\internet\selectreactor.py", line 18, in <module>
from twisted.internet import posixbase
File "twisted\internet\posixbase.py", line 24, in <module>
from twisted.internet import error, udp, tcp
File "twisted\internet\tcp.py", line 29, in <module>
from twisted.internet._newtls import (
File "twisted\internet\_newtls.py", line 21, in <module>
from twisted.protocols.tls import TLSMemoryBIOFactory, TLSMemoryBIOProtocol
File "twisted\protocols\tls.py", line 40, in <module>
from OpenSSL.SSL import Error, ZeroReturnError, WantReadError
File "build\bdist.win32\egg\OpenSSL\__init__.py", line 8, in <module>
File "build\bdist.win32\egg\OpenSSL\rand.py", line 11, in <module>
File "build\bdist.win32\egg\OpenSSL\_util.py", line 4, in <module>
File "C:\Python27\lib\site-packages\cryptography\hazmat\bindings\openssl\binding.py", l3, in __init__
self._ensure_ffi_initialized()
File "C:\Python27\lib\site-packages\cryptography\hazmat\bindings\openssl\binding.py", l9, in _ensure_ffi_initialized libraries)
File "C:\Python27\lib\site-packages\cryptography\hazmat\bindings\utils.py", line 77, ind_ffi
ext_package="cryptography",
File "C:\Python27\lib\site-packages\cffi\api.py", line 341, in verify
lib = self.verifier.load_library()
File "C:\Python27\lib\site-packages\cffi\verifier.py", line 75, in load_library
return self._load_library()
File "C:\Python27\lib\site-packages\cffi\verifier.py", line 151, in _load_library
return self._vengine.load_library()
File "C:\Python27\lib\site-packages\cffi\vengine_cpy.py", line 138, in load_library
raise ffiplatform.VerificationError(error)
cffi.ffiplatform.VerificationError: importing 'C:\\Python27\\lib\\site-packages\\cryptogr
\_Cryptography_cffi_48bbf0ebx93c91939.pyd': DLL load failed: The operating system cannot
1.**
I just reinstall cryptography to make it work.
pip uninstall cryptography
pip install cryptography
I had this problem due to another (older?) version of libeay32.dll and ssleay32.dll being on the PATH before those of my own OpenSSL 1.0.1g installation. I recommend you use Process Monitor from SysInternals to monitor python.exe to see where it is actually loading your OpenSSL DLLs from.
In my case, the offender was some Intel components at C:\Program Files (x86)\Intel\iCLS Client\ that came with my drivers. After moving my own OpenSSL bin directory earlier in the PATH environment variable, everything worked as expected.
this is due to _Cryptography_cffi_48bbf0ebx93c91939.pyd depends on openssl dll ssleay32.dll and libeay32.dll. after you compile openssl on windows, u can copy these 2 files to system32.
#crazyzh1984's method is a little complex, but his supplement that posted at the bottom of his answer is very useful. I download the "Win32 OpenSSL v1.0.1g Light" at http://slproweb.com/products/Win32OpenSSL.html and then I could install pyOpenSSL successfully.
#lambokini is right，but i'cant comment on the answer, so come this one.
First download openssl source from http://www.openssl.org/
Second start "Visual Studio Command Prompt", compile and install openssl follow install guide(INSTALL.W32 or INSTALL.W64).
Then add "[openssl install path]\bin" to the environment variable "path",
and you can delete ssleay32.dll and libeay32.dll under system32.
Notice: dll will be load from the first place it seached.
for exmple: Path=xxx;d:\PHP5;d:\openssl\bin;
if ssleay32.dll and libeay32.dll also appears under PHP5,
then python will load that one.
I'll write my own take on this due to my extreme frustration and hopes that this might help some other poor chap with his issues getting scrapy to work on windows..
Had a similar issues with faulty libeay32.dll and ssleay32.dll, or so it would seem, but confirmed it with Process Monitor.
It all went to hell after those dll's were found in System32:
So i wasnt suffering from wrong PATHs (but i did change their order so system32 comes before C:\Program Files (x86)\Intel\iCLS Client\.
Next i used this site the other guys here pointed to. Ive used Win32 OpenSSL v1.0.2j Light, since the newer one (1.1.0b) didnt generate these dlls for me for some reason.
Generate them in a different bin folder so you see they are new, and then copy them to system32. Voila, scrapy command line works.
Also, as of today, anaconda's scrapy 1.1.2 does not create a key file for running scrapy commands, ive opened and closed a question about this issue here.
As others have said above, there are two .dll files that are loaded instead of the openssl that is downloaded with Scrapy. Those two .dll files are not compatible.
I resolved this by placing the downloaded .dll files (i used Anaconda to install scrapy) as a higher priority in the Environment Variables as shown in the attached image.
In Environment Variables, within System Variables and the "Path" i placed the Anaconda3 location as shown above at the top and therefore makes it the priority when python comes to load openssl
As already mentioned in the other answers, the issue is caused by the two files ssleay32.dll and libeay32.dll.
The error occurred when importing OpenSSL in Python
>>> from OpenSSL import crypto, SSL
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "c:\Python27\lib\site-packages\OpenSSL\__init__.py", line 8, in <module>
from OpenSSL import crypto, SSL
ImportError: DLL load failed: The operating system cannot run %1.
Here's how I solved the issue on Windows 10. If you have git for Windows (download here) you can just copy these two files from Git\mingw64\bin\ to some location in your path. What worked for me was to copy them to c:\Windows (or else to the folder from which you're starting Python).
To find the locations of the dll files type at a cmd prompt:
>where libeay32.dll
C:\Program Files (x86)\Intel\TXE Components\TCS\libeay32.dll
...
C:\Users\me\AppData\Local\Programs\Git\mingw64\bin\libeay32.dll
I ran across this error today on a Windows 7 system. The problem for me was similar to #user2314737, #voetsjoeba, and #Olegp. libeay32.dll and ssleay32.dll where in folders related to Intel that were in the Windows path environmental variable.
Unfortunately, my system is locked (work computer), so I could not move the files or alter the system's path variable. Instead, I manually removed the Intel related items from the path variable that Python accesses using:
import os
os.environ['path'] = ';'.join(
filter(lambda x: 'intel' not in x.lower(), os.environ['path'].split(';'))
)
import OpenSSL
I was able to build and install Pydoop without errors, so, for example, I can do the following:
>>> import pydoop
>>> pydoop.__version__
'0.10.0'
However, when I try to import main Pydoop modules such as pipes or hdfs I'm getting ImportError:
>>> import pydoop.hdfs
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "pydoop/hdfs/__init__.py", line 79, in <module>
from fs import hdfs, default_is_local
File "pydoop/hdfs/fs.py", line 28, in <module>
hdfs_ext = pydoop.import_version_specific_module("_hdfs")
File "pydoop/__init__.py", line 111, in import_version_specific_module
return import_module(complete_mod_name(name))
File "/usr/lib/python2.7/importlib/__init__.py", line 37, in import_module
__import__(name)
ImportError: No module named _hdfs_2_0_0_cdh_4_3_0
In addition, when I try to use pydoop script I'm getting such a hint:
...
ImportError: /usr/local/lib/python2.7/dist-packages/pydoop/_pipes_2_0_0_cdh_4_3_0.so: undefined symbol: BIO_s_mem
BIO_s_mem is a symbol from libssl (OpenSSL), so I guess Pydoop can't find this shared library. I made sure it is available, ends with .so (as opposed to, say, .so.1) and is in LD_LIBRARY_PATH.
So what may be the reason for this error? How can I fix it (build options? environment variables?)
Any help is appreciated.
What OS version are you using? Try setting LD_PRELOAD to the path of your libssl, e.g.:
export LD_PRELOAD=/lib/x86_64-linux-gnu/libssl.so.1.0.0
Not sure about the pipes error, but I ran into your issue with _hdfs_2_0_0_cdh_4_3_0 (mine was a different version of hadoop, but I believe the problem is similar).
The setup.py script seems to want to make an egg file in /usr/local/lib/python2.7/dist-packages for pydoop, but the setup requires that it just be a folder (which will have that _hdfs_2_0_0_cdh_4_3_0.so file in it).
The solution is pretty simple: just delete /usr/local/lib/python2.7/dist-packages/pydoop-0.11.1.egg-info or the equivalent for you version.
I am trying to make an executable with python, py2exe using Enthought. The program.py starts with:
import pandas as pd
import pyper as pr
r=pr.R(use_pandas=True)
The setup.py is as below:
from distutils.core import setup
import py2exe
import matplotlib
import sys
opts={"py2exe":{"bundle_files"}}
sys.argv.append('py2exe')
opts = {'py2exe': {"bundle_files" : 3, "includes" : [ "matplotlib.backends", "matplotlib.backends.backend_qt4agg", "pylab","numpy", "matplotlib.backends.backend_tkagg"], 'excludes':['_gtkagg', '_tkagg', '_agg2', '_cairo', '_cocoaagg', '_fltkagg', '_gtk', '_gtkcairo'], 'dll_excludes': ['libgdk-win32-2.0-0.dll', 'libgobject-2.0-0.dll']}}
setup(console=['program.py'],zipfile=None,options=opts,data_files=matplotlib.get_py2exe_datafiles())
After running:python setup.py py2exe the two folders dist and build were created. But when launched program.exe I got this:
Dll load failed: The specified module could not be found
Traceback (most recent call last):
File "program.py", line 1, in module
File "pandas\__init__.pyc", line 6, in module
File "pandas\hashtable.pyc", line 12, in module
File "pandas\hashtable.pyc", line 10, in module
File "numpy.pxd.", line 157, in init pandas.hastable (pandas\hastable.c:19547)
File "numpy\__init__.pyc", line 143, in module
File "numpy\lib\add_newdocs.pyc", line 9, in module
File "numpy\lib\__init__.pyc", line 13, in module
File "numpy\lib\polynomial.pyc", line 17, in module
File "numpy\linalg\__init__.pyc", line 48, in module
File "numpy\linalg\linalg.pyc", line 23, in module
File "numpy\linalg\lapack_lite.pyc", line 12, in module
File "numpy\linalg\lapack_lite.pyc", line 10, in __load
ImportError: DLL load failed: The specified module could not be found**
I'm using Canopy 1.1.0 with Pandas 0.12.0-2 on a 64-Windows 7
Thanks for any help.
It looks as though the .exe built by py2exe may be missing the Intel MKL DLLs that come with Canopy (EPD). Depending on which version of Canopy (EPD) you have, the NumPy DLLs may be linked against the Intel MKL DLLs. Those DLLs should be in C:\Python27\Scripts directory (or the corresponding Scripts directory if you've installed EPD in a different place) -- look for DLLs whose filename starts with mk2.
EDIT: The path above works only for older EPD versions. For a more recent version of Canopy, the DLLS are typically located somewhere like C:\Users\<user>\AppData\Local\Enthought\Canopy\App\appdata\canopy-1.1.0-1371-win-x86_64\Scripts