Unable to install pyodbc on Linux

PythonLinuxCentosPyodbc

Python Problem Overview


I am running Linux (2.6.18-164.15.1.el5.centos.plus) and trying to install pyodbc. I am doing pip install pyodbc and get a very long list of errors, which end in

> error: command 'gcc' failed with exit status 1

I looked in /root/.pip/pip.log and saw the following:

> InstallationError: Command /usr/local/bin/python -c "import setuptools; file='/home/build/pyodbc/setup.py'; execfile('/home/build/pyodbc/setup.py')" install --single-version-externally-managed --record /tmp/pip-7MS9Vu-record/install-record.txt failed with error code 1

Has anybody had a similar issue installing pyodbc?

Python Solutions


Solution 1 - Python

On Ubuntu, you'll need to install unixodbc-dev:

sudo apt-get install unixodbc-dev

Install pip by using this command:

sudo apt-get install python-pip

once that is installed, you should be able to install pyodbc successfully:

pip install pyodbc

Solution 2 - Python

I resolved my issue by following correct directions on pyodbc - Building wiki which states:

> On Linux, pyodbc is typically built using the unixODBC headers, so you will need unixODBC and its headers installed. On a RedHat/CentOS/Fedora box, this means you would need to install unixODBC-devel: > > yum install unixODBC-devel

Solution 3 - Python

Adding one more answer on this question. For Linux Debian Stretch release you would need to install the following dependencies:

apt-get update
apt-get install unixodbc-dev
pip install pyodbc

On Debian Slim some users mentioned they needed to add g++

apt-get update
apt-get install g++ unixodbc-dev
pip install pyodbc

Solution 4 - Python

Struggled with the same issue

After running: sudo apt-get install unixodbc-dev

I was able to pip install pyodbc

Solution 5 - Python

Follow below steps to install pyodbc in any redhat version

yum install unixODBC unixODBC-devel
yum install gcc-c++
yum install python-devel
pip install pyodbc

Solution 6 - Python

I have referenced this question several times, and gone on to actually find the answer I was looking for here: pyodbc wiki

> To avoid gcc error on Ubuntu Linux, I did: > > sudo aptitude install g++ > > I also installed the following 2 packages from Synaptic: > > - python-dev > > - tdsodbc

Solution 7 - Python

Execute the following commands (tested on centos 6.5):

yum install install unixodbc-dev
yum install gcc-c++
yum install python-devel
pip install --allow-external pyodbc --allow-unverified pyodbc pyodbc

Solution 8 - Python

According to official Microsoft docs for Ubuntu 18.04 you should run next commands:

sudo su 
curl https://packages.microsoft.com/keys/microsoft.asc | apt-key add -
curl https://packages.microsoft.com/config/ubuntu/18.04/prod.list > /etc/apt/sources.list.d/mssql-release.list
apt-get update
ACCEPT_EULA=Y apt-get install msodbcsql17
exit

If you are using python3.7, it is very important to run:

sudo apt-get install python3.7-dev

Solution 9 - Python

A easy way to install pyodbc is by using 'conda'. As conda automatically installs required dependencies including unixodbc.

conda --ugrade all (optional)

then conda install pyodbc

it will install following packages:

libgfortran-ng: 7.2.0-hdf63c60_3 defaults mkl: 2018.0.3-1 defaults mkl_fft: 1.0.2-py36_0 conda-forge mkl_random: 1.0.1-py36_0 conda-forge numpy-base: 1.14.5-py36hdbf6ddf_0 defaults pyodbc: 4.0.17-py36_0 conda-forge unixodbc: 2.3.4-1 conda-forge

Solution 10 - Python

I had the same problem on CentOS 5.5

In addition to installing unixODBC-devel I also had to install gcc-c++

yum install gcc-c++

Solution 11 - Python

In my case (Amazon Linux AMI) none of the above worked. The following worked (idea from here):

  • Find the path to the file cc1plus. For me it was in /usr/libexec/gcc/x86_64-amazon-linux/4.8.5/cc1plus. For you it may vary a bit. Try ls -l /usr/libexec/gcc to find the proper directory name and go ahead.
  • Find directories in your path: echo $PATH (for me it was /sbin:/bin:/usr/sbin:/usr/bin:/opt/aws/bin)
  • Put a link to cc1plus in one of the directories in your PATH: sudo ln -s /PATH/TO/cc1plus /DIRinPATH/
    For example in my case:
    sudo ln -s /usr/libexec/gcc/x86_64-amazon-linux/4.8.5/cc1plus /usr/bin/

Solution 12 - Python

I needed all that, but I also needed python devel installed:

sudo yum install python-devel

Solution 13 - Python

How about installing pyobdc from zip file? From How to connect to Microsoft Sql Server from Ubuntu using pyODBC:

> Download source vs apt-get > > The apt-get utility in Ubuntu does have a version of pyODBC. (version 2.1.7).
However, it is badly out-of-date (2.1.7 vs 3.0.6) and may not work well with the newer versions of unixODBC and freetds.
This is especially important if you are trying to connect to later versions of Microsoft Sql Server (2008 onwards).
It is recommended that you use the latest versions of unixODBC, freetds and pyODBC when working with the latest Microsoft Sql Server instead of relying on packages in apt-get.

Solution 14 - Python

I know this is an old question, but the maintainer has a pyodbc GitHub Repo.

I also found a very good example for installing FreeTDS and setting up the config files.


Following the instructions on the GitHub docs seems to me to always be the best option. As of February, 2018, for CentOs7 (they have all flavors at the link) they say:

# Add the RHEL 6 library for Centos-7 of MSSQL driver. Centos7 uses RHEL-6 Libraries.
sudo su 
curl https://packages.microsoft.com/config/rhel/6/prod.repo > /etc/yum.repos.d/mssql-release.repo
exit

# Uninstall if already installed Unix ODBC driver
sudo yum remove unixODBC-utf16 unixODBC-utf16-devel #to avoid conflicts

# Install the  msodbcsql unixODBC-utf16 unixODBC-utf16-devel driver
sudo ACCEPT_EULA=Y yum install msodbcsql

#optional: for bcp and sqlcmd
sudo ACCEPT_EULA=Y yum install mssql-tools
echo 'export PATH="$PATH:/opt/mssql-tools/bin"' >> ~/.bash_profile
echo 'export PATH="$PATH:/opt/mssql-tools/bin"' >> ~/.bashrc
source ~/.bashrc

# optional: for unixODBC development headers
sudo yum install unixODBC-devel

# the Microsoft driver expects unixODBC to be here /usr/lib64/libodbc.so.1, so add soft links to the '.so.2' files
cd /usr/lib64
sudo ln -s libodbccr.so.2   libodbccr.so.1
sudo ln -s libodbcinst.so.2 libodbcinst.so.1
sudo ln -s libodbc.so.2     libodbc.so.1

# Set the path for unixODBC
export ODBCINI=/usr/local/etc/odbc.ini
export ODBCSYSINI=/usr/local/etc
source ~/.bashrc

# Prepare a temp file for defining the DSN to your database server
vi /home/user/odbcadd.txt

[MyMSSQLServer]
Driver      = ODBC Driver 13 for SQL Server
Description = My MS SQL Server
Trace       = No
Server      = 10.100.1.10

# register the SQL Server database DSN information in /etc/odbc.ini
sudo odbcinst -i -s -f /home/user/odbcadd.txt -l

# check the DSN installation with:
odbcinst -j
cat /etc/odbc.ini

# should contain a section called [MyMSSQLServer]

# install the python driver for database connection
pip install pyodbc

Solution 15 - Python

For archlinux/manjaro:

sudo pacman -S unixodbc

then:

sudo pip install pyodbc

or:

pip install pyodbc

You can upgrade your pip wheel setuptools before installing pyodbc (it won't affect the pyodbc installation) also with:

sudo python -m pip install --upgrade pip wheel setuptools

or

python -m pip install --upgrade pip wheel setuptools

Solution 16 - Python

I used this:

yum install unixODBC.x86_64

Depending on the version of centos could change the package, you can search like this:

yum search unixodbc

Solution 17 - Python

I faced with same issue. For python3.6.8 and ubuntu 16.04 none of above did not help me.

sudo apt-get install python3.6-dev

This solved my problem.

Solution 18 - Python

These 2 commands from here worked for me in RHEL 8

sudo dnf install redhat-rpm-config gcc-c++ python3-devel unixODBC-devel
pip3 install --user pyodbc

Attributions

All content for this solution is sourced from the original question on Stackoverflow.

The content on this page is licensed under the Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) license.

Content TypeOriginal AuthorOriginal Content on Stackoverflow
QuestionIgorGanapolskyView Question on Stackoverflow
Solution 1 - PythonJim ClouseView Answer on Stackoverflow
Solution 2 - PythonIgorGanapolskyView Answer on Stackoverflow
Solution 3 - PythonAlexis.RollandView Answer on Stackoverflow
Solution 4 - PythonCornelis KuijpersView Answer on Stackoverflow
Solution 5 - PythonJeevan ChaitanyaView Answer on Stackoverflow
Solution 6 - PythoncammilView Answer on Stackoverflow
Solution 7 - Pythonuser2550587View Answer on Stackoverflow
Solution 8 - PythonrfedorovView Answer on Stackoverflow
Solution 9 - PythonSalahView Answer on Stackoverflow
Solution 10 - PythonAndy MadgeView Answer on Stackoverflow
Solution 11 - PythonLoMaPhView Answer on Stackoverflow
Solution 12 - PythonIsaacView Answer on Stackoverflow
Solution 13 - Pythonjarppa79View Answer on Stackoverflow
Solution 14 - PythonMike WilliamsonView Answer on Stackoverflow
Solution 15 - PythonThái LêView Answer on Stackoverflow
Solution 16 - PythonOriana MaitaView Answer on Stackoverflow
Solution 17 - PythonSarper KumcuView Answer on Stackoverflow
Solution 18 - PythonMadat SardarliView Answer on Stackoverflow