In igraph/rigraph: Network Analysis and Visualization

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

This page details common problems encountered when trying to install igraph. If you did not find an answer to your problem here, feel free to ask for help on https://igraph.discourse.group/. When you do so, be sure to state:

1. the output of sessionInfo()
2. where you obtained R from
3. the complete output from running install.packages("igraph") in a fresh R session

Cannot compile igraph from sources on Windows or macOS

Do not compile igraph from sources, unless you know what you are doing! It is much more convenient to use the binaries from CRAN instead. These can be installed using install.packages("igraph").

Usually, it takes a few days for binary releases to become available on CRAN after each new igraph source release. During this period, install.packages() will give you a choice about using the latest source release or the previous binary release. Just choose to use the binary release, i.e. respond "no" to "Do you want to install from sources?"

> install.packages('igraph')

  There is a binary version available but the source
  version is later:
       binary source needs_compilation
igraph  1.2.7  1.2.8              TRUE

Do you want to install from sources the package which needs compilation? (Yes/no/cancel) no

CRAN provides Windows and macOS binaries only for the last two minor releases of R (e.g. 4.5 and 4.4), but not for older ones (e.g. 4.3). Make sure that you are using one of these supported R versions. Refer to https://r-project.org/ to find out what the latest R version is at the moment.

Note that the pre-compiled binaries on CRAN are only compatible with the R distribution provided by CRAN itself. If you obtained R from different sources, such as Homebrew or MacPorts on macOS, you will not be able to use these, and the only way to install packages will be to compile them from sources.

I still want to compile igraph from sources

If you decide to try to compile from sources anyway, basic requirements are listed below. It is assumed that you are comfortable compiling software from sources and resolving basic issues as they arise. Otherwise, please use the binaries.

• Make sure you have a compatible toolchain installed, including a Fortran compiler. On macOS, make sure you have Xcode and gfortran from here. For Windows, you will need Rtools.
• On Windows, see the instructions for installing Rtools for your version of R. Check the "Using pacman" link in the Rtools 4.0 instructions. You will find a copyable pacman command for installing all dependencies.
• On Windows, make sure you have the GLPK C library installed. This is not the same as the Rglpk R package. Instructions for installing libraries for Rtools4 on Windows are here

Cannot compile igraph from sources on Linux

Since CRAN does not provide binary packages for Linux, Linux users usually need to compile igraph from sources. In order to do so, make sure that you have all the prerequisites:

• C, C++ and Fortran compilers. On Debian-based distros, use sudo apt install build-essential gfortran. On Fedora, use yum install gcc gcc-c++ gcc-gfortran.
• Development packages for glpk and libxml2. On Debian-based distros, use sudo apt install libglpk-dev libxml2-dev. On Fedora, use yum install glpk-devel libxml2-devel.

If you have an Anaconda environment active, deactivate it before compiling igraph. Having conda environments active are known to cause a number of issues, both during compilation and later during usage.

Set up your Anaconda installation so no environment is active by default. Make sure that Anaconda's bin directory is not present in your PATH environment variable. Do not add this directory to the path manually. Instead, let Anaconda manage it using conda init, as recommended by Anaconda. Only Anaconda's condabin, but not bin, should be present in the PATH.

fatal error: glpk.h: No such file or directory

If you are using Windows or macOS, please refer to Cannot compile igraph from sources on Windows or macOS.

If you are using Linux, please refer to Cannot compile igraph from sources on Linux.

libgfortran.so.4: cannot open shared object file: No such file or directory

This problem occurs when there are multiple incompatible gfortran versions on your machine. Most commonly, one of them comes from an active Anaconda environment. Always deactivate all Anaconda environments using conda deactivate before installing igraph using install.packages().

Set up your Anaconda installation so no environment is active by default. Make sure that Anaconda's bin directory is not present in your PATH environment variable. Do not add this directory to the path manually. Instead, let Anaconda manage it using conda init, as recommended by Anaconda. Only Anaconda's condabin, but not bin, should be present in the PATH.

libmkl_rt.so.1: cannot open shared object file: No such file or directory

This issue typically occurs when an Anaconda environment containing conflicting BLAS libraries is active. Please refer to the instructions in the section "libgfortran.so.4: cannot open shared object file".

libglpk.so.40: cannot open shared object file: No such file or directory

This error will occur on Linux systems when igraph was compiled with GLPK support, but GLPK is not currently installed on the system. To resolve it, install GLPK. On Debian-based systems, simply use sudo apt install libglpk40.

This may happen when a binary version of igraph is installed from https://packagemanager.rstudio.com. These binaries assume that GLPK is present on your system.

GLPK is not available, Unimplemented function call

This error occurs when calling an igraph function that relies on GLPK, but igraph was compiled without GLPK support. This cannot happen in igraph 2.0.3 and later. If you are compiling an earlier version, please refer to "Cannot compile igraph from sources on Linux" for instructions on compiling igraph with GLPK support.

igraph/rigraph documentation built on June 13, 2025, 1:44 p.m.