# ABSTRACT: Tips for installing Pinto package Pinto::Manual::Installing; #------------------------------------------------------------------------------ # VERSION #------------------------------------------------------------------------------ 1; __END__ =pod =encoding UTF-8 =for :stopwords Jeffrey Ryan Thalhammer =head1 NAME Pinto::Manual::Installing - Tips for installing Pinto =head1 VERSION version 0.14 =head1 SYNOPSIS For the impatient... curl -L http://getpinto.stratopan.com | bash source ~/opt/local/pinto/etc/bashrc And then possibly... echo source ~/opt/local/pinto/etc/bashrc >> ~/.bashrc =head1 APPLICATION VERSUS LIBRARY VERSUS SERVER For most situations, Pinto is more like an I than a I. It is a tool that you use to develop and manage your code, but Pinto itself is not I your code. Pinto also has a lot of dependencies, some of which may conflict with or complicate your code. Pinto can also serve as the backend supporting a daemonized Starman server exposed to the wilds of the internet. Doing so opens the door to many additional security concerns. We suggest below some practices we hope will serve to minimize the risks of doing so. =head1 INSTALLING AS AN APPLICATION For the reasons above, I recommend installing Pinto as a stand-alone application in its own sandbox. That way, it doesn't pollute your environment with its dependencies. Nor will you pollute Pinto with changes to your environment, so Pinto will function even when your other environment dependencies are broken. And hopefully, you can use Pinto to help fix whatever broke! =head2 Step 1: Run the pinto installer script The installer script at L is mostly just a wrapper around L, which installs pinto in a self-contained directory: # If you use curl... curl -L http://getpinto.stratopan.com | bash # If you use wget... wget -O - http://getpinto.stratopan.com | bash All the dependent modules will come from a curated repository on L. These aren't always the latest versions of things, but they are versions that I know will work. =head2 Step 2: Set up the pinto environment The pinto installer generates a setup script for you. By default, it is located at F<~/opt/local/pinto/etc/bashrc>. To load that setup into your current shell, just give this command: source ~/opt/local/pinto/etc/bashrc To make these settings part of your everyday shell environment, just add that last command to your F<~/.profile> or F<~/.bashrc> or whatever setup file is appropriate for your shell. If you wish to customize any of the other environment variables that pinto uses, you can place those commands in F<~/.pintorc>. If that file exists, the setup script will source them as well. See L for a list of the relevant environment variables. =head1 INSTALLING AS A SERVER If you will be running the pintod daemon exposed to the internet, it is suggested that you assume root privileges and proceed as follows: (1) create a pinto user like so: adduser --system --home /opt/local/pinto --shell /bin/false \ --disabled-login --group pinto (2) set some environmental variables: export PINTO_HOME=/opt/local/pinto export PINTO_REPOSITORY_ROOT=/var/pinto check that the exports took with `env`. (3) run the installer as described above, and source the environmental variables to facilitate the steps of setting up the repository. (4) choose an authentication backend and install it like so: cpanm -L $PINTO_HOME Authen::Simple::Kerberos to review your options see L. (5) choose an appropriate startup script and install it: cp $PINTO_HOME/etc/init.d/pintod.debian /etc/init.d/pintod update-rc.d pintod start 50 2 3 4 5 . stop 20 0 1 6 . =head2 AN IMPORTANT NOTE ABOUT SECURITY Currently daemonizing the pintod server will run the starman workers as root. We hope to soon have the pintod daemon drop its privileges after initiating the master and before spawning the workers, so that the workers will run as the pinto user. Until that feature is in place, pinto repository administrators are urged to keep their installations safely behind firewalls, protected from the potentially hostile user. =head1 INSTALLING AS A LIBRARY If you're going to be hacking on Pinto itself, or want to try building on the API directly, then you can install Pinto straight into your development environment, just like you would do for any other module. Just beware that Pinto has lots of dependencies. And if you subsequently upgrade any of those dependencies to something that breaks Pinto, then you might find yourself in a pickle. The whole point of Pinto is to help you manage your dependencies, so if you break Pinto, it won't be able to help you. =head1 OTHER INSTALLATION OPTIONS Naturally, installation procedures will vary from one environment to another. If this procedure doesn't work for you, or if you'd like to suggest a procedure for a different environment (e.g. Windows, Perlbrew, Strawberry Perl, etc.), then please contact me. Your contributions would be greatly appreciated. =head1 SEE ALSO L L L (the library) L (the command) =head1 AUTHOR Jeffrey Ryan Thalhammer =head1 COPYRIGHT AND LICENSE This software is copyright (c) 2015 by Jeffrey Ryan Thalhammer. This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself. =cut