<html>
<head>
<link rel="stylesheet" href="style.css" type="text/css">
</head>
<body>
<hr>
<h1>The Hydra Buildfarm User Manual</h1>
<h2>Draft (Version 0.1)</h2>
<p/>
<h3>Eelco Dolstra and Eelco Visser</h3>
<p/>
Delft University of Technology
<p/>
Department of Software Technology
<p/>
Copyright 2008 Eelco Dolstra
<hr>
<h2>Chapter 1. Introduction</h2>
<h3>1.1. About Hydra</h3>
Hydra is a buildfarm system based on the Nix software deployment
system.
<p/>
... advantages ...
<p/>
<p/>
<p/>
<h3>1.2. Hydra at nixos.org</h3>
The nixos.org installation of Hydra runs at
<blockquote>
<a href="http://hydra.nixos.org/">http://hydra.nixos.org</a>
</blockquote>
That installation is used to build software components from the
<a href="http://nixos.org">Nix</a>,
<a href="http://nixos.org/nixos">NixOS</a>,
<a href="http://strategoxt.org">Stratego/XT</a>,
and related projects.
If you are one of the developers on those projects, it is likely that
you will be using the NixOS Hydra server in some way. If you need to
administer automatic builds for your project, you should pull the
right strings to get an account on the server. This manual will tell
you how to set up new projects and build jobs within those projects
and write a release.nix file to describe the build process of your
project to Hydra. You can skip Chapter 2.
<p/>
If your project does not yet have automatic builds within the NixOS
Hydra server, it may actually be eligible. We are in the process of
setting up a large buildfarm that should be able to support open
source and academic software projects. Get in touch.
<h3>1.3. Hydra on your own buildfarm</h3>
If you need to run your own Hydra installation, Chapter 2 explains
how to download and install the system on your own server.
<h2>Chapter 2. Installation and Configuration</h2>
This chapter explains how to install Hydra on your own buildfarm server.
We assume that you have
<h3>2.1. Platform Requirements</h3>
To run Hydra you need a Linux server with at least a considerable
amount of diskspace to store builds. A multi-core machine is not a
waste since Hydra can schedule multiple simultaneous build jobs.
<p/>
Of course we think it is a good idea to use the <a
href="http://nixos.org/nixos">NixOS</a> Linux distribution for your
buildfarm server. But this is not a requirement. The Nix software
deployment system can be installed on any Linux distribution in
parallel to the regular package management system. Thus, you can use
Hydra on a Suse, Fedora, or Ubuntu system.
<p/>
Hydra on Windows??
<h3>2.2. Getting Nix</h3>
If your server runs NixOS you are all set to continue with
installation of Hydra. Otherwise you first need to install Nix.
The <a href="http://nixos.org/releases/nix/nix-0.12">latest stable release</a> is Nix 0.12.
Installation instructions can be found in the
<a href="http://nixos.org/releases/nix/nix-0.12/manual/">Nix User's Guide</a>.
<p/>
Why Nix ...
<h3>2.3. Installation</h2>
To install Hydra, get the most recent 'closure' available from
<blockquote>
<a href="http://hydra.nixos.org/releases/hydra/unstable">http://hydra.nixos.org/releases/hydra/unstable</a>
</blockquote>
And follow the instructions that are revealed by clicking [help].
<pre class="programlisting">
$ gunzip < hydra-build.closure.gz | nix-store --import
</pre>
This unpacks the closure and imports its components into the Nix store.
<pre class="programlisting">
$ nix-env -i /nix/store/...-hydra-build
</pre>
This makes the tools in the Hydra package available in your Nix user
environment.
<p/>
Command completion should then reveal a number of tools related to
hydra installed:
<pre class="programlisting">
$ hydra_<tab>
hydra_build.pl hydra_fastcgi.pl hydra_scheduler.pl
hydra_cgi.pl hydra_init.pl hydra_server.pl
hydra_create.pl hydra_queue_runner.pl hydra_test.pl
</pre>
<h3>2.4. Configuration</h3>
The Hydra software is installed in the Nix store, but to run it needs
a directory for storing the database, logs, and session data. In your
<code>.bashrc</code> or similar configuration file define:
<pre class="programlisting">
export HYDRA_DATA=/usr/local/hydra
</pre>
and make sure that you actually create that directory. (Of course, you
can use another directory, but then remember to also substitute that
name in the commands below.)
<p/>
Run <code>hydra_init.pl</code> to initialize the database
<pre class="programlisting">
$ hydra_init.pl
</pre>
Run <code>hydra_server.pl</code> to start the webserver at <a href="http://localhost:3000">http://localhost:3000</a>
<pre class="programlisting">
$ hydra_server.pl
</pre>
Also start the scheduler, which monitors the source repositories and
adds builds to the queue, and the runner, which executes jobs in the
queue.
<pre class="programlisting">
$ hydra_scheduler.pl
$ hydra_queue_runner.pl
</pre>
Now your Hydra server should be up and running and the web interface
operational.
<h3>2.5. User Administration</h3>
To be able to add jobs and create projects you need to register users
in the Hydra database. In the current version, the web interface does
not yet support user administration. Use the following command to add
a new user to the database.
<pre class="programlisting">
$ sqlite3 /usr/local/hydra/hydra.sqlite "insert into Users(userName, emailAddress, password) values('eelco', 'blablah@example.org', '$(echo -n foobar | sha1sum | cut -c1-40)');"
</pre>
where <code>eelco</code> is the username, and <code>foobar</code> the
password. (Make sure to use other values!)
<p/>
To give this user administrator privileges, follow this up by:
<pre class="programlisting">
$ sqlite3 /usr/local/hydra/hydra.sqlite "insert into UserRoles(userName, role) values('eelco', 'admin');"
</pre>
Now you should be able to create a project using the Hydra web interface.
<h2>Chapter 3. Creating Projects</h2>
The next step is to add projects to the buildfarm. We follow the
example of the patchelf project at hydra.nixos.org. Note that the
error messages provided as feedback by the webinterface can be
somewhat unfriendly in the current version.
<p/>
<a href="http://localhost:3000/login">Login</a>
to the webinterface of your Hydra installation using
the username and password you inserted in the database.
Then follow the
'<a href="http://localhost:3000/createproject">Create Project</a>'
link to create a new project.
<h3>3.1. General information</h3>
A project definition consists of some general information
The general information of a project are mainly its name and
owner. Here's what we fill in for the patchelf project:
<pre class="programlisting">
Identifier: patchelf
Display name: Patchelf
Description: A tool for modifying ELF binaries
Owner: eelco
Enabled: Yes
</pre>
The <code>Identifier</code> should be a unique name (it is the primary
database key for the project table in the database). If you try to
create a project with an already existing identifier you'd get an
error message such as:
<pre>
I'm very sorry, but an error occurred:
DBIx::Class::ResultSet::create(): DBI Exception: DBD::SQLite::st execute failed: column name is not unique(19) at dbdimp.c line 402
</pre>
So try to create the project after entering just the general
information to figure out if you have chosen a unique name.
Jobsets can be added once the project has been created.
<p/>
Once create there should be an entry for the project in the sidebar. Go to
the project page for the
<a href="http://localhost:3000/project/patchelf">Patchelf</a>
project.
<h3>Jobsets</h3>
A project can consist of multiple `jobsets', separate tasks that can
be built separately, but may depend on each other (without cyclic
dependencies, of course). Go to the
<a href="http://localhost:3000/project/patchelf/edit">Edit</a>
page of the Patchelf project and 'Add a new jobset'
by providing the following 'Information':
<pre>
Identifier: trunk
Description: Trunk
Nix expression: release.nix in input patchelfSrc
</pre>
This states that in order to build the 'Trunk' jobset, the Nix
expression in the file 'release.nix', which can be obtained from input
'patchelfSrc', should be evaluated. (We'll have a look at release.nix
later.)
<p/>
To realize a job we probably need a number of inputs, which can be
declared in the table below. As many inputs as required can be added.
For patchelf we declare the following inputs.
<pre>
nixpkgs 'CVS checkout' https://svn.nixos.org/repos/nix/nixpkgs/trunk
</pre>
<pre>
officialRelease Boolean false
</pre>
<pre>
patchelfSrc Subversion checkout https://svn.nixos.org/repos/nix/patchelf/trunk
</pre>
<pre>
system String value "i686-linux"
</pre>
nixpkgs
officialRelease
patchelfSrc
system
<h3>Release Set</h3>
there must be one primary job
check the radio button of exactly one job
https://svn.nixos.org/repos/nix/nixpkgs/trunk
<h3>Building Jobs</h3>
<h3>release.nix</h3>
- Voorbeelden van Nix expressies voor Hydra:
https://svn.nixos.org/repos/nix/patchelf/trunk/release.nix
https://svn.nixos.org/repos/nix/nix/trunk/release.nix
https://svn.nixos.org/repos/nix/hydra/trunk/release.nix
<h3>Building on the command line</h3>
Overigens zijn die helemaal niet Hydra-specifiek, je kunt ze gewoon vanaf de
command line bouwen, bijv. als je een patchelf checkout hebt (met een nixpkgs
checkout in ../nixpkgs):
$ nix-build release.nix -A rpm_fedora10i386
<h3>Release Set</h3>
</body>
</html>