.. _overview:

=====================

Installation overview

=====================

Some overview and some details that can help understanding the options when

installing Kallithea.

1. **Prepare environment and external dependencies.**

    Kallithea needs:

    * A filesystem where the Mercurial and Git repositories can be stored.

    * A database where meta data can be stored.

    * A Python environment where the Kallithea application and its dependencies

      can be installed.

    * A web server that can host the Kallithea web application using the WSGI

      API.

2. **Install Kallithea software.**

    This makes the ``kallithea-cli`` command line tool available.

3. **Create low level configuration file.**

    Use ``kallithea-cli config-create`` to create a ``.ini`` file with database

    connection info, mail server information, configuration for the specified

    web server, etc.

4. **Populate the database.**

    Use ``kallithea-cli db-create`` with the ``.ini`` file to create the

    database schema and insert the most basic information: the location of the

    repository store and an initial local admin user.

5. **Configure the web server.**

    The web server must invoke the WSGI entrypoint for the Kallithea software

    using the ``.ini`` file (and thus the database). This makes the web

    application available so the local admin user can log in and tweak the

    configuration further.

6. **Configure users.**

    The initial admin user can create additional local users, or configure how

    users can be created and authenticated from other user directories.

See the subsequent sections, the separate OS-specific instructions, and

:ref:`setup` for details on these steps.

Python environment

------------------

**Kallithea** is written entirely in Python_ and requires Python version

3.6 or higher.

Given a Python installation, there are different ways of providing the

environment for running Python applications. Each of them pretty much

corresponds to a ``site-packages`` directory somewhere where packages can be

installed.

Kallithea itself can be run from source or be installed, but even when running

from source, there are some dependencies that must be installed in the Python

environment used for running Kallithea.

- Packages *could* be installed in Python's ``site-packages`` directory ... but

  that would require running pip_ as root and it would be hard to uninstall or

  upgrade and is probably not a good idea unless using a package manager.

- Packages could also be installed in ``~/.local`` ... but that is probably

  only a good idea if using a dedicated user per application or instance.

- Finally, it can be installed in a virtualenv. That is a very lightweight

  "container" where each Kallithea instance can get its own dedicated and

  self-contained virtual environment.

We recommend using virtualenv for installing Kallithea.

Locale environment

------------------

In order to ensure a correct functioning of Kallithea with respect to non-ASCII

characters in user names, file paths, commit messages, etc., it is very

important that Kallithea is run with a correct `locale` configuration.

On Unix, environment variables like ``LANG`` or ``LC_ALL`` can specify a language (like

``en_US``) and encoding (like ``UTF-8``) to use for code points outside the ASCII

range. The flexibility of supporting multiple encodings of Unicode has the flip

side of having to specify which encoding to use - especially for Mercurial.

It depends on the OS distribution and system configuration which locales are

available. For example, some Docker containers based on Debian default to only

supporting the ``C`` language, while other Linux environments have ``en_US`` but not

``C``. The ``locale -a`` command will show which values are available on the

current system. Regardless of the actual language, you should normally choose a

locale that has the ``UTF-8`` encoding (note that spellings ``utf8``, ``utf-8``,

``UTF8``, ``UTF-8`` are all referring to the same thing)

For technical reasons, the locale configuration **must** be provided in the

environment in which Kallithea runs - it cannot be specified in the ``.ini`` file.

How to practically do this depends on the web server that is used and the way it

is started. For example, gearbox is often started by a normal user, either

manually or via a script. In this case, the required locale environment

variables can be provided directly in that user's environment or in the script.

However, web servers like Apache are often started at boot via an init script or

service file. Modifying the environment for this case would thus require

root/administrator privileges. Moreover, that environment would dictate the

settings for all web services running under that web server, Kallithea being

just one of them. Specifically in the case of Apache with ``mod_wsgi``, the

locale can be set for a specific service in its ``WSGIDaemonProcess`` directive,

using the ``lang`` parameter.

Installation methods

--------------------

Kallithea must be installed on a server. Kallithea is installed in a Python

environment so it can use packages that are installed there and make itself

available for other packages.

Two different cases will pretty much cover the options for how it can be

installed.

- The Kallithea source repository can be cloned and used -- it is kept stable and

  can be used in production. The Kallithea maintainers use the development

  branch in production. The advantage of installation from source and regularly

  updating it is that you take advantage of the most recent improvements. Using

  it directly from a DVCS also means that it is easy to track local customizations.

  Running ``pip install -e .`` in the source will use pip to install the

  necessary dependencies in the Python environment and create a

  ``.../site-packages/Kallithea.egg-link`` file there that points at the Kallithea

  source.

- Kallithea can also be installed from ready-made packages using a package manager.

  The official released versions are available on PyPI_ and can be downloaded and

  installed with all dependencies using ``pip install kallithea``.

  With this method, Kallithea is installed in the Python environment as any

  other package, usually as a ``.../site-packages/Kallithea-X-py3.8.egg/``

  directory with Python files and everything else that is needed.

  (``pip install kallithea`` from a source tree will do pretty much the same

  but build the Kallithea package itself locally instead of downloading it.)

.. note::

Web server

----------

Kallithea is (primarily) a WSGI_ application that must be run from a web

server that serves WSGI applications over HTTP.

Kallithea itself is not serving HTTP (or HTTPS); that is the web server's

responsibility. Kallithea does however need to know its own user facing URL

(protocol, address, port and path) for each HTTP request. Kallithea will

usually use its own HTML/cookie based authentication but can also be configured

to use web server authentication.

There are several web server options:

- Kallithea uses the Gearbox_ tool as command line interface. Gearbox provides

  ``gearbox serve`` as a convenient way to launch a Python WSGI / web server

  from the command line. That is perfect for development and evaluation.

  Actual use in production might have different requirements and need extra

  work to make it manageable as a scalable system service.

  Gearbox comes with its own built-in web server for development but Kallithea

  defaults to using Waitress_. Gunicorn_ and Gevent_ are also options. These

  web servers have different limited feature sets.

  The web server used by ``gearbox serve`` is configured in the ``.ini`` file.

  Create it with ``config-create`` using for example ``http_server=waitress``

  to get a configuration starting point for your choice of web server.

  (Gearbox will do like ``paste`` and use the WSGI application entry point

  ``kallithea.config.middleware:make_app`` as specified in ``setup.py``.)

- `Apache httpd`_ can serve WSGI applications directly using mod_wsgi_ and a

  simple Python file with the necessary configuration. This is a good option if

  Apache is an option.

- uWSGI_ is also a full web server with built-in WSGI module. Use

  ``config-create`` with ``http_server=uwsgi`` to get a ``.ini`` file with

  uWSGI configuration.

- IIS_ can also server WSGI applications directly using isapi-wsgi_.

- A `reverse HTTP proxy <https://en.wikipedia.org/wiki/Reverse_proxy>`_

  can be put in front of another web server which has WSGI support.

  Such a layered setup can be complex but might in some cases be the right

  option, for example to standardize on one internet-facing web server, to add

  encryption or special authentication or for other security reasons, to

  provide caching of static files, or to provide load balancing or fail-over.

  Nginx_, Varnish_ and HAProxy_ are often used for this purpose, often in front

  of a ``gearbox serve`` that somehow is wrapped as a service.

The best option depends on what you are familiar with and the requirements for

performance and stability. Also, keep in mind that Kallithea mainly is serving

dynamically generated pages from a relatively slow Python process. Kallithea is

also often used inside organizations with a limited amount of users and thus no

continuous hammering from the internet.

.. _Python: http://www.python.org/

.. _Gunicorn: http://gunicorn.org/

.. _Gevent: http://www.gevent.org/

.. _Waitress: http://waitress.readthedocs.org/en/latest/

.. _Gearbox: http://turbogears.readthedocs.io/en/latest/turbogears/gearbox.html

.. _PyPI: https://pypi.python.org/pypi

.. _Apache httpd: http://httpd.apache.org/

.. _mod_wsgi: https://code.google.com/p/modwsgi/

.. _isapi-wsgi: https://github.com/hexdump42/isapi-wsgi

.. _uWSGI: https://uwsgi-docs.readthedocs.org/en/latest/

.. _nginx: http://nginx.org/en/

.. _iis: http://en.wikipedia.org/wiki/Internet_Information_Services

.. _pip: http://en.wikipedia.org/wiki/Pip_%28package_manager%29

.. _WSGI: http://en.wikipedia.org/wiki/Web_Server_Gateway_Interface

.. _HAProxy: http://www.haproxy.org/

.. _Varnish: https://www.varnish-cache.org/

.. _npm: https://www.npmjs.com/

.. _setup:

=====

Setup

=====

Setting up Kallithea

--------------------

First, you will need to create a Kallithea configuration file. Run the

following command to do so::

    kallithea-cli config-create my.ini

This will create the file ``my.ini`` in the current directory. This

configuration file contains the various settings for Kallithea, e.g.

proxy port, email settings, usage of static files, cache, Celery

settings, and logging. Extra settings can be specified like::

    kallithea-cli config-create my.ini host=8.8.8.8 "[handler_console]" formatter=color_formatter

Next, you need to create the databases used by Kallithea. It is recommended to

use PostgreSQL or SQLite (default). If you choose a database other than the

default, ensure you properly adjust the database URL in your ``my.ini``

configuration file to use this other database. Kallithea currently supports

PostgreSQL, SQLite and MariaDB/MySQL databases. Create the database by running

the following command::

    kallithea-cli db-create -c my.ini

This will prompt you for a "root" path. This "root" path is the location where

Kallithea will store all of its repositories on the current machine. After

entering this "root" path ``db-create`` will also prompt you for a username

and password for the initial admin account which ``db-create`` sets

up for you.

The ``db-create`` values can also be given on the command line.

Example::

    kallithea-cli db-create -c my.ini --user=nn --password=secret --email=nn@example.com --repos=/srv/repos

The ``db-create`` command will create all needed tables and an

admin account. When choosing a root path you can either use a new

empty location, or a location which already contains existing

repositories. If you choose a location which contains existing

repositories Kallithea will add all of the repositories at the chosen

location to its database.  (Note: make sure you specify the correct

path to the root).

.. note:: the given path for Mercurial_ repositories **must** be write

          accessible for the application. It's very important since

          the Kallithea web interface will work without write access,

          but when trying to do a push it will fail with permission

          denied errors unless it has write access.

    kallithea-cli front-end-build

You are now ready to use Kallithea. To run it simply execute::

    gearbox serve -c my.ini

- This command runs the Kallithea server. The web app should be available at

  http://127.0.0.1:5000. The IP address and port is configurable via the

  configuration file created in the previous step.

- Log in to Kallithea using the admin account created when running ``db-create``.

- The default permissions on each repository is read, and the owner is admin.

  Remember to update these if needed.

- In the admin panel you can toggle LDAP, anonymous, and permissions

  settings, as well as edit more advanced options on users and

  repositories.

Internationalization (i18n support)

-----------------------------------

The Kallithea web interface is automatically displayed in the user's preferred

language, as indicated by the browser. Thus, different users may see the

application in different languages. If the requested language is not available

(because the translation file for that language does not yet exist or is

incomplete), English is used.

If you want to disable automatic language detection and instead configure a

fixed language regardless of user preference, set ``i18n.enabled = false`` and

specify another language by setting ``i18n.lang`` in the Kallithea

configuration file.

Using Kallithea with SSH

------------------------

Kallithea supports repository access via SSH key based authentication.

This means:

- repository URLs like ``ssh://kallithea@example.com/name/of/repository``

- all network traffic for both read and write happens over the SSH protocol on

  port 22, without using HTTP/HTTPS nor the Kallithea WSGI application

- encryption and authentication protocols are managed by the system's ``sshd``

  process, with all users using the same Kallithea system user (e.g.

  ``kallithea``) when connecting to the SSH server, but with users' public keys

  in the Kallithea system user's `.ssh/authorized_keys` file granting each user

  sandboxed access to the repositories.

- users and admins can manage SSH public keys in the web UI

- in their SSH client configuration, users can configure how the client should

  control access to their SSH key - without passphrase, with passphrase, and

  optionally with passphrase caching in the local shell session (``ssh-agent``).

  This is standard SSH functionality, not something Kallithea provides or

  interferes with.

- network communication between client and server happens in a bidirectional

  stateful stream, and will in some cases be faster than HTTP/HTTPS with several

  stateless round-trips.

.. note:: At this moment, repository access via SSH has been tested on Unix

    only. Windows users that care about SSH are invited to test it and report

    problems, ideally contributing patches that solve these problems.

Users and admins can upload SSH public keys (e.g. ``.ssh/id_rsa.pub``) through

the web interface. The server's ``.ssh/authorized_keys`` file is automatically

maintained with an entry for each SSH key. Each entry will tell ``sshd`` to run

``kallithea-cli`` with the ``ssh-serve`` sub-command and the right Kallithea user ID

when encountering the corresponding SSH key.

To enable SSH repository access, Kallithea must be configured with the path to

the ``.ssh/authorized_keys`` file for the Kallithea user, and the path to the

``kallithea-cli`` command. Put something like this in the ``.ini`` file::

    ssh_enabled = true

    ssh_authorized_keys = /home/kallithea/.ssh/authorized_keys

    kallithea_cli_path = /srv/kallithea/venv/bin/kallithea-cli

The SSH service must be running, and the Kallithea user account must be active

(not necessarily with password access, but public key access must be enabled),

all file permissions must be set as sshd wants it, and ``authorized_keys`` must

be writeable by the Kallithea user.

.. note:: The ``authorized_keys`` file will be rewritten from scratch on

    each update. If it already exists with other data, Kallithea will not

    overwrite the existing ``authorized_keys``, and the server process will

    instead throw an exception. The system administrator thus cannot ssh

    directly to the Kallithea user but must use su/sudo from another account.

    If ``/home/kallithea/.ssh/`` (the directory of the path specified in the

    ``ssh_authorized_keys`` setting of the ``.ini`` file) does not exist as a

    directory, Kallithea will attempt to create it. If that path exists but is

    *not* a directory, or is not readable-writable-executable by the server

    process, the server process will raise an exception each time it attempts to

    write the ``authorized_keys`` file.

.. note:: It is possible to configure the SSH server to look for authorized

   keys in multiple files, for example reserving ``ssh/authorized_keys`` to be

   used for normal SSH and with Kallithea using

   ``.ssh/authorized_keys_kallithea``. In ``/etc/ssh/sshd_config`` set

   ``AuthorizedKeysFile .ssh/authorized_keys .ssh/authorized_keys_kallithea``

   and restart sshd, and in ``my.ini`` set ``ssh_authorized_keys =

   /home/kallithea/.ssh/authorized_keys_kallithea``. Note that this new

   location will apply to all system users, and that multiple entries for the

   same SSH key will shadow each other.

.. warning:: The handling of SSH access is steered directly by the command

    specified in the ``authorized_keys`` file. There is no interaction with the

    web UI.  Once SSH access is correctly configured and enabled, it will work

    regardless of whether the Kallithea web process is actually running. Hence,

    if you want to perform repository or server maintenance and want to fully

    disable all access to the repositories, disable SSH access by setting

    ``ssh_enabled = false`` in the correct ``.ini`` file (i.e. the ``.ini`` file

    specified in the ``authorized_keys`` file.)

The ``authorized_keys`` file can be updated manually with ``kallithea-cli

ssh-update-authorized-keys -c my.ini``. This command is not needed in normal

operation but is for example useful after changing SSH-related settings in the

``.ini`` file or renaming that file. (The path to the ``.ini`` file is used in

the generated ``authorized_keys`` file).

Setting up Whoosh full text search

----------------------------------

Kallithea provides full text search of repositories using `Whoosh`__.

.. __: https://whoosh.readthedocs.io/en/latest/

For an incremental index build, run::

    kallithea-cli index-create -c my.ini

For a full index rebuild, run::

    kallithea-cli index-create -c my.ini --full

The ``--repo-location`` option allows the location of the repositories to be overridden;

usually, the location is retrieved from the Kallithea database.

The ``--index-only`` option can be used to limit the indexed repositories to a comma-separated list::

    kallithea-cli index-create -c my.ini --index-only=vcs,kallithea

To keep your index up-to-date it is necessary to do periodic index builds;

for this, it is recommended to use a crontab entry. Example::

    0  3  *  *  *  /path/to/virtualenv/bin/kallithea-cli index-create -c /path/to/kallithea/my.ini

When using incremental mode (the default), Whoosh will check the last

modification date of each file and add it to be reindexed if a newer file is

available. The indexing daemon checks for any removed files and removes them

from index.

If you want to rebuild the index from scratch, you can use the ``-f`` flag as above,

or in the admin panel you can check the "build from scratch" checkbox.

Integration with issue trackers

-------------------------------

Kallithea provides a simple integration with issue trackers. It's possible

to define a regular expression that will match an issue ID in commit messages,

and have that replaced with a URL to the issue.

This is achieved with following three variables in the ini file::

    issue_pat = #(\d+)

    issue_server_link = https://issues.example.com/{repo}/issue/\1

    issue_sub =

``issue_pat`` is the regular expression describing which strings in

commit messages will be treated as issue references. The expression can/should

have one or more parenthesized groups that can later be referred to in

``issue_server_link`` and ``issue_sub`` (see below). If you prefer, named groups

can be used instead of simple parenthesized groups.

If the pattern should only match if it is preceded by whitespace, add the

following string before the actual pattern: ``(?:^|(?<=\s))``.

If the pattern should only match if it is followed by whitespace, add the

following string after the actual pattern: ``(?:$|(?=\s))``.

These expressions use lookbehind and lookahead assertions of the Python regular

expression module to avoid the whitespace to be part of the actual pattern,

otherwise the link text will also contain that whitespace.

Matched issue references are replaced with the link specified in

``issue_server_link``, in which any backreferences are resolved. Backreferences

can be ``\1``, ``\2``, ... or for named groups ``\g<groupname>``.

The special token ``{repo}`` is replaced with the full repository path

(including repository groups), while token ``{repo_name}`` is replaced with the