MAAS follows the Launchpad Python Style Guide, except where it gets
Launchpad specific, and where it talks about method naming. MAAS
instead adopts PEP-8 naming in all cases, so method names should
usually use the lowercase_with_underscores form.
There's a configure-lxd-profile script in utilities, that will set up a LXD profile that is configured properly.
You can grab MAAS's code manually from Launchpad but Git makes it easy to fetch the last version of the code. First of all, install Git:
$ sudo apt install git
Then go into the directory where you want the code to reside and run:
$ git clone https://git.launchpad.net/maas && cd maas
MAAS depends on Postgres, isc-dhcp, bind9, and many other packages. To install everything that's needed for running and developing MAAS, run:
$ make install-dependencies
Careful: this will apt-get install many packages on your system,
via sudo. It may prompt you for your password.
This will install bind9. As a result you will have an extra daemon
running. If you are a developer and don't intend to run BIND locally,
you can disable the daemon by inserting exit 1 at the top of
/etc/default/bind9. The package still needs to be installed for
tests though.
Python development dependencies are pulled automatically from
PyPI when buildout runs. (buildout will be automatically
configured to create a cache, in order to improve build times.
See utilities/configure-buildout.)
Javascript development dependencies are pulled automatically from
npm when make runs. (npm will be automatically
configured to use a cache, in order to improve build times.)
You will want to adjust your git repository of lp:maas some before you start making changes to the code. This includes setting up your own copy of the repository and making your changes in branches.
First you will want to rename the origin remote to upstream and create a new origin in your namespace.
$ git remote rename origin upstream $ git remote add origin git+ssh://{launchpad-id}@git.launchpad.net/~{launchpad-id}/maas
Now you can make a branch and start making changes.
$ git checkout -b new-branch
Once you have made the changes you want, you should commit and push the branch to your origin.
$ git commit -m "My change" -a $ git push origin new-branch
Now you can view that branch on Launchpad and propose it to the maas repository.
Once the branch has been merged and your done with it you can update your git repository to remove the branch.
$ git fetch upstream $ git checkout master $ git merge upstream/master $ git branch -d new-branch
The PyCharm IDE is a useful tool when developing MAAS. The MAAS team does not
endorse any particular IDE, but .idea project files are included with
MAAS, so PyCharm is an easy choice.
To run the whole suite:
$ make test
To run tests at a lower level of granularity:
$ ./bin/test.region src/maasserver/tests/test_api.py $ ./bin/test.region src/maasserver/tests/test_api.py:AnonymousEnlistmentAPITest
The test runner is nose, so you can pass in options like
--with-coverage and --nocapture (short option: -s). The
latter is essential when using pdb so that stdout is not
adulterated.
Note
When running make test through ssh from a machine with locales
that are not set up on the machine that runs the tests, some tests
will fail with a MismatchError and an "unsupported locale
setting" message. Running locale-gen for the missing locales or
changing your locales on your workstation to ones present on the
server will solve the issue.
Pass the --with-subunit flag to any of the test runners (e.g.
bin/test.rack) to produce a subunit stream of test results. This
may be useful for parallelising test runs, or to allow later analysis of
a test run. The optional --subunit-fd flag can be used to direct the
results to a different file descriptor, to ensure a clean stream.
The JavaScript tests are run using Karma. Chromium and PhantomJS are the default browser but any browser supported by Karma can be used to run the tests.:
$ ./bin/test.js
If you want to run the JavaScript tests in debug mode so you can inspect the code inside of a running browser you can launch Karma manually.:
$ ./bin/karma start src/maastesting/karma.conf.js --browsers Chrome --no-single-run
When MAAS is installed from packaging it can help to enable debugging features to triage issues.
By default MAAS only logs HTTP 500 - INTERNAL_SERVER_ERROR into the regiond.log. To enable logging of all exceptions even exceptions where MAAS will return the correct HTTP status code.:
$ sudo sed -i 's/DEBUG = False/DEBUG = True/g' \ > /usr/lib/python3/dist-packages/maasserver/djangosettings/settings.py $ sudo service maas-regiond restart
It can help when debugging to run regiond a foreground process so you can interact with the regiond by placing a breakpoint in the code. Once you have placed a breakpoint into the code you want to inspect you can start the regiond process in the foreground.:
$ sudo service maas-regiond stop $ sudo -u maas -H \ > DJANGO_SETTINGS_MODULE=maasserver.djangosettings.settings \ > twistd3 --nodaemon --pidfile= maas-regiond
Note
By default a MAAS installation runs 4 regiond processes at the same time. This will change it to only run 1 process in the foreground. This should only be used for debugging. Once finished the breakpoint should be removed and maas-regiond service should be started.
It can help when debugging to run rackd a foreground process so you can interact with the rackd by placing a breakpoint in the code. Once you have placed a breakpoint into the code you want to inspect you can start the rackd process in the foreground.:
$ sudo service maas-rackd stop $ sudo -u maas -H /usr/bin/authbind --deep /usr/bin/twistd3 --nodaemon --pidfile= maas-rackd
Access to the database is configured in
src/maasserver/djangosettings/development.py.
The Makefile or the test suite sets up a development database
cluster inside your branch. It lives in the db directory, which
gets created on demand. You'll want to shut it down before deleting a
branch; see below.
First, set up the project. This fetches all the required dependencies
and sets up some useful commands in bin/:
$ make
Create the database cluster and initialise the development database:
$ make syncdb
Optionally, if all you want to do is to take a look around the UI and API, without interacting with real machines or VMs, populate your database with the sample data:
$ make sampledata
You can login as a simple user using the test account (username: 'test', password: 'test') or the admin account (username: 'admin', password: 'test').
If you want to interact with real machines or VMs, it's better to use the snap. Instead of building a real snap, though, you can use 'snapcraft prime' to create the prime directory. That has all the contents of the snap, but it's in a plain directory insted of in a squashfs image. Using a directory is better for testing, since you can change the files in there and not rebuild the snap.
There's a sync-dev-snap make target to automate this:
$ make sync-dev-snap
The sync-dev-snap target creates a clean copy of your working tree (so
that you don't have to run 'make clean' before building the snap) in
build/dev-snap and creates the snap directory in build/dev-snap/prime.
You can now install the snap:
$ sudo snap try --devmode build/dev-snap/prime
Note that 'snap try' is used instead of 'snap install'. The maas snap should now be installed:
$ snap list Name Version Rev Developer Notes core 16-2.27.5 2774 canonical core maas 2.3.0~alpha3-6225-gaa05ba6-snap x1 devmode,try
Next you need to initialize the snap, just like you would normally do:
$ sudo maas init
And now you're ready to make changes to the code. After you've change
some source files and want to test them out, run the sync-dev-snap
target again:
$ make sync-dev-snap
You should now see that you files were synced to the prime directory. If you changed JS and HTML files only, you should see that changes straight away by just reloading the browser. If you changed Python files, you need to restart MAAS:
$ sudo service snap.maas.supervisor restart
VMs or even real machines can now PXE boot off your development snap. But of course, you need to set up the networking first. If you want to do some simple testing, the easiest is to create a networking in virt-manager that has NAT, but doesn't provide DHCP. If the name of the bridge that got created is virbr1, you can expose it to your container as eth1 using the following config:
- eth1:
- name: eth1 nictype: bridged parent: virbr1 type: nic
Of course, you also need to configure that eth1 interface. Since MAAS is the one providing DHCP, you need to give it a static address on the network you created. For example:
auto eth1 iface eth1 inet static address 192.168.100.2 netmask 255.255.255.0
Note that your LXD host will have the .1 address and will act as a gateway for your VMs.
To shut down the database cluster and clean up all other generated files in your branch:
$ make clean
To use PXE booting, each cluster controller needs to download several files relating to PXE booting. This process is automated, but it does not start by default.
First create a superuser and start all MAAS services:
$ bin/maas-region createadmin $ make run
Substitute your own email. The command will prompt for a choice of password.
Next, get the superuser's API key on the account preferences page in the web UI, and use it to log into MAAS at the command-line:
$ bin/maas login dev http://localhost:5240/MAAS/
Start downloading PXE boot resources:
$ bin/maas dev node-groups import-boot-images
This sends jobs to each cluster controller, asking each to download the boot resources they require. This may download dozens or hundreds of megabytes, so it may take a while. To save bandwidth, set an HTTP proxy beforehand:
$ bin/maas dev maas set-config name=http_proxy value=http://...
You will need to run the built-in TFTP server on the real TFTP port (69) if you want to boot some real hardware. By default, it's set to start up on port 5244 for testing purposes. To make it run on port 69, set the MAAS_TFTP_PORT environment variable before running make run/start:
export MAAS_TFTP_PORT=69
Then you need install and configure the authbind, so that your user can bind to port 69:
* Install the ``authbind``package: $ sudo apt install authbind * Create a file ``/etc/authbind/byport/69`` that is *executable* by the user running MAAS. $ sudo touch /etc/authbind/byport/69 $ sudo chown $USER /etc/authbind/byport/69 $ sudo chmod u+x /etc/authbind/byport/69
Now when starting up the MAAS development webserver, "make run" and "make start" will detect authbind's presence and use it automatically.
There's a BIND daemon that is started up as part of the development service but it runs on port 5246 by default. If you want to make it run as a real DNS server on the box then set the MAAS_BIND_PORT environment variable before running make run/start:
export MAAS_BIND_PORT=53
Then as for TFTP above, create an authbind authorisation:
$ sudo touch /etc/authbind/byport/53 $ sudo chown $USER /etc/authbind/byport/53 $ sudo chmod u+x /etc/authbind/byport/53
and run as normal.
The cluster also needs authbind as it needs to bind a socket on UDP port 68 for DHCP probing:
$ sudo touch /etc/authbind/byport/68 $ sudo chown $USER /etc/authbind/byport/68 $ sudo chmod u+x /etc/authbind/byport/68
If you omit this, nothing else will break, but you will get an error in the cluster log because it can't bind to the port.
MAAS requires a properly configured DHCP server so it can boot machines using PXE. MAAS can work with its own instance of the ISC DHCP server, if you install the maas-dhcp package:
$ sudo apt install maas-dhcp
Note that maas-dhcpd service definition referencese the maas-rackd service, which won't be present if you run a development service. To workaround edit /lib/systemd/system/maas-dhcp.service and comment out this line:
BindsTo=maas-rackd.service
The development environment uses daemontools to manage the various
services that are required. These are all defined in subdirectories in
services/.
There are familiar service-like commands:
$ make start $ make status $ make restart $ make stop
The latter is a dependency of distclean so just running make
distclean when you've finished with your branch is enough to stop
everything.
Individual services can be manipulated too:
$ make services/rackd/@start
The @<action> pattern works for any of the services.
There's an additional special action, run:
$ make run
This starts all services up and tails their log files. When you're
done, kill tail (e.g. Ctrl-c), and all the services will be
stopped.
However, when used with individual services:
$ make services/regiond/@run
it does something even cooler. First it shuts down the service, then
it restarts it in the foreground so you can see the logs in the
console. More importantly, it allows you to use pdb, for example.
A note of caution: some of the services have slightly different behaviour when run in the foreground:
- regiond (the webapp service) will be run with its auto-reloading enabled.
There's a convenience target for hacking regiond that starts everything up, but with regiond in the foreground:
$ make run+regiond
Apparently Django needs a lot of debugging ;)
Since MAAS is distributed mainly as an Ubuntu package, all runtime
dependencies should be packaged, and we should develop with the
packaged version if possible. All dependencies, from a package or not,
need to be added to setup.py and buildout.cfg, and the version
specified in versions.cfg (allowed-picked-version is disabled,
hence buildout must be given precise version information).
If it is a development-only dependency (i.e. only needed for the test suite, or
for developers' convenience), simply running buildout like this will make
the necessary updates to versions.cfg:
$ ./bin/buildout -v buildout:allow-picked-versions=true
When creating a new source file, a Python module or test for example,
always start with the appropriate template from the templates
directory.
MAAS uses Django to manage changes to the database schema.
Be sure to have a look at Django's migration documentation before you make any change.
Once you've made a model change (i.e. a change to a file in
src/<application>/models/*.py) you have to run Django's makemigrations
command to create a migration file that will be stored in
src/<application>/migrations/builtin/.
Note that if you want to add a new model class you'll need to import it
in src/<application>/models/__init__.py
Generate the migration script with:
$ ./bin/maas-region makemigrations --name description_of_the_change maasserver
This will generate a migration module named
src/maasserver/migrations/builtin/<auto_number>_description_of_the_change.py.
Don't forget to add that file to the project with:
$ git add src/maasserver/migrations/builtin/<auto_number>_description_of_the_change.py
To apply that migration, run:
$ make syncdb
If you need to perform data migration, very much in the same way, you will need
to run Django's makemigrations command. For instance, if you want to perform
changes to the maasserver application, run:
$ ./bin/maas-region makemigrations --empty --name description_of_the_change maasserver
This will generate a migration module named
src/maasserver/migrations/builtin/<auto_number>_description_of_the_change.py.
You will need to edit that file and fill the operations list with the
options that need to be performed. Again, don't forget to add that file to the
project:
$ git add src/maasserver/migrations/builtin/<auto_number>_description_of_the_change.py
Once the operations have been added, apply that migration with:
$ make syncdb
Previous version before MAAS 2.0 used South to perform database migrations. To support upgrading from any previous version of MAAS before 2.0 the South migrations are run. On upgrade of MAAS those migrations will be run before the new Django migrations are run. On a fresh installation of MAAS the South migrations will be skipped because the Django migrations already provide the entire schema in the initial migration. All of this logic is performed on upgrade by the dbupgrade command.:
$ bin/maas-region dbupgrade
In some testing case you might need to always run the South migrations before the Django migrations on a clean database. Using the always-south option on the dbupgrade command allows this testing scenario.:
$ bin/maas-region dbupgrade --always-south
Note
When the South migrations run they are actually being ran under Django 1.6
and South that is provided in the MAAS source code in a tarball. Located
at src/maasserver/migrations/south/django16_south.tar.gz this file is
extracted into a temporary folder and imported by MAAS to run the South
migrations.
If you need to get an interactive psql prompt, you can use dbshell:
$ bin/maas-region dbshell
If you need to do the same thing with a version of MAAS you have installed from the package, you can use:
$ sudo maas-region dbshell --installed
You can use the \dt command to list the tables in the MAAS database. You
can also execute arbitrary SQL. For example::
maasdb=# select system_id, hostname from maasserver_node;
system_id | hostname
-------------------------------------------+--------------------
node-709703ec-c304-11e4-804c-00163e32e5b5 | gross-debt.local
node-7069401a-c304-11e4-a64e-00163e32e5b5 | round-attack.local
(2 rows)
If you need to view the SQL queries that are performed during a test, the LogSQL fixture can be used to output all the queries during the test.:
from maasserver.fixture import LogSQL self.useFixture(LogSQL())
Sometimes you need to see where in the code that query was performed.:
from maasserver.fixture import LogSQL self.useFixture(LogSQL(include_stacktrace=True))
Use reST with the convention for headings as used in the Python documentation.
Use the Bazaar Copyright Updater:
bzr branch lp:bzr-update-copyright ~/.bazaar/plugins/update_copyright make copyright
Then commit any changes.