* Add support for paramater trusted, valid options are 'true' and false.
defaults to false. True sets the value to trusted=yes.
trusted=yes can be set to indicate that packages from this source are
always authenticated even if the Release file is not signed or the
signature can't be checked.
* Update documentation
Add "oldstable" to the default update origins to ensure
the updates keep working after wheezy+1 gets released
See unattended-upgrades 0.79.5+wheezy1 and https://bugs.debian.org/711826
You can feed the command the long key, but it truncates it to add the
key. This causes issues due to the short-key collision with the
puppetlabs key. So, test with a different key on debian 6.
- fix spec tests to include osfamily fact
- add spec tests to verify current default behavior unimpacted.
- manage the update-stamp file in puppet via content rather than a served file.
- update custom fact to return -1 if the file doesn't exist
- add spec test for custom fact
- refactor to use a variable vs a collector/override
- document parameters a bit more verbosely
- remove empty unconstrained fact
- Add osfamily fact to backports tests to facilitate functional tests on non-debian hosts
when updating or installing newer packages with apt::force and you have changed previous
configuration files aptitude or apt-get will prompt what to do. You can suppress that
by pre-define the action with cfg_files parameter (new, old or unchanged and its backward
compatible if not defined). With a second optional parameter cfg_missing you can force
your provider to install missing configuration files as well.
Signed-off-by: Martin Seener <martin@seener.de>
apt::force: Changed selectors used in force.pp to case statements; refs #module-1306
Signed-off-by: Martin Seener <martin@seener.de>
apt::force: rspec: fixed the failing tests and added validate_re for cfg_files and validate_bool for cfg_missing. Also removed default values for both case statements and only allow pre-defined values or true/false. Furthermore enhanced the README refs #module-1306
Was able to fix the failing rspec tests for the patch.
Thanks to Morgan Haskel.
Signed-off-by: Martin Seener <martin@seener.de>
Despite the puppetlabs-stdlib documentation says validation_re supports 3 arguments the tests failed telling that only 2 are supported. Fixed this by removing the 3 optional argument; refs #modules-1306
Signed-off-by: Martin Seener <martin.seener@barzahlen.de>
apt::force: updated readme refs #module-1306
Signed-off-by: Martin Seener <martin@seener.de>
A lot of the tests were testing things that really should be tested via
unit tests, so those were deleted and unit tests will be revamped to
make sure they are covering everything they need to be covering.
Conflicts:
spec/acceptance/unattended_upgrade_spec.rb
A lot of the tests were testing things that really should be tested via
unit tests, so those were deleted and unit tests will be revamped to
make sure they are covering everything they need to be covering.
fix for default debian installations
all files in /etc/apt/preferences without _ will be silently ignore according to debian manpage. Addionally its not a good idea to write versionnumber in filename cause there is no way to delete this files if you increase versionumber
Update source_spec.rb
add a way to include debsrc only (useful for debian/ubuntu build server ... jenkins ect)
Update source_spec.rb
var rename
Update source.list.erb
add include_deb "switch"
Update source.pp
"include_deb" defaultvalue = true
Update hold_spec.rb
change the name of the preferences file (hold)
Update source_spec.rb
Update README.md
Doku: 'include_deb' included next to 'include_src' in examples
Update README.md
typo
New fact was added that matched a regex breaking the always_apt_update
tests. Updated the tests to check for the apt_update exec, not just the
string apt_update.
Adds spec test
If lab-release is not installed, then the end user sees a confusing/ vague message
Error: Unsupported lsbdistid () at /modules/apt/manifests/params.pp:52
It is common for docker containers to not include this package by default
After fix, the user sees a friendlier message if lab-release is not installed
Error: Unable to determine lsbdistid, is lsb-release installed? at /modules/apt/manifests/params.pp:52
Making use of the apt-check command from the 'update-notifier-common'
package (if available) display the number of available updates, number of
security updates as well as the update package names.
Quoting https://wiki.debian.org/LTS
Official security support for Debian GNU/Linux 6.0
(code name "Squeeze") has ended on 31 May 2014.
However long term support for the distribution
is going to be extended until February 2016,
i.e. five years after the initial release.
See https://wiki.debian.org/LTS for more details.
As some places dont have port 11371 open, they are required to use URL as
key_server instead of domain name therefore adding the capability to use URL or
domain name as key_server parameter
Ubuntu 14.04 ships with apt 0.9.15, has a ``fancy progress bar'', which
is a green bar that shows at the bottom of the terminal showing progress
throughout install.
This patch enables the progress bar, which is usually done by running
echo 'Dpkg::Progress-Fancy "1";' > /etc/apt/apt.conf.d/99progressbar
This commit changes the proxy file name to be more consistent with other files
in `apt.conf.d`. The old file (`apt.conf.d/proxy`) is removed.
Tests has been updated.
Because Squeeze is now oldstable we need to add an oldstable line too
otherwise security updates won't be picked up. This is still because we
can't match on codename.
In APT preferences files the only allowed comments are lines that start
with `Explanation:`, commented lines that start with a # trigger a
myriad of interesting bugs. This is considered a feature of APT.
Because we're only ever writing a single file at a time with only a #
comment at the top we were getting away with this but it shouldn't be
there in the first place.
The default configuration we were writing for Debian was only working
for Squeeze, from Wheezy and onwards this wasn't working anymore. This
has to do with the fact that we should now be using Origins-Pattern
according to the unattended-upgrades docs. However, Ubuntu didn't
entirely get with the program yet...
This change reflects the defaults that unattended-upgrade installs on
every platform we support. In order to do so the unattended-upgrades
Debian archive for Squeeze, Wheezy, Lucid, Precise and Trusty were
downloaded and the default /etc/apt/apt.conf.d/50unattended-upgrades
checked for its content with regard to using Allow-Origins or
Origins-Pattern.
Fixes#277
The module used to always pin backports to a priority of 200. This
default is still retained but is now configurable.
Additionally the default is now an Integer, not a 'quoted Integer' and
the tests have been updated to reflect this. This matters for future
parser as it will now kick people if they pass in a stringified integer
as priority.
I am aware this can be done with `dpkg --set-selections`, `apt-mark`
or `ensure => 'held'` on a package resource. The changes to the README
include the full rationale for wanting another mechanism.
Introducing a totally rewritten and tested apt::key. This commit also
patches the spec's of apt::source because it was passing in data that
is no longer allowed by the new validation rules in apt::key.
It does its best to not touch any other specs and where we touch them
only minimally to ensure that we're not introducing breaking changes.
We already had a feature to manage and purge entries in preferences.d
but not the preferences file in /etc/apt. This commit adds that
capability.
Fixes#199
Lucid (10.04) has `add-apt-repository` but it doesn't accept any
options. The define defaulted to `-y` but this changes that on lucid.
This was made 7 months ago, so apparently no one cares about 10.04 any
more.
This commits introduces:
* The apt_key type;
* The apt_key provider;
* Unit tests for the type;
* Beaker/acceptance tests for the type/provider.
The idea behind apt_key is that apt::key will simply become a wrapper
that uses apt_key. Being a native type/provider apt_key is a lot less
error prone than the current exec behaviour of apt::key and adds a few
nice bonuses like inventory capabilities for mcollective users.
We actually expect an extra space. The previous build failed because a test is issued for location='', which indeed results in 2 spaces between the architecture specification and the release. According to the sources.list man page a location is always required though (unlike the missing/empty location in the :default_params of the source_spec test).
The two forms of APT preferences records (general & specific) can now be
completely and not partially defined. All distribution properties can be passed
as resource parameters. This change is totally backward-compatible.
This is necessary when required_packages contains GPG keys that are used for
authenticating other packages. Tested with package ubuntu-cloud-keyring which
is included in Ubuntu main and used by the Ubuntu Cloud Archive.
I think the same problem applies to other *-keyring packages as well.
