| Age | Commit message (Collapse) | Author |
|---|
|
Some containers lack /dev/console, so when multi_log attempts to open
that device and write to it directly things can start going haywire.
Here we address this problem by sending console-bound output to stdout
and letting init take care of getting it to the console instead.
We already configure upstart with "console output", so we need only
change systemd to use "journal+console".
The one reason that 'console output' might not be sufficient is if
the user redirected output with 'output'. Ie:
output:
init: "> /var/log/my-cloud-init.log"
Would then mean all output would go there, and anything that
*needed* to go to the console (and was explicitly using multi_log for
that purpose) would not get there.
|
|
|
|
exceptions, causing nasty things to happen on instances that boot with
selinux=0. The fix is easy: simply consult is_selinux_enabled() first.
|
|
that device and write to it directly things can start going haywire.
Here we address this problem by sending console-bound output to stdout
and letting init take care of getting it to the console instead.
We already configure upstart with "console output", so we need only
change systemd to use "journal+console".
|
|
|
|
It appears that udelta could
have been left undefined or left
defined as a string "N/A" and then
put threw a float formatter previously.
Fix that by ensure its set to a default
and put strong checking to make sure it
is a float before using float formatting.
|
|
|
|
reading /proc/uptime is going to be slower, and no reason to do it on most
things. Better to only do it when you suspect maybe a need for it.
|
|
The reason for this is that more and more things I was wanting to be able to
see how long they took. This puts that time logic into a single place. It
also supports (by default) reading from /proc/uptime as the timing mechanism.
While that is almost certainly slower than time.time(), it does give
millisecond granularity and is not affected by 'ntpdate' having
run in between the two events.
|
|
This adds a datasource designed to work on Joyent cloud (SmartOS).
|
|
initialized.
Added a mapping of attributes between cloud-init and smartos.
|
|
|
|
|
|
|
|
Instead of previously having merging which was not backwards compatible
with the 0.7.1 and prior methods this patch works to ensure said backwards
compatible while at the same time making the new merging functionality
work in a more customizable manner.
|
|
Couple of things here:
* do not re-try on user-data (404 means 'not here')
* re-generate headers on retry requests
LP: #1172742
|
|
|
|
This patch fixes issues in Fedora 18 (and upcoming RHEL 7) which are
present due to their use of systemd:
- store locale configuration in /etc/locale.conf
- store hostname in /etc/hostname
- use a symlink for /etc/localtime (prior code would set the timezone
but corrupt data in /usr/share/zoneinfo due to presence of symlink)
It also contains fixes for issues unrelated to systemd adoption:
- explicitly scan /dev/sr0 with blkid in order to get the optical drive
in the blkid cache. This prevents an issue on systems running 2.6
kernels (such as RHEL 6) in which config disks on some devices won't
be detected unless the device has previously been queried.
(For reference, see https://patchwork.kernel.org/patch/1770241/)
- append a newline when rewriting sysconfig files, as this is customary
text configuration file formatting and is expected by some parsers
(such as the ifcfg-rh plugin for NetworkManager)
|
|
* cloudinit/distros/parsers/resolv_conf.py
added some pylint overrides with 'plXXXXX' syntax.
example: # pl51222 pylint: disable=E0102
The pl51222 there means: http://www.logilab.org/ticket/51222
This specific issue is present in 12.04 pylint, but not 13.04.
* pylint doesn't like the requests special handling we have.
which makes sense as it is only checking versus one specific version.
* general pep8 and pylint cleanups.
|
|
|
|
a.) appease pylint on raring, as it doesn't like subprocess
pylint: 0.26.0-1ubuntu1
This is mentioned in comments at http://www.logilab.org/ticket/46273
b.) tests/unittests/test_util.py:
the mountinfo lines are longer than 80 chars.
Just disable long lines complaints for this file.
|
|
|
|
As reported in bug 1154599, I'm seeing this on my desktop system:
$ python -c \
'from cloudinit import util; print util.is_resolvable("brickies.neiit")'
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "cloudinit/util.py", line 865, in is_resolvable
socket.SOCK_STREAM, socket.AI_CANONNAME)
LP: #1154599
|
|
|
|
Refactor the parsing portion of util.get_mount_info() into a new
util.parse_mount_info() method. Now util.get_mount_info() opens
/proc/$$/mountinfo, splits on newlines and passes the lines to
util.parse_mount_info().
|
|
|
|
|
|
|
|
|
|
|
|
LP: #1136936
|
|
|
|
|
|
|
|
|
|
|
|
Augmenting the package version support to be available when specifying
extra packages to be installed at boot via the 'packages:' yaml key. This
change also improves type checking and add a configuration example to the
docs.
|
|
This change adds the ability to provide specific package versions to
Distro.install_packages and subsequently Distro.package_command. In order
to effectively use Distro.install_packages, one is now able to pass a
variety of formats in order to easily manage package requirements. These
are examples of what can be passed:
- "package"
- ["package1","package2"]
- ("package",)
- ("package", "version")
- [("package1",)("package2",)]
- [("package1", "version1"),("package2","version2")]
This change also adds the option to install a specific version for the
puppet configuration module. This is especially important here as
successful puppet deployments are highly reliant on specific puppet
versions.
|
|
previously, there was an attempt in the config drive source to limit
the source device to a "full block device" rather than a partition.
This was done by a simplistic approach of checking that the last
character of the name was not a number. That was filtering out
CD-rom devices (sr0).
Now, we have a bit more sophisticated approach to that same problem.
We filter out block devices that have a 'partition' entry in
/sys/class/block/DEVICE_NAME/partition .
LP: #1100545
|
|
|
|
|
|
These changes were pulled out of the previous merge (cc_yum_add_repo)
as they were unrelated there. Re-applying them here.
|
|
Use only util methods for reading/loading/appending/peeking
at files since it is likely soon that we will add a new
way of adjusting the root of files read, also it is useful
for debugging to track what is being read/written in a central
fashion.
|
|
1. Move out the old helpers that provided oop access/reading/writing
to various standard conf files and place those in parsers instead.
2. Unify the 'update_hostname' which varied very little between distros
and make it generic so that subclasses can only provide a couple of
functions to obtain the hostname updating functionality
3. Implement that new set of functions in rhel/debian
4. Use the new parsers chop_comment function for similar use
cases as well as add a new utils make header function that
can be used for configuration files that are newly generated
to use (less duplication here of this same thing being done
in multiple places.
5. Add in a distro '_apply_hostname' which calls out to the 'hostname'
program to set the system hostname (more duplication elimination).
6. Make the 'constant' filenames being written to for configuration
by the various distros be instance members instead of string
constants 'sprinkled' throughout the code
|
|
preparation for having a new branch
for all of these changes.
|
|
|
|
the previous 'user' and make those locations
go through the new distros functions to select
the default user or the user list (depending on usage).
Adjust the tests to check the new 'default' field
that signifies the default user + test the new method
to extract just the default user from a normalized
user dictionary.
|
|
In 0.7.0 we started using yaml.safe_load to load data rather than
yaml.load. Some producers (namely, ubuntu MAAS created) have produced
cloud-config data in the past that included python unicode types.
This creates a specialized safe_loader that is basically safe_load +
support for python unicode.
|
|
responsible only for creating users and groups and
normalizing a input configuration into a normalized
format that splits up the user list, the group list
and the default user listsand let the add user/group config
module handle calling those methods to add its own users/groups
and the default user (if any).
2. Also add in tests for this normalization process to ensure
that it is pretty bug free and works with the different types
of formats that users/groups/defaults + options can take.
|
|
at files since it is likely soon that we will add a new
way of adjusting the root of files read, also it is useful
for debugging to track what is being read/written in a central
fashion.
|
