| Age | Commit message (Collapse) | Author |
|---|
|
this adds 'power-state-change' config module that allows the user to
specify in cloud-config syntax that a system reboot or shutdown should occur
after cloud-init is done.
|
|
|
|
It seems like its possible that boto 2.5.2 and below have the lazy loading
metadata dictionary so as a precaution we will always take the hit of
unlazying the metadata dictionary by traversing it which in the non-lazy
dictionary case has no effect (its marginal). This also removes the need
to check the boto version and the dependency on setup tools just for this
case.
|
|
The sleep is added here simply to not completely spin cpu on waiting
for the parent pid to die.
the allowing of errno 3 is because I was getting this. I dont have
a perfect explanation, but I suspect that the 'open' was actually
getting this back from the /proc filesystem after the pid had died.
Possibly in the window between when the 'open' was done and the 'read()'
was done.
|
|
It seems like its possible that boto 2.5.2 and below have
the lazy loading metadata dictionary so as a precaution
we will always take the hit of unlazying the metadata dictionary
by traversing it which in the non-lazy dictionary case has
no effect (its marginal). This also removes the need to check
the boto version and the dependency on setup tools just for
this case.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
This allows the user to easily run stuff even after cloud-init-final
has finished. The initial reason for it is to be able to run
/sbin/poweroff and not have cloud-init complain loudly that it is
being killed.
LP: #1064665
|
|
|
|
Fix how the comparison of a fqdn and its aliases was done
via sorting instead of existence checking which is the better
way to check if a alias already exists as well as cleanup
the new files pep8/pylint issues.
LP: #1078097
|
|
Previous commit set up an object for checking "None" in the stages class.
Here, change that to be None. The benefit of the DataSourceNone was so that
the stages was generally functional even with that "None" object.
The negative affect of that is that you could go down the error path much
longer when using it. If we use None type, then self.datasource.<something>
will fail immediately and loudly.
|
|
This just replaces the portions of stages.py that were checking or
setting self.datasource to a DataSourceNone to use a static/global
datasource. And then, check for None is done against that.
|
|
This does a couple things:
* separates out the 'normalize_public_keys' from the DataSource's get_public_ssh_keys
* uses that from config-drive datasource
* supports config drive v1 or v2 public-keys
* adds a test.
LP: #1077700
|
|
|
|
'public_keys'
|
|
|
|
|
|
other datasources populate 'public-keys' rather than 'public_keys'
and there is a more complete handler in the base DataSource.
So, to take advantage of that, have DataSourceConfigDrive copy
public_keys to public-keys, and remove the 'get_public_ssh_keys'
from the DataSourcEConfigDrive.
LP: #1077700
|
|
These changes were pulled out of the previous merge (cc_yum_add_repo)
as they were unrelated there. Re-applying them here.
|
|
Add in a configuration module that can write out the yum.repo format for
those that want to hook into different repos for installing.
|
|
|
|
|
|
|
|
This migrator module does a few things
* fixes filename markers that were written with a '-' in them to have
an '_' instead
* support renaming modules. Explicitly this handles the name change
of 'apt-update-upgrade' to 'apt-configure' and
'package-update-upgrade-install'
Also, just be more consistent everywhere where writing semaphore/marker
files and use '_' instead of '-' (canon_sem_name).
|
|
|
|
when that group configuration is a dict => members.
LP: #1077245
|
|
|
|
where rhel uses the fqdn for its config while
ubuntu uses the short hostname.
|
|
|
|
the fully qualified domain name should end
up in /etc/sysconfig/network by passing the
fqdn to the update and set hostname methods
and using it accordingly.
LP: #1076759
|
|
of the init class are reset so that when they are
regenerated that they will use the updated data
instead of using previous data (since they weren't reset).
LP: #1076811
|
|
is actually a valid value returned.
2. Adjust variable naming
|
|
when applying the canon routine.
2. Add in a function to migrate legacy
semaphores to new semaphores.
|
|
that can be used to aid in the moving
of older versions of cloud-inits data
to newer versions of cloud-inits data.
1. Move the semaphores for the current instance
to there canonicalized names and use the
canonicalized in the file 'locking' code
|
|
2. Added example cloud-config
3. Added functioning test for yum config
|
|
|
|
1. Remove the usage of the path.join function now that all code should
be going through the util file methods (and they can be mocked out as
needed).
2. Adjust all occurences of the above join function to either not use it
or replace it with the standard os.path.join (which can also be mocked
out as needed)
3. Fix pylint from complaining about the tests folder 'helpers.py' not
being found.
4. Add a pylintrc file that is used instead of the options hidden in the
'run_pylint' tool.
|
|
|
|
|
|
|
|
attempt the unicode-escape path.
|
