| Age | Commit message (Collapse) | Author |
|---|
|
flight.
|
|
|
|
|
|
* In Py3, pass universal_newlines to subprocess.Popen()
|
|
The sources.DataSource class has method defined as:
def get_hostname(self, fqdn=False, resolve_ip=False)
Make the parameter list for this method in DataSourceDigitalOcean
and DataSourceGCE consistent with superclass sources.DataSource.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
might have to rewrite this for Python 2.6.
Disable Cepko tests (test_cs_util.py) since they are essentially worthless.
Convert test_azure to unittest.mock.
|
|
on linux we can read dmi information from /sys rather than using
dmidecode binary, and thus removing a dependency.
|
|
|
|
|
|
to be behind trunk.
`tox -e py27` passes full test suite. Now to work on replacing mocker.
|
|
|
|
|
|
|
|
2. Cheetah is not compatible with Python 3.
|
|
* Use universal_newlines in setup.py so it will work properly in Python 3.
* Fix a pyflakes complaint in setup.py
* Add a simple MANIFEST.in
|
|
|
|
on RHEL, we were writing to persistent configuration the fqdn, but
invoking 'hostname' on the first boot with just the shortname. On 'reboot',
then the hostname would differ.
Now, whatever we write, invoke hostname with.
Also remove some duplicate code.
LP: #1246485
|
|
|
|
This fixes a race condition that can cause cloud-init output to be spit out
over the login prompt on the console when booting under systemd.
|
|
correct race-free ordering
|
|
display any getty on the console so that the cloud-init output doesn't get
written on top of a login prompt.
|
|
Google Compute Engine fqdn hostnames are usually longer than 64 characters.
This causes issues with many tools (often Java based).
Note that per gethostname(2):
POSIX.1-2001 guarantees that "Host names (not including the terminating null
byte) are limed to HOST_NAME_MAX bytes". On Linux, HOST_NAME_MAX is defined
with the value 64.
LP: #1383794
|
|
|
|
|
|
MBR uses block sizes, which is what the current (apparently portable) code was
producing. GPT uses sectors to determine partition size.
|
|
|
|
This includes moving some shared logic in to check_partition_layout.
|
|
|
|
Enable user-data encoding support for GCE. Extended and updated tests to
support checking the user-data encoding.
User can now pass in user-data encoded in base64 and indicate they've
done so by adding a tag 'user-data-encoding' with value 'base64'.
LP: #1404311
|
|
just so we don't come to rely on it being set to 'base64' or ''.
rather the default case is that it wont be present at all.
|
|
|
|
--ignore was being called with ',E121,E...' rather than
'E121,E...'.
that resulted in odd behavior, missing the pep8 errors that are fixed
here.
|
|
|
|
|
|
|
|
this was broken previously when user-data and vender-data were
brought together.
|
|
- Handle ipv6 route information not existing
gracefully (for systems that don't have it)
- Fix the getgateway function (broken due to
ipv4/ipv6 keys now existing in route info)
- Separate the centering of the route info ipv4
information from the centering of the ipv6 information
so that this looks prettier...
- Use try: except: else instead of settings value to
None and then later checking for None (more pythonic
this way)
|
|
|
|
|
|
After the routeinfo function started to return
a dictionary containing ipv4 and ipv6 information
we now need to make sure we search the appropriate
key.
|
|
Instead of failing when IPv6 information is not found we
should be more tolerant of said information not existing
so that we behave like the pre IPv6 addition.
|
|
Currently the rpm building process that cloud-init provides is
not working correctly. This adjusts the spec file, the setup.py
file and the distro files to ensure that it continues to work
as expected.
|
|
|
