summaryrefslogtreecommitdiff
path: root/cloudinit
AgeCommit message (Collapse)Author
2015-02-23Always close the password server connection, even on failure.Daniel Watkins
2015-02-20Split CloudStack password handling out to separate class.Daniel Watkins
2015-02-20Minor formatting clean-up in CloudStack DS.Daniel Watkins
2015-02-18Set an explicit timeout when fetching CloudStack passwords.Daniel Watkins
2015-02-18Failing to fetch a CloudStack password should never fail the whole DS.Daniel Watkins
There might be some CloudStack deployments without the :8080 password server, and there's no reason the rest of the data source can't be used for them.
2015-02-17Add explanatory comment.Daniel Watkins
2015-02-17Fetch and use passwords from CloudStack virtual router.Daniel Watkins
2015-02-17Clean up imports in DataSourceCloudStack.py.Daniel Watkins
2015-02-13fix usage of python2 'print'Scott Moser
2015-02-13support for managing GPT partitionsScott Moser
Specifically, this is to support Azure's G-series VMs (which come with disks up to 6500GB).
2015-02-12mount_cb: fix scoping of an exception in python3Scott Moser
2015-02-11include exception in error again.Scott Moser
it is admittedly not clear, but 'exc' should be definied if mountpoint is not.
2015-02-11pickle contents: be careful loading and storing pickle to be binaryScott Moser
2015-02-11provide default final message in jinja to avoid WARN in logScott Moser
2015-02-11Open /dev/console in text mode (so we don't have to encode strings to write ↵Daniel Watkins
them).
2015-02-11Fix reference to non-existent variable.Daniel Watkins
2015-02-10pep8Scott Moser
2015-02-10pep8Scott Moser
2015-02-11fix random_seed moduleScott Moser
2015-02-10use encode_textScott Moser
2015-02-10fix parse_qs usageScott Moser
2015-02-10fix use of 'letters' and translateScott Moser
2015-01-27Remove debugging turd.Barry Warsaw
2015-01-27Respond to review:Barry Warsaw
- Refactor "fully" decoding the payload of a text/* part. In Python 3, decode=True only means to decode according to Content-Transfer-Encoding, not according to any charset in the Content-Type header. So do that.
2015-01-27Respond to review:Barry Warsaw
- Refactor both the base64 encoding and decoding into utility functions. Also: - Mechanically fix some other broken untested code.
2015-01-27Respond to review:Barry Warsaw
- Just use util.load_file() instead of yet another way to open and read the file.
2015-01-27Remove a comment turd.Barry Warsaw
2015-01-27Remove some unused code.Barry Warsaw
2015-01-27Respond to review:Barry Warsaw
- Remove str() wrappers to second argument to write_files() where it is no longer necessary. Also: Fixed a couple of other octal literals which clearly weren't being tested.
2015-01-26Repair the Python 2.6 tests.Barry Warsaw
2015-01-26Trunk merged and ported.Barry Warsaw
2015-01-26Port the MAAS code to oauthlib.Barry Warsaw
2015-01-26Python 3 tests pass, except for skips.Barry Warsaw
2015-01-26Another handling of b64decode.Barry Warsaw
Also, restore Python 2 compatibility.
2015-01-26* More str/bytes fixes.Barry Warsaw
* Temporarily skip the MAAS tests in py3 since they need to be ported to oauthlib.
2015-01-26More Python 3 test fixes.Barry Warsaw
2015-01-26Avoid a nose bug when running under the test suite and no exception is inBarry Warsaw
flight.
2015-01-26Fix a few string/bytes problems with Python 3.Barry Warsaw
2015-01-26More test repairs.Barry Warsaw
2015-01-23* Fix the filter() imports.Barry Warsaw
* In Py3, pass universal_newlines to subprocess.Popen()
2015-01-22Low hanging Python 3 fruit.Barry Warsaw
2015-01-22merge from trunkScott Moser
2015-01-21Largely merge lp:~harlowja/cloud-init/py2-3 albeit manually because it seemedBarry Warsaw
to be behind trunk. `tox -e py27` passes full test suite. Now to work on replacing mocker.
2015-01-21Use either syspath or dmidecode based on the availability.Ben Howard
2015-01-21More octal literal fixes.Barry Warsaw
2015-01-21Fix file modes to be Python 2/3 compatible.Barry Warsaw
2015-01-20New Azure disk_setup default.Daniel Watkins
2015-01-16hostname: apply hostname same as is writtenScott Moser
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
2015-01-16pep8 fixesScott Moser
2015-01-14Drop reliance on dmidecode executable.Ben Howard