almost all of our current credential values are strings, but under the
v1 Credential model, `authorize` is boolean
additionally, if a field is specified with no type, fall back to
`string` as a default (this is almost always what people creating custom
types will want)
see: #6406
this provides error messages keyed by input fields, so that instead of
e.g.,
{
'inputs': ['Invalid certificate or key: u'XYZ']
}
...you get:
{
'inputs': {
'ssh_key_data': ['Invalid certificate or key: u'XYZ']
}
}
Includes /api/v1/ compatability for error message format. Requests to
/api/v1/ will get:
{'ssh_key_data': ['Invalid certificate or key: u'XYZ']}
* A credential is considered an insights credential if a project of
scm_type 'insights' is associated.
* all non-insights Projects that point at an insights credential are
nulled
* We've switched from Q() to Host.objects.filter() and you can't do
~(Host.objects.filter()). We'd need to create an AST to track the
operations in order to perform the boolean logic ~ and we don't have
time for that now.
* make user_capability for 'start' more simple
* provide error messages if relaunch not possible
* check for prompted extra_credentials
* add extra_credentials permission checks in orphan mode
* rampart_groups_setup_playbook:
Updating changelog for Instance Groups
Fix an incorrect reference on instance group jobs list
Purge remaining references to rampart groups
Simplify can_access for instance groups on job templates
Adding Instance Group permissions and tests
Increase test coverage for task scheduler inventory updates
Exit logic fixes for instance group tools
View Fixes for instance groups
new view to allow associations but no creations
Updating acceptance documentation and system docs
Updating unit tests for task manager refactoring
Update views and serializers to support instance group (ramparts)
Implementing models for instance groups, updating task manager
Updating the setup playbook to support instance group installation
Add nginx to server start and switch back to first tmux win
Fix an issue where the local queue wouldn't use the rabbitmq name
`clean_ssh_key_data` and `clean_ssh_key_unlock` no longer work because
they're not actual fields on `model.Credential` anymore. This change
refactors/moves their validation to a place that works (and makes more
sense).
* Before, the boolean logic operators were performed against the Q()
objects, iteratively. Now, boolean logic is done after
Host.objects.filter(Q()). This results in the wanted and expected
results.
* fixed a problem where the SCM last revision was not getting reset
* change the 400 editing error to just reset the last revision
* handle `source_vars` in the same way as custom scripts
* block unallowed `source_vars` in the validator
* hide POST in OPTIONS for inventory source sublist if not allowed
