* Ensure "fresh bits" in dependent jobs
- run IUs in same context and dependent PU
- review docs for update-on-launch expectations
* Avoid customer variable precedence upgrade issues
- only allow SCM types to have overwrite_vars=true
- do not run other cloud sources through the backported script
* Make extra sure to echo stdout and stderr in event of error
* allow for filtering Jobs and JobTemplates by v1 `cloud_credential` and
`network_credential` fields
* properly validate uniqueness of `extra_credentials` types
see: #5807
on InventorySource model
* scm_project -> source_project
on InventorySourceSerializer
* scm_inventories -> scm_inventory_sources
on InventoryUpdate model
* scm_project_update -> source_project_update
The following fields:
* (Job | JobTemplate).cloud_credential
* (Job | JobTemplate).network_credential
...are replaced by M2M relationships:
* Job.extra_credentials
* JobTemplate.extra_credentials
Includes support for task execution with multiple cloud credentials.
see: #5807
* Remove system tracking piggybacked most-recent-fact-gathering
* Add explicit per-jobtemplate configurable fact gathering
* Continue to support system tracking
* Fixed a bug where we weren't ack()ing system tracking rabbit msgs
* remove support for loading from executable and static files
* instead use ansible-inventory with fallback to backport
* provide private file dir in task manager for cred injection
* durable management of tmp dirs for user scripts
* new 'scm' source choice for scm-type
* update SCM inventory docs to new reality
This separates file parsing logic that was mixed
in with other important code inside of the
inventory import command.
The logic around MemObject data structures was
moved to utils, and the file parsing was moved
to a legacy module. As of this commit, that
module can operate within the Tower environment
but it will be removed.
Also refactor the loggers to fix old bug and
work inside of the different contexts - the
Loader classes, mem objects, and hopefully
the inventory modules eventually.
* release_3.1.3: (52 commits)
ack fact scan messages
making ldap user/group search fields into codemirror instances
removing UI parsing for LDAP User and Group Search fields
Allow exception view to accept all valid HTTP methods.
Restore ability of parsing extra_vars string for provisioning callback.
Fix up backup/restore role broken in f7a8e45809758322d9ee41c5305850dd70ed5faf
Stop / start ansible-tower-service during restores
value_to_python should encode lookup fields as ascii
fix brace interpolation on standard out pane
Adjust some hardcoded usages of 'awx' to use 'aw_user' and 'aw_group'.
Pull Spanish updates from Zanata
Temporarily grant awx user createdb role
Stop giving ownership of backups to postgres
don't display chunked lines'
Add dropdown li truncation with ellipsis
CTiT -> adhoc modules should allow the user to add new modules
Remove task that was replacing the supervisor systemd tmp file
Fix failing supervisorctl commands on RH-based distros
Give ownership of the supervisor socket to awx
Setting for external log emissions cert verification
...
Credentials now have a required CredentialType, which defines inputs
(i.e., username, password) and injectors (i.e., assign the username to
SOME_ENV_VARIABLE at job runtime)
This commit only implements the model changes necessary to support the
new inputs model, and includes code for the credential serializer that
allows backwards-compatible support for /api/v1/credentials/; tasks.py
still needs to be updated to actually respect CredentialType injectors.
This change *will* break the UI for credentials (because it needs to be
updated to use the new v2 endpoint).
see: #5877
see: #5876
see: #5805
