* This also adds fields to the instance view for tracking cpu and
memory usage as well as information on what the capacity ranges are
* Also adds a flag for enabling/disabling instances which removes them
from all queues and has them stop processing new work
* The capacity is now based almost exclusively on some value relative
to forks
* capacity_adjustment allows you to commit an instance to a certain
amount of forks, cpu focused or memory focused
* Each job run adds a single fork overhead (that's the reasoning
behind the +1)
* Switch policy router queue to not be "tower" so that we don't
fall into a chicken/egg scenario
* Show fixed policy list in serializer so a user can determine if
an instance is manually managed
* Change IG membership mixin to not directly handle applying topology
changes. Instead it just makes sure the policy instance list is
accurate
* Add create/delete hooks for instances and groups to trigger policy
re-evaluation
* Update policy algorithm for fairer distribution
* Fix an issue where CELERY_ROUTES wasn't renamed after celery/django
upgrade
* Update unit tests to be more explicit
* Update count calculations used by algorithm to only consider
non-manual instances
* Adding unit tests and fixture
* Don't propagate logging messages from awx.main.tasks and
awx.main.scheduler
* Use advisory lock to prevent policy eval conflicts
* Allow updating instance groups from view
* Associating/Disassociating an instance with a group
* Triggering a topology rebuild on that change
* Force rabbitmq cleanup of offline nodes
* Automatically check for dependent service startup
* Fetch and set hostname for celery so it doesn't clobber other
celeries
* Rely on celery init signals to dyanmically set listen queues
* Removing old total_capacity instance manager property
this commit allows schedule `rrule` strings to include local timezone
information via TZID=NNNNN; occurrences are _generated_ in the local
time specific by the user (or UTC, if e.g., DTSTART:YYYYMMDDTHHMMSSZ)
while Schedule.next_run, Schedule.dtstart, and Schedule.dtend will be
stored in the UTC equivalent (i.e., the scheduler will still do math on
"what to run next" based on UTC datetimes).
in addition to this change, there is now a new API endpoint,
`/api/v2/schedules/preview/`, which takes an rrule and shows the next
10 occurrences in local and UTC time.
see: https://github.com/ansible/ansible-tower/issues/823
related: https://github.com/dateutil/dateutil/issues/614
this approach totally removes the process of reading and writing stdout
files on the local file system at settings.JOBOUTPUT_ROOT when jobs are
run; now stdout content is only written on-demand as it's fetched for
the deprecated `stdout` endpoint
see: https://github.com/ansible/awx/issues/200
* introduces three new models: `ProjectUpdateEvent`,
`InventoryUpdateEvent`, and `SystemJobEvent`
* simplifies the stdout callback management in `tasks.py` - now _all_
job run types capture and emit events to the callback receiver
* supports stdout reconstruction from events for stdout downloads for
_all_ job types
* configures `ProjectUpdate` runs to configure the awx display callback
(so we can capture real playbook events for `project_update.yml`)
* ProjectUpdate, InventoryUpdate, and SystemJob runs no longer write
text blobs to the deprecated `main_unifiedjob.result_stdout_text` column
see: https://github.com/ansible/awx/issues/200
display_extra_vars was not taking a copy of the data before
acting on it - this causes a bug where the activity stream will
modify the existing object on the model. That leads to new data
not being accepted.
Also moved the processing of extra_data to prior to the accept
or ignore kwargs logic so that we pass the right (post-encryption)
form of the variables.
This data often (in the case of inventory updates) represents large data
blobs (5+MB per job run). Storing it on the polymorphic base class
table, `main_unifiedjob`, causes it to be automatically fetched on every
query (and every polymorphic join) against that table, which can result
in _very_ poor performance for awx across the board. Django offers
`defer()`, but it's quite complicated to sprinkle this everywhere (and
easy to get wrong/introduce side effects related to our RBAC and usage
of polymorphism).
This change moves the field definition to a separate unmanaged model
(which references the same underlying `main_unifiedjob` table) and adds
a proxy for fetching the data as needed
see https://github.com/ansible/awx/issues/200
