Compare commits

...

228 Commits

Author SHA1 Message Date
Hao Liu
dd9160135d Prune dangle image periodically (#14957)
Prune dangle image periodically

pairs with https://github.com/ansible/ansible-runner/pull/1342

this fix the problem of us forcefully remove images when setting changing ee image that's being used in a job causing the job to fail
2024-03-12 10:57:57 -04:00
Chris Meyers
ad96a92fa7 Align Orign and Host header (#14970)
* Align Orign and Host header

* Before this change the Host: header was runserver. Seems to be set by
  nginx upstream flow.
* After this change we explicitly set the Host: header
* More about CSRF checks ...
  CSRF checks that Origin == Host. Think about how the browser works.

  <browser goes to awx.com>
  "I'm executing javascript that I downloaded from awx.com (ORIGIN) and
  I'm making an XHR POST request to awx.com (HOST)"
  Server verifies; Host: header == Origin: header; OK!

  vs. the malicious case.

  <hacker injects javascript code into google.com>
  <browser goes to google.com>
  "I'm executing javascript that I downloaded from google.com (ORIGIN)
  and I'm making an XHR POST request to awx.com (HOST)"
  Server verifies; Host: header != Origin: header; NOT OK!

* Update awx/settings/development.py

---------

Co-authored-by: Hao Liu <44379968+TheRealHaoLiu@users.noreply.github.com>
2024-03-11 17:06:09 -04:00
David O Neill
ca8085fe7e English string validation to error code validation 2024-03-11 20:07:16 +00:00
Hao Liu
b076cb00a9 Revert "Implement project pulling from Azure DevOps using Service Pri… (#14977)
Revert "Implement project pulling from Azure DevOps using Service Principals (#14628)"

This reverts commit 2e2cd7f2de.
2024-03-11 14:05:24 +00:00
John Westcott IV
ee9eac15dc Upgrade to postgres:15 (#14230)
* Upgrade to postgres:15
* Changed postgres:15 to quay.io/sclorg/postgresql-15-c9s
2024-03-07 16:27:03 -05:00
Hao Liu
3f2f7b75a6 [developer productivity improvement] Running awx components in vscode debugger (#14942)
Enable VSCode debugger integration when attaching VSCode to with AWX docker-compose development environment container

- add debugpy launch target in `.vscode/launch.json` to enable launching awx processes with debugpy
- add vscode tasks in `.vscode/tasks.json` to facilitate shutting down corresponding supervisord managed processes while launching process with debugpy
- modify nginx conf to add django runserver as fallback to uwsgi (enable launching API server via debugpy)
2024-03-07 19:31:50 +00:00
Dave
b71645f3b1 AAP-12273 remove incorrect sentence conjugation (#14946)
AAP-12273 remove incorrect sentance conjugation

Co-authored-by: David O Neill <daoneill@redhat.com>
2024-03-07 14:04:11 -05:00
Hao Liu
eb300252b8 Fix awx-autoreload in dev environment (#14968)
Fix awx-autoreload, recent change made autoreload no longer take the command parameter
2024-03-07 16:33:23 +00:00
Patrick Uiterwijk
2e2cd7f2de Implement project pulling from Azure DevOps using Service Principals (#14628)
* Credential Lookup with multiple types
Allow looking up a credential with one of multiple type IDs.

* Allow Azure cred for SCM
Allow selecting an Azure Resource Manager credential for Git-based SCMs.
This is in order to enable using Azure Service Principals for project updates.

* Implement Azure Service Principal Git
This adds support for using an Azure Service Principal for project updates.

---------

Signed-off-by: Patrick Uiterwijk <patrick@puiterwijk.org>
2024-03-07 10:07:03 -05:00
Hao Liu
727278aaa3 Add pip>=21.3 to dev requirement to install django-ansible-base in editable mode (#14961)
Add  pip>=21.3 to dev requirement required for installing django-ansible-base in editable mode

https://peps.python.org/pep-0660/

PEP 660 – Editable installs for pyproject.toml based builds (wheel based)
2024-03-06 21:28:41 -05:00
Michael Abashian
81825ab755 Bump axios UI dep to 1.6.z (#14954)
* Bump axios UI dep to 1.6.z

* Add proxy-from-env license
2024-03-06 16:03:49 -05:00
Helen Bailey
7f2a1b6b03 Add terraform state inventory source (#14840)
* Add terraform state inventory source
* Update inventory source plugin test
Signed-off-by: Helen Bailey <hebailey@redhat.com>
2024-03-06 20:27:52 +00:00
Hao Liu
1b56d94d30 In development environment not auto-reload explicitly STOPPED processes (#14958)
Not auto-reload explicitly STOPPED processes

In development/debug workflow sometime we explicitly STOP processes this will make sure auto-reload does not start them back up
2024-03-06 20:22:44 +00:00
Shane McDonald
e1e32c971c Allow for manually starting workflow to build devel images (#14955) 2024-03-06 17:53:05 +00:00
Alan Rominger
a4a2fabc01 Add test for utils method is_testing 2024-03-05 11:38:14 +00:00
Hao Liu
b7b7bfa520 Fix test that fail on rerun due to expecting exact IDs (#14943)
Fix test that fail on rerun

due to expecting exact IDs
2024-03-01 12:37:17 -05:00
jessicamack
887604317e Integrate resources API in Controller (#14896)
* add resources api to controller

* update setting

models are not the source of truth in AWX

* Force creation of ServiceID object in tests

* fix typo

* settings fix for CI

---------

Co-authored-by: Alan Rominger <arominge@redhat.com>
2024-03-01 11:18:35 -05:00
Chris Meyers
d35d8b6ed7 fails when ran with vscode debugger
Tried to dig as to why we ever needed this and could not find the answer. We removed it and ran all the tests and the tests passed so assuming it's no longer needed.
2024-02-29 15:22:41 -05:00
Hao Liu
ec28eff7f7 Convert swagger release fixture to env var (#14940)
`pytest awx/main/tests/docs --release=$(VERSION_TARGET)`
where --release is required breaks test discovery and running in vscode (from within the container)
2024-02-29 14:55:02 -05:00
Cesar Francisco San Nicolas Martinez
a5d17539c6 Removing Podman to use Docker again in the collection ci (#14938)
Removing Podman var to use Docker again in the collection ci
2024-02-28 14:56:49 +01:00
Thanhnguyet Vo
a49d894cf1 Added missing AWS secret management lookup creds. 2024-02-28 04:16:59 +00:00
Chris Meyers
b3466d4449 Make JWT the first auth class and default
* No harm in adding it to the list. If a JWT auth header is provided,
  then process it (valid or not). If a JWT is not provided, move on to
  the next auth.
2024-02-27 15:09:16 -05:00
Hao Liu
237adc6150 Publish multi-arch versioned awx-ee
dependent on https://github.com/ansible/awx-ee/pull/235
2024-02-27 08:21:51 +00:00
Hao Liu
09b028ee3c Publish multi-arch manifest of awx (#14929)
Promote multi-arch awx manifest
2024-02-26 17:05:57 -05:00
Hao Liu
fb83bfbc31 Fix ui_next banner (#14928) 2024-02-26 14:20:42 -05:00
Hao Liu
88e406e121 Fix CVEs and bump receptorctl (#14925)
CVE-2023-47627
CVE-2023-49083
CVE-2023-41040
CVE-2024-22195
CVE-2023-46137
2024-02-26 15:48:38 +00:00
Thanhnguyet Vo
59d0bcc63f Fixed some misc errors in illustrations and header formatting 2024-02-22 13:04:37 +00:00
Hao Liu
3fb3125bc3 Send QUIT to worker before dying (#14913)
Fix deadlock scenario where dispatcher child process stuck in reading from queue loop after dispatcher parent process decided to quit

Co-authored-by: Alan Rominger <arominge@redhat.com>
2024-02-21 16:08:43 -05:00
Alan Rominger
d70c6b9474 Reset another to test-playbooks 2024-02-21 15:11:12 +00:00
Alan Rominger
5549516a37 Reset these tests back to test-playbooks 2024-02-21 15:11:12 +00:00
Alan Rominger
14ac91a8a2 Swap repos test fix 2024-02-21 15:11:12 +00:00
Alan Rominger
d5753818a0 Stop using the bulky test-playbooks in tests where possible 2024-02-21 15:11:12 +00:00
Chad Ferman
33010a2e02 added # -*-coding:utf-8-*- to test if this fixes issues with users being unable to have Japanese, Chinese and Korean Characters in email messages 2024-02-21 14:50:46 +00:00
Alan Rominger
14454cc670 Fix playbook errors found by debugging 2024-02-21 14:36:41 +00:00
Alan Rominger
7ab2bca16e Fix missing var name change 2024-02-21 14:36:41 +00:00
Alan Rominger
f0f655f2c3 Style consistency for task 'when' 2024-02-21 14:36:41 +00:00
Brian Coca
4286d411a7 fix project_update role/collection install
- avoid looping
  - avoid using multiple files, only one should be provided and processed per type
  - use first_found and variables to locate existing file
  - skip if no file exists
2024-02-21 14:36:41 +00:00
James Talton
06ad32ed8e Enhance the dashboard job summary endpoint to contain canceled and error job counts
Signed-off-by: James Talton <jtalton@redhat.com>
2024-02-21 14:12:59 +00:00
Alan Rominger
1ebff23232 Do not rely on unreliable dir output and use exists query 2024-02-21 13:43:54 +00:00
Alan Rominger
700de14c76 Make the migration middleware faster, second attempt 2024-02-21 13:43:54 +00:00
Thanhnguyet Vo
8605e339df Deleted duplicate graphics that were converted to drawio. 2024-02-21 13:10:41 +00:00
Thanhnguyet Vo
e50954ce40 Fixed graphics, illustrations, tables, examples, sizing 2024-02-21 13:10:41 +00:00
Hao Liu
7caca60308 Multi-arch build for AWX images in ghcr.io (#14899)
build amd64 and ARM image for
- awx
- awx_devel
- awx_kube_devel
2024-02-20 17:17:31 -05:00
Cesar Francisco San Nicolas Martinez
f4e13af056 Add support for terraform credentials in awxkit (#14902) 2024-02-20 13:42:25 +00:00
Jérémy Lal
decdb56288 Fix mistake in french message 2024-02-20 12:33:59 +00:00
Fdubois97
bcd4c2e8ef Fix: Ajout de traduction sur la langue FR 2024-02-20 12:32:18 +00:00
Thaïs DOUCET
d663066ac5 Fix typo in french message
Typo in "Launch Template" french message

Signed-off-by: Thaïs DOUCET <tdoucet@amiltone.com>
2024-02-20 12:21:31 +00:00
Sasa Jovicic
1ceebb275c Fix: login rerouting only works on the user's current tab (PR:#14399)
Signed-off-by: Sasa Jovicic <sjovicic@anexia-it.com>
2024-02-20 12:16:13 +00:00
César Francisco San Nicolás Martínez
f78ba282a6 Ability to user awxkit with websocket custom urls 2024-02-20 11:58:18 +00:00
Mikhail Yohman
81d88df757 Add YAML tab for Job Output event modal. 2024-02-19 16:45:46 +00:00
Hao Liu
0bdb01a9e9 Allow dev image to build on fork (#14898)
* Allow dev image to build on fork

Fix uppercase repo owner error in CI
example
```
Run docker pull ghcr.io/TheRealHaoLiu/awx_devel:devel
  docker pull ghcr.io/TheRealHaoLiu/awx_devel:devel
  shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0}
  env:
    LC_ALL: C.UTF-8
    CI_GITHUB_TOKEN: ***
    DEV_DOCKER_OWNER: TheRealHaoLiu
    COMPOSE_TAG: devel
    py_version: 3
invalid reference format: repository name must be lowercase
```

---------

Co-authored-by: Rick Elrod <rick@elrod.me>
2024-02-19 16:19:59 +00:00
Alan Rominger
cd91fbf59f Label any changes to requirements folder with dependencies label 2024-02-19 15:56:53 +00:00
Michael Abashian
f240e640e5 Run prettier 2024-02-19 14:48:36 +00:00
ivarmu
46f489185e fixes problems with workflow nodes information section 2024-02-19 14:48:36 +00:00
Thanhnguyet Vo
dbb80fb7e3 Updated release notes so they don't need to be revised so often. 2024-02-19 12:18:57 +00:00
Seth Foster
cb3d357ce1 Disable install_bundle endpoint for ingress node
As we do for control nodes, disable the
install_bundle endpoint for ingress nodes.

This can be done by checking if instance managed
is True.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-19 11:00:06 +00:00
Chris Meyers
dfa4db9266 Add tests for websocket endpoints
* authorized/not authorized tests for wsrelay endpoint
* not authorized test for web browser websockets
* skeleton of a test for authorized web browser websockets
2024-02-17 18:37:53 -05:00
Chris Meyers
6906a88dc9 Add pytest-asyncio to test channels websockets 2024-02-17 18:37:53 -05:00
Alan Rominger
1f7be9258c Remove tower_legacy module_utils that appears unused (#14421)
* Remove tower_legacy module that appears unused

* Update license details
2024-02-16 16:02:09 -05:00
Jake Jackson
dcce024424 Update release doc to check for awxkit tar files. (#14892)
add a step to release process to confirm tar file is published
2024-02-16 18:51:58 +00:00
Jeff Bradberry
79d7179c72 Fix the persistent breakage when cleaning up branches
The github workflow that we have set up for branch deletion doesn't work:

- the `on: delete` event does not support the `branches:` filter
- the `mode` flag for the aws_s3 module does not have `delete` as one
  of the options.  The proper option appears to be `delobj`.
2024-02-15 15:16:18 -05:00
Alan Rominger
4d80f886e0 Revert "Drop cython dep" (#14884)
* Revert "Remove cython lib"

This reverts commit 46f816e7a4.

* Revert "WIP consider droping cython dep"

This reverts commit 54b32c10f0.

* Update Cython comment
2024-02-15 11:58:17 -05:00
TVo
5179333185 Added mesh ingress content to instances chapter. (#14854)
* Added mesh ingress content to instances chapter.

* Changes to incorp feedback from @TheRealHaoLiu.

* Added mesh ingress content to instances chapter.

* Line 75

Co-authored-by: Seth Foster <fosterseth@users.noreply.github.com>

* Line 117

Co-authored-by: Seth Foster <fosterseth@users.noreply.github.com>

* Line 126

Co-authored-by: Seth Foster <fosterseth@users.noreply.github.com>

* Wording changes and update graphics

---------

Co-authored-by: Seth Foster <fosterseth@users.noreply.github.com>
2024-02-15 07:19:52 -07:00
Alan Rominger
362e11aaf2 Respect old downtime setting name if user has already set it 2024-02-15 12:34:24 +00:00
Hao Liu
decff01fa4 Update command for sos-report
wsbroadcast has been renamed to wsrelay
2024-02-15 12:32:05 +00:00
Daniel Gonçalves
a14cc8199d Add python 3.12 dependency (#14869)
Signed-off-by: Daniel Gonçalves <daniel.gonc@lves.fr>
2024-02-14 15:12:51 -05:00
Chris Meyers
b6436826f6 Support websocket per-endpoint auth
* Channels doesn't really give you an interface to support per-endpoint
  auth ... so this adds one.
* The web browser and node <--> node communication have different auth
  needs.
2024-02-14 15:11:52 -05:00
Chris Meyers
2109b5039e Fixes "Task was destroyed but it is pending"
* asyncio.gather expects *tasks .. gotta unpack the list
2024-02-14 15:11:52 -05:00
root
b6f9b73418 adding iputils 2024-02-14 17:01:15 +00:00
Christian M. Adams
40a8a3cb2f Add dockerx make target for building awx for arm64
Signed-off-by: Christian M. Adams <chadams@redhat.com>
2024-02-14 16:01:36 +00:00
David O Neill
19f80c0a26 GH13983 - Add additional check for bad templates 2024-02-14 15:49:26 +00:00
Christian Adams
5d1bb2125e Update .github/workflows/stage.yml
Co-authored-by: Hao Liu <44379968+TheRealHaoLiu@users.noreply.github.com>
2024-02-14 15:44:21 +00:00
Christian Adams
99c512bcef Update .github/workflows/stage.yml
Co-authored-by: Hao Liu <44379968+TheRealHaoLiu@users.noreply.github.com>
2024-02-14 15:44:21 +00:00
Christian M. Adams
ed0329f5db Build multi-arch awx-operator images when releasing 2024-02-14 15:44:21 +00:00
Chris Meyers
dd53345397 shhhhhhhhhhhhhhh
Hopefully silence some setuptools
2024-02-14 15:38:31 +00:00
Chris Meyers
f66cde51d7 More locked down websocket path
* Previously, the nginx location would match on /foo/websocket... or
  /foo/api/websocket... Now, we require these two paths to start at the
  root i.e. <host>/websocket/... /api/websocket/...
* Note: We now also require an ending / and do NOT support
  <host>/websocket_foobar but DO support <host>/websocket/foobar. This
  was always the intended behavior. We want to keep
  <host>/api/websocket/... "open" and routing to daphne in case we want
  to add more websocket urls in the future.
2024-02-14 13:50:51 +00:00
thedoubl3j
d1c31687fc remove the ldap volume when cleaning all volumes 2024-02-14 12:33:42 +00:00
Jeff Bradberry
38424487f1 Unbreak the pip-compile command when multiple files are passed in (#14875) 2024-02-13 15:59:04 -05:00
Hao Liu
b0565e9937 Switch to docker_compose_v2 in tools playbook (#14872)
Switch to docker_compose_v2

Fix
```
"Configuration error - kwargs_from_env() got an unexpected keyword argument 'ssl_version'"}
```
2024-02-13 13:05:33 -05:00
Hao Liu
44d85b589c Retries on vault on seal (#14873)
Sometime we tried to unseal when vault is not ready yet
2024-02-13 13:05:23 -05:00
Alan Rominger
46f816e7a4 Remove cython lib 2024-02-13 14:45:28 +00:00
Alan Rominger
54b32c10f0 WIP consider droping cython dep 2024-02-13 14:45:28 +00:00
Alan Rominger
20202054cc Use lowercase password 2024-02-13 14:36:39 +00:00
Alan Rominger
e84e2962d0 Support DB configs where PASSWORD is not used 2024-02-13 14:36:39 +00:00
Chris Meyers
2259047527 Websockets now use rest_framework configed auth
* Always support cookies, session, and also allow rest_framework
  configured auth methods over the browser websocket.
* The node -> node websocket auth remains locked down and unchanged
2024-02-13 12:05:25 +00:00
Chris Meyers
f429ef6ca7 Allow connecting to websockets via api/websocket/
* Before, we just allowed websockets on <host>/websocket/. With this
  change, they can now come from <host>/api/websocket/
2024-02-13 12:02:44 +00:00
Hao Liu
4b637c1319 gitignore pyenv python-version file
pyenv local can be use to set directory specific python version in `.python-version` file in the directory

Signed-off-by: Hao Liu <44379968+TheRealHaoLiu@users.noreply.github.com>
2024-02-13 12:00:57 +00:00
Alan Rominger
4c41f6b018 Ability to use updater script to pin dev requirements (#14644)
* Add a dev option for updater script to pin CI reqs

* Avoid removing git links for dev requirements

* Add dev to primary options

* Fix up sanitize git switch
2024-02-12 11:57:59 -05:00
Jesse Wattenbarger
3ae72219b4 Change parsing of docker info in dev build
This is a non-functional change. The way os_info is populated with docker info
and grep 'Operating System' breaks on podman and likely in other places. This
makes it work on both podman and docker, and it will continue to return the
exact same strings everywhere else.
2024-02-12 16:40:48 +00:00
jessicamack
402c29dc52 Switch mailing list to forum (#14600)
* Switch mailing list to forum

* add link to community

Co-authored-by: Hao Liu <44379968+TheRealHaoLiu@users.noreply.github.com>

* use correct channel name

Co-authored-by: Hao Liu <44379968+TheRealHaoLiu@users.noreply.github.com>

---------

Co-authored-by: Hao Liu <44379968+TheRealHaoLiu@users.noreply.github.com>
2024-02-12 15:51:31 +00:00
Alan Rominger
8eb4a9a2a0 Update location of logstash build context (#14676) 2024-02-12 15:49:29 +00:00
Jeff Bradberry
36f3b46726 Avoid using SmartFilter during migrations (#14786)
Our migrations that touch roles tend to bring in our real models via
migration_utils.set_current_apps_for_migrations, and that can have
some undesirable side-effects.
2024-02-12 15:48:58 +00:00
Bikouo Aubin
55c6a319dc Add new credential type to support Terraform backend configuration (#14828)
* Add new credential type to support configuration of Terraform Backend

* Fix unit tests
2024-02-12 15:47:24 +00:00
Dave
56b6a07f6e Remove json serialization for notify validation (#14847)
* Remove json serialization for notify validation

* Update serializers.py
2024-02-12 15:43:43 +00:00
Jake Jackson
519fd22bec Add ldap support to vault container in docker dev environment (#14777)
* add ldap_auth mount and configure it

* added in key engines, userpass auth method, still needs testing

* add policies and fix ldap_user

* start awx automation for vault demo and move ldap

* update docs with new flags/new credentials
2024-02-09 15:19:17 -05:00
TVo
2e5306ae8e Added LDAP support for HashiCorp Vault lookup credential (#14833)
* Added LDAP support for HashiCorp Vault lookup credential

* Added LDAP support for HashiCorp Vault lookup credential

* Replaced graphics and updated missing fields.

* Added LDAP support for HashiCorp Vault lookup credential

* Replaced graphics and updated missing fields.

* Incorporated review feedback from @thedoubl3j and @djyasin.
2024-02-09 10:17:14 -07:00
Cesar Francisco San Nicolas Martinez
068e6acbd5 Fix the way we are passing the awxkit base path to resources (#14862) 2024-02-09 15:31:52 +01:00
Seth Foster
f9a23a5645 Remove enable button for hop nodes (#14861)
Enabled/disabled does not apply to hop nodes,
since hop nodes don't run jobs.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-08 22:10:10 -05:00
Seth Foster
40150a2be8 Fix UI peers_from_control_nodes (#14858)
* Fix UI peers_from_control_nodes

Fixes bug where peers_from_control_node was
greyed out in UI.

Additional changes:
- Make edit instance button only show for instances
with managed = False
- Make remove instance button only show for instances
with managed = False
- InstanceList selectable only for instances with
managed = False

---------

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-08 14:13:11 -05:00
TVo
b79aa5b1ed Removed erroneous line for basic auth 2024-02-08 12:45:47 -05:00
Jeff Bradberry
b3aeb962ce Fix the test_export_system_auditor collection test 2024-02-07 15:55:19 -05:00
Jeff Bradberry
2300b8fddf Fix linting problem 2024-02-07 15:55:19 -05:00
Lorenzo Tanganelli
3a3284b5df rework with check if POST exists 2024-02-07 15:55:19 -05:00
Lorenzo Tanganelli
2359004cc1 fix decription extraction on export 2024-02-07 15:55:19 -05:00
Lorenzo Tanganelli
694d7e98e7 leave $encrypted$ on export
add encrypted removal from import when object not exists
2024-02-07 14:29:30 -05:00
Julen Landa Alustiza
8c9c02c975 awxkit: allow to modify api base url (#14835)
Signed-off-by: Julen Landa Alustiza <jlanda@redhat.com>
2024-02-07 12:26:42 +01:00
Chris Meyers
8a902debd5 Per-service metrics http server
* Organize metrics into their respective service
* Server per-service metrics on a per-service http server
* Increase prometheus client usage over our custom metrics fields
2024-02-05 15:17:24 -05:00
Seth Foster
6dcaa09dfb UI rename Endpoints to Listener Addresses
Listener Addresses is a better name to
emphasize these are routable addresses to
reach a listener service on the node.

Also removed expand toggle on the listener
addresses list items, as the expanded mode
had no additional information.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
21fd6af0f9 InstanceLink unique constraint source and target
Prevent creating InstanceLinks with duplicate
source and target pairings.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
eeae1d59d4 Disable health check button if managed
Also, update ui screen tests to expect
injecting "listener_port: null" if
listener_port is empty

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
a252d0ae33 Fix UI lint by running npm prettier
Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
48971411cc Protocol blank if no canonical address
Make protocol be blank on instance if there
is no canonical address for this instance.

It was defaulting to "tcp" before.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
083c05f12a Prevent duplicating instance links
In receptor address post-save method:
- Fixed detecting if address was missing
a link from control nodes
- Use InstanceLink create_or_update to prevent
adding duplicate InstanceLink source and target
peers

In instance serializer create_or_update,
delete receptor addresses first before doing
instance create or update. This ensures that we don't
trigger unnecessary post-save methods that might
attempt to manipulate receptor addresses that
will just be removed later.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
b558397b67 Remove redundant tests
test_listener_port
test_peers_from_control_nodes
test_peers_from_control_nodes_without_listener_port

are covered in the following tests:

test_no_op
test_creates_canonical_address
test_deletes_canonical_address
test_updates_canonical_address
test_canonical_address_validation_error

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
904c6001e9 If managed, cannot modify peers_from_control_nodes
Adds validation to prevent changing
peers_from_control_nodes if instance managed=True

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
818e11dfdc Test inspect_established_receptor_connections
Add functional test case for inspecting
established receptor connections.

InstanceLink starts in ADDING state, and should
move to ESTABLISHED state if the connection
is detected in the receptor status output.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Jeff Bradberry
7fc13a0569 Write tests around the two special instance serializer fields
and all of the cases that they might be in.
2024-02-02 10:37:41 -05:00
Jeff Bradberry
92c693f14e Break out peer validation into its own method 2024-02-02 10:37:41 -05:00
Jeff Bradberry
f2417f0ed2 Make the peer validation more compact
and reuse information.
2024-02-02 10:37:41 -05:00
Jeff Bradberry
8f22188116 Use a select_related to build the peers queryset in the install bundle
Since the relationship is ReceptorAddress -> Instance,
prefetch_related isn't necessary.
2024-02-02 10:37:41 -05:00
Jeff Bradberry
05502c0af8 Placeholder FIXMEs for things of concern 2024-02-02 10:37:41 -05:00
Jeff Bradberry
957ce59bf7 Use Counter to find duplicate peer relationships
Gives a bit more readability.
2024-02-02 10:37:41 -05:00
Jeff Bradberry
cc4cc37d46 'managed' is a read-only field on InstanceSerializer
so we don't need complex logic to compare an incoming to existing value.
2024-02-02 10:37:41 -05:00
Jeff Bradberry
1e254c804c The listener port cannot be disabled when setting peers_from_control_nodes
If the port is explicitly set to null (causing any ReceptorAddress to
be deleted), then that's a validation error.

If the port is left off but a ReceptorAddress doesn't already exist,
we should not infer a port number and that is also a validation error.
2024-02-02 10:37:41 -05:00
Seth Foster
1b44bebed3 Peers_from_control_nodes requires listener port
Adds validation and a unit test to ensure:

- peers_from_control_nodes=True should fail if
listener_port is not set
- peers_from_control_nodes=False should be NOOP if
listener_port is not set

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
a4cf55bdba Require receptor collection 2.0.3
This release supports ingress hop node, and
is backwards compatible with previous AWX
version.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
c333d0e82f Prevent modifying peers on managed node
Add validation to prevent any managed node
from modifying "peers" through the API

Peering from these nodes should be handled
by setting peers_from_control_nodes only.

Managed nodes are control nodes and
ingress hop nodes.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
David O Neill
b093c89a84 Form hardening and node type exclusion
Disabled add/edit/remove for managed nodes
Tightened validation on Peers from control nodes
2024-02-02 10:37:41 -05:00
Jeff Bradberry
f98493aa61 Support wss as ws-listener in the Receptor config 2024-02-02 10:37:41 -05:00
Hao Liu
c36d2b0485 Update requirements.yml 2024-02-02 10:37:41 -05:00
Jeff Bradberry
8ddb604bf1 Template the listener protocol into the receptor install bundle (#14792)
Previously we were hard-coding tcp, now we need to also support ws/wss.
2024-02-02 10:37:41 -05:00
Seth Foster
cd9dd43be7 Make InstanceLink target non-nullable
InstanceLink target should not be null.

Should be safe to set to null=False, because we have
a custom RunPython method to explicitly set
target to a proper key.

Also, add new test to test_migrations
which ensures data integrity after migrating
the receptor address model changes.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Jeff Bradberry
82323390a7 Reconstitute migration file
Do the things as a proper 3-step process -- add new stuff, copy the data, drop the old.
2024-02-02 10:37:41 -05:00
Seth Foster
4c5ac1d3da Add management command to remove address
Adds remove_receptor_address to delete a
receptor address from the database

Also, enforce that only 1 canonical address
can be added to an instance via
the add_receptor_address command.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
David O Neill
9c06370e33 InstanceAdd sends null for port_listener 2024-02-02 10:37:41 -05:00
David O Neill
449b95d1eb Fix remaning tests, removed unused code 2024-02-02 10:37:41 -05:00
David O Neill
1712540c8e Updates for receptor reaslese to ui for protocol and is_managed 2024-02-02 10:37:41 -05:00
Seth Foster
7cf639d8eb Add migration to support InstanceLink changes
- Add forwards method to create a receptor address
for any existing Instance that has listener_port defined

- Add forwards method to modify each InstanceLink object
that changes target to the newly created receptor addresses

This migration was implemented as follows:

1. Add a target_new to InstanceLink which is a foreign key
to ReceptorAddress
2. create receptor addresses
3. link to these receptor addresses using the target_new field
4. rename target_new to target
5. drop listener_port and peers_from_control_nodes from Instance

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
dbfcc40d7c Only create receptor address if port is defined
If a Instance endpoint is patched with
{"peers_from_control_nodes" True}

but a listener_port is not defined on the instance,
or is part of the patch payload, do not create a
receptor address.

Only update or create a receptor address if listener_port
is set, either in the payload or already on the instance.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Hao Liu
73d2c92ae3 Fix condition for creating receptor_address
If listener_port is not explicitly defined don't create a receptor_address
2024-02-02 10:37:41 -05:00
Seth Foster
24a4242147 Add protocol to receptor address serializer
Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
92ce85b688 Remove unused warnings import
Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
9531f8377a Ensure register_peers target is ReceptorAddress
command has 3 "targets", all of which should be a
ReceptorAddress object

- peers, disconnect, exact

Add logic to make sure each entry in those lists
are receptor addresses.

When creating InstanceLink objects, make sure
target is ReceptorAddress, not an Instance.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Hao Liu
15a16b3dd1 Update bootstrap_development.sh 2024-02-02 10:37:41 -05:00
Seth Foster
a37e7bf147 Add canonical=True when creating ReceptorAddress in tests
Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Hao Liu
a2fcd2f97a Fix ui-lint error 2024-02-02 10:37:41 -05:00
Hao Liu
c394ffdd19 Fix lint error, remove unused import 2024-02-02 10:37:41 -05:00
Seth Foster
69102cf265 Remove receptor_address module from collection
After removing CRUD from receptor addresses, we need
to remove the module.

- remove receptor_address module
- Add listener_port to instance module
- Add peers_from_control_nodes to instance module

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
a188798543 Make canonical field default to False
Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
60108ebd10 Rename migration dependency
Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Jeff Bradberry
8c7c00451a Join across the InstanceLink.target to the underlying Instance
to avoid having to do a lookup for every instance in the list for the
mesh visualizer endpoint.
2024-02-02 10:37:41 -05:00
Seth Foster
7a1ed406da Remove CRUD for Receptor Addresses
Removes ability to directly create and delete
receptor addresses for a given node.

Instead, receptor addresses are created automatically
if listener_port is set on the Instance.

For example patching "hop" instance

with {"listener_port": 6667}

will create a canonical receptor address with port
6667.

Likewise, peers_from_control_nodes on the instance
sets the peers_from_control_nodes on the canonical
address (if listener port is also set).

protocol is a read-only field that simply reflects
the canonical address protocol.

Other Changes:

- rename k8s_routable to is_internal
- add protocol to ReceptorAddress
- remove peers_from_control_nodes and listener_port
from Instance model

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
César Francisco San Nicolás Martínez
f916ffe1e9 Adjust migration names and dependencies 2024-02-02 10:37:41 -05:00
César Francisco San Nicolás Martínez
901dbd697e Comment unused dependency 2024-02-02 10:37:41 -05:00
David O Neill
d8b4a9825e Cleanup
- Remove peer selection on add and edit instance
- Added conconcial name and order comlums names same on endpoints and
  peers
- Other cleanup items
- rename name to instance name
2024-02-02 10:37:41 -05:00
Hao Liu
6db66c5f81 Fix provision instance not respecting protocol 2024-02-02 10:37:41 -05:00
David O Neill
82ad7dcf40 Mesh UI support
- add endpoint
- delete endpoint (wip)
- associate
- disassociate
2024-02-02 10:37:41 -05:00
David O Neill
93500f9fea Fix lint trailing whitespace 2024-02-02 10:37:41 -05:00
Seth Foster
9ba70c151d Add canonical receptor address
Creates a non-deletable address that acts as
the "main" address for this instance.

All other addresses for that instance must
be non-canonical.

When listener_port on an instance is set, automatically
create a canonical receptor address where:
  - address is hostname of instance
  - port is listener_port
  - canonical is True

Additionally, protocol field is added to instance to
denote the receptor listener protocol to use (ws, tcp).

The receptor config listener information is derived from
the listener_port and protocol information. Having a
canonical address that mirrors the listener_port ensures that
an address exists that matches the receptor config information.

Other changes:
- Add managed field to receptor address.
If managed is True, no fields on on this address can be edited
via the API.
If canonical is True, only the address cannot be edited.

- Add managed field to instance. If managed is True, users
cannot set node_state to deprovisioning (i.e. cannot delete node)

This change to our mechanism to prevent users from deleting
the mesh ingress hop node.

- Field is_internal is now renamed to k8s_routable

- Add reverse_peers on instance which is a list of instance IDs
that peer to this instance (via an address)

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
David O Neill
46dc61253f UI Updates for receptor peering 2024-02-02 10:37:41 -05:00
Seth Foster
6cb2cd18b0 Fix proper indent to instance module
Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
5d1dd8ec41 Fix inconsistent tab width
Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
9f69daf787 Add choices to module protocol field
Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
16ece5de7e Remove unused variables and imports
Addresses flake8 and linting failures

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
ab0e9265c5 Add search fields to views
Make receptoraddress list views
searchable by "address"

Other changes:
- Add help text to source and target of the
InstanceLink model

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
04cbbbccfa Update awx_collection to support ReceptorAddress
- Add receptor_address module which allows
users to create addresses for instances

- Update awx_collection functional and integration
tests to support new peering design

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
d1cacf64de Add functional and unit tests
Updated existing tests to support the
ReceptorAddress model

- cannot peer to self
- cannot peer to node that is already peered to me
- cannot peer to node more than once (via 2+ addresses)
- cannot set is_internal True

Other changes:
Change post save signal to only call
schedule_write_receptor_config() when an actual change is detected.

Make functional tests more robust by
checking for specific validation error in the
response.

I.e. instead of just checking for 400, just for 400
and that the error message corresponds to the
validation we are testing for.

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
5385eb0fb3 Add validation when setting peers
- cannot peer to self
- cannot peer to instance that is already peered to self

Other changes:
- ReceptorAddress protocol field restricted to choices: tcp, ws, wss
- fix awx-manage list_instances when instance.last_seen is None
- InstanceLink make source and target unique together
- Add help text to the ReceptorAddress fields

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
7d7503279d Add API validation when creating ReceptorAddress
- websocket_path can only be set if protocol is ws
- is_internal must be False
- only 1 address per instance can have
peers_from_control_nodes set to True

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Hao Liu
d860d1d91b Temp change to aid dev
Temporarily switch to feature branch for receptor-collection in instance_install_bundle to make dev easier
2024-02-02 10:37:41 -05:00
Seth Foster
3a17c45b64 Register_peers support for receptor_addresses
register_peers has inputs:

source: source instance
peers: list of instances the source should peer to

InstanceLink "target" is now expected to be a ReceptorAddress

For each peer, we can just use the first receptor address. If
multiple receptor addresses exist, throw a command error.

Currently this command is only used on VM-deployments, where
there is only a single receptor address per instance, so this
should work fine.

Other changes:
drop listener_port field from Instance. Listener port is now just
"port" on ReceptorAddress

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
bca68bcdf1 Add install bundle support
group_vars all.yaml changes:
- peer entry has two fields, address and port
- receptor_port is inferred from the first
receptor_address entry that uses protocol tcp

other changes:
ActivityStream now records when receptor_addresses
are peered to

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
c32f234ebb Add peers_from_control_nodes to ReceptorAddress
- write_receptor_config peers to ReceptorAddress entries
that have peers_from_control_nodes enabled

- peers_from_control_nodes and listener_port removed from Instance model

- peers_from_control_nodes added to ReceptorAddress model

- ReceptorAddress is now unique by address and protocol combination

- Write receptor config task is dispatched upon ReceptorAddress creation
or deletion, and when control node is first created

- InstanceLinkSerializer adds a target_address field and has logic
to grab the instance hostname associated with the peered ReceptorAddress

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
5cb3d3b078 Update receptor conf when address changes
Add post save and post delete hooks to
call write_receptor_config when
a receptor address is added / removed.

Add peers_from_control_nodes to
provision_instance

Signed-off-by: Seth Foster <fosterbseth@gmail.com>
2024-02-02 10:37:41 -05:00
Seth Foster
5199cc5246 Add ReceptorAddress to root urls
- Add database contraints to make sure addresses
are unique
If port is defined:
address, port, protocol, websocket_path are unique together

if port is not defined:
address, protocol, websocket_path are unique together

- Allow deleting address via API
- Add ReceptorAddressAccess to determine permissions
- awx-manage add_receptor_address returns changed: True
if successful
2024-02-02 10:37:41 -05:00
Hao Liu
387e877485 Connect from controlplane node to mesh ingress 2024-02-02 10:37:41 -05:00
Seth Foster
d54c5934ff Add support for inbound hop nodes 2024-02-02 10:37:41 -05:00
Chris Meyers
2fa5116197 Project updates do not run against hosts
* The project update event has no host_id or host_name because they
  only run on localhost.
2024-02-01 14:45:16 -05:00
Chris Meyers
527755d986 Always get host from event data
* Regardless of if there is a host map or not, get the host that Ansible
  reports and assign it to the Events' host_name
2024-01-31 09:42:01 -05:00
Chris Meyers
f9c0b97c53 Avoid EDA dev env port conflict
* Not many, if any, folks use the notebook feature. It kind of goes in
  and out of popularity. We've used it in the past when we work on
  features that require visualization (i.e. network graphs, workflows).
  Might as well keep it around in case we use it again.
2024-01-30 11:17:30 -05:00
TVo
65655f84de Added section for using private image for default EE. (#14815)
* Added hop node information and beefed up exec nodes section.

* Made updates to the Topology viewer & dis/associate peers

* Added hop node information and beefed up exec nodes section.

* Made updates to the Topology viewer & dis/associate peers

* Fixed build and image rendering issues.

* Replaced graphic with many many nodes!

* Updated images with latest

* Incorporated review inputs from @fosterseth.

* Incorporated more review feedback from @fosterseth

* Resized graphic

* Resized graphic again

* Reduced image sizes to scale better on different browsers.

* Added section for using private image for default EE.

* Deleted .md file for execution nodes due to migration to RTD.
2024-01-29 17:48:53 -07:00
Elijah DeLee
9aa3d5584a fix nginx append slash to respect proxy
This is already fixed in awx-operator.
See a534c856db/roles/installer/templates/configmaps/config.yaml.j2 (L215)
This just makes it so a development environment can also work correctly
behind a proxy

Fixes problem of
GET to https://$PROXY/something/awx/v2/me
rewritten to https://$AWX/something/awx/v2/me/ (which doesn't exist)

instead path is correctly rewritten as https://$PROXY/something/awx/v2/me/
2024-01-29 15:30:16 -05:00
TVo
266e31d71a Added hop node information and beefed up exec nodes section. (#14787)
* Added hop node information and beefed up exec nodes section.

* Made updates to the Topology viewer & dis/associate peers

* Added hop node information and beefed up exec nodes section.

* Made updates to the Topology viewer & dis/associate peers

* Fixed build and image rendering issues.

* Replaced graphic with many many nodes!

* Updated images with latest

* Incorporated review inputs from @fosterseth.

* Incorporated more review feedback from @fosterseth

* Resized graphic

* Resized graphic again

* Reduced image sizes to scale better on different browsers.
2024-01-27 14:23:32 -07:00
Alan Rominger
a1bbe75aed Adopt new rules from black upgrade (#14809) 2024-01-26 12:54:44 -05:00
Sandra McCann
695f1cf892 Replaced old tower docs link with new AWX docs link (#14801) 2024-01-25 17:23:48 -07:00
Chris Meyers
0ab103d8c4 Get that new AWX DAB hotness 2024-01-25 15:45:18 -05:00
Chris Meyers
9ac1c0f6c2 Specify docker network when multiple networks
* We introduced multi networks to our docker env. The code this replaces
  would return both networks' ip addresses concatinated i.e.
  '192.168.2.1192.168.2.3'.
2024-01-25 14:47:02 -05:00
Lila Yasin
2e168d8177 Add userpass and LDAP support for HashiCorp vault credential_plugin (#14654)
* Add username and password to handle_auth and update exception message

Revise naming of ldap username and password

* Add url for LDAP and userpass to method_auth

* Add information regarding LDAP and username and password to credential plugins documentation

Revise ldap_auth to userpass_auth and revised exception to better reflect functionality

* Revise method_auth to ensure certs can be used with username and ensure namespace functionality is not hindered
2024-01-25 09:50:13 -05:00
Kristof Wevers
d4f7bfef18 feat: Add retries to requests sessions
Every so often we get connection timed out errors towards our HCP Vault
endpoint. This is usually when a larger number of jobs is running
simultaneously. Considering requests for other jobs do still succeed this
is probably load related and adding a retry should help in making this a
bit more robust.
2024-01-24 15:45:54 -05:00
dependabot[bot]
985a8d499d Bump jinja2 from 3.1.2 to 3.1.3 in /docs/docsite
Bumps [jinja2](https://github.com/pallets/jinja) from 3.1.2 to 3.1.3.
- [Release notes](https://github.com/pallets/jinja/releases)
- [Changelog](https://github.com/pallets/jinja/blob/main/CHANGES.rst)
- [Commits](https://github.com/pallets/jinja/compare/3.1.2...3.1.3)

---
updated-dependencies:
- dependency-name: jinja2
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
2024-01-24 15:36:19 -05:00
Chris Meyers
e3b52f0169 Join the service-mesh docker network
* Put the awx node(s) on a service-mesh docker network so they can be
  proxied to. Also put all the other containers on an explicit awx
  network otherwise they can not talk to each other. We might could be
  more surgical about what containers we put on awx but I just added all
  of them.
2024-01-24 10:34:44 -05:00
jessicamack
f69f600cff Refer to the ansible repo for django-ansible-base requirement (#14793)
* update to the proper repo

* refer to devel
2024-01-22 10:29:47 -05:00
TVo
74cd23be5c Fixed/updated URL for “Passing Variables on the Command Line" link. (#14763) 2024-01-19 11:41:17 -07:00
jessicamack
209747d88e Update for django-ansible-base split (#14783)
* update paths and names

* temp to get tests passing

* fix typo
2024-01-19 12:30:32 -05:00
Alan Rominger
d91da39f81 New setting for pg_notify listener DB settings, add keepalive (#14755) 2024-01-17 13:44:04 -05:00
Michael Tipton
5cd029df96 Add secure flag option for userLoggedIn cookie if SESSION_COOKIE_SECU… (#14762)
Add secure flag option for userLoggedIn cookie if SESSION_COOKIE_SECURE set to True
2024-01-17 09:36:06 -05:00
Michael Abashian
5a93a519f6 Fix linting error in SubscriptionUsageChart 2024-01-16 14:12:55 -05:00
jessicamack
5f5cd960d5 Add django-ansible-base settings (#14768)
add ansible base settings
2024-01-16 15:55:59 +00:00
Jeff Bradberry
42701f32fe Build the source distribution bundle to also upload to PyPI
This has been broken since 20.0.1.
2024-01-11 15:49:07 -05:00
Hao Liu
30d4df788f Update dependency django-ansible-base (#14752) 2024-01-10 11:05:57 -05:00
Cameron McLaughlin
1bcd71a8ac Update execution environment documentation link (fixes #14690) (#14741)
Update execution enviorment documentation link (fixes #14690)

Signed-off-by: Cameron McLaughlin <auatr7@protonmail.com>
2024-01-05 15:47:55 -07:00
Patrick Uiterwijk
43be90f051 Add support for Bitbucket Data Center webhooks (#14674)
Add support for receiving webhooks from Bitbucket Data Center, and add support for posting build statuses back

Note that this is very explicitly only for Bitbucket Data Center.
The entire webhook format and API is entirely different for Bitbucket Cloud.
2024-01-05 09:34:29 -05:00
TVo
bb1922cdbb Update conf.py to 2024 (#14743)
* Update conf.py to 2023

* Update awxkit/awxkit/cli/docs/source/conf.py

Co-authored-by: Hao Liu <44379968+TheRealHaoLiu@users.noreply.github.com>

---------

Co-authored-by: Hao Liu <44379968+TheRealHaoLiu@users.noreply.github.com>
2024-01-04 12:38:21 -07:00
Martin Slemr
403f545071 Fix port conflicts when running other Ansible dev environments (#14701)
AAP: Docker port conflicts
2024-01-04 09:10:55 -05:00
Nenodema
a06a2a883c Adding "address" property 2024-01-03 15:08:18 -05:00
Keith Grant
2529fdcfd7 Persist schedule prompt on launch fields when editing (#14736)
* persist schedule prompt on launch fields when editing

* Merge job template default credentials with schedule overrides in schedule prompt

* rename vars for clarity

* handle undefined defaultCredentials

---------

Co-authored-by: Michael Abashian <mabashia@redhat.com>
2023-12-21 14:46:49 -05:00
loh
19dff9c2d1 Fix twilio_backend.py to send SMS to multiple destinations. (#14656)
AWX only sends Twilio notifications to one destination with the current version of code, but this is a bug. Fixed this bug for sending SMS to multiple destinations.
2023-12-20 15:31:47 -05:00
Chris Meyers
2a6cf032f8 refactor awxkit import code
* Move awxkit import code into a pytest fixture to better control when
  the import happens
* Ensure /awx_devel/awxkit is added to sys path before awxkit import
  runs
2023-12-18 12:00:50 -05:00
Chris Meyers
6119b33a50 add awx collection export tests
* Basic export tests
* Added test that highlights a problem with running Schedule exports as
  non-root user. We rely on the POST key in the OPTIONS response to
  determine the fields to export for a resource. The POST key is not
  present if a user does NOT have create privileges.
* Fixed up forwarding all headers from the API server back to the test
  code. This was causing a problem in awxkit code that checks for
  allowed HTTP Verbs in the headers.
2023-12-18 12:00:50 -05:00
John Westcott IV
aacf9653c5 Use filtering/sorting from django-ansible-base (#14726)
* Move filtering to DAB

* add comment to trigger building a new image

Signed-off-by: jessicamack <jmack@redhat.com>

* remove unneeded comment

Signed-off-by: jessicamack <jmack@redhat.com>

* remove unused imports

Signed-off-by: jessicamack <jmack@redhat.com>

* change mock import

Signed-off-by: jessicamack <jmack@redhat.com>

---------

Signed-off-by: jessicamack <jmack@redhat.com>
Co-authored-by: jessicamack <jmack@redhat.com>
2023-12-18 10:05:02 -05:00
Alan Rominger
325f5250db Narrow the actor types accepted for RBAC evaluations (#14709)
* Narrow the scope of RBAC evaluations

* Update tests for RBAC method changes

* Simplify querset for credentials in org

* Fix call pattern to pass in team role obj
2023-12-14 21:30:47 -05:00
Alan Rominger
b14518c1e5 Simplify RBAC get_roles_on_resource method (#14710)
* Simplify RBAC get_roles_on_resource method

* Fix bug

* Fix query type bug
2023-12-14 10:42:26 -05:00
Hao Liu
6440e3cb55 Send SIGKILL to rsyslog if hard cancellation is needed 2023-12-14 10:41:48 -05:00
Hao Liu
b5f6aac3aa Correct misuse of stdxxx_event_enabled
Not every log messages need to be emitted as a event!
2023-12-14 10:41:48 -05:00
Hao Liu
6e5e1c8fff Recover rsyslog from 4xx error
Due to https://github.com/ansible/awx/issues/7560

'omhttp' module for rsyslog will completely stop forwarding message to external log aggregator after receiving a 4xx error from the external log aggregator

This PR is an "workaround" for this problem by restarting rsyslogd after detecting that rsyslog received a 4xx error
2023-12-14 10:41:48 -05:00
Hao Liu
bf42c63c12 Remove superwatcher from docker-compose dev (#14708)
When making changes to the application sometime you can accidentally cause FATAL state and cause the dev container to crash which will remove any ephemeral changes that you have made and is ANNOYING!
2023-12-13 14:26:53 -05:00
Avi Layani
df24cb692b Adding hosts bulk deletion feature (#14462)
* Adding hosts bulk deletion feature

Signed-off-by: Avi Layani <alayani@redhat.com>

* fix the type of the argument

Signed-off-by: Avi Layani <alayani@redhat.com>

* fixing activity_entry tracking

Signed-off-by: Avi Layani <alayani@redhat.com>

* Revert "fixing activity_entry tracking"

This reverts commit c8eab52c2ccc5abe215d56d1704ba1157e5fbbd0.
Since the bulk_delete is not related to an inventory, only hosts which
can be from different inventories.

* get only needed vars to reduce memory consumption

Signed-off-by: Avi Layani <alayani@redhat.com>

* filtering the data to reduce memory increase the number of queries

Signed-off-by: Avi Layani <alayani@redhat.com>

* update the activity stream for inventories

Signed-off-by: Avi Layani <alayani@redhat.com>

* fix the changes dict initialiazation

Signed-off-by: Avi Layani <alayani@redhat.com>

---------

Signed-off-by: Avi Layani <alayani@redhat.com>
2023-12-13 10:28:31 -06:00
jessicamack
0d825a744b Update setuptools-scm (#14716)
* properly format requirement

* upgrade setuptools_scm

* Revert "properly format requirement"

This reverts commit 4c8792950f.

* test ansible-runner package upgrade

* Revert "test ansible-runner package upgrade"

This reverts commit ba4b74f2bb.
2023-12-11 17:11:04 -05:00
Marliana Lara
5e48bf091b Fix undefined error in settings/logging/edit form (#14715)
Fix undefined error in logging settings edit form
2023-12-11 10:58:30 -05:00
Alan Rominger
1294cec92c Fix updater bug due to missing newline at EOF (#14713) 2023-12-08 16:51:17 +00:00
jessicamack
dae12ee1b8 Remove incorrectly formatted line from requirements.txt (#14714)
remove git+ line
2023-12-08 11:06:17 -05:00
jessicamack
b091f6cf79 Add django-ansible-base (#14705)
* add django-ansible-base

Signed-off-by: jessicamack <jmack@redhat.com>

* add licenses

* add django-ansible-base

Signed-off-by: jessicamack <jmack@redhat.com>

* add licenses

* apply patch to fix permissions issue

---------

Signed-off-by: jessicamack <jmack@redhat.com>
2023-12-07 11:45:44 -05:00
Don Naro
fe564c5fad Dependabot for docsite requirements (#14670) 2023-12-06 15:11:44 -05:00
Tyler Muir
eb3bc84461 remove unnecessary required flags for saml backend (#14666)
Signed-off-by: Tyler Muir <tylergmuir@gmail.com>
2023-12-06 15:08:54 -05:00
Andrew Austin
6aa2997dce Add TLS certificate auth for HashiCorp Vault (#14534)
* Add TLS certificate auth for HashiCorp Vault

Add support for AWX to authenticate with HashiCorp Vault using
TLS client certificates.

Also updates the documentation for the HashiCorp Vault secret management
plugins to include both the new TLS options and the missing Kubernetes
auth method options.

Signed-off-by: Andrew Austin <aaustin@redhat.com>

* Refactor docker-compose vault for TLS cert auth

Add TLS configuration to the docker-compose Vault configuration and
use that method by default in vault plumbing.

This ensures that the result of bringing up the docker-compose stack
with vault enabled and running the plumb-vault playbook is a fully
working credential retrieval setup using TLS client cert authentication.

Signed-off-by: Andrew Austin <aaustin@redhat.com>

* Remove incorrect trailing space

Co-authored-by: Hao Liu <44379968+TheRealHaoLiu@users.noreply.github.com>

* Make vault init idempotent

- improve error handling for vault_initialization
- ignore error if vault cert auth is already configured
- removed unused register

* Add VAULT_TLS option

Make TLS for HashiCorp Vault optional and configurable via VAULT_TLS env var

* Add retries for vault init

Sometime it took longer for vault to fully come up and init will fail

---------

Signed-off-by: Andrew Austin <aaustin@redhat.com>
Co-authored-by: Hao Liu <44379968+TheRealHaoLiu@users.noreply.github.com>
Co-authored-by: Hao Liu <haoli@redhat.com>
2023-12-06 19:12:15 +00:00
Don Naro
dd00bbba42 separate tox calls in readthedocs config (#14673) 2023-12-06 17:17:12 +00:00
Rick Elrod
fe6bac6d9e [CI] Reduce GHA timeouts from 6h default (#14704)
* [CI] Reduce GHA timeouts from 6h default

The goal here is to never interfere with a real run (so most of the
timeout-minutes values seem rather high) but to avoid having 6h long
runs if something goes crazy and never ends.

Signed-off-by: Rick Elrod <rick@elrod.me>

* Do bash hackery instead

Signed-off-by: Rick Elrod <rick@elrod.me>

---------

Signed-off-by: Rick Elrod <rick@elrod.me>
2023-12-06 14:43:45 +00:00
jainnikhil30
87abbd4b10 Fix the bulk Job Launch Integration test in awx collection (#14702)
* fix the integration tests
2023-12-06 18:50:33 +05:30
296 changed files with 5845 additions and 2202 deletions

View File

@@ -11,6 +11,12 @@ runs:
shell: bash
run: echo py_version=`make PYTHON_VERSION` >> $GITHUB_ENV
- name: Set lower case owner name
shell: bash
run: echo "OWNER_LC=${OWNER,,}" >> $GITHUB_ENV
env:
OWNER: '${{ github.repository_owner }}'
- name: Log in to registry
shell: bash
run: |
@@ -18,11 +24,11 @@ runs:
- name: Pre-pull latest devel image to warm cache
shell: bash
run: docker pull ghcr.io/${{ github.repository_owner }}/awx_devel:${{ github.base_ref }}
run: docker pull ghcr.io/${OWNER_LC}/awx_devel:${{ github.base_ref }}
- name: Build image for current source checkout
shell: bash
run: |
DEV_DOCKER_TAG_BASE=ghcr.io/${{ github.repository_owner }} \
DEV_DOCKER_TAG_BASE=ghcr.io/${OWNER_LC} \
COMPOSE_TAG=${{ github.base_ref }} \
make docker-compose-build

View File

@@ -35,7 +35,7 @@ runs:
- name: Start AWX
shell: bash
run: |
DEV_DOCKER_TAG_BASE=ghcr.io/${{ github.repository_owner }} \
DEV_DOCKER_OWNER=${{ github.repository_owner }} \
COMPOSE_TAG=${{ github.base_ref }} \
COMPOSE_UP_OPTS="-d" \
make docker-compose
@@ -43,10 +43,14 @@ runs:
- name: Update default AWX password
shell: bash
run: |
while [[ "$(curl -s -o /dev/null -w ''%{http_code}'' -k https://localhost:8043/api/v2/ping/)" != "200" ]]
do
echo "Waiting for AWX..."
sleep 5
SECONDS=0
while [[ "$(curl -s -o /dev/null -w ''%{http_code}'' -k https://localhost:8043/api/v2/ping/)" != "200" ]]; do
if [[ $SECONDS -gt 600 ]]; then
echo "Timing out, AWX never came up"
exit 1
fi
echo "Waiting for AWX..."
sleep 5
done
echo "AWX is up, updating the password..."
docker exec -i tools_awx_1 sh <<-EOSH
@@ -67,7 +71,7 @@ runs:
id: data
shell: bash
run: |
AWX_IP=$(docker inspect -f '{{range.NetworkSettings.Networks}}{{.IPAddress}}{{end}}' tools_awx_1)
AWX_IP=$(docker inspect -f '{{.NetworkSettings.Networks._sources_awx.IPAddress}}' tools_awx_1)
ADMIN_TOKEN=$(docker exec -i tools_awx_1 awx-manage create_oauth2_token --user admin)
echo "ip=$AWX_IP" >> $GITHUB_OUTPUT
echo "admin_token=$ADMIN_TOKEN" >> $GITHUB_OUTPUT

10
.github/dependabot.yml vendored Normal file
View File

@@ -0,0 +1,10 @@
version: 2
updates:
- package-ecosystem: "pip"
directory: "docs/docsite/"
schedule:
interval: "weekly"
open-pull-requests-limit: 2
labels:
- "docs"
- "dependencies"

View File

@@ -15,5 +15,4 @@
"dependencies":
- any: ["awx/ui/package.json"]
- any: ["requirements/*.txt"]
- any: ["requirements/requirements.in"]
- any: ["requirements/*"]

View File

@@ -11,6 +11,7 @@ jobs:
common-tests:
name: ${{ matrix.tests.name }}
runs-on: ubuntu-latest
timeout-minutes: 60
permissions:
packages: write
contents: read
@@ -49,6 +50,7 @@ jobs:
dev-env:
runs-on: ubuntu-latest
timeout-minutes: 60
steps:
- uses: actions/checkout@v3
@@ -63,6 +65,7 @@ jobs:
awx-operator:
runs-on: ubuntu-latest
timeout-minutes: 60
steps:
- name: Checkout awx
uses: actions/checkout@v3
@@ -112,6 +115,7 @@ jobs:
collection-sanity:
name: awx_collection sanity
runs-on: ubuntu-latest
timeout-minutes: 30
strategy:
fail-fast: false
steps:
@@ -123,14 +127,11 @@ jobs:
- name: Run sanity tests
run: make test_collection_sanity
env:
# needed due to cgroupsv2. This is fixed, but a stable release
# with the fix has not been made yet.
ANSIBLE_TEST_PREFER_PODMAN: 1
collection-integration:
name: awx_collection integration
runs-on: ubuntu-latest
timeout-minutes: 60
strategy:
fail-fast: false
matrix:
@@ -182,6 +183,7 @@ jobs:
collection-integration-coverage-combine:
name: combine awx_collection integration coverage
runs-on: ubuntu-latest
timeout-minutes: 10
needs:
- collection-integration
strategy:

View File

@@ -3,27 +3,50 @@ name: Build/Push Development Images
env:
LC_ALL: "C.UTF-8" # prevent ERROR: Ansible could not initialize the preferred locale: unsupported locale setting
on:
workflow_dispatch:
push:
branches:
- devel
- release_*
- feature_*
jobs:
push:
if: endsWith(github.repository, '/awx') || startsWith(github.ref, 'refs/heads/release_')
push-development-images:
runs-on: ubuntu-latest
timeout-minutes: 120
permissions:
packages: write
contents: read
strategy:
fail-fast: false
matrix:
build-targets:
- image-name: awx_devel
make-target: docker-compose-buildx
- image-name: awx_kube_devel
make-target: awx-kube-dev-buildx
- image-name: awx
make-target: awx-kube-buildx
steps:
- name: Skipping build of awx image for non-awx repository
run: |
echo "Skipping build of awx image for non-awx repository"
exit 0
if: matrix.build-targets.image-name == 'awx' && !endsWith(github.repository, '/awx')
- uses: actions/checkout@v3
- name: Get python version from Makefile
run: echo py_version=`make PYTHON_VERSION` >> $GITHUB_ENV
- name: Set up QEMU
uses: docker/setup-qemu-action@v3
- name: Set lower case owner name
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v3
- name: Set GITHUB_ENV variables
run: |
echo "OWNER_LC=${OWNER,,}" >>${GITHUB_ENV}
echo "DEV_DOCKER_TAG_BASE=ghcr.io/${OWNER,,}" >> $GITHUB_ENV
echo "COMPOSE_TAG=${GITHUB_REF##*/}" >> $GITHUB_ENV
echo py_version=`make PYTHON_VERSION` >> $GITHUB_ENV
env:
OWNER: '${{ github.repository_owner }}'
@@ -36,23 +59,19 @@ jobs:
run: |
echo "${{ secrets.GITHUB_TOKEN }}" | docker login ghcr.io -u ${{ github.actor }} --password-stdin
- name: Pre-pull image to warm build cache
run: |
docker pull ghcr.io/${OWNER_LC}/awx_devel:${GITHUB_REF##*/} || :
docker pull ghcr.io/${OWNER_LC}/awx_kube_devel:${GITHUB_REF##*/} || :
docker pull ghcr.io/${OWNER_LC}/awx:${GITHUB_REF##*/} || :
- name: Setup node and npm
uses: actions/setup-node@v2
with:
node-version: '16.13.1'
if: matrix.build-targets.image-name == 'awx'
- name: Build images
- name: Prebuild UI for awx image (to speed up build process)
run: |
DEV_DOCKER_TAG_BASE=ghcr.io/${OWNER_LC} COMPOSE_TAG=${GITHUB_REF##*/} make docker-compose-build
DEV_DOCKER_TAG_BASE=ghcr.io/${OWNER_LC} COMPOSE_TAG=${GITHUB_REF##*/} make awx-kube-dev-build
DEV_DOCKER_TAG_BASE=ghcr.io/${OWNER_LC} COMPOSE_TAG=${GITHUB_REF##*/} make awx-kube-build
sudo apt-get install gettext
make ui-release
make ui-next
if: matrix.build-targets.image-name == 'awx'
- name: Push development images
- name: Build and push AWX devel images
run: |
docker push ghcr.io/${OWNER_LC}/awx_devel:${GITHUB_REF##*/}
docker push ghcr.io/${OWNER_LC}/awx_kube_devel:${GITHUB_REF##*/}
- name: Push AWX k8s image, only for upstream and feature branches
run: docker push ghcr.io/${OWNER_LC}/awx:${GITHUB_REF##*/}
if: endsWith(github.repository, '/awx')
make ${{ matrix.build-targets.make-target }}

View File

@@ -6,6 +6,7 @@ jobs:
docsite-build:
name: docsite test build
runs-on: ubuntu-latest
timeout-minutes: 30
steps:
- uses: actions/checkout@v3

View File

@@ -2,13 +2,12 @@
name: Feature branch deletion cleanup
env:
LC_ALL: "C.UTF-8" # prevent ERROR: Ansible could not initialize the preferred locale: unsupported locale setting
on:
delete:
branches:
- feature_**
on: delete
jobs:
push:
branch_delete:
if: ${{ github.event.ref_type == 'branch' && startsWith(github.event.ref, 'feature_') }}
runs-on: ubuntu-latest
timeout-minutes: 20
permissions:
packages: write
contents: read
@@ -21,6 +20,4 @@ jobs:
run: |
ansible localhost -c local, -m command -a "{{ ansible_python_interpreter + ' -m pip install boto3'}}"
ansible localhost -c local -m aws_s3 \
-a "bucket=awx-public-ci-files object=${GITHUB_REF##*/}/schema.json mode=delete permission=public-read"
-a "bucket=awx-public-ci-files object=${GITHUB_REF##*/}/schema.json mode=delobj permission=public-read"

View File

@@ -13,6 +13,7 @@ permissions:
jobs:
triage:
runs-on: ubuntu-latest
timeout-minutes: 20
name: Label Issue
steps:
@@ -26,6 +27,7 @@ jobs:
community:
runs-on: ubuntu-latest
timeout-minutes: 20
name: Label Issue - Community
steps:
- uses: actions/checkout@v3

View File

@@ -14,6 +14,7 @@ permissions:
jobs:
triage:
runs-on: ubuntu-latest
timeout-minutes: 20
name: Label PR
steps:
@@ -25,6 +26,7 @@ jobs:
community:
runs-on: ubuntu-latest
timeout-minutes: 20
name: Label PR - Community
steps:
- uses: actions/checkout@v3

View File

@@ -10,6 +10,7 @@ jobs:
if: github.repository_owner == 'ansible' && endsWith(github.repository, 'awx')
name: Scan PR description for semantic versioning keywords
runs-on: ubuntu-latest
timeout-minutes: 20
permissions:
packages: write
contents: read

View File

@@ -15,6 +15,7 @@ jobs:
promote:
if: endsWith(github.repository, '/awx')
runs-on: ubuntu-latest
timeout-minutes: 90
steps:
- name: Checkout awx
uses: actions/checkout@v3
@@ -65,7 +66,7 @@ jobs:
- name: Build awxkit and upload to pypi
run: |
git reset --hard
cd awxkit && python3 setup.py bdist_wheel
cd awxkit && python3 setup.py sdist bdist_wheel
twine upload \
-r ${{ env.pypi_repo }} \
-u ${{ secrets.PYPI_USERNAME }} \
@@ -82,11 +83,15 @@ jobs:
- name: Re-tag and promote awx image
run: |
docker pull ghcr.io/${{ github.repository }}:${{ github.event.release.tag_name }}
docker tag ghcr.io/${{ github.repository }}:${{ github.event.release.tag_name }} quay.io/${{ github.repository }}:${{ github.event.release.tag_name }}
docker tag ghcr.io/${{ github.repository }}:${{ github.event.release.tag_name }} quay.io/${{ github.repository }}:latest
docker push quay.io/${{ github.repository }}:${{ github.event.release.tag_name }}
docker push quay.io/${{ github.repository }}:latest
docker pull ghcr.io/${{ github.repository_owner }}/awx-ee:${{ github.event.release.tag_name }}
docker tag ghcr.io/${{ github.repository_owner }}/awx-ee:${{ github.event.release.tag_name }} quay.io/${{ github.repository_owner }}/awx-ee:${{ github.event.release.tag_name }}
docker push quay.io/${{ github.repository_owner }}/awx-ee:${{ github.event.release.tag_name }}
docker buildx imagetools create \
ghcr.io/${{ github.repository }}:${{ github.event.release.tag_name }} \
--tag quay.io/${{ github.repository }}:${{ github.event.release.tag_name }}
docker buildx imagetools create \
ghcr.io/${{ github.repository }}:${{ github.event.release.tag_name }} \
--tag quay.io/${{ github.repository }}:latest
- name: Re-tag and promote awx-ee image
run: |
docker buildx imagetools create \
ghcr.io/${{ github.repository_owner }}/awx-ee:${{ github.event.release.tag_name }} \
--tag quay.io/${{ github.repository_owner }}/awx-ee:${{ github.event.release.tag_name }}

View File

@@ -23,6 +23,7 @@ jobs:
stage:
if: endsWith(github.repository, '/awx')
runs-on: ubuntu-latest
timeout-minutes: 90
permissions:
packages: write
contents: write
@@ -85,27 +86,33 @@ jobs:
-e push=yes \
-e awx_official=yes
- name: Log in to GHCR
run: |
echo ${{ secrets.GITHUB_TOKEN }} | docker login ghcr.io -u ${{ github.actor }} --password-stdin
- name: Log into registry ghcr.io
uses: docker/login-action@343f7c4344506bcbf9b4de18042ae17996df046d # v3.0.0
with:
registry: ghcr.io
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
- name: Log in to Quay
run: |
echo ${{ secrets.QUAY_TOKEN }} | docker login quay.io -u ${{ secrets.QUAY_USER }} --password-stdin
- name: Log into registry quay.io
uses: docker/login-action@343f7c4344506bcbf9b4de18042ae17996df046d # v3.0.0
with:
registry: quay.io
username: ${{ secrets.QUAY_USER }}
password: ${{ secrets.QUAY_TOKEN }}
- name: tag awx-ee:latest with version input
run: |
docker pull quay.io/ansible/awx-ee:latest
docker tag quay.io/ansible/awx-ee:latest ghcr.io/${{ github.repository_owner }}/awx-ee:${{ github.event.inputs.version }}
docker push ghcr.io/${{ github.repository_owner }}/awx-ee:${{ github.event.inputs.version }}
docker buildx imagetools create \
quay.io/ansible/awx-ee:latest \
--tag ghcr.io/${{ github.repository_owner }}/awx-ee:${{ github.event.inputs.version }}
- name: Build and stage awx-operator
- name: Stage awx-operator image
working-directory: awx-operator
run: |
BUILD_ARGS="--build-arg DEFAULT_AWX_VERSION=${{ github.event.inputs.version }} \
--build-arg OPERATOR_VERSION=${{ github.event.inputs.operator_version }}" \
IMAGE_TAG_BASE=ghcr.io/${{ github.repository_owner }}/awx-operator \
VERSION=${{ github.event.inputs.operator_version }} make docker-build docker-push
BUILD_ARGS="--build-arg DEFAULT_AWX_VERSION=${{ github.event.inputs.version}} \
--build-arg OPERATOR_VERSION=${{ github.event.inputs.operator_version }}" \
IMG=ghcr.io/${{ github.repository_owner }}/awx-operator:${{ github.event.inputs.operator_version }} \
make docker-buildx
- name: Run test deployment with awx-operator
working-directory: awx-operator

View File

@@ -9,6 +9,7 @@ jobs:
name: Update Dependabot Prs
if: contains(github.event.pull_request.labels.*.name, 'dependencies') && contains(github.event.pull_request.labels.*.name, 'component:ui')
runs-on: ubuntu-latest
timeout-minutes: 20
steps:
- name: Checkout branch

View File

@@ -13,6 +13,7 @@ on:
jobs:
push:
runs-on: ubuntu-latest
timeout-minutes: 60
permissions:
packages: write
contents: read

3
.gitignore vendored
View File

@@ -169,3 +169,6 @@ awx/ui_next/build
# Docs build stuff
docs/docsite/build/
_readthedocs/
# Pyenv
.python-version

View File

@@ -10,6 +10,7 @@ build:
3.11
commands:
- pip install --user tox
- python3 -m tox -e docs
- python3 -m tox -e docs --notest -v
- python3 -m tox -e docs --skip-pkg-install -q
- mkdir -p _readthedocs/html/
- mv docs/docsite/build/html/* _readthedocs/html/

113
.vscode/launch.json vendored Normal file
View File

@@ -0,0 +1,113 @@
{
"version": "0.2.0",
"configurations": [
{
"name": "run_ws_heartbeat",
"type": "debugpy",
"request": "launch",
"program": "manage.py",
"args": ["run_ws_heartbeat"],
"django": true,
"preLaunchTask": "stop awx-ws-heartbeat",
"postDebugTask": "start awx-ws-heartbeat"
},
{
"name": "run_cache_clear",
"type": "debugpy",
"request": "launch",
"program": "manage.py",
"args": ["run_cache_clear"],
"django": true,
"preLaunchTask": "stop awx-cache-clear",
"postDebugTask": "start awx-cache-clear"
},
{
"name": "run_callback_receiver",
"type": "debugpy",
"request": "launch",
"program": "manage.py",
"args": ["run_callback_receiver"],
"django": true,
"preLaunchTask": "stop awx-receiver",
"postDebugTask": "start awx-receiver"
},
{
"name": "run_dispatcher",
"type": "debugpy",
"request": "launch",
"program": "manage.py",
"args": ["run_dispatcher"],
"django": true,
"preLaunchTask": "stop awx-dispatcher",
"postDebugTask": "start awx-dispatcher"
},
{
"name": "run_rsyslog_configurer",
"type": "debugpy",
"request": "launch",
"program": "manage.py",
"args": ["run_rsyslog_configurer"],
"django": true,
"preLaunchTask": "stop awx-rsyslog-configurer",
"postDebugTask": "start awx-rsyslog-configurer"
},
{
"name": "run_cache_clear",
"type": "debugpy",
"request": "launch",
"program": "manage.py",
"args": ["run_cache_clear"],
"django": true,
"preLaunchTask": "stop awx-cache-clear",
"postDebugTask": "start awx-cache-clear"
},
{
"name": "run_wsrelay",
"type": "debugpy",
"request": "launch",
"program": "manage.py",
"args": ["run_wsrelay"],
"django": true,
"preLaunchTask": "stop awx-wsrelay",
"postDebugTask": "start awx-wsrelay"
},
{
"name": "daphne",
"type": "debugpy",
"request": "launch",
"program": "/var/lib/awx/venv/awx/bin/daphne",
"args": ["-b", "127.0.0.1", "-p", "8051", "awx.asgi:channel_layer"],
"django": true,
"preLaunchTask": "stop awx-daphne",
"postDebugTask": "start awx-daphne"
},
{
"name": "runserver(uwsgi alternative)",
"type": "debugpy",
"request": "launch",
"program": "manage.py",
"args": ["runserver", "127.0.0.1:8052"],
"django": true,
"preLaunchTask": "stop awx-uwsgi",
"postDebugTask": "start awx-uwsgi"
},
{
"name": "runserver_plus(uwsgi alternative)",
"type": "debugpy",
"request": "launch",
"program": "manage.py",
"args": ["runserver_plus", "127.0.0.1:8052"],
"django": true,
"preLaunchTask": "stop awx-uwsgi and install Werkzeug",
"postDebugTask": "start awx-uwsgi"
},
{
"name": "shell_plus",
"type": "debugpy",
"request": "launch",
"program": "manage.py",
"args": ["shell_plus"],
"django": true,
},
]
}

100
.vscode/tasks.json vendored Normal file
View File

@@ -0,0 +1,100 @@
{
"version": "2.0.0",
"tasks": [
{
"label": "start awx-cache-clear",
"type": "shell",
"command": "supervisorctl start tower-processes:awx-cache-clear"
},
{
"label": "stop awx-cache-clear",
"type": "shell",
"command": "supervisorctl stop tower-processes:awx-cache-clear"
},
{
"label": "start awx-daphne",
"type": "shell",
"command": "supervisorctl start tower-processes:awx-daphne"
},
{
"label": "stop awx-daphne",
"type": "shell",
"command": "supervisorctl stop tower-processes:awx-daphne"
},
{
"label": "start awx-dispatcher",
"type": "shell",
"command": "supervisorctl start tower-processes:awx-dispatcher"
},
{
"label": "stop awx-dispatcher",
"type": "shell",
"command": "supervisorctl stop tower-processes:awx-dispatcher"
},
{
"label": "start awx-receiver",
"type": "shell",
"command": "supervisorctl start tower-processes:awx-receiver"
},
{
"label": "stop awx-receiver",
"type": "shell",
"command": "supervisorctl stop tower-processes:awx-receiver"
},
{
"label": "start awx-rsyslog-configurer",
"type": "shell",
"command": "supervisorctl start tower-processes:awx-rsyslog-configurer"
},
{
"label": "stop awx-rsyslog-configurer",
"type": "shell",
"command": "supervisorctl stop tower-processes:awx-rsyslog-configurer"
},
{
"label": "start awx-rsyslogd",
"type": "shell",
"command": "supervisorctl start tower-processes:awx-rsyslogd"
},
{
"label": "stop awx-rsyslogd",
"type": "shell",
"command": "supervisorctl stop tower-processes:awx-rsyslogd"
},
{
"label": "start awx-uwsgi",
"type": "shell",
"command": "supervisorctl start tower-processes:awx-uwsgi"
},
{
"label": "stop awx-uwsgi",
"type": "shell",
"command": "supervisorctl stop tower-processes:awx-uwsgi"
},
{
"label": "stop awx-uwsgi and install Werkzeug",
"type": "shell",
"command": "pip install Werkzeug; supervisorctl stop tower-processes:awx-uwsgi"
},
{
"label": "start awx-ws-heartbeat",
"type": "shell",
"command": "supervisorctl start tower-processes:awx-ws-heartbeat"
},
{
"label": "stop awx-ws-heartbeat",
"type": "shell",
"command": "supervisorctl stop tower-processes:awx-ws-heartbeat"
},
{
"label": "start awx-wsrelay",
"type": "shell",
"command": "supervisorctl start tower-processes:awx-wsrelay"
},
{
"label": "stop awx-wsrelay",
"type": "shell",
"command": "supervisorctl stop tower-processes:awx-wsrelay"
}
]
}

View File

@@ -22,7 +22,7 @@ recursive-exclude awx/settings local_settings.py*
include tools/scripts/request_tower_configuration.sh
include tools/scripts/request_tower_configuration.ps1
include tools/scripts/automation-controller-service
include tools/scripts/failure-event-handler
include tools/scripts/rsyslog-4xx-recovery
include tools/scripts/awx-python
include awx/playbooks/library/mkfifo.py
include tools/sosreport/*

View File

@@ -10,7 +10,7 @@ KIND_BIN ?= $(shell which kind)
CHROMIUM_BIN=/tmp/chrome-linux/chrome
GIT_BRANCH ?= $(shell git rev-parse --abbrev-ref HEAD)
MANAGEMENT_COMMAND ?= awx-manage
VERSION ?= $(shell $(PYTHON) tools/scripts/scm_version.py)
VERSION ?= $(shell $(PYTHON) tools/scripts/scm_version.py 2> /dev/null)
# ansible-test requires semver compatable version, so we allow overrides to hack it
COLLECTION_VERSION ?= $(shell $(PYTHON) tools/scripts/scm_version.py | cut -d . -f 1-3)
@@ -43,6 +43,8 @@ PROMETHEUS ?= false
GRAFANA ?= false
# If set to true docker-compose will also start a hashicorp vault instance
VAULT ?= false
# If set to true docker-compose will also start a hashicorp vault instance with TLS enabled
VAULT_TLS ?= false
# If set to true docker-compose will also start a tacacs+ instance
TACACS ?= false
@@ -61,7 +63,7 @@ RECEPTOR_IMAGE ?= quay.io/ansible/receptor:devel
SRC_ONLY_PKGS ?= cffi,pycparser,psycopg,twilio
# These should be upgraded in the AWX and Ansible venv before attempting
# to install the actual requirements
VENV_BOOTSTRAP ?= pip==21.2.4 setuptools==65.6.3 setuptools_scm[toml]==7.0.5 wheel==0.38.4
VENV_BOOTSTRAP ?= pip==21.2.4 setuptools==65.6.3 setuptools_scm[toml]==8.0.4 wheel==0.38.4
NAME ?= awx
@@ -73,6 +75,9 @@ SDIST_TAR_FILE ?= $(SDIST_TAR_NAME).tar.gz
I18N_FLAG_FILE = .i18n_built
## PLATFORMS defines the target platforms for the manager image be build to provide support to multiple
PLATFORMS ?= linux/amd64,linux/arm64 # linux/ppc64le,linux/s390x
.PHONY: awx-link clean clean-tmp clean-venv requirements requirements_dev \
develop refresh adduser migrate dbchange \
receiver test test_unit test_coverage coverage_html \
@@ -211,8 +216,6 @@ collectstatic:
fi; \
$(PYTHON) manage.py collectstatic --clear --noinput > /dev/null 2>&1
DEV_RELOAD_COMMAND ?= supervisorctl restart tower-processes:*
uwsgi: collectstatic
@if [ "$(VENV_BASE)" ]; then \
. $(VENV_BASE)/awx/bin/activate; \
@@ -220,7 +223,7 @@ uwsgi: collectstatic
uwsgi /etc/tower/uwsgi.ini
awx-autoreload:
@/awx_devel/tools/docker-compose/awx-autoreload /awx_devel/awx "$(DEV_RELOAD_COMMAND)"
@/awx_devel/tools/docker-compose/awx-autoreload /awx_devel/awx
daphne:
@if [ "$(VENV_BASE)" ]; then \
@@ -300,7 +303,7 @@ swagger: reports
@if [ "$(VENV_BASE)" ]; then \
. $(VENV_BASE)/awx/bin/activate; \
fi; \
(set -o pipefail && py.test $(PYTEST_ARGS) awx/conf/tests/functional awx/main/tests/functional/api awx/main/tests/docs --release=$(VERSION_TARGET) | tee reports/$@.report)
(set -o pipefail && py.test $(PYTEST_ARGS) awx/conf/tests/functional awx/main/tests/functional/api awx/main/tests/docs | tee reports/$@.report)
check: black
@@ -528,13 +531,16 @@ docker-compose-sources: .git/hooks/pre-commit
-e enable_prometheus=$(PROMETHEUS) \
-e enable_grafana=$(GRAFANA) \
-e enable_vault=$(VAULT) \
-e vault_tls=$(VAULT_TLS) \
-e enable_tacacs=$(TACACS) \
$(EXTRA_SOURCES_ANSIBLE_OPTS)
$(EXTRA_SOURCES_ANSIBLE_OPTS)
docker-compose: awx/projects docker-compose-sources
ansible-galaxy install --ignore-certs -r tools/docker-compose/ansible/requirements.yml;
ansible-playbook -i tools/docker-compose/inventory tools/docker-compose/ansible/initialize_containers.yml \
-e enable_vault=$(VAULT);
-e enable_vault=$(VAULT) \
-e vault_tls=$(VAULT_TLS) \
-e enable_ldap=$(LDAP);
$(DOCKER_COMPOSE) -f tools/docker-compose/_sources/docker-compose.yml $(COMPOSE_OPTS) up $(COMPOSE_UP_OPTS) --remove-orphans
docker-compose-credential-plugins: awx/projects docker-compose-sources
@@ -581,12 +587,27 @@ docker-compose-build: Dockerfile.dev
--build-arg BUILDKIT_INLINE_CACHE=1 \
--cache-from=$(DEV_DOCKER_TAG_BASE)/awx_devel:$(COMPOSE_TAG) .
.PHONY: docker-compose-buildx
## Build awx_devel image for docker compose development environment for multiple architectures
docker-compose-buildx: Dockerfile.dev
- docker buildx create --name docker-compose-buildx
docker buildx use docker-compose-buildx
- docker buildx build \
--push \
--build-arg BUILDKIT_INLINE_CACHE=1 \
--cache-from=$(DEV_DOCKER_TAG_BASE)/awx_devel:$(COMPOSE_TAG) \
--platform=$(PLATFORMS) \
--tag $(DEVEL_IMAGE_NAME) \
-f Dockerfile.dev .
- docker buildx rm docker-compose-buildx
docker-clean:
-$(foreach container_id,$(shell docker ps -f name=tools_awx -aq && docker ps -f name=tools_receptor -aq),docker stop $(container_id); docker rm -f $(container_id);)
-$(foreach image_id,$(shell docker images --filter=reference='*/*/*awx_devel*' --filter=reference='*/*awx_devel*' --filter=reference='*awx_devel*' -aq),docker rmi --force $(image_id);)
docker-clean-volumes: docker-compose-clean docker-compose-container-group-clean
docker volume rm -f tools_awx_db tools_vault_1 tools_grafana_storage tools_prometheus_storage $(docker volume ls --filter name=tools_redis_socket_ -q)
docker volume rm -f tools_awx_db tools_vault_1 tools_ldap_1 tools_grafana_storage tools_prometheus_storage $(docker volume ls --filter name=tools_redis_socket_ -q)
docker-refresh: docker-clean docker-compose
@@ -608,9 +629,6 @@ clean-elk:
docker rm tools_elasticsearch_1
docker rm tools_kibana_1
psql-container:
docker run -it --net tools_default --rm postgres:12 sh -c 'exec psql -h "postgres" -p "5432" -U postgres'
VERSION:
@echo "awx: $(VERSION)"
@@ -643,6 +661,21 @@ awx-kube-build: Dockerfile
--build-arg HEADLESS=$(HEADLESS) \
-t $(DEV_DOCKER_TAG_BASE)/awx:$(COMPOSE_TAG) .
## Build multi-arch awx image for deployment on Kubernetes environment.
awx-kube-buildx: Dockerfile
- docker buildx create --name awx-kube-buildx
docker buildx use awx-kube-buildx
- docker buildx build \
--push \
--build-arg VERSION=$(VERSION) \
--build-arg SETUPTOOLS_SCM_PRETEND_VERSION=$(VERSION) \
--build-arg HEADLESS=$(HEADLESS) \
--platform=$(PLATFORMS) \
--tag $(DEV_DOCKER_TAG_BASE)/awx:$(COMPOSE_TAG) \
-f Dockerfile .
- docker buildx rm awx-kube-buildx
.PHONY: Dockerfile.kube-dev
## Generate Docker.kube-dev for awx_kube_devel image
Dockerfile.kube-dev: tools/ansible/roles/dockerfile/templates/Dockerfile.j2
@@ -659,6 +692,18 @@ awx-kube-dev-build: Dockerfile.kube-dev
--cache-from=$(DEV_DOCKER_TAG_BASE)/awx_kube_devel:$(COMPOSE_TAG) \
-t $(DEV_DOCKER_TAG_BASE)/awx_kube_devel:$(COMPOSE_TAG) .
## Build and push multi-arch awx_kube_devel image for development on local Kubernetes environment.
awx-kube-dev-buildx: Dockerfile.kube-dev
- docker buildx create --name awx-kube-dev-buildx
docker buildx use awx-kube-dev-buildx
- docker buildx build \
--push \
--build-arg BUILDKIT_INLINE_CACHE=1 \
--cache-from=$(DEV_DOCKER_TAG_BASE)/awx_kube_devel:$(COMPOSE_TAG) \
--platform=$(PLATFORMS) \
--tag $(DEV_DOCKER_TAG_BASE)/awx_kube_devel:$(COMPOSE_TAG) \
-f Dockerfile.kube-dev .
- docker buildx rm awx-kube-dev-buildx
kind-dev-load: awx-kube-dev-build
$(KIND_BIN) load docker-image $(DEV_DOCKER_TAG_BASE)/awx_kube_devel:$(COMPOSE_TAG)

View File

@@ -7,7 +7,7 @@ AWX provides a web-based user interface, REST API, and task engine built on top
To install AWX, please view the [Install guide](./INSTALL.md).
To learn more about using AWX, and Tower, view the [Tower docs site](http://docs.ansible.com/ansible-tower/index.html).
To learn more about using AWX, view the [AWX docs site](https://ansible.readthedocs.io/projects/awx/en/latest/).
The AWX Project Frequently Asked Questions can be found [here](https://www.ansible.com/awx-project-faq).

View File

@@ -154,10 +154,12 @@ def manage():
from django.conf import settings
from django.core.management import execute_from_command_line
# enforce the postgres version is equal to 12. if not, then terminate program with exit code of 1
# enforce the postgres version is a minimum of 12 (we need this for partitioning); if not, then terminate program with exit code of 1
# In the future if we require a feature of a version of postgres > 12 this should be updated to reflect that.
# The return of connection.pg_version is something like 12013
if not os.getenv('SKIP_PG_VERSION_CHECK', False) and not MODE == 'development':
if (connection.pg_version // 10000) < 12:
sys.stderr.write("Postgres version 12 is required\n")
sys.stderr.write("At a minimum, postgres version 12 is required\n")
sys.exit(1)
if len(sys.argv) >= 2 and sys.argv[1] in ('version', '--version'): # pragma: no cover

View File

@@ -1,450 +0,0 @@
# Copyright (c) 2015 Ansible, Inc.
# All Rights Reserved.
# Python
import re
import json
from functools import reduce
# Django
from django.core.exceptions import FieldError, ValidationError, FieldDoesNotExist
from django.db import models
from django.db.models import Q, CharField, IntegerField, BooleanField, TextField, JSONField
from django.db.models.fields.related import ForeignObjectRel, ManyToManyField, ForeignKey
from django.db.models.functions import Cast
from django.contrib.contenttypes.models import ContentType
from django.contrib.contenttypes.fields import GenericForeignKey
from django.utils.encoding import force_str
from django.utils.translation import gettext_lazy as _
# Django REST Framework
from rest_framework.exceptions import ParseError, PermissionDenied
from rest_framework.filters import BaseFilterBackend
# AWX
from awx.main.utils import get_type_for_model, to_python_boolean
from awx.main.utils.db import get_all_field_names
class TypeFilterBackend(BaseFilterBackend):
"""
Filter on type field now returned with all objects.
"""
def filter_queryset(self, request, queryset, view):
try:
types = None
for key, value in request.query_params.items():
if key == 'type':
if ',' in value:
types = value.split(',')
else:
types = (value,)
if types:
types_map = {}
for ct in ContentType.objects.filter(Q(app_label='main') | Q(app_label='auth', model='user')):
ct_model = ct.model_class()
if not ct_model:
continue
ct_type = get_type_for_model(ct_model)
types_map[ct_type] = ct.pk
model = queryset.model
model_type = get_type_for_model(model)
if 'polymorphic_ctype' in get_all_field_names(model):
types_pks = set([v for k, v in types_map.items() if k in types])
queryset = queryset.filter(polymorphic_ctype_id__in=types_pks)
elif model_type in types:
queryset = queryset
else:
queryset = queryset.none()
return queryset
except FieldError as e:
# Return a 400 for invalid field names.
raise ParseError(*e.args)
def get_fields_from_path(model, path):
"""
Given a Django ORM lookup path (possibly over multiple models)
Returns the fields in the line, and also the revised lookup path
ex., given
model=Organization
path='project__timeout'
returns tuple of fields traversed as well and a corrected path,
for special cases we do substitutions
([<IntegerField for timeout>], 'project__timeout')
"""
# Store of all the fields used to detect repeats
field_list = []
new_parts = []
for name in path.split('__'):
if model is None:
raise ParseError(_('No related model for field {}.').format(name))
# HACK: Make project and inventory source filtering by old field names work for backwards compatibility.
if model._meta.object_name in ('Project', 'InventorySource'):
name = {'current_update': 'current_job', 'last_update': 'last_job', 'last_update_failed': 'last_job_failed', 'last_updated': 'last_job_run'}.get(
name, name
)
if name == 'type' and 'polymorphic_ctype' in get_all_field_names(model):
name = 'polymorphic_ctype'
new_parts.append('polymorphic_ctype__model')
else:
new_parts.append(name)
if name in getattr(model, 'PASSWORD_FIELDS', ()):
raise PermissionDenied(_('Filtering on password fields is not allowed.'))
elif name == 'pk':
field = model._meta.pk
else:
name_alt = name.replace("_", "")
if name_alt in model._meta.fields_map.keys():
field = model._meta.fields_map[name_alt]
new_parts.pop()
new_parts.append(name_alt)
else:
field = model._meta.get_field(name)
if isinstance(field, ForeignObjectRel) and getattr(field.field, '__prevent_search__', False):
raise PermissionDenied(_('Filtering on %s is not allowed.' % name))
elif getattr(field, '__prevent_search__', False):
raise PermissionDenied(_('Filtering on %s is not allowed.' % name))
if field in field_list:
# Field traversed twice, could create infinite JOINs, DoSing Tower
raise ParseError(_('Loops not allowed in filters, detected on field {}.').format(field.name))
field_list.append(field)
model = getattr(field, 'related_model', None)
return field_list, '__'.join(new_parts)
def get_field_from_path(model, path):
"""
Given a Django ORM lookup path (possibly over multiple models)
Returns the last field in the line, and the revised lookup path
ex.
(<IntegerField for timeout>, 'project__timeout')
"""
field_list, new_path = get_fields_from_path(model, path)
return (field_list[-1], new_path)
class FieldLookupBackend(BaseFilterBackend):
"""
Filter using field lookups provided via query string parameters.
"""
RESERVED_NAMES = ('page', 'page_size', 'format', 'order', 'order_by', 'search', 'type', 'host_filter', 'count_disabled', 'no_truncate', 'limit')
SUPPORTED_LOOKUPS = (
'exact',
'iexact',
'contains',
'icontains',
'startswith',
'istartswith',
'endswith',
'iendswith',
'regex',
'iregex',
'gt',
'gte',
'lt',
'lte',
'in',
'isnull',
'search',
)
# A list of fields that we know can be filtered on without the possibility
# of introducing duplicates
NO_DUPLICATES_ALLOW_LIST = (CharField, IntegerField, BooleanField, TextField)
def get_fields_from_lookup(self, model, lookup):
if '__' in lookup and lookup.rsplit('__', 1)[-1] in self.SUPPORTED_LOOKUPS:
path, suffix = lookup.rsplit('__', 1)
else:
path = lookup
suffix = 'exact'
if not path:
raise ParseError(_('Query string field name not provided.'))
# FIXME: Could build up a list of models used across relationships, use
# those lookups combined with request.user.get_queryset(Model) to make
# sure user cannot query using objects he could not view.
field_list, new_path = get_fields_from_path(model, path)
new_lookup = new_path
new_lookup = '__'.join([new_path, suffix])
return field_list, new_lookup
def get_field_from_lookup(self, model, lookup):
'''Method to match return type of single field, if needed.'''
field_list, new_lookup = self.get_fields_from_lookup(model, lookup)
return (field_list[-1], new_lookup)
def to_python_related(self, value):
value = force_str(value)
if value.lower() in ('none', 'null'):
return None
else:
return int(value)
def value_to_python_for_field(self, field, value):
if isinstance(field, models.BooleanField):
return to_python_boolean(value)
elif isinstance(field, (ForeignObjectRel, ManyToManyField, GenericForeignKey, ForeignKey)):
try:
return self.to_python_related(value)
except ValueError:
raise ParseError(_('Invalid {field_name} id: {field_id}').format(field_name=getattr(field, 'name', 'related field'), field_id=value))
else:
return field.to_python(value)
def value_to_python(self, model, lookup, value):
try:
lookup.encode("ascii")
except UnicodeEncodeError:
raise ValueError("%r is not an allowed field name. Must be ascii encodable." % lookup)
field_list, new_lookup = self.get_fields_from_lookup(model, lookup)
field = field_list[-1]
needs_distinct = not all(isinstance(f, self.NO_DUPLICATES_ALLOW_LIST) for f in field_list)
# Type names are stored without underscores internally, but are presented and
# and serialized over the API containing underscores so we remove `_`
# for polymorphic_ctype__model lookups.
if new_lookup.startswith('polymorphic_ctype__model'):
value = value.replace('_', '')
elif new_lookup.endswith('__isnull'):
value = to_python_boolean(value)
elif new_lookup.endswith('__in'):
items = []
if not value:
raise ValueError('cannot provide empty value for __in')
for item in value.split(','):
items.append(self.value_to_python_for_field(field, item))
value = items
elif new_lookup.endswith('__regex') or new_lookup.endswith('__iregex'):
try:
re.compile(value)
except re.error as e:
raise ValueError(e.args[0])
elif new_lookup.endswith('__iexact'):
if not isinstance(field, (CharField, TextField)):
raise ValueError(f'{field.name} is not a text field and cannot be filtered by case-insensitive search')
elif new_lookup.endswith('__search'):
related_model = getattr(field, 'related_model', None)
if not related_model:
raise ValueError('%s is not searchable' % new_lookup[:-8])
new_lookups = []
for rm_field in related_model._meta.fields:
if rm_field.name in ('username', 'first_name', 'last_name', 'email', 'name', 'description', 'playbook'):
new_lookups.append('{}__{}__icontains'.format(new_lookup[:-8], rm_field.name))
return value, new_lookups, needs_distinct
else:
if isinstance(field, JSONField):
new_lookup = new_lookup.replace(field.name, f'{field.name}_as_txt')
value = self.value_to_python_for_field(field, value)
return value, new_lookup, needs_distinct
def filter_queryset(self, request, queryset, view):
try:
# Apply filters specified via query_params. Each entry in the lists
# below is (negate, field, value).
and_filters = []
or_filters = []
chain_filters = []
role_filters = []
search_filters = {}
needs_distinct = False
# Can only have two values: 'AND', 'OR'
# If 'AND' is used, an item must satisfy all conditions to show up in the results.
# If 'OR' is used, an item just needs to satisfy one condition to appear in results.
search_filter_relation = 'OR'
for key, values in request.query_params.lists():
if key in self.RESERVED_NAMES:
continue
# HACK: make `created` available via API for the Django User ORM model
# so it keep compatibility with other objects which exposes the `created` attr.
if queryset.model._meta.object_name == 'User' and key.startswith('created'):
key = key.replace('created', 'date_joined')
# HACK: Make job event filtering by host name mostly work even
# when not capturing job event hosts M2M.
if queryset.model._meta.object_name == 'JobEvent' and key.startswith('hosts__name'):
key = key.replace('hosts__name', 'or__host__name')
or_filters.append((False, 'host__name__isnull', True))
# Custom __int filter suffix (internal use only).
q_int = False
if key.endswith('__int'):
key = key[:-5]
q_int = True
# RBAC filtering
if key == 'role_level':
role_filters.append(values[0])
continue
# Search across related objects.
if key.endswith('__search'):
if values and ',' in values[0]:
search_filter_relation = 'AND'
values = reduce(lambda list1, list2: list1 + list2, [i.split(',') for i in values])
for value in values:
search_value, new_keys, _ = self.value_to_python(queryset.model, key, force_str(value))
assert isinstance(new_keys, list)
search_filters[search_value] = new_keys
# by definition, search *only* joins across relations,
# so it _always_ needs a .distinct()
needs_distinct = True
continue
# Custom chain__ and or__ filters, mutually exclusive (both can
# precede not__).
q_chain = False
q_or = False
if key.startswith('chain__'):
key = key[7:]
q_chain = True
elif key.startswith('or__'):
key = key[4:]
q_or = True
# Custom not__ filter prefix.
q_not = False
if key.startswith('not__'):
key = key[5:]
q_not = True
# Convert value(s) to python and add to the appropriate list.
for value in values:
if q_int:
value = int(value)
value, new_key, distinct = self.value_to_python(queryset.model, key, value)
if distinct:
needs_distinct = True
if '_as_txt' in new_key:
fname = next(item for item in new_key.split('__') if item.endswith('_as_txt'))
queryset = queryset.annotate(**{fname: Cast(fname[:-7], output_field=TextField())})
if q_chain:
chain_filters.append((q_not, new_key, value))
elif q_or:
or_filters.append((q_not, new_key, value))
else:
and_filters.append((q_not, new_key, value))
# Now build Q objects for database query filter.
if and_filters or or_filters or chain_filters or role_filters or search_filters:
args = []
for n, k, v in and_filters:
if n:
args.append(~Q(**{k: v}))
else:
args.append(Q(**{k: v}))
for role_name in role_filters:
if not hasattr(queryset.model, 'accessible_pk_qs'):
raise ParseError(_('Cannot apply role_level filter to this list because its model does not use roles for access control.'))
args.append(Q(pk__in=queryset.model.accessible_pk_qs(request.user, role_name)))
if or_filters:
q = Q()
for n, k, v in or_filters:
if n:
q |= ~Q(**{k: v})
else:
q |= Q(**{k: v})
args.append(q)
if search_filters and search_filter_relation == 'OR':
q = Q()
for term, constrains in search_filters.items():
for constrain in constrains:
q |= Q(**{constrain: term})
args.append(q)
elif search_filters and search_filter_relation == 'AND':
for term, constrains in search_filters.items():
q_chain = Q()
for constrain in constrains:
q_chain |= Q(**{constrain: term})
queryset = queryset.filter(q_chain)
for n, k, v in chain_filters:
if n:
q = ~Q(**{k: v})
else:
q = Q(**{k: v})
queryset = queryset.filter(q)
queryset = queryset.filter(*args)
if needs_distinct:
queryset = queryset.distinct()
return queryset
except (FieldError, FieldDoesNotExist, ValueError, TypeError) as e:
raise ParseError(e.args[0])
except ValidationError as e:
raise ParseError(json.dumps(e.messages, ensure_ascii=False))
class OrderByBackend(BaseFilterBackend):
"""
Filter to apply ordering based on query string parameters.
"""
def filter_queryset(self, request, queryset, view):
try:
order_by = None
for key, value in request.query_params.items():
if key in ('order', 'order_by'):
order_by = value
if ',' in value:
order_by = value.split(',')
else:
order_by = (value,)
default_order_by = self.get_default_ordering(view)
# glue the order by and default order by together so that the default is the backup option
order_by = list(order_by or []) + list(default_order_by or [])
if order_by:
order_by = self._validate_ordering_fields(queryset.model, order_by)
# Special handling of the type field for ordering. In this
# case, we're not sorting exactly on the type field, but
# given the limited number of views with multiple types,
# sorting on polymorphic_ctype.model is effectively the same.
new_order_by = []
if 'polymorphic_ctype' in get_all_field_names(queryset.model):
for field in order_by:
if field == 'type':
new_order_by.append('polymorphic_ctype__model')
elif field == '-type':
new_order_by.append('-polymorphic_ctype__model')
else:
new_order_by.append(field)
else:
for field in order_by:
if field not in ('type', '-type'):
new_order_by.append(field)
queryset = queryset.order_by(*new_order_by)
return queryset
except FieldError as e:
# Return a 400 for invalid field names.
raise ParseError(*e.args)
def get_default_ordering(self, view):
ordering = getattr(view, 'ordering', None)
if isinstance(ordering, str):
return (ordering,)
return ordering
def _validate_ordering_fields(self, model, order_by):
for field_name in order_by:
# strip off the negation prefix `-` if it exists
prefix = ''
path = field_name
if field_name[0] == '-':
prefix = field_name[0]
path = field_name[1:]
try:
field, new_path = get_field_from_path(model, path)
new_path = '{}{}'.format(prefix, new_path)
except (FieldError, FieldDoesNotExist) as e:
raise ParseError(e.args[0])
yield new_path

View File

@@ -30,12 +30,13 @@ from rest_framework.permissions import IsAuthenticated
from rest_framework.renderers import StaticHTMLRenderer
from rest_framework.negotiation import DefaultContentNegotiation
from ansible_base.rest_filters.rest_framework.field_lookup_backend import FieldLookupBackend
from ansible_base.lib.utils.models import get_all_field_names
# AWX
from awx.api.filters import FieldLookupBackend
from awx.main.models import UnifiedJob, UnifiedJobTemplate, User, Role, Credential, WorkflowJobTemplateNode, WorkflowApprovalTemplate
from awx.main.access import optimize_queryset
from awx.main.utils import camelcase_to_underscore, get_search_fields, getattrd, get_object_or_400, decrypt_field, get_awx_version
from awx.main.utils.db import get_all_field_names
from awx.main.utils.licensing import server_product_name
from awx.main.views import ApiErrorView
from awx.api.serializers import ResourceAccessListElementSerializer, CopySerializer
@@ -90,7 +91,7 @@ class LoggedLoginView(auth_views.LoginView):
ret = super(LoggedLoginView, self).post(request, *args, **kwargs)
if request.user.is_authenticated:
logger.info(smart_str(u"User {} logged in from {}".format(self.request.user.username, request.META.get('REMOTE_ADDR', None))))
ret.set_cookie('userLoggedIn', 'true')
ret.set_cookie('userLoggedIn', 'true', secure=getattr(settings, 'SESSION_COOKIE_SECURE', False))
ret.setdefault('X-API-Session-Cookie-Name', getattr(settings, 'SESSION_COOKIE_NAME', 'awx_sessionid'))
return ret
@@ -106,7 +107,7 @@ class LoggedLogoutView(auth_views.LogoutView):
original_user = getattr(request, 'user', None)
ret = super(LoggedLogoutView, self).dispatch(request, *args, **kwargs)
current_user = getattr(request, 'user', None)
ret.set_cookie('userLoggedIn', 'false')
ret.set_cookie('userLoggedIn', 'false', secure=getattr(settings, 'SESSION_COOKIE_SECURE', False))
if (not current_user or not getattr(current_user, 'pk', True)) and current_user != original_user:
logger.info("User {} logged out.".format(original_user.username))
return ret

View File

@@ -6,7 +6,7 @@ import copy
import json
import logging
import re
from collections import OrderedDict
from collections import Counter, OrderedDict
from datetime import timedelta
from uuid import uuid4
@@ -43,6 +43,8 @@ from rest_framework.utils.serializer_helpers import ReturnList
# Django-Polymorphic
from polymorphic.models import PolymorphicModel
from ansible_base.lib.utils.models import get_type_for_model
# AWX
from awx.main.access import get_user_capabilities
from awx.main.constants import ACTIVE_STATES, CENSOR_VALUE
@@ -80,6 +82,7 @@ from awx.main.models import (
Project,
ProjectUpdate,
ProjectUpdateEvent,
ReceptorAddress,
RefreshToken,
Role,
Schedule,
@@ -99,10 +102,9 @@ from awx.main.models import (
CLOUD_INVENTORY_SOURCES,
)
from awx.main.models.base import VERBOSITY_CHOICES, NEW_JOB_TYPE_CHOICES
from awx.main.models.rbac import get_roles_on_resource, role_summary_fields_generator
from awx.main.models.rbac import role_summary_fields_generator, RoleAncestorEntry
from awx.main.fields import ImplicitRoleField
from awx.main.utils import (
get_type_for_model,
get_model_for_type,
camelcase_to_underscore,
getattrd,
@@ -635,7 +637,7 @@ class BaseSerializer(serializers.ModelSerializer, metaclass=BaseSerializerMetacl
exclusions = self.get_validation_exclusions(self.instance)
obj = self.instance or self.Meta.model()
for k, v in attrs.items():
if k not in exclusions:
if k not in exclusions and k != 'canonical_address_port':
setattr(obj, k, v)
obj.full_clean(exclude=exclusions)
# full_clean may modify values on the instance; copy those changes
@@ -2201,6 +2203,99 @@ class BulkHostCreateSerializer(serializers.Serializer):
return return_data
class BulkHostDeleteSerializer(serializers.Serializer):
hosts = serializers.ListField(
allow_empty=False,
max_length=100000,
write_only=True,
help_text=_('List of hosts ids to be deleted, e.g. [105, 130, 131, 200]'),
)
class Meta:
model = Host
fields = ('hosts',)
def validate(self, attrs):
request = self.context.get('request', None)
max_hosts = settings.BULK_HOST_MAX_DELETE
# Validating the number of hosts to be deleted
if len(attrs['hosts']) > max_hosts:
raise serializers.ValidationError(
{
"ERROR": 'Number of hosts exceeds system setting BULK_HOST_MAX_DELETE',
"BULK_HOST_MAX_DELETE": max_hosts,
"Hosts_count": len(attrs['hosts']),
}
)
# Getting list of all host objects, filtered by the list of the hosts to delete
attrs['host_qs'] = Host.objects.get_queryset().filter(pk__in=attrs['hosts']).only('id', 'inventory_id', 'name')
# Converting the queryset data in a dict. to reduce the number of queries when
# manipulating the data
attrs['hosts_data'] = attrs['host_qs'].values()
if len(attrs['host_qs']) == 0:
error_hosts = {host: "Hosts do not exist or you lack permission to delete it" for host in attrs['hosts']}
raise serializers.ValidationError({'hosts': error_hosts})
if len(attrs['host_qs']) < len(attrs['hosts']):
hosts_exists = [host['id'] for host in attrs['hosts_data']]
failed_hosts = list(set(attrs['hosts']).difference(hosts_exists))
error_hosts = {host: "Hosts do not exist or you lack permission to delete it" for host in failed_hosts}
raise serializers.ValidationError({'hosts': error_hosts})
# Getting all inventories that the hosts can be in
inv_list = list(set([host['inventory_id'] for host in attrs['hosts_data']]))
# Checking that the user have permission to all inventories
errors = dict()
for inv in Inventory.objects.get_queryset().filter(pk__in=inv_list):
if request and not request.user.is_superuser:
if request.user not in inv.admin_role:
errors[inv.name] = "Lack permissions to delete hosts from this inventory."
if errors != {}:
raise PermissionDenied({"inventories": errors})
# check the inventory type only if the user have permission to it.
errors = dict()
for inv in Inventory.objects.get_queryset().filter(pk__in=inv_list):
if inv.kind != '':
errors[inv.name] = "Hosts can only be deleted from manual inventories."
if errors != {}:
raise serializers.ValidationError({"inventories": errors})
attrs['inventories'] = inv_list
return attrs
def delete(self, validated_data):
result = {"hosts": dict()}
changes = {'deleted_hosts': dict()}
for inventory in validated_data['inventories']:
changes['deleted_hosts'][inventory] = list()
for host in validated_data['hosts_data']:
result["hosts"][host["id"]] = f"The host {host['name']} was deleted"
changes['deleted_hosts'][host["inventory_id"]].append({"host_id": host["id"], "host_name": host["name"]})
try:
validated_data['host_qs'].delete()
except Exception as e:
raise serializers.ValidationError({"detail": _(f"cannot delete hosts, host deletion error {e}")})
request = self.context.get('request', None)
for inventory in validated_data['inventories']:
activity_entry = ActivityStream.objects.create(
operation='update',
object1='inventory',
changes=json.dumps(changes['deleted_hosts'][inventory]),
actor=request.user,
)
activity_entry.inventory.add(inventory)
return result
class GroupTreeSerializer(GroupSerializer):
children = serializers.SerializerMethodField()
@@ -2664,6 +2759,17 @@ class ResourceAccessListElementSerializer(UserSerializer):
if 'summary_fields' not in ret:
ret['summary_fields'] = {}
team_content_type = ContentType.objects.get_for_model(Team)
content_type = ContentType.objects.get_for_model(obj)
def get_roles_on_resource(parent_role):
"Returns a string list of the roles a parent_role has for current obj."
return list(
RoleAncestorEntry.objects.filter(ancestor=parent_role, content_type_id=content_type.id, object_id=obj.id)
.values_list('role_field', flat=True)
.distinct()
)
def format_role_perm(role):
role_dict = {'id': role.id, 'name': role.name, 'description': role.description}
try:
@@ -2679,7 +2785,7 @@ class ResourceAccessListElementSerializer(UserSerializer):
else:
# Singleton roles should not be managed from this view, as per copy/edit rework spec
role_dict['user_capabilities'] = {'unattach': False}
return {'role': role_dict, 'descendant_roles': get_roles_on_resource(obj, role)}
return {'role': role_dict, 'descendant_roles': get_roles_on_resource(role)}
def format_team_role_perm(naive_team_role, permissive_role_ids):
ret = []
@@ -2705,12 +2811,9 @@ class ResourceAccessListElementSerializer(UserSerializer):
else:
# Singleton roles should not be managed from this view, as per copy/edit rework spec
role_dict['user_capabilities'] = {'unattach': False}
ret.append({'role': role_dict, 'descendant_roles': get_roles_on_resource(obj, team_role)})
ret.append({'role': role_dict, 'descendant_roles': get_roles_on_resource(team_role)})
return ret
team_content_type = ContentType.objects.get_for_model(Team)
content_type = ContentType.objects.get_for_model(obj)
direct_permissive_role_ids = Role.objects.filter(content_type=content_type, object_id=obj.id).values_list('id', flat=True)
all_permissive_role_ids = Role.objects.filter(content_type=content_type, object_id=obj.id).values_list('ancestors__id', flat=True)
@@ -5074,16 +5177,21 @@ class NotificationTemplateSerializer(BaseSerializer):
body = messages[event].get('body', {})
if body:
try:
rendered_body = (
sandbox.ImmutableSandboxedEnvironment(undefined=DescriptiveUndefined).from_string(body).render(JobNotificationMixin.context_stub())
)
potential_body = json.loads(rendered_body)
if not isinstance(potential_body, dict):
error_list.append(
_("Webhook body for '{}' should be a json dictionary. Found type '{}'.".format(event, type(potential_body).__name__))
)
except json.JSONDecodeError as exc:
error_list.append(_("Webhook body for '{}' is not a valid json dictionary ({}).".format(event, exc)))
sandbox.ImmutableSandboxedEnvironment(undefined=DescriptiveUndefined).from_string(body).render(JobNotificationMixin.context_stub())
# https://github.com/ansible/awx/issues/14410
# When rendering something such as "{{ job.id }}"
# the return type is not a dict, unlike "{{ job_metadata }}" which is a dict
# potential_body = json.loads(rendered_body)
# if not isinstance(potential_body, dict):
# error_list.append(
# _("Webhook body for '{}' should be a json dictionary. Found type '{}'.".format(event, type(potential_body).__name__))
# )
except Exception as exc:
error_list.append(_("Webhook body for '{}' is not valid. The following gave an error ({}).".format(event, exc)))
if error_list:
raise serializers.ValidationError(error_list)
@@ -5356,17 +5464,25 @@ class ScheduleSerializer(LaunchConfigurationBaseSerializer, SchedulePreviewSeria
class InstanceLinkSerializer(BaseSerializer):
class Meta:
model = InstanceLink
fields = ('id', 'url', 'related', 'source', 'target', 'link_state')
fields = ('id', 'related', 'source', 'target', 'target_full_address', 'link_state')
source = serializers.SlugRelatedField(slug_field="hostname", queryset=Instance.objects.all())
target = serializers.SlugRelatedField(slug_field="hostname", queryset=Instance.objects.all())
target = serializers.SerializerMethodField()
target_full_address = serializers.SerializerMethodField()
def get_related(self, obj):
res = super(InstanceLinkSerializer, self).get_related(obj)
res['source_instance'] = self.reverse('api:instance_detail', kwargs={'pk': obj.source.id})
res['target_instance'] = self.reverse('api:instance_detail', kwargs={'pk': obj.target.id})
res['target_address'] = self.reverse('api:receptor_address_detail', kwargs={'pk': obj.target.id})
return res
def get_target(self, obj):
return obj.target.instance.hostname
def get_target_full_address(self, obj):
return obj.target.get_full_address()
class InstanceNodeSerializer(BaseSerializer):
class Meta:
@@ -5374,6 +5490,29 @@ class InstanceNodeSerializer(BaseSerializer):
fields = ('id', 'hostname', 'node_type', 'node_state', 'enabled')
class ReceptorAddressSerializer(BaseSerializer):
full_address = serializers.SerializerMethodField()
class Meta:
model = ReceptorAddress
fields = (
'id',
'url',
'address',
'port',
'protocol',
'websocket_path',
'is_internal',
'canonical',
'instance',
'peers_from_control_nodes',
'full_address',
)
def get_full_address(self, obj):
return obj.get_full_address()
class InstanceSerializer(BaseSerializer):
show_capabilities = ['edit']
@@ -5382,11 +5521,17 @@ class InstanceSerializer(BaseSerializer):
jobs_running = serializers.IntegerField(help_text=_('Count of jobs in the running or waiting state that are targeted for this instance'), read_only=True)
jobs_total = serializers.IntegerField(help_text=_('Count of all jobs that target this instance'), read_only=True)
health_check_pending = serializers.SerializerMethodField()
peers = serializers.SlugRelatedField(many=True, required=False, slug_field="hostname", queryset=Instance.objects.all())
peers = serializers.PrimaryKeyRelatedField(
help_text=_('Primary keys of receptor addresses to peer to.'), many=True, required=False, queryset=ReceptorAddress.objects.all()
)
reverse_peers = serializers.SerializerMethodField()
listener_port = serializers.IntegerField(source='canonical_address_port', required=False, allow_null=True)
peers_from_control_nodes = serializers.BooleanField(source='canonical_address_peers_from_control_nodes', required=False)
protocol = serializers.SerializerMethodField()
class Meta:
model = Instance
read_only_fields = ('ip_address', 'uuid', 'version')
read_only_fields = ('ip_address', 'uuid', 'version', 'managed', 'reverse_peers')
fields = (
'id',
'hostname',
@@ -5417,10 +5562,13 @@ class InstanceSerializer(BaseSerializer):
'managed_by_policy',
'node_type',
'node_state',
'managed',
'ip_address',
'listener_port',
'peers',
'reverse_peers',
'listener_port',
'peers_from_control_nodes',
'protocol',
)
extra_kwargs = {
'node_type': {'initial': Instance.Types.EXECUTION, 'default': Instance.Types.EXECUTION},
@@ -5442,16 +5590,54 @@ class InstanceSerializer(BaseSerializer):
def get_related(self, obj):
res = super(InstanceSerializer, self).get_related(obj)
res['receptor_addresses'] = self.reverse('api:instance_receptor_addresses_list', kwargs={'pk': obj.pk})
res['jobs'] = self.reverse('api:instance_unified_jobs_list', kwargs={'pk': obj.pk})
res['instance_groups'] = self.reverse('api:instance_instance_groups_list', kwargs={'pk': obj.pk})
if obj.node_type in [Instance.Types.EXECUTION, Instance.Types.HOP]:
res['install_bundle'] = self.reverse('api:instance_install_bundle', kwargs={'pk': obj.pk})
res['peers'] = self.reverse('api:instance_peers_list', kwargs={"pk": obj.pk})
res['instance_groups'] = self.reverse('api:instance_instance_groups_list', kwargs={'pk': obj.pk})
if obj.node_type in [Instance.Types.EXECUTION, Instance.Types.HOP] and not obj.managed:
res['install_bundle'] = self.reverse('api:instance_install_bundle', kwargs={'pk': obj.pk})
if self.context['request'].user.is_superuser or self.context['request'].user.is_system_auditor:
if obj.node_type == 'execution':
res['health_check'] = self.reverse('api:instance_health_check', kwargs={'pk': obj.pk})
return res
def create_or_update(self, validated_data, obj=None, create=True):
# create a managed receptor address if listener port is defined
port = validated_data.pop('listener_port', -1)
peers_from_control_nodes = validated_data.pop('peers_from_control_nodes', -1)
# delete the receptor address if the port is explicitly set to None
if obj and port == None:
obj.receptor_addresses.filter(address=obj.hostname).delete()
if create:
instance = super(InstanceSerializer, self).create(validated_data)
else:
instance = super(InstanceSerializer, self).update(obj, validated_data)
instance.refresh_from_db() # instance canonical address lookup is deferred, so needs to be reloaded
# only create or update if port is defined in validated_data or already exists in the
# canonical address
# this prevents creating a receptor address if peers_from_control_nodes is in
# validated_data but a port is not set
if (port != None and port != -1) or instance.canonical_address_port:
kwargs = {}
if port != -1:
kwargs['port'] = port
if peers_from_control_nodes != -1:
kwargs['peers_from_control_nodes'] = peers_from_control_nodes
if kwargs:
kwargs['canonical'] = True
instance.receptor_addresses.update_or_create(address=instance.hostname, defaults=kwargs)
return instance
def create(self, validated_data):
return self.create_or_update(validated_data, create=True)
def update(self, obj, validated_data):
return self.create_or_update(validated_data, obj, create=False)
def get_summary_fields(self, obj):
summary = super().get_summary_fields(obj)
@@ -5461,6 +5647,16 @@ class InstanceSerializer(BaseSerializer):
return summary
def get_reverse_peers(self, obj):
return Instance.objects.prefetch_related('peers').filter(peers__in=obj.receptor_addresses.all()).values_list('id', flat=True)
def get_protocol(self, obj):
# note: don't create a different query for receptor addresses, as this is prefetched on the View for optimization
for addr in obj.receptor_addresses.all():
if addr.canonical:
return addr.protocol
return ""
def get_consumed_capacity(self, obj):
return obj.consumed_capacity
@@ -5474,47 +5670,20 @@ class InstanceSerializer(BaseSerializer):
return obj.health_check_pending
def validate(self, attrs):
def get_field_from_model_or_attrs(fd):
return attrs.get(fd, self.instance and getattr(self.instance, fd) or None)
def check_peers_changed():
'''
return True if
- 'peers' in attrs
- instance peers matches peers in attrs
'''
return self.instance and 'peers' in attrs and set(self.instance.peers.all()) != set(attrs['peers'])
# Oddly, using 'source' on a DRF field populates attrs with the source name, so we should rename it back
if 'canonical_address_port' in attrs:
attrs['listener_port'] = attrs.pop('canonical_address_port')
if 'canonical_address_peers_from_control_nodes' in attrs:
attrs['peers_from_control_nodes'] = attrs.pop('canonical_address_peers_from_control_nodes')
if not self.instance and not settings.IS_K8S:
raise serializers.ValidationError(_("Can only create instances on Kubernetes or OpenShift."))
node_type = get_field_from_model_or_attrs("node_type")
peers_from_control_nodes = get_field_from_model_or_attrs("peers_from_control_nodes")
listener_port = get_field_from_model_or_attrs("listener_port")
peers = attrs.get('peers', [])
if peers_from_control_nodes and node_type not in (Instance.Types.EXECUTION, Instance.Types.HOP):
raise serializers.ValidationError(_("peers_from_control_nodes can only be enabled for execution or hop nodes."))
if node_type in [Instance.Types.CONTROL, Instance.Types.HYBRID]:
if check_peers_changed():
raise serializers.ValidationError(
_("Setting peers manually for control nodes is not allowed. Enable peers_from_control_nodes on the hop and execution nodes instead.")
)
if not listener_port and peers_from_control_nodes:
raise serializers.ValidationError(_("Field listener_port must be a valid integer when peers_from_control_nodes is enabled."))
if not listener_port and self.instance and self.instance.peers_from.exists():
raise serializers.ValidationError(_("Field listener_port must be a valid integer when other nodes peer to it."))
for peer in peers:
if peer.listener_port is None:
raise serializers.ValidationError(_("Field listener_port must be set on peer ") + peer.hostname + ".")
if not settings.IS_K8S:
if check_peers_changed():
raise serializers.ValidationError(_("Cannot change peers."))
# cannot enable peers_from_control_nodes if listener_port is not set
if attrs.get('peers_from_control_nodes'):
port = attrs.get('listener_port', -1) # -1 denotes missing, None denotes explicit null
if (port is None) or (port == -1 and self.instance and self.instance.canonical_address is None):
raise serializers.ValidationError(_("Cannot enable peers_from_control_nodes if listener_port is not set."))
return super().validate(attrs)
@@ -5534,8 +5703,8 @@ class InstanceSerializer(BaseSerializer):
raise serializers.ValidationError(_("Can only change the state on Kubernetes or OpenShift."))
if value != Instance.States.DEPROVISIONING:
raise serializers.ValidationError(_("Can only change instances to the 'deprovisioning' state."))
if self.instance.node_type not in (Instance.Types.EXECUTION, Instance.Types.HOP):
raise serializers.ValidationError(_("Can only deprovision execution or hop nodes."))
if self.instance.managed:
raise serializers.ValidationError(_("Cannot deprovision managed nodes."))
else:
if value and value != Instance.States.INSTALLED:
raise serializers.ValidationError(_("Can only create instances in the 'installed' state."))
@@ -5554,18 +5723,48 @@ class InstanceSerializer(BaseSerializer):
def validate_listener_port(self, value):
"""
Cannot change listener port, unless going from none to integer, and vice versa
If instance is managed, cannot change listener port at all
"""
if value and self.instance and self.instance.listener_port and self.instance.listener_port != value:
raise serializers.ValidationError(_("Cannot change listener port."))
if self.instance:
canonical_address_port = self.instance.canonical_address_port
if value and canonical_address_port and canonical_address_port != value:
raise serializers.ValidationError(_("Cannot change listener port."))
if self.instance.managed and value != canonical_address_port:
raise serializers.ValidationError(_("Cannot change listener port for managed nodes."))
return value
def validate_peers(self, value):
# cannot peer to an instance more than once
peers_instances = Counter(p.instance_id for p in value)
if any(count > 1 for count in peers_instances.values()):
raise serializers.ValidationError(_("Cannot peer to the same instance more than once."))
if self.instance:
instance_addresses = set(self.instance.receptor_addresses.all())
setting_peers = set(value)
peers_changed = set(self.instance.peers.all()) != setting_peers
if not settings.IS_K8S and peers_changed:
raise serializers.ValidationError(_("Cannot change peers."))
if self.instance.managed and peers_changed:
raise serializers.ValidationError(_("Setting peers manually for managed nodes is not allowed."))
# cannot peer to self
if instance_addresses & setting_peers:
raise serializers.ValidationError(_("Instance cannot peer to its own address."))
# cannot peer to an instance that is already peered to this instance
if instance_addresses:
for p in setting_peers:
if set(p.instance.peers.all()) & instance_addresses:
raise serializers.ValidationError(_(f"Instance {p.instance.hostname} is already peered to this instance."))
return value
def validate_peers_from_control_nodes(self, value):
"""
Can only enable for K8S based deployments
"""
if value and not settings.IS_K8S:
raise serializers.ValidationError(_("Can only be enabled on Kubernetes or Openshift."))
if self.instance and self.instance.managed and self.instance.canonical_address_peers_from_control_nodes != value:
raise serializers.ValidationError(_("Cannot change peers_from_control_nodes for managed nodes."))
return value

View File

@@ -0,0 +1,22 @@
# Bulk Host Delete
This endpoint allows the client to delete multiple hosts from inventories.
They may do this by providing a list of hosts ID's to be deleted.
Example:
{
"hosts": [1, 2, 3, 4, 5]
}
Return data:
{
"hosts": {
"1": "The host a1 was deleted",
"2": "The host a2 was deleted",
"3": "The host a3 was deleted",
"4": "The host a4 was deleted",
"5": "The host a5 was deleted",
}
}

View File

@@ -17,19 +17,18 @@ custom_worksign_public_keyfile: receptor/work_public_key.pem
custom_tls_certfile: receptor/tls/receptor.crt
custom_tls_keyfile: receptor/tls/receptor.key
custom_ca_certfile: receptor/tls/ca/mesh-CA.crt
receptor_protocol: 'tcp'
{% if instance.listener_port %}
{% if listener_port %}
receptor_protocol: {{ listener_protocol }}
receptor_listener: true
receptor_port: {{ instance.listener_port }}
receptor_port: {{ listener_port }}
{% else %}
receptor_listener: false
{% endif %}
{% if peers %}
receptor_peers:
{% for peer in peers %}
- host: {{ peer.host }}
port: {{ peer.port }}
protocol: tcp
- address: {{ peer.address }}
protocol: {{ peer.protocol }}
{% endfor %}
{% endif %}
{% verbatim %}

View File

@@ -1,4 +1,4 @@
---
collections:
- name: ansible.receptor
version: 2.0.2
version: 2.0.3

View File

@@ -10,6 +10,7 @@ from awx.api.views import (
InstanceInstanceGroupsList,
InstanceHealthCheck,
InstancePeersList,
InstanceReceptorAddressesList,
)
from awx.api.views.instance_install_bundle import InstanceInstallBundle
@@ -21,6 +22,7 @@ urls = [
re_path(r'^(?P<pk>[0-9]+)/instance_groups/$', InstanceInstanceGroupsList.as_view(), name='instance_instance_groups_list'),
re_path(r'^(?P<pk>[0-9]+)/health_check/$', InstanceHealthCheck.as_view(), name='instance_health_check'),
re_path(r'^(?P<pk>[0-9]+)/peers/$', InstancePeersList.as_view(), name='instance_peers_list'),
re_path(r'^(?P<pk>[0-9]+)/receptor_addresses/$', InstanceReceptorAddressesList.as_view(), name='instance_receptor_addresses_list'),
re_path(r'^(?P<pk>[0-9]+)/install_bundle/$', InstanceInstallBundle.as_view(), name='instance_install_bundle'),
]

View File

@@ -0,0 +1,17 @@
# Copyright (c) 2017 Ansible, Inc.
# All Rights Reserved.
from django.urls import re_path
from awx.api.views import (
ReceptorAddressesList,
ReceptorAddressDetail,
)
urls = [
re_path(r'^$', ReceptorAddressesList.as_view(), name='receptor_addresses_list'),
re_path(r'^(?P<pk>[0-9]+)/$', ReceptorAddressDetail.as_view(), name='receptor_address_detail'),
]
__all__ = ['urls']

View File

@@ -36,6 +36,7 @@ from awx.api.views import (
from awx.api.views.bulk import (
BulkView,
BulkHostCreateView,
BulkHostDeleteView,
BulkJobLaunchView,
)
@@ -84,6 +85,7 @@ from .oauth2_root import urls as oauth2_root_urls
from .workflow_approval_template import urls as workflow_approval_template_urls
from .workflow_approval import urls as workflow_approval_urls
from .analytics import urls as analytics_urls
from .receptor_address import urls as receptor_address_urls
v2_urls = [
re_path(r'^$', ApiV2RootView.as_view(), name='api_v2_root_view'),
@@ -152,7 +154,9 @@ v2_urls = [
re_path(r'^workflow_approvals/', include(workflow_approval_urls)),
re_path(r'^bulk/$', BulkView.as_view(), name='bulk'),
re_path(r'^bulk/host_create/$', BulkHostCreateView.as_view(), name='bulk_host_create'),
re_path(r'^bulk/host_delete/$', BulkHostDeleteView.as_view(), name='bulk_host_delete'),
re_path(r'^bulk/job_launch/$', BulkJobLaunchView.as_view(), name='bulk_job_launch'),
re_path(r'^receptor_addresses/', include(receptor_address_urls)),
]

View File

@@ -1,10 +1,11 @@
from django.urls import re_path
from awx.api.views.webhooks import WebhookKeyView, GithubWebhookReceiver, GitlabWebhookReceiver
from awx.api.views.webhooks import WebhookKeyView, GithubWebhookReceiver, GitlabWebhookReceiver, BitbucketDcWebhookReceiver
urlpatterns = [
re_path(r'^webhook_key/$', WebhookKeyView.as_view(), name='webhook_key'),
re_path(r'^github/$', GithubWebhookReceiver.as_view(), name='webhook_receiver_github'),
re_path(r'^gitlab/$', GitlabWebhookReceiver.as_view(), name='webhook_receiver_gitlab'),
re_path(r'^bitbucket_dc/$', BitbucketDcWebhookReceiver.as_view(), name='webhook_receiver_bitbucket_dc'),
]

View File

@@ -272,16 +272,24 @@ class DashboardJobsGraphView(APIView):
success_query = user_unified_jobs.filter(status='successful')
failed_query = user_unified_jobs.filter(status='failed')
canceled_query = user_unified_jobs.filter(status='canceled')
error_query = user_unified_jobs.filter(status='error')
if job_type == 'inv_sync':
success_query = success_query.filter(instance_of=models.InventoryUpdate)
failed_query = failed_query.filter(instance_of=models.InventoryUpdate)
canceled_query = canceled_query.filter(instance_of=models.InventoryUpdate)
error_query = error_query.filter(instance_of=models.InventoryUpdate)
elif job_type == 'playbook_run':
success_query = success_query.filter(instance_of=models.Job)
failed_query = failed_query.filter(instance_of=models.Job)
canceled_query = canceled_query.filter(instance_of=models.Job)
error_query = error_query.filter(instance_of=models.Job)
elif job_type == 'scm_update':
success_query = success_query.filter(instance_of=models.ProjectUpdate)
failed_query = failed_query.filter(instance_of=models.ProjectUpdate)
canceled_query = canceled_query.filter(instance_of=models.ProjectUpdate)
error_query = error_query.filter(instance_of=models.ProjectUpdate)
end = now()
interval = 'day'
@@ -297,10 +305,12 @@ class DashboardJobsGraphView(APIView):
else:
return Response({'error': _('Unknown period "%s"') % str(period)}, status=status.HTTP_400_BAD_REQUEST)
dashboard_data = {"jobs": {"successful": [], "failed": []}}
dashboard_data = {"jobs": {"successful": [], "failed": [], "canceled": [], "error": []}}
succ_list = dashboard_data['jobs']['successful']
fail_list = dashboard_data['jobs']['failed']
canceled_list = dashboard_data['jobs']['canceled']
error_list = dashboard_data['jobs']['error']
qs_s = (
success_query.filter(finished__range=(start, end))
@@ -318,6 +328,22 @@ class DashboardJobsGraphView(APIView):
.annotate(agg=Count('id', distinct=True))
)
data_f = {item['d']: item['agg'] for item in qs_f}
qs_c = (
canceled_query.filter(finished__range=(start, end))
.annotate(d=Trunc('finished', interval, tzinfo=end.tzinfo))
.order_by()
.values('d')
.annotate(agg=Count('id', distinct=True))
)
data_c = {item['d']: item['agg'] for item in qs_c}
qs_e = (
error_query.filter(finished__range=(start, end))
.annotate(d=Trunc('finished', interval, tzinfo=end.tzinfo))
.order_by()
.values('d')
.annotate(agg=Count('id', distinct=True))
)
data_e = {item['d']: item['agg'] for item in qs_e}
start_date = start.replace(hour=0, minute=0, second=0, microsecond=0)
for d in itertools.count():
@@ -326,6 +352,8 @@ class DashboardJobsGraphView(APIView):
break
succ_list.append([time.mktime(date.timetuple()), data_s.get(date, 0)])
fail_list.append([time.mktime(date.timetuple()), data_f.get(date, 0)])
canceled_list.append([time.mktime(date.timetuple()), data_c.get(date, 0)])
error_list.append([time.mktime(date.timetuple()), data_e.get(date, 0)])
return Response(dashboard_data)
@@ -337,12 +365,20 @@ class InstanceList(ListCreateAPIView):
search_fields = ('hostname',)
ordering = ('id',)
def get_queryset(self):
qs = super().get_queryset().prefetch_related('receptor_addresses')
return qs
class InstanceDetail(RetrieveUpdateAPIView):
name = _("Instance Detail")
model = models.Instance
serializer_class = serializers.InstanceSerializer
def get_queryset(self):
qs = super().get_queryset().prefetch_related('receptor_addresses')
return qs
def update_raw_data(self, data):
# these fields are only valid on creation of an instance, so they unwanted on detail view
data.pop('node_type', None)
@@ -375,13 +411,37 @@ class InstanceUnifiedJobsList(SubListAPIView):
class InstancePeersList(SubListAPIView):
name = _("Instance Peers")
name = _("Peers")
model = models.ReceptorAddress
serializer_class = serializers.ReceptorAddressSerializer
parent_model = models.Instance
model = models.Instance
serializer_class = serializers.InstanceSerializer
parent_access = 'read'
search_fields = {'hostname'}
relationship = 'peers'
search_fields = ('address',)
class InstanceReceptorAddressesList(SubListAPIView):
name = _("Receptor Addresses")
model = models.ReceptorAddress
parent_key = 'instance'
parent_model = models.Instance
serializer_class = serializers.ReceptorAddressSerializer
search_fields = ('address',)
class ReceptorAddressesList(ListAPIView):
name = _("Receptor Addresses")
model = models.ReceptorAddress
serializer_class = serializers.ReceptorAddressSerializer
search_fields = ('address',)
class ReceptorAddressDetail(RetrieveAPIView):
name = _("Receptor Address Detail")
model = models.ReceptorAddress
serializer_class = serializers.ReceptorAddressSerializer
parent_model = models.Instance
relationship = 'receptor_addresses'
class InstanceInstanceGroupsList(InstanceGroupMembershipMixin, SubListCreateAttachDetachAPIView):
@@ -742,8 +802,8 @@ class TeamActivityStreamList(SubListAPIView):
qs = self.request.user.get_queryset(self.model)
return qs.filter(
Q(team=parent)
| Q(project__in=models.Project.accessible_objects(parent, 'read_role'))
| Q(credential__in=models.Credential.accessible_objects(parent, 'read_role'))
| Q(project__in=models.Project.accessible_objects(parent.member_role, 'read_role'))
| Q(credential__in=models.Credential.accessible_objects(parent.member_role, 'read_role'))
)
@@ -1397,7 +1457,7 @@ class OrganizationCredentialList(SubListCreateAPIView):
self.check_parent_access(organization)
user_visible = models.Credential.accessible_objects(self.request.user, 'read_role').all()
org_set = models.Credential.accessible_objects(organization.admin_role, 'read_role').all()
org_set = models.Credential.objects.filter(organization=organization)
if self.request.user.is_superuser or self.request.user.is_system_auditor:
return org_set

View File

@@ -34,6 +34,7 @@ class BulkView(APIView):
'''List top level resources'''
data = OrderedDict()
data['host_create'] = reverse('api:bulk_host_create', request=request)
data['host_delete'] = reverse('api:bulk_host_delete', request=request)
data['job_launch'] = reverse('api:bulk_job_launch', request=request)
return Response(data)
@@ -72,3 +73,20 @@ class BulkHostCreateView(GenericAPIView):
result = serializer.create(serializer.validated_data)
return Response(result, status=status.HTTP_201_CREATED)
return Response(serializer.errors, status=status.HTTP_400_BAD_REQUEST)
class BulkHostDeleteView(GenericAPIView):
permission_classes = [IsAuthenticated]
model = Host
serializer_class = serializers.BulkHostDeleteSerializer
allowed_methods = ['GET', 'POST', 'OPTIONS']
def get(self, request):
return Response({"detail": "Bulk delete hosts with this endpoint"}, status=status.HTTP_200_OK)
def post(self, request):
serializer = serializers.BulkHostDeleteSerializer(data=request.data, context={'request': request})
if serializer.is_valid():
result = serializer.delete(serializer.validated_data)
return Response(result, status=status.HTTP_201_CREATED)
return Response(serializer.errors, status=status.HTTP_400_BAD_REQUEST)

View File

@@ -124,10 +124,19 @@ def generate_inventory_yml(instance_obj):
def generate_group_vars_all_yml(instance_obj):
# get peers
peers = []
for instance in instance_obj.peers.all():
peers.append(dict(host=instance.hostname, port=instance.listener_port))
all_yaml = render_to_string("instance_install_bundle/group_vars/all.yml", context=dict(instance=instance_obj, peers=peers))
for addr in instance_obj.peers.select_related('instance'):
peers.append(dict(address=addr.get_full_address(), protocol=addr.protocol))
context = dict(instance=instance_obj, peers=peers)
canonical_addr = instance_obj.canonical_address
if canonical_addr:
context['listener_port'] = canonical_addr.port
protocol = canonical_addr.protocol if canonical_addr.protocol != 'wss' else 'ws'
context['listener_protocol'] = protocol
all_yaml = render_to_string("instance_install_bundle/group_vars/all.yml", context=context)
# convert consecutive newlines with a single newline
return re.sub(r'\n+', '\n', all_yaml)

View File

@@ -17,7 +17,7 @@ class MeshVisualizer(APIView):
def get(self, request, format=None):
data = {
'nodes': InstanceNodeSerializer(Instance.objects.all(), many=True).data,
'links': InstanceLinkSerializer(InstanceLink.objects.select_related('target', 'source'), many=True).data,
'links': InstanceLinkSerializer(InstanceLink.objects.select_related('target__instance', 'source'), many=True).data,
}
return Response(data)

View File

@@ -84,6 +84,7 @@ class ApiVersionRootView(APIView):
data['ping'] = reverse('api:api_v2_ping_view', request=request)
data['instances'] = reverse('api:instance_list', request=request)
data['instance_groups'] = reverse('api:instance_group_list', request=request)
data['receptor_addresses'] = reverse('api:receptor_addresses_list', request=request)
data['config'] = reverse('api:api_v2_config_view', request=request)
data['settings'] = reverse('api:setting_category_list', request=request)
data['me'] = reverse('api:user_me_list', request=request)

View File

@@ -1,4 +1,4 @@
from hashlib import sha1
from hashlib import sha1, sha256
import hmac
import logging
import urllib.parse
@@ -99,14 +99,31 @@ class WebhookReceiverBase(APIView):
def get_signature(self):
raise NotImplementedError
def must_check_signature(self):
return True
def is_ignored_request(self):
return False
def check_signature(self, obj):
if not obj.webhook_key:
raise PermissionDenied
if not self.must_check_signature():
logger.debug("skipping signature validation")
return
mac = hmac.new(force_bytes(obj.webhook_key), msg=force_bytes(self.request.body), digestmod=sha1)
logger.debug("header signature: %s", self.get_signature())
hash_alg, expected_digest = self.get_signature()
if hash_alg == 'sha1':
mac = hmac.new(force_bytes(obj.webhook_key), msg=force_bytes(self.request.body), digestmod=sha1)
elif hash_alg == 'sha256':
mac = hmac.new(force_bytes(obj.webhook_key), msg=force_bytes(self.request.body), digestmod=sha256)
else:
logger.debug("Unsupported signature type, supported: sha1, sha256, received: {}".format(hash_alg))
raise PermissionDenied
logger.debug("header signature: %s", expected_digest)
logger.debug("calculated signature: %s", force_bytes(mac.hexdigest()))
if not hmac.compare_digest(force_bytes(mac.hexdigest()), self.get_signature()):
if not hmac.compare_digest(force_bytes(mac.hexdigest()), expected_digest):
raise PermissionDenied
@csrf_exempt
@@ -118,6 +135,10 @@ class WebhookReceiverBase(APIView):
obj = self.get_object()
self.check_signature(obj)
if self.is_ignored_request():
# This was an ignored request type (e.g. ping), don't act on it
return Response({'message': _("Webhook ignored")}, status=status.HTTP_200_OK)
event_type = self.get_event_type()
event_guid = self.get_event_guid()
event_ref = self.get_event_ref()
@@ -186,7 +207,7 @@ class GithubWebhookReceiver(WebhookReceiverBase):
if hash_alg != 'sha1':
logger.debug("Unsupported signature type, expected: sha1, received: {}".format(hash_alg))
raise PermissionDenied
return force_bytes(signature)
return hash_alg, force_bytes(signature)
class GitlabWebhookReceiver(WebhookReceiverBase):
@@ -214,15 +235,73 @@ class GitlabWebhookReceiver(WebhookReceiverBase):
return "{}://{}/api/v4/projects/{}/statuses/{}".format(parsed.scheme, parsed.netloc, project['id'], self.get_event_ref())
def get_signature(self):
return force_bytes(self.request.META.get('HTTP_X_GITLAB_TOKEN') or '')
def check_signature(self, obj):
if not obj.webhook_key:
raise PermissionDenied
token_from_request = force_bytes(self.request.META.get('HTTP_X_GITLAB_TOKEN') or '')
# GitLab only returns the secret token, not an hmac hash. Use
# the hmac `compare_digest` helper function to prevent timing
# analysis by attackers.
if not hmac.compare_digest(force_bytes(obj.webhook_key), self.get_signature()):
if not hmac.compare_digest(force_bytes(obj.webhook_key), token_from_request):
raise PermissionDenied
class BitbucketDcWebhookReceiver(WebhookReceiverBase):
service = 'bitbucket_dc'
ref_keys = {
'repo:refs_changed': 'changes.0.toHash',
'mirror:repo_synchronized': 'changes.0.toHash',
'pr:opened': 'pullRequest.toRef.latestCommit',
'pr:from_ref_updated': 'pullRequest.toRef.latestCommit',
'pr:modified': 'pullRequest.toRef.latestCommit',
}
def get_event_type(self):
return self.request.META.get('HTTP_X_EVENT_KEY')
def get_event_guid(self):
return self.request.META.get('HTTP_X_REQUEST_ID')
def get_event_status_api(self):
# https://<bitbucket-base-url>/rest/build-status/1.0/commits/<commit-hash>
if self.get_event_type() not in self.ref_keys.keys():
return
if self.get_event_ref() is None:
return
any_url = None
if 'actor' in self.request.data:
any_url = self.request.data['actor'].get('links', {}).get('self')
if any_url is None and 'repository' in self.request.data:
any_url = self.request.data['repository'].get('links', {}).get('self')
if any_url is None:
return
any_url = any_url[0].get('href')
if any_url is None:
return
parsed = urllib.parse.urlparse(any_url)
return "{}://{}/rest/build-status/1.0/commits/{}".format(parsed.scheme, parsed.netloc, self.get_event_ref())
def is_ignored_request(self):
return self.get_event_type() not in [
'repo:refs_changed',
'mirror:repo_synchronized',
'pr:opened',
'pr:from_ref_updated',
'pr:modified',
]
def must_check_signature(self):
# Bitbucket does not sign ping requests...
return self.get_event_type() != 'diagnostics:ping'
def get_signature(self):
header_sig = self.request.META.get('HTTP_X_HUB_SIGNATURE')
if not header_sig:
logger.debug("Expected signature missing from header key HTTP_X_HUB_SIGNATURE")
raise PermissionDenied
hash_alg, signature = header_sig.split('=')
return hash_alg, force_bytes(signature)

View File

@@ -7,8 +7,10 @@ import json
# Django
from django.db import models
from ansible_base.lib.utils.models import prevent_search
# AWX
from awx.main.models.base import CreatedModifiedModel, prevent_search
from awx.main.models.base import CreatedModifiedModel
from awx.main.utils import encrypt_field
from awx.conf import settings_registry

View File

@@ -1,6 +1,7 @@
# Python
import contextlib
import logging
import psycopg
import threading
import time
import os
@@ -13,7 +14,7 @@ from django.conf import settings, UserSettingsHolder
from django.core.cache import cache as django_cache
from django.core.exceptions import ImproperlyConfigured, SynchronousOnlyOperation
from django.db import transaction, connection
from django.db.utils import Error as DBError, ProgrammingError
from django.db.utils import DatabaseError, ProgrammingError
from django.utils.functional import cached_property
# Django REST Framework
@@ -80,18 +81,26 @@ def _ctit_db_wrapper(trans_safe=False):
logger.debug('Obtaining database settings in spite of broken transaction.')
transaction.set_rollback(False)
yield
except DBError as exc:
except ProgrammingError as e:
# Exception raised for programming errors
# Examples may be table not found or already exists,
# this generally means we can't fetch Tower configuration
# because the database hasn't actually finished migrating yet;
# this is usually a sign that a service in a container (such as ws_broadcast)
# has come up *before* the database has finished migrating, and
# especially that the conf.settings table doesn't exist yet
# syntax error in the SQL statement, wrong number of parameters specified, etc.
if trans_safe:
level = logger.warning
if isinstance(exc, ProgrammingError):
if 'relation' in str(exc) and 'does not exist' in str(exc):
# this generally means we can't fetch Tower configuration
# because the database hasn't actually finished migrating yet;
# this is usually a sign that a service in a container (such as ws_broadcast)
# has come up *before* the database has finished migrating, and
# especially that the conf.settings table doesn't exist yet
level = logger.debug
level(f'Database settings are not available, using defaults. error: {str(exc)}')
logger.debug(f'Database settings are not available, using defaults. error: {str(e)}')
else:
logger.exception('Error modifying something related to database settings.')
except DatabaseError as e:
if trans_safe:
cause = e.__cause__
if cause and hasattr(cause, 'sqlstate'):
sqlstate = cause.sqlstate
sqlstate_str = psycopg.errors.lookup(sqlstate)
logger.error('SQL Error state: {} - {}'.format(sqlstate, sqlstate_str))
else:
logger.exception('Error modifying something related to database settings.')
finally:

View File

@@ -20,11 +20,12 @@ from rest_framework.exceptions import ParseError, PermissionDenied
# Django OAuth Toolkit
from awx.main.models.oauth import OAuth2Application, OAuth2AccessToken
from ansible_base.lib.utils.validation import to_python_boolean
# AWX
from awx.main.utils import (
get_object_or_400,
get_pk_from_dict,
to_python_boolean,
get_licenser,
)
from awx.main.models import (
@@ -56,6 +57,7 @@ from awx.main.models import (
Project,
ProjectUpdate,
ProjectUpdateEvent,
ReceptorAddress,
Role,
Schedule,
SystemJob,
@@ -2429,6 +2431,29 @@ class InventoryUpdateEventAccess(BaseAccess):
return False
class ReceptorAddressAccess(BaseAccess):
"""
I can see receptor address records whenever I can access the instance
"""
model = ReceptorAddress
def filtered_queryset(self):
return self.model.objects.filter(Q(instance__in=Instance.accessible_pk_qs(self.user, 'read_role')))
@check_superuser
def can_add(self, data):
return False
@check_superuser
def can_change(self, obj, data):
return False
@check_superuser
def can_delete(self, obj):
return False
class SystemJobEventAccess(BaseAccess):
"""
I can only see manage System Jobs events if I'm a super user

View File

@@ -2,7 +2,7 @@
import logging
# AWX
from awx.main.analytics.subsystem_metrics import Metrics
from awx.main.analytics.subsystem_metrics import DispatcherMetrics, CallbackReceiverMetrics
from awx.main.dispatch.publish import task
from awx.main.dispatch import get_task_queuename
@@ -11,4 +11,5 @@ logger = logging.getLogger('awx.main.scheduler')
@task(queue=get_task_queuename)
def send_subsystem_metrics():
Metrics().send_metrics()
DispatcherMetrics().send_metrics()
CallbackReceiverMetrics().send_metrics()

View File

@@ -419,7 +419,7 @@ def _events_table(since, full_path, until, tbl, where_column, project_job_create
resolved_action,
resolved_role,
-- '-' operator listed here:
-- https://www.postgresql.org/docs/12/functions-json.html
-- https://www.postgresql.org/docs/15/functions-json.html
-- note that operator is only supported by jsonb objects
-- https://www.postgresql.org/docs/current/datatype-json.html
(CASE WHEN event = 'playbook_on_stats' THEN {event_data} - 'artifact_data' END) as playbook_on_stats,

View File

@@ -1,10 +1,15 @@
import itertools
import redis
import json
import time
import logging
import prometheus_client
from prometheus_client.core import GaugeMetricFamily, HistogramMetricFamily
from prometheus_client.registry import CollectorRegistry
from django.conf import settings
from django.apps import apps
from django.http import HttpRequest
from rest_framework.request import Request
from awx.main.consumers import emit_channel_notification
from awx.main.utils import is_testing
@@ -13,6 +18,30 @@ root_key = settings.SUBSYSTEM_METRICS_REDIS_KEY_PREFIX
logger = logging.getLogger('awx.main.analytics')
class MetricsNamespace:
def __init__(self, namespace):
self._namespace = namespace
class MetricsServerSettings(MetricsNamespace):
def port(self):
return settings.METRICS_SUBSYSTEM_CONFIG['server'][self._namespace]['port']
class MetricsServer(MetricsServerSettings):
def __init__(self, namespace, registry):
MetricsNamespace.__init__(self, namespace)
self._registry = registry
def start(self):
try:
# TODO: addr for ipv6 ?
prometheus_client.start_http_server(self.port(), addr='localhost', registry=self._registry)
except Exception:
logger.error(f"MetricsServer failed to start for service '{self._namespace}.")
raise
class BaseM:
def __init__(self, field, help_text):
self.field = field
@@ -148,76 +177,40 @@ class HistogramM(BaseM):
return output_text
class Metrics:
def __init__(self, auto_pipe_execute=False, instance_name=None):
class Metrics(MetricsNamespace):
# metric name, help_text
METRICSLIST = []
_METRICSLIST = [
FloatM('subsystem_metrics_pipe_execute_seconds', 'Time spent saving metrics to redis'),
IntM('subsystem_metrics_pipe_execute_calls', 'Number of calls to pipe_execute'),
FloatM('subsystem_metrics_send_metrics_seconds', 'Time spent sending metrics to other nodes'),
]
def __init__(self, namespace, auto_pipe_execute=False, instance_name=None, metrics_have_changed=True, **kwargs):
MetricsNamespace.__init__(self, namespace)
self.pipe = redis.Redis.from_url(settings.BROKER_URL).pipeline()
self.conn = redis.Redis.from_url(settings.BROKER_URL)
self.last_pipe_execute = time.time()
# track if metrics have been modified since last saved to redis
# start with True so that we get an initial save to redis
self.metrics_have_changed = True
self.metrics_have_changed = metrics_have_changed
self.pipe_execute_interval = settings.SUBSYSTEM_METRICS_INTERVAL_SAVE_TO_REDIS
self.send_metrics_interval = settings.SUBSYSTEM_METRICS_INTERVAL_SEND_METRICS
# auto pipe execute will commit transaction of metric data to redis
# at a regular interval (pipe_execute_interval). If set to False,
# the calling function should call .pipe_execute() explicitly
self.auto_pipe_execute = auto_pipe_execute
Instance = apps.get_model('main', 'Instance')
if instance_name:
self.instance_name = instance_name
elif is_testing():
self.instance_name = "awx_testing"
else:
self.instance_name = Instance.objects.my_hostname()
self.instance_name = settings.CLUSTER_HOST_ID # Same as Instance.objects.my_hostname() BUT we do not need to import Instance
# metric name, help_text
METRICSLIST = [
SetIntM('callback_receiver_events_queue_size_redis', 'Current number of events in redis queue'),
IntM('callback_receiver_events_popped_redis', 'Number of events popped from redis'),
IntM('callback_receiver_events_in_memory', 'Current number of events in memory (in transfer from redis to db)'),
IntM('callback_receiver_batch_events_errors', 'Number of times batch insertion failed'),
FloatM('callback_receiver_events_insert_db_seconds', 'Total time spent saving events to database'),
IntM('callback_receiver_events_insert_db', 'Number of events batch inserted into database'),
IntM('callback_receiver_events_broadcast', 'Number of events broadcast to other control plane nodes'),
HistogramM(
'callback_receiver_batch_events_insert_db', 'Number of events batch inserted into database', settings.SUBSYSTEM_METRICS_BATCH_INSERT_BUCKETS
),
SetFloatM('callback_receiver_event_processing_avg_seconds', 'Average processing time per event per callback receiver batch'),
FloatM('subsystem_metrics_pipe_execute_seconds', 'Time spent saving metrics to redis'),
IntM('subsystem_metrics_pipe_execute_calls', 'Number of calls to pipe_execute'),
FloatM('subsystem_metrics_send_metrics_seconds', 'Time spent sending metrics to other nodes'),
SetFloatM('task_manager_get_tasks_seconds', 'Time spent in loading tasks from db'),
SetFloatM('task_manager_start_task_seconds', 'Time spent starting task'),
SetFloatM('task_manager_process_running_tasks_seconds', 'Time spent processing running tasks'),
SetFloatM('task_manager_process_pending_tasks_seconds', 'Time spent processing pending tasks'),
SetFloatM('task_manager__schedule_seconds', 'Time spent in running the entire _schedule'),
IntM('task_manager__schedule_calls', 'Number of calls to _schedule, after lock is acquired'),
SetFloatM('task_manager_recorded_timestamp', 'Unix timestamp when metrics were last recorded'),
SetIntM('task_manager_tasks_started', 'Number of tasks started'),
SetIntM('task_manager_running_processed', 'Number of running tasks processed'),
SetIntM('task_manager_pending_processed', 'Number of pending tasks processed'),
SetIntM('task_manager_tasks_blocked', 'Number of tasks blocked from running'),
SetFloatM('task_manager_commit_seconds', 'Time spent in db transaction, including on_commit calls'),
SetFloatM('dependency_manager_get_tasks_seconds', 'Time spent loading pending tasks from db'),
SetFloatM('dependency_manager_generate_dependencies_seconds', 'Time spent generating dependencies for pending tasks'),
SetFloatM('dependency_manager__schedule_seconds', 'Time spent in running the entire _schedule'),
IntM('dependency_manager__schedule_calls', 'Number of calls to _schedule, after lock is acquired'),
SetFloatM('dependency_manager_recorded_timestamp', 'Unix timestamp when metrics were last recorded'),
SetIntM('dependency_manager_pending_processed', 'Number of pending tasks processed'),
SetFloatM('workflow_manager__schedule_seconds', 'Time spent in running the entire _schedule'),
IntM('workflow_manager__schedule_calls', 'Number of calls to _schedule, after lock is acquired'),
SetFloatM('workflow_manager_recorded_timestamp', 'Unix timestamp when metrics were last recorded'),
SetFloatM('workflow_manager_spawn_workflow_graph_jobs_seconds', 'Time spent spawning workflow tasks'),
SetFloatM('workflow_manager_get_tasks_seconds', 'Time spent loading workflow tasks from db'),
# dispatcher subsystem metrics
SetIntM('dispatcher_pool_scale_up_events', 'Number of times local dispatcher scaled up a worker since startup'),
SetIntM('dispatcher_pool_active_task_count', 'Number of active tasks in the worker pool when last task was submitted'),
SetIntM('dispatcher_pool_max_worker_count', 'Highest number of workers in worker pool in last collection interval, about 20s'),
SetFloatM('dispatcher_availability', 'Fraction of time (in last collection interval) dispatcher was able to receive messages'),
]
# turn metric list into dictionary with the metric name as a key
self.METRICS = {}
for m in METRICSLIST:
for m in itertools.chain(self.METRICSLIST, self._METRICSLIST):
self.METRICS[m.field] = m
# track last time metrics were sent to other nodes
@@ -230,7 +223,7 @@ class Metrics:
m.reset_value(self.conn)
self.metrics_have_changed = True
self.conn.delete(root_key + "_lock")
for m in self.conn.scan_iter(root_key + '_instance_*'):
for m in self.conn.scan_iter(root_key + '-' + self._namespace + '_instance_*'):
self.conn.delete(m)
def inc(self, field, value):
@@ -297,7 +290,7 @@ class Metrics:
def send_metrics(self):
# more than one thread could be calling this at the same time, so should
# acquire redis lock before sending metrics
lock = self.conn.lock(root_key + '_lock')
lock = self.conn.lock(root_key + '-' + self._namespace + '_lock')
if not lock.acquire(blocking=False):
return
try:
@@ -307,9 +300,10 @@ class Metrics:
payload = {
'instance': self.instance_name,
'metrics': serialized_metrics,
'metrics_namespace': self._namespace,
}
# store the serialized data locally as well, so that load_other_metrics will read it
self.conn.set(root_key + '_instance_' + self.instance_name, serialized_metrics)
self.conn.set(root_key + '-' + self._namespace + '_instance_' + self.instance_name, serialized_metrics)
emit_channel_notification("metrics", payload)
self.previous_send_metrics.set(current_time)
@@ -331,14 +325,14 @@ class Metrics:
instances_filter = request.query_params.getlist("node")
# get a sorted list of instance names
instance_names = [self.instance_name]
for m in self.conn.scan_iter(root_key + '_instance_*'):
for m in self.conn.scan_iter(root_key + '-' + self._namespace + '_instance_*'):
instance_names.append(m.decode('UTF-8').split('_instance_')[1])
instance_names.sort()
# load data, including data from the this local instance
instance_data = {}
for instance in instance_names:
if len(instances_filter) == 0 or instance in instances_filter:
instance_data_from_redis = self.conn.get(root_key + '_instance_' + instance)
instance_data_from_redis = self.conn.get(root_key + '-' + self._namespace + '_instance_' + instance)
# data from other instances may not be available. That is OK.
if instance_data_from_redis:
instance_data[instance] = json.loads(instance_data_from_redis.decode('UTF-8'))
@@ -357,6 +351,120 @@ class Metrics:
return output_text
class DispatcherMetrics(Metrics):
METRICSLIST = [
SetFloatM('task_manager_get_tasks_seconds', 'Time spent in loading tasks from db'),
SetFloatM('task_manager_start_task_seconds', 'Time spent starting task'),
SetFloatM('task_manager_process_running_tasks_seconds', 'Time spent processing running tasks'),
SetFloatM('task_manager_process_pending_tasks_seconds', 'Time spent processing pending tasks'),
SetFloatM('task_manager__schedule_seconds', 'Time spent in running the entire _schedule'),
IntM('task_manager__schedule_calls', 'Number of calls to _schedule, after lock is acquired'),
SetFloatM('task_manager_recorded_timestamp', 'Unix timestamp when metrics were last recorded'),
SetIntM('task_manager_tasks_started', 'Number of tasks started'),
SetIntM('task_manager_running_processed', 'Number of running tasks processed'),
SetIntM('task_manager_pending_processed', 'Number of pending tasks processed'),
SetIntM('task_manager_tasks_blocked', 'Number of tasks blocked from running'),
SetFloatM('task_manager_commit_seconds', 'Time spent in db transaction, including on_commit calls'),
SetFloatM('dependency_manager_get_tasks_seconds', 'Time spent loading pending tasks from db'),
SetFloatM('dependency_manager_generate_dependencies_seconds', 'Time spent generating dependencies for pending tasks'),
SetFloatM('dependency_manager__schedule_seconds', 'Time spent in running the entire _schedule'),
IntM('dependency_manager__schedule_calls', 'Number of calls to _schedule, after lock is acquired'),
SetFloatM('dependency_manager_recorded_timestamp', 'Unix timestamp when metrics were last recorded'),
SetIntM('dependency_manager_pending_processed', 'Number of pending tasks processed'),
SetFloatM('workflow_manager__schedule_seconds', 'Time spent in running the entire _schedule'),
IntM('workflow_manager__schedule_calls', 'Number of calls to _schedule, after lock is acquired'),
SetFloatM('workflow_manager_recorded_timestamp', 'Unix timestamp when metrics were last recorded'),
SetFloatM('workflow_manager_spawn_workflow_graph_jobs_seconds', 'Time spent spawning workflow tasks'),
SetFloatM('workflow_manager_get_tasks_seconds', 'Time spent loading workflow tasks from db'),
# dispatcher subsystem metrics
SetIntM('dispatcher_pool_scale_up_events', 'Number of times local dispatcher scaled up a worker since startup'),
SetIntM('dispatcher_pool_active_task_count', 'Number of active tasks in the worker pool when last task was submitted'),
SetIntM('dispatcher_pool_max_worker_count', 'Highest number of workers in worker pool in last collection interval, about 20s'),
SetFloatM('dispatcher_availability', 'Fraction of time (in last collection interval) dispatcher was able to receive messages'),
]
def __init__(self, *args, **kwargs):
super().__init__(settings.METRICS_SERVICE_DISPATCHER, *args, **kwargs)
class CallbackReceiverMetrics(Metrics):
METRICSLIST = [
SetIntM('callback_receiver_events_queue_size_redis', 'Current number of events in redis queue'),
IntM('callback_receiver_events_popped_redis', 'Number of events popped from redis'),
IntM('callback_receiver_events_in_memory', 'Current number of events in memory (in transfer from redis to db)'),
IntM('callback_receiver_batch_events_errors', 'Number of times batch insertion failed'),
FloatM('callback_receiver_events_insert_db_seconds', 'Total time spent saving events to database'),
IntM('callback_receiver_events_insert_db', 'Number of events batch inserted into database'),
IntM('callback_receiver_events_broadcast', 'Number of events broadcast to other control plane nodes'),
HistogramM(
'callback_receiver_batch_events_insert_db', 'Number of events batch inserted into database', settings.SUBSYSTEM_METRICS_BATCH_INSERT_BUCKETS
),
SetFloatM('callback_receiver_event_processing_avg_seconds', 'Average processing time per event per callback receiver batch'),
]
def __init__(self, *args, **kwargs):
super().__init__(settings.METRICS_SERVICE_CALLBACK_RECEIVER, *args, **kwargs)
def metrics(request):
m = Metrics()
return m.generate_metrics(request)
output_text = ''
for m in [DispatcherMetrics(), CallbackReceiverMetrics()]:
output_text += m.generate_metrics(request)
return output_text
class CustomToPrometheusMetricsCollector(prometheus_client.registry.Collector):
"""
Takes the metric data from redis -> our custom metric fields -> prometheus
library metric fields.
The plan is to get rid of the use of redis, our custom metric fields, and
to switch fully to the prometheus library. At that point, this translation
code will be deleted.
"""
def __init__(self, metrics_obj, *args, **kwargs):
super().__init__(*args, **kwargs)
self._metrics = metrics_obj
def collect(self):
my_hostname = settings.CLUSTER_HOST_ID
instance_data = self._metrics.load_other_metrics(Request(HttpRequest()))
if not instance_data:
logger.debug(f"No metric data not found in redis for metric namespace '{self._metrics._namespace}'")
return None
host_metrics = instance_data.get(my_hostname)
for _, metric in self._metrics.METRICS.items():
entry = host_metrics.get(metric.field)
if not entry:
logger.debug(f"{self._metrics._namespace} metric '{metric.field}' not found in redis data payload {json.dumps(instance_data, indent=2)}")
continue
if isinstance(metric, HistogramM):
buckets = list(zip(metric.buckets, entry['counts']))
buckets = [[str(i[0]), str(i[1])] for i in buckets]
yield HistogramMetricFamily(metric.field, metric.help_text, buckets=buckets, sum_value=entry['sum'])
else:
yield GaugeMetricFamily(metric.field, metric.help_text, value=entry)
class CallbackReceiverMetricsServer(MetricsServer):
def __init__(self):
registry = CollectorRegistry(auto_describe=True)
registry.register(CustomToPrometheusMetricsCollector(DispatcherMetrics(metrics_have_changed=False)))
super().__init__(settings.METRICS_SERVICE_CALLBACK_RECEIVER, registry)
class DispatcherMetricsServer(MetricsServer):
def __init__(self):
registry = CollectorRegistry(auto_describe=True)
registry.register(CustomToPrometheusMetricsCollector(CallbackReceiverMetrics(metrics_have_changed=False)))
super().__init__(settings.METRICS_SERVICE_DISPATCHER, registry)
class WebsocketsMetricsServer(MetricsServer):
def __init__(self):
registry = CollectorRegistry(auto_describe=True)
# registry.register()
super().__init__(settings.METRICS_SERVICE_WEBSOCKETS, registry)

View File

@@ -827,6 +827,16 @@ register(
category_slug='bulk',
)
register(
'BULK_HOST_MAX_DELETE',
field_class=fields.IntegerField,
default=250,
label=_('Max number of hosts to allow to be deleted in a single bulk action'),
help_text=_('Max number of hosts to allow to be deleted in a single bulk action'),
category=_('Bulk Actions'),
category_slug='bulk',
)
register(
'UI_NEXT',
field_class=fields.BooleanField,

View File

@@ -14,7 +14,7 @@ __all__ = [
'STANDARD_INVENTORY_UPDATE_ENV',
]
CLOUD_PROVIDERS = ('azure_rm', 'ec2', 'gce', 'vmware', 'openstack', 'rhv', 'satellite6', 'controller', 'insights')
CLOUD_PROVIDERS = ('azure_rm', 'ec2', 'gce', 'vmware', 'openstack', 'rhv', 'satellite6', 'controller', 'insights', 'terraform')
PRIVILEGE_ESCALATION_METHODS = [
('sudo', _('Sudo')),
('su', _('Su')),

View File

@@ -106,7 +106,7 @@ class RelayConsumer(AsyncJsonWebsocketConsumer):
if group == "metrics":
message = json.loads(message['text'])
conn = redis.Redis.from_url(settings.BROKER_URL)
conn.set(settings.SUBSYSTEM_METRICS_REDIS_KEY_PREFIX + "_instance_" + message['instance'], message['metrics'])
conn.set(settings.SUBSYSTEM_METRICS_REDIS_KEY_PREFIX + "-" + message['metrics_namespace'] + "_instance_" + message['instance'], message['metrics'])
else:
await self.channel_layer.group_send(group, message)

View File

@@ -58,7 +58,7 @@ aim_inputs = {
'id': 'object_property',
'label': _('Object Property'),
'type': 'string',
'help_text': _('The property of the object to return. Default: Content Ex: Username, Address, etc.'),
'help_text': _('The property of the object to return. Available properties: Username, Password and Address.'),
},
{
'id': 'reason',
@@ -111,8 +111,12 @@ def aim_backend(**kwargs):
object_property = 'Content'
elif object_property.lower() == 'username':
object_property = 'UserName'
elif object_property.lower() == 'password':
object_property = 'Content'
elif object_property.lower() == 'address':
object_property = 'Address'
elif object_property not in res:
raise KeyError('Property {} not found in object'.format(object_property))
raise KeyError('Property {} not found in object, available properties: Username, Password and Address'.format(object_property))
else:
object_property = object_property.capitalize()

View File

@@ -41,6 +41,34 @@ base_inputs = {
'secret': True,
'help_text': _('The Secret ID for AppRole Authentication'),
},
{
'id': 'client_cert_public',
'label': _('Client Certificate'),
'type': 'string',
'multiline': True,
'help_text': _(
'The PEM-encoded client certificate used for TLS client authentication.'
' This should include the certificate and any intermediate certififcates.'
),
},
{
'id': 'client_cert_private',
'label': _('Client Certificate Key'),
'type': 'string',
'multiline': True,
'secret': True,
'help_text': _('The certificate private key used for TLS client authentication.'),
},
{
'id': 'client_cert_role',
'label': _('TLS Authentication Role'),
'type': 'string',
'multiline': False,
'help_text': _(
'The role configured in Hashicorp Vault for TLS client authentication.'
' If not provided, Hashicorp Vault may assign roles based on the certificate used.'
),
},
{
'id': 'namespace',
'label': _('Namespace name (Vault Enterprise only)'),
@@ -59,6 +87,20 @@ base_inputs = {
' see https://www.vaultproject.io/docs/auth/kubernetes#configuration'
),
},
{
'id': 'username',
'label': _('Username'),
'type': 'string',
'secret': False,
'help_text': _('Username for user authentication.'),
},
{
'id': 'password',
'label': _('Password'),
'type': 'string',
'secret': True,
'help_text': _('Password for user authentication.'),
},
{
'id': 'default_auth_path',
'label': _('Path to Auth'),
@@ -157,19 +199,25 @@ hashi_ssh_inputs['required'].extend(['public_key', 'role'])
def handle_auth(**kwargs):
token = None
if kwargs.get('token'):
token = kwargs['token']
elif kwargs.get('username') and kwargs.get('password'):
token = method_auth(**kwargs, auth_param=userpass_auth(**kwargs))
elif kwargs.get('role_id') and kwargs.get('secret_id'):
token = method_auth(**kwargs, auth_param=approle_auth(**kwargs))
elif kwargs.get('kubernetes_role'):
token = method_auth(**kwargs, auth_param=kubernetes_auth(**kwargs))
elif kwargs.get('client_cert_public') and kwargs.get('client_cert_private'):
token = method_auth(**kwargs, auth_param=client_cert_auth(**kwargs))
else:
raise Exception('Either token or AppRole/Kubernetes authentication parameters must be set')
raise Exception('Token, Username/Password, AppRole, Kubernetes, or TLS authentication parameters must be set')
return token
def userpass_auth(**kwargs):
return {'username': kwargs['username'], 'password': kwargs['password']}
def approle_auth(**kwargs):
return {'role_id': kwargs['role_id'], 'secret_id': kwargs['secret_id']}
@@ -181,6 +229,10 @@ def kubernetes_auth(**kwargs):
return {'role': kwargs['kubernetes_role'], 'jwt': jwt}
def client_cert_auth(**kwargs):
return {'name': kwargs.get('client_cert_role')}
def method_auth(**kwargs):
# get auth method specific params
request_kwargs = {'json': kwargs['auth_param'], 'timeout': 30}
@@ -193,13 +245,25 @@ def method_auth(**kwargs):
cacert = kwargs.get('cacert', None)
sess = requests.Session()
sess.mount(url, requests.adapters.HTTPAdapter(max_retries=5))
# Namespace support
if kwargs.get('namespace'):
sess.headers['X-Vault-Namespace'] = kwargs['namespace']
request_url = '/'.join([url, 'auth', auth_path, 'login']).rstrip('/')
if kwargs['auth_param'].get('username'):
request_url = request_url + '/' + (kwargs['username'])
with CertFiles(cacert) as cert:
request_kwargs['verify'] = cert
resp = sess.post(request_url, **request_kwargs)
# TLS client certificate support
if kwargs.get('client_cert_public') and kwargs.get('client_cert_private'):
# Add client cert to requests Session before making call
with CertFiles(kwargs['client_cert_public'], key=kwargs['client_cert_private']) as client_cert:
sess.cert = client_cert
resp = sess.post(request_url, **request_kwargs)
else:
# Make call without client certificate
resp = sess.post(request_url, **request_kwargs)
resp.raise_for_status()
token = resp.json()['auth']['client_token']
return token
@@ -220,6 +284,7 @@ def kv_backend(**kwargs):
}
sess = requests.Session()
sess.mount(url, requests.adapters.HTTPAdapter(max_retries=5))
sess.headers['Authorization'] = 'Bearer {}'.format(token)
# Compatibility header for older installs of Hashicorp Vault
sess.headers['X-Vault-Token'] = token
@@ -290,6 +355,7 @@ def ssh_backend(**kwargs):
request_kwargs['json']['valid_principals'] = kwargs['valid_principals']
sess = requests.Session()
sess.mount(url, requests.adapters.HTTPAdapter(max_retries=5))
sess.headers['Authorization'] = 'Bearer {}'.format(token)
if kwargs.get('namespace'):
sess.headers['X-Vault-Namespace'] = kwargs['namespace']

View File

@@ -93,6 +93,26 @@ class PubSub(object):
self.conn.close()
def create_listener_connection():
conf = settings.DATABASES['default'].copy()
conf['OPTIONS'] = conf.get('OPTIONS', {}).copy()
# Modify the application name to distinguish from other connections the process might use
conf['OPTIONS']['application_name'] = get_application_name(settings.CLUSTER_HOST_ID, function='listener')
# Apply overrides specifically for the listener connection
for k, v in settings.LISTENER_DATABASES.get('default', {}).items():
conf[k] = v
for k, v in settings.LISTENER_DATABASES.get('default', {}).get('OPTIONS', {}).items():
conf['OPTIONS'][k] = v
# Allow password-less authentication
if 'PASSWORD' in conf:
conf['OPTIONS']['password'] = conf.pop('PASSWORD')
connection_data = f"dbname={conf['NAME']} host={conf['HOST']} user={conf['USER']} port={conf['PORT']}"
return psycopg.connect(connection_data, autocommit=True, **conf['OPTIONS'])
@contextmanager
def pg_bus_conn(new_connection=False, select_timeout=None):
'''
@@ -106,12 +126,7 @@ def pg_bus_conn(new_connection=False, select_timeout=None):
'''
if new_connection:
conf = settings.DATABASES['default'].copy()
conf['OPTIONS'] = conf.get('OPTIONS', {}).copy()
# Modify the application name to distinguish from other connections the process might use
conf['OPTIONS']['application_name'] = get_application_name(settings.CLUSTER_HOST_ID, function='listener')
connection_data = f"dbname={conf['NAME']} host={conf['HOST']} user={conf['USER']} password={conf['PASSWORD']} port={conf['PORT']}"
conn = psycopg.connect(connection_data, autocommit=True, **conf['OPTIONS'])
conn = create_listener_connection()
else:
if pg_connection.connection is None:
pg_connection.connect()

View File

@@ -162,13 +162,13 @@ class AWXConsumerRedis(AWXConsumerBase):
class AWXConsumerPG(AWXConsumerBase):
def __init__(self, *args, schedule=None, **kwargs):
super().__init__(*args, **kwargs)
self.pg_max_wait = settings.DISPATCHER_DB_DOWNTIME_TOLERANCE
self.pg_max_wait = getattr(settings, 'DISPATCHER_DB_DOWNTOWN_TOLLERANCE', settings.DISPATCHER_DB_DOWNTIME_TOLERANCE)
# if no successful loops have ran since startup, then we should fail right away
self.pg_is_down = True # set so that we fail if we get database errors on startup
init_time = time.time()
self.pg_down_time = init_time - self.pg_max_wait # allow no grace period
self.last_cleanup = init_time
self.subsystem_metrics = s_metrics.Metrics(auto_pipe_execute=False)
self.subsystem_metrics = s_metrics.DispatcherMetrics(auto_pipe_execute=False)
self.last_metrics_gather = init_time
self.listen_cumulative_time = 0.0
if schedule:
@@ -214,7 +214,10 @@ class AWXConsumerPG(AWXConsumerBase):
# bypasses pg_notify for scheduled tasks
self.dispatch_task(body)
self.pg_is_down = False
if self.pg_is_down:
logger.info('Dispatcher listener connection established')
self.pg_is_down = False
self.listen_start = time.time()
return self.scheduler.time_until_next_run()
@@ -256,6 +259,12 @@ class AWXConsumerPG(AWXConsumerBase):
current_downtime = time.time() - self.pg_down_time
if current_downtime > self.pg_max_wait:
logger.exception(f"Postgres event consumer has not recovered in {current_downtime} s, exiting")
# Sending QUIT to multiprocess queue to signal workers to exit
for worker in self.pool.workers:
try:
worker.quit()
except Exception:
logger.exception(f"Error sending QUIT to worker {worker}")
raise
# Wait for a second before next attempt, but still listen for any shutdown signals
for i in range(10):
@@ -267,6 +276,12 @@ class AWXConsumerPG(AWXConsumerBase):
except Exception:
# Log unanticipated exception in addition to writing to stderr to get timestamps and other metadata
logger.exception('Encountered unhandled error in dispatcher main loop')
# Sending QUIT to multiprocess queue to signal workers to exit
for worker in self.pool.workers:
try:
worker.quit()
except Exception:
logger.exception(f"Error sending QUIT to worker {worker}")
raise

View File

@@ -72,7 +72,7 @@ class CallbackBrokerWorker(BaseWorker):
def __init__(self):
self.buff = {}
self.redis = redis.Redis.from_url(settings.BROKER_URL)
self.subsystem_metrics = s_metrics.Metrics(auto_pipe_execute=False)
self.subsystem_metrics = s_metrics.CallbackReceiverMetrics(auto_pipe_execute=False)
self.queue_pop = 0
self.queue_name = settings.CALLBACK_QUEUE
self.prof = AWXProfiler("CallbackBrokerWorker")

View File

@@ -5,6 +5,7 @@
import copy
import json
import re
import sys
import urllib.parse
from jinja2 import sandbox, StrictUndefined
@@ -406,11 +407,13 @@ class SmartFilterField(models.TextField):
# https://docs.python.org/2/library/stdtypes.html#truth-value-testing
if not value:
return None
value = urllib.parse.unquote(value)
try:
SmartFilter().query_from_string(value)
except RuntimeError as e:
raise models.base.ValidationError(e)
# avoid doing too much during migrations
if 'migrate' not in sys.argv:
value = urllib.parse.unquote(value)
try:
SmartFilter().query_from_string(value)
except RuntimeError as e:
raise models.base.ValidationError(e)
return super(SmartFilterField, self).get_prep_value(value)

View File

@@ -0,0 +1,53 @@
# Copyright (c) 2015 Ansible, Inc.
# All Rights Reserved
from django.core.management.base import BaseCommand
from awx.main.models import Instance, ReceptorAddress
def add_address(**kwargs):
try:
instance = Instance.objects.get(hostname=kwargs.pop('instance'))
kwargs['instance'] = instance
if kwargs.get('canonical') and instance.receptor_addresses.filter(canonical=True).exclude(address=kwargs['address']).exists():
print(f"Instance {instance.hostname} already has a canonical address, skipping")
return False
# if ReceptorAddress already exists with address, just update
# otherwise, create new ReceptorAddress
addr, _ = ReceptorAddress.objects.update_or_create(address=kwargs.pop('address'), defaults=kwargs)
print(f"Successfully added receptor address {addr.get_full_address()}")
return True
except Exception as e:
print(f"Error adding receptor address: {e}")
return False
class Command(BaseCommand):
"""
Internal controller command.
Register receptor address to an already-registered instance.
"""
help = "Add receptor address to an instance."
def add_arguments(self, parser):
parser.add_argument('--instance', dest='instance', required=True, type=str, help="Instance hostname this address is added to")
parser.add_argument('--address', dest='address', required=True, type=str, help="Receptor address")
parser.add_argument('--port', dest='port', type=int, help="Receptor listener port")
parser.add_argument('--websocket_path', dest='websocket_path', type=str, default="", help="Path for websockets")
parser.add_argument('--is_internal', action='store_true', help="If true, address only resolvable within the Kubernetes cluster")
parser.add_argument('--protocol', type=str, default='tcp', choices=['tcp', 'ws', 'wss'], help="Protocol to use for the Receptor listener")
parser.add_argument('--canonical', action='store_true', help="If true, address is the canonical address for the instance")
parser.add_argument('--peers_from_control_nodes', action='store_true', help="If true, control nodes will peer to this address")
def handle(self, **options):
address_options = {
k: options[k]
for k in ('instance', 'address', 'port', 'websocket_path', 'is_internal', 'protocol', 'peers_from_control_nodes', 'canonical')
if options[k]
}
changed = add_address(**address_options)
if changed:
print("(changed: True)")

View File

@@ -55,7 +55,7 @@ class Command(BaseCommand):
capacity = f' capacity={x.capacity}' if x.node_type != 'hop' else ''
version = f" version={x.version or '?'}" if x.node_type != 'hop' else ''
heartbeat = f' heartbeat="{x.last_seen:%Y-%m-%d %H:%M:%S}"' if x.capacity or x.node_type == 'hop' else ''
heartbeat = f' heartbeat="{x.last_seen:%Y-%m-%d %H:%M:%S}"' if x.last_seen else ''
print(f'\t{color}{x.hostname}{capacity} node_type={x.node_type}{version}{heartbeat}{end_color}')
print()

View File

@@ -25,20 +25,17 @@ class Command(BaseCommand):
def add_arguments(self, parser):
parser.add_argument('--hostname', dest='hostname', type=str, help="Hostname used during provisioning")
parser.add_argument('--listener_port', dest='listener_port', type=int, help="Receptor listener port")
parser.add_argument('--node_type', type=str, default='hybrid', choices=['control', 'execution', 'hop', 'hybrid'], help="Instance Node type")
parser.add_argument('--uuid', type=str, help="Instance UUID")
def _register_hostname(self, hostname, node_type, uuid, listener_port):
def _register_hostname(self, hostname, node_type, uuid):
if not hostname:
if not settings.AWX_AUTO_DEPROVISION_INSTANCES:
raise CommandError('Registering with values from settings only intended for use in K8s installs')
from awx.main.management.commands.register_queue import RegisterQueue
(changed, instance) = Instance.objects.register(
ip_address=os.environ.get('MY_POD_IP'), listener_port=listener_port, node_type='control', node_uuid=settings.SYSTEM_UUID
)
(changed, instance) = Instance.objects.register(ip_address=os.environ.get('MY_POD_IP'), node_type='control', node_uuid=settings.SYSTEM_UUID)
RegisterQueue(settings.DEFAULT_CONTROL_PLANE_QUEUE_NAME, 100, 0, [], is_container_group=False).register()
RegisterQueue(
settings.DEFAULT_EXECUTION_QUEUE_NAME,
@@ -51,16 +48,17 @@ class Command(BaseCommand):
max_concurrent_jobs=settings.DEFAULT_EXECUTION_QUEUE_MAX_CONCURRENT_JOBS,
).register()
else:
(changed, instance) = Instance.objects.register(hostname=hostname, node_type=node_type, node_uuid=uuid, listener_port=listener_port)
(changed, instance) = Instance.objects.register(hostname=hostname, node_type=node_type, node_uuid=uuid)
if changed:
print("Successfully registered instance {}".format(hostname))
else:
print("Instance already registered {}".format(instance.hostname))
self.changed = changed
@transaction.atomic
def handle(self, **options):
self.changed = False
self._register_hostname(options.get('hostname'), options.get('node_type'), options.get('uuid'), options.get('listener_port'))
self._register_hostname(options.get('hostname'), options.get('node_type'), options.get('uuid'))
if self.changed:
print("(changed: True)")

View File

@@ -1,9 +1,7 @@
import warnings
from django.core.management.base import BaseCommand, CommandError
from django.db import transaction
from awx.main.models import Instance, InstanceLink
from awx.main.models import Instance, InstanceLink, ReceptorAddress
class Command(BaseCommand):
@@ -28,7 +26,9 @@ class Command(BaseCommand):
def handle(self, **options):
# provides a mapping of hostname to Instance objects
nodes = Instance.objects.in_bulk(field_name='hostname')
nodes = Instance.objects.all().in_bulk(field_name='hostname')
# provides a mapping of address to ReceptorAddress objects
addresses = ReceptorAddress.objects.all().in_bulk(field_name='address')
if options['source'] not in nodes:
raise CommandError(f"Host {options['source']} is not a registered instance.")
@@ -39,6 +39,14 @@ class Command(BaseCommand):
if options['exact'] is not None and options['disconnect']:
raise CommandError("The option --disconnect may not be used with --exact.")
# make sure each target has a receptor address
peers = options['peers'] or []
disconnect = options['disconnect'] or []
exact = options['exact'] or []
for peer in peers + disconnect + exact:
if peer not in addresses:
raise CommandError(f"Peer {peer} does not have a receptor address.")
# No 1-cycles
for collection in ('peers', 'disconnect', 'exact'):
if options[collection] is not None and options['source'] in options[collection]:
@@ -47,9 +55,12 @@ class Command(BaseCommand):
# No 2-cycles
if options['peers'] or options['exact'] is not None:
peers = set(options['peers'] or options['exact'])
incoming = set(InstanceLink.objects.filter(target=nodes[options['source']]).values_list('source__hostname', flat=True))
if options['source'] in addresses:
incoming = set(InstanceLink.objects.filter(target=addresses[options['source']]).values_list('source__hostname', flat=True))
else:
incoming = set()
if peers & incoming:
warnings.warn(f"Source node {options['source']} should not link to nodes already peering to it: {peers & incoming}.")
raise CommandError(f"Source node {options['source']} should not link to nodes already peering to it: {peers & incoming}.")
if options['peers']:
missing_peers = set(options['peers']) - set(nodes)
@@ -60,7 +71,7 @@ class Command(BaseCommand):
results = 0
for target in options['peers']:
_, created = InstanceLink.objects.update_or_create(
source=nodes[options['source']], target=nodes[target], defaults={'link_state': InstanceLink.States.ESTABLISHED}
source=nodes[options['source']], target=addresses[target], defaults={'link_state': InstanceLink.States.ESTABLISHED}
)
if created:
results += 1
@@ -70,9 +81,9 @@ class Command(BaseCommand):
if options['disconnect']:
results = 0
for target in options['disconnect']:
if target not in nodes: # Be permissive, the node might have already been de-registered.
if target not in addresses: # Be permissive, the node might have already been de-registered.
continue
n, _ = InstanceLink.objects.filter(source=nodes[options['source']], target=nodes[target]).delete()
n, _ = InstanceLink.objects.filter(source=nodes[options['source']], target=addresses[target]).delete()
results += n
print(f"{results} peer links removed from the database.")
@@ -81,11 +92,11 @@ class Command(BaseCommand):
additions = 0
with transaction.atomic():
peers = set(options['exact'])
links = set(InstanceLink.objects.filter(source=nodes[options['source']]).values_list('target__hostname', flat=True))
removals, _ = InstanceLink.objects.filter(source=nodes[options['source']], target__hostname__in=links - peers).delete()
links = set(InstanceLink.objects.filter(source=nodes[options['source']]).values_list('target__address', flat=True))
removals, _ = InstanceLink.objects.filter(source=nodes[options['source']], target__instance__hostname__in=links - peers).delete()
for target in peers - links:
_, created = InstanceLink.objects.update_or_create(
source=nodes[options['source']], target=nodes[target], defaults={'link_state': InstanceLink.States.ESTABLISHED}
source=nodes[options['source']], target=addresses[target], defaults={'link_state': InstanceLink.States.ESTABLISHED}
)
if created:
additions += 1

View File

@@ -0,0 +1,26 @@
# Copyright (c) 2015 Ansible, Inc.
# All Rights Reserved
from django.core.management.base import BaseCommand
from awx.main.models import ReceptorAddress
class Command(BaseCommand):
"""
Internal controller command.
Delete a receptor address.
"""
help = "Add receptor address to an instance."
def add_arguments(self, parser):
parser.add_argument('--address', dest='address', type=str, help="Receptor address to remove")
def handle(self, **options):
deleted = ReceptorAddress.objects.filter(address=options['address']).delete()
if deleted[0]:
print(f"Successfully removed {options['address']}")
print("(changed: True)")
else:
print(f"Did not remove {options['address']}, not found")

View File

@@ -3,6 +3,7 @@
from django.conf import settings
from django.core.management.base import BaseCommand
from awx.main.analytics.subsystem_metrics import CallbackReceiverMetricsServer
from awx.main.dispatch.control import Control
from awx.main.dispatch.worker import AWXConsumerRedis, CallbackBrokerWorker
@@ -25,6 +26,9 @@ class Command(BaseCommand):
print(Control('callback_receiver').status())
return
consumer = None
CallbackReceiverMetricsServer().start()
try:
consumer = AWXConsumerRedis(
'callback_receiver',

View File

@@ -10,6 +10,7 @@ from awx.main.dispatch import get_task_queuename
from awx.main.dispatch.control import Control
from awx.main.dispatch.pool import AutoscalePool
from awx.main.dispatch.worker import AWXConsumerPG, TaskWorker
from awx.main.analytics.subsystem_metrics import DispatcherMetricsServer
logger = logging.getLogger('awx.main.dispatch')
@@ -62,6 +63,8 @@ class Command(BaseCommand):
consumer = None
DispatcherMetricsServer().start()
try:
queues = ['tower_broadcast_all', 'tower_settings_change', get_task_queuename()]
consumer = AWXConsumerPG('dispatcher', TaskWorker(), queues, AutoscalePool(min_workers=4), schedule=settings.CELERYBEAT_SCHEDULE)

View File

@@ -16,6 +16,7 @@ from awx.main.analytics.broadcast_websocket import (
RelayWebsocketStatsManager,
safe_name,
)
from awx.main.analytics.subsystem_metrics import WebsocketsMetricsServer
from awx.main.wsrelay import WebSocketRelayManager
@@ -91,6 +92,8 @@ class Command(BaseCommand):
return host_stats
def handle(self, *arg, **options):
WebsocketsMetricsServer().start()
# it's necessary to delay this import in case
# database migrations are still running
from awx.main.models.ha import Instance

View File

@@ -115,7 +115,14 @@ class InstanceManager(models.Manager):
return node[0]
raise RuntimeError("No instance found with the current cluster host id")
def register(self, node_uuid=None, hostname=None, ip_address="", listener_port=None, node_type='hybrid', defaults=None):
def register(
self,
node_uuid=None,
hostname=None,
ip_address="",
node_type='hybrid',
defaults=None,
):
if not hostname:
hostname = settings.CLUSTER_HOST_ID
@@ -161,9 +168,6 @@ class InstanceManager(models.Manager):
if instance.node_type != node_type:
instance.node_type = node_type
update_fields.append('node_type')
if instance.listener_port != listener_port:
instance.listener_port = listener_port
update_fields.append('listener_port')
if update_fields:
instance.save(update_fields=update_fields)
return (True, instance)
@@ -174,11 +178,13 @@ class InstanceManager(models.Manager):
create_defaults = {
'node_state': Instance.States.INSTALLED,
'capacity': 0,
'managed': True,
}
if defaults is not None:
create_defaults.update(defaults)
uuid_option = {'uuid': node_uuid if node_uuid is not None else uuid.uuid4()}
if node_type == 'execution' and 'version' not in create_defaults:
create_defaults['version'] = RECEPTOR_PENDING
instance = self.create(hostname=hostname, ip_address=ip_address, listener_port=listener_port, node_type=node_type, **create_defaults, **uuid_option)
instance = self.create(hostname=hostname, ip_address=ip_address, node_type=node_type, **create_defaults, **uuid_option)
return (True, instance)

View File

@@ -5,11 +5,12 @@ import logging
import threading
import time
import urllib.parse
from pathlib import Path
from django.conf import settings
from django.contrib.auth import logout
from django.contrib.auth.models import User
from django.db.migrations.executor import MigrationExecutor
from django.db.migrations.recorder import MigrationRecorder
from django.db import connection
from django.shortcuts import redirect
from django.apps import apps
@@ -17,9 +18,11 @@ from django.utils.deprecation import MiddlewareMixin
from django.utils.translation import gettext_lazy as _
from django.urls import reverse, resolve
from awx.main import migrations
from awx.main.utils.named_url_graph import generate_graph, GraphNode
from awx.conf import fields, register
from awx.main.utils.profiling import AWXProfiler
from awx.main.utils.common import memoize
logger = logging.getLogger('awx.main.middleware')
@@ -198,9 +201,22 @@ class URLModificationMiddleware(MiddlewareMixin):
request.path_info = new_path
@memoize(ttl=20)
def is_migrating():
latest_number = 0
latest_name = ''
for migration_path in Path(migrations.__path__[0]).glob('[0-9]*.py'):
try:
migration_number = int(migration_path.name.split('_', 1)[0])
except ValueError:
continue
if migration_number > latest_number:
latest_number = migration_number
latest_name = migration_path.name[: -len('.py')]
return not MigrationRecorder(connection).migration_qs.filter(app='main', name=latest_name).exists()
class MigrationRanCheckMiddleware(MiddlewareMixin):
def process_request(self, request):
executor = MigrationExecutor(connection)
plan = executor.migration_plan(executor.loader.graph.leaf_nodes())
if bool(plan) and getattr(resolve(request.path), 'url_name', '') != 'migrations_notran':
if is_migrating() and getattr(resolve(request.path), 'url_name', '') != 'migrations_notran':
return redirect(reverse("ui:migrations_notran"))

View File

@@ -0,0 +1,52 @@
# Generated by Django 4.2.6 on 2023-11-16 21:00
from django.db import migrations, models
class Migration(migrations.Migration):
dependencies = [
('main', '0187_hop_nodes'),
]
operations = [
migrations.AlterField(
model_name='job',
name='webhook_service',
field=models.CharField(
blank=True,
choices=[('github', 'GitHub'), ('gitlab', 'GitLab'), ('bitbucket_dc', 'BitBucket DataCenter')],
help_text='Service that webhook requests will be accepted from',
max_length=16,
),
),
migrations.AlterField(
model_name='jobtemplate',
name='webhook_service',
field=models.CharField(
blank=True,
choices=[('github', 'GitHub'), ('gitlab', 'GitLab'), ('bitbucket_dc', 'BitBucket DataCenter')],
help_text='Service that webhook requests will be accepted from',
max_length=16,
),
),
migrations.AlterField(
model_name='workflowjob',
name='webhook_service',
field=models.CharField(
blank=True,
choices=[('github', 'GitHub'), ('gitlab', 'GitLab'), ('bitbucket_dc', 'BitBucket DataCenter')],
help_text='Service that webhook requests will be accepted from',
max_length=16,
),
),
migrations.AlterField(
model_name='workflowjobtemplate',
name='webhook_service',
field=models.CharField(
blank=True,
choices=[('github', 'GitHub'), ('gitlab', 'GitLab'), ('bitbucket_dc', 'BitBucket DataCenter')],
help_text='Service that webhook requests will be accepted from',
max_length=16,
),
),
]

View File

@@ -0,0 +1,150 @@
# Generated by Django 4.2.6 on 2024-01-19 19:24
import django.core.validators
from django.db import migrations, models
import django.db.models.deletion
def create_receptor_addresses(apps, schema_editor):
"""
If listener_port was defined on an instance, create a receptor address for it
"""
Instance = apps.get_model('main', 'Instance')
ReceptorAddress = apps.get_model('main', 'ReceptorAddress')
for instance in Instance.objects.exclude(listener_port=None):
ReceptorAddress.objects.create(
instance=instance,
address=instance.hostname,
port=instance.listener_port,
peers_from_control_nodes=instance.peers_from_control_nodes,
protocol='tcp',
is_internal=False,
canonical=True,
)
def link_to_receptor_addresses(apps, schema_editor):
"""
Modify each InstanceLink to point to the newly created
ReceptorAddresses, using the new target field
"""
InstanceLink = apps.get_model('main', 'InstanceLink')
for link in InstanceLink.objects.all():
link.target = link.target_old.receptor_addresses.get()
link.save()
class Migration(migrations.Migration):
dependencies = [
('main', '0188_add_bitbucket_dc_webhook'),
]
operations = [
migrations.CreateModel(
name='ReceptorAddress',
fields=[
('id', models.AutoField(auto_created=True, primary_key=True, serialize=False, verbose_name='ID')),
('address', models.CharField(help_text='Routable address for this instance.', max_length=255)),
(
'port',
models.IntegerField(
default=27199,
help_text='Port for the address.',
validators=[django.core.validators.MinValueValidator(0), django.core.validators.MaxValueValidator(65535)],
),
),
('websocket_path', models.CharField(blank=True, default='', help_text='Websocket path.', max_length=255)),
(
'protocol',
models.CharField(
choices=[('tcp', 'TCP'), ('ws', 'WS'), ('wss', 'WSS')],
default='tcp',
help_text="Protocol to use for the Receptor listener, 'tcp', 'wss', or 'ws'.",
max_length=10,
),
),
('is_internal', models.BooleanField(default=False, help_text='If True, only routable within the Kubernetes cluster.')),
('canonical', models.BooleanField(default=False, help_text='If True, this address is the canonical address for the instance.')),
(
'peers_from_control_nodes',
models.BooleanField(default=False, help_text='If True, control plane cluster nodes should automatically peer to it.'),
),
],
),
migrations.RemoveConstraint(
model_name='instancelink',
name='source_and_target_can_not_be_equal',
),
migrations.RenameField(
model_name='instancelink',
old_name='target',
new_name='target_old',
),
migrations.AlterUniqueTogether(
name='instancelink',
unique_together=set(),
),
migrations.AddField(
model_name='instance',
name='managed',
field=models.BooleanField(default=False, editable=False, help_text='If True, this instance is managed by the control plane.'),
),
migrations.AlterField(
model_name='instancelink',
name='source',
field=models.ForeignKey(help_text='The source instance of this peer link.', on_delete=django.db.models.deletion.CASCADE, to='main.instance'),
),
migrations.AddField(
model_name='receptoraddress',
name='instance',
field=models.ForeignKey(on_delete=django.db.models.deletion.CASCADE, related_name='receptor_addresses', to='main.instance'),
),
migrations.AddField(
model_name='activitystream',
name='receptor_address',
field=models.ManyToManyField(blank=True, to='main.receptoraddress'),
),
migrations.AddConstraint(
model_name='receptoraddress',
constraint=models.UniqueConstraint(fields=('address',), name='unique_receptor_address', violation_error_message='Receptor address must be unique.'),
),
migrations.AddField(
model_name='instancelink',
name='target',
field=models.ForeignKey(
help_text='The target receptor address of this peer link.', null=True, on_delete=django.db.models.deletion.CASCADE, to='main.receptoraddress'
),
),
migrations.RunPython(create_receptor_addresses),
migrations.RunPython(link_to_receptor_addresses),
migrations.RemoveField(
model_name='instance',
name='peers_from_control_nodes',
),
migrations.RemoveField(
model_name='instance',
name='listener_port',
),
migrations.RemoveField(
model_name='instancelink',
name='target_old',
),
migrations.AlterField(
model_name='instance',
name='peers',
field=models.ManyToManyField(related_name='peers_from', through='main.InstanceLink', to='main.receptoraddress'),
),
migrations.AlterField(
model_name='instancelink',
name='target',
field=models.ForeignKey(
help_text='The target receptor address of this peer link.', on_delete=django.db.models.deletion.CASCADE, to='main.receptoraddress'
),
),
migrations.AddConstraint(
model_name='instancelink',
constraint=models.UniqueConstraint(
fields=('source', 'target'), name='unique_source_target', violation_error_message='Field source and target must be unique together.'
),
),
]

View File

@@ -0,0 +1,59 @@
# Generated by Django 4.2.6 on 2024-02-15 20:51
from django.db import migrations, models
class Migration(migrations.Migration):
dependencies = [
('main', '0189_inbound_hop_nodes'),
]
operations = [
migrations.AlterField(
model_name='inventorysource',
name='source',
field=models.CharField(
choices=[
('file', 'File, Directory or Script'),
('constructed', 'Template additional groups and hostvars at runtime'),
('scm', 'Sourced from a Project'),
('ec2', 'Amazon EC2'),
('gce', 'Google Compute Engine'),
('azure_rm', 'Microsoft Azure Resource Manager'),
('vmware', 'VMware vCenter'),
('satellite6', 'Red Hat Satellite 6'),
('openstack', 'OpenStack'),
('rhv', 'Red Hat Virtualization'),
('controller', 'Red Hat Ansible Automation Platform'),
('insights', 'Red Hat Insights'),
('terraform', 'Terraform State'),
],
default=None,
max_length=32,
),
),
migrations.AlterField(
model_name='inventoryupdate',
name='source',
field=models.CharField(
choices=[
('file', 'File, Directory or Script'),
('constructed', 'Template additional groups and hostvars at runtime'),
('scm', 'Sourced from a Project'),
('ec2', 'Amazon EC2'),
('gce', 'Google Compute Engine'),
('azure_rm', 'Microsoft Azure Resource Manager'),
('vmware', 'VMware vCenter'),
('satellite6', 'Red Hat Satellite 6'),
('openstack', 'OpenStack'),
('rhv', 'Red Hat Virtualization'),
('controller', 'Red Hat Ansible Automation Platform'),
('insights', 'Red Hat Insights'),
('terraform', 'Terraform State'),
],
default=None,
max_length=32,
),
),
]

View File

@@ -76,7 +76,7 @@ class azure_rm(PluginFileInjector):
user_filters = []
old_filterables = [
('resource_groups', 'resource_group'),
('tags', 'tags')
('tags', 'tags'),
# locations / location would be an entry
# but this would conflict with source_regions
]

View File

@@ -6,12 +6,15 @@ from django.conf import settings # noqa
from django.db import connection
from django.db.models.signals import pre_delete # noqa
from ansible_base.lib.utils.models import prevent_search
# AWX
from awx.main.models.base import BaseModel, PrimordialModel, prevent_search, accepts_json, CLOUD_INVENTORY_SOURCES, VERBOSITY_CHOICES # noqa
from awx.main.models.base import BaseModel, PrimordialModel, accepts_json, CLOUD_INVENTORY_SOURCES, VERBOSITY_CHOICES # noqa
from awx.main.models.unified_jobs import UnifiedJob, UnifiedJobTemplate, StdoutMaxBytesExceeded # noqa
from awx.main.models.organization import Organization, Profile, Team, UserSessionMembership # noqa
from awx.main.models.credential import Credential, CredentialType, CredentialInputSource, ManagedCredentialType, build_safe_env # noqa
from awx.main.models.projects import Project, ProjectUpdate # noqa
from awx.main.models.receptor_address import ReceptorAddress # noqa
from awx.main.models.inventory import ( # noqa
CustomInventoryScript,
Group,

View File

@@ -77,6 +77,7 @@ class ActivityStream(models.Model):
notification_template = models.ManyToManyField("NotificationTemplate", blank=True)
notification = models.ManyToManyField("Notification", blank=True)
label = models.ManyToManyField("Label", blank=True)
receptor_address = models.ManyToManyField("ReceptorAddress", blank=True)
role = models.ManyToManyField("Role", blank=True)
instance = models.ManyToManyField("Instance", blank=True)
instance_group = models.ManyToManyField("InstanceGroup", blank=True)

View File

@@ -12,9 +12,11 @@ from django.utils.text import Truncator
from django.utils.translation import gettext_lazy as _
from django.core.exceptions import ValidationError
from ansible_base.lib.utils.models import prevent_search
# AWX
from awx.api.versioning import reverse
from awx.main.models.base import prevent_search, AD_HOC_JOB_TYPE_CHOICES, VERBOSITY_CHOICES, VarsDictProperty
from awx.main.models.base import AD_HOC_JOB_TYPE_CHOICES, VERBOSITY_CHOICES, VarsDictProperty
from awx.main.models.events import AdHocCommandEvent, UnpartitionedAdHocCommandEvent
from awx.main.models.unified_jobs import UnifiedJob
from awx.main.models.notifications import JobNotificationMixin, NotificationTemplate

View File

@@ -15,7 +15,6 @@ from awx.main.utils import encrypt_field, parse_yaml_or_json
from awx.main.constants import CLOUD_PROVIDERS
__all__ = [
'prevent_search',
'VarsDictProperty',
'BaseModel',
'CreatedModifiedModel',
@@ -384,23 +383,6 @@ class NotificationFieldsModel(BaseModel):
notification_templates_started = models.ManyToManyField("NotificationTemplate", blank=True, related_name='%(class)s_notification_templates_for_started')
def prevent_search(relation):
"""
Used to mark a model field or relation as "restricted from filtering"
e.g.,
class AuthToken(BaseModel):
user = prevent_search(models.ForeignKey(...))
sensitive_data = prevent_search(models.CharField(...))
The flag set by this function is used by
`awx.api.filters.FieldLookupBackend` to block fields and relations that
should not be searchable/filterable via search query params
"""
setattr(relation, '__prevent_search__', True)
return relation
def accepts_json(relation):
"""
Used to mark a model field as allowing JSON e.g,. JobTemplate.extra_vars

View File

@@ -953,6 +953,25 @@ ManagedCredentialType(
},
)
ManagedCredentialType(
namespace='bitbucket_dc_token',
kind='token',
name=gettext_noop('Bitbucket Data Center HTTP Access Token'),
managed=True,
inputs={
'fields': [
{
'id': 'token',
'label': gettext_noop('Token'),
'type': 'string',
'secret': True,
'help_text': gettext_noop('This token needs to come from your user settings in Bitbucket'),
}
],
'required': ['token'],
},
)
ManagedCredentialType(
namespace='insights',
kind='insights',
@@ -1197,6 +1216,26 @@ ManagedCredentialType(
},
)
ManagedCredentialType(
namespace='terraform',
kind='cloud',
name=gettext_noop('Terraform backend configuration'),
managed=True,
inputs={
'fields': [
{
'id': 'configuration',
'label': gettext_noop('Backend configuration'),
'type': 'string',
'secret': True,
'multiline': True,
'help_text': gettext_noop('Terraform backend config as Hashicorp configuration language.'),
},
],
'required': ['configuration'],
},
)
class CredentialInputSource(PrimordialModel):
class Meta:

View File

@@ -122,3 +122,11 @@ def kubernetes_bearer_token(cred, env, private_data_dir):
env['K8S_AUTH_SSL_CA_CERT'] = to_container_path(path, private_data_dir)
else:
env['K8S_AUTH_VERIFY_SSL'] = 'False'
def terraform(cred, env, private_data_dir):
handle, path = tempfile.mkstemp(dir=os.path.join(private_data_dir, 'env'))
with os.fdopen(handle, 'w') as f:
os.chmod(path, stat.S_IRUSR | stat.S_IWUSR)
f.write(cred.get_input('configuration'))
env['TF_BACKEND_CONFIG_FILE'] = to_container_path(path, private_data_dir)

View File

@@ -124,8 +124,6 @@ class BasePlaybookEvent(CreatedModifiedModel):
'parent_uuid',
'start_line',
'end_line',
'host_id',
'host_name',
'verbosity',
]
WRAPUP_EVENT = 'playbook_on_stats'
@@ -473,7 +471,7 @@ class JobEvent(BasePlaybookEvent):
An event/message logged from the callback when running a job.
"""
VALID_KEYS = BasePlaybookEvent.VALID_KEYS + ['job_id', 'workflow_job_id', 'job_created']
VALID_KEYS = BasePlaybookEvent.VALID_KEYS + ['job_id', 'workflow_job_id', 'job_created', 'host_id', 'host_name']
JOB_REFERENCE = 'job_id'
objects = DeferJobCreatedManager()

View File

@@ -5,7 +5,7 @@ from decimal import Decimal
import logging
import os
from django.core.validators import MinValueValidator, MaxValueValidator
from django.core.validators import MinValueValidator
from django.db import models, connection
from django.db.models.signals import post_save, post_delete
from django.dispatch import receiver
@@ -17,6 +17,8 @@ from django.db.models import Sum, Q
import redis
from solo.models import SingletonModel
from ansible_base.lib.utils.models import prevent_search
# AWX
from awx import __version__ as awx_application_version
from awx.main.utils import is_testing
@@ -24,7 +26,7 @@ from awx.api.versioning import reverse
from awx.main.fields import ImplicitRoleField
from awx.main.managers import InstanceManager, UUID_DEFAULT
from awx.main.constants import JOB_FOLDER_PREFIX
from awx.main.models.base import BaseModel, HasEditsMixin, prevent_search
from awx.main.models.base import BaseModel, HasEditsMixin
from awx.main.models.rbac import (
ROLE_SINGLETON_SYSTEM_ADMINISTRATOR,
ROLE_SINGLETON_SYSTEM_AUDITOR,
@@ -32,6 +34,7 @@ from awx.main.models.rbac import (
from awx.main.models.unified_jobs import UnifiedJob
from awx.main.utils.common import get_corrected_cpu, get_cpu_effective_capacity, get_corrected_memory, get_mem_effective_capacity
from awx.main.models.mixins import RelatedJobsMixin, ResourceMixin
from awx.main.models.receptor_address import ReceptorAddress
# ansible-runner
from ansible_runner.utils.capacity import get_cpu_count, get_mem_in_bytes
@@ -62,8 +65,19 @@ class HasPolicyEditsMixin(HasEditsMixin):
class InstanceLink(BaseModel):
source = models.ForeignKey('Instance', on_delete=models.CASCADE, related_name='+')
target = models.ForeignKey('Instance', on_delete=models.CASCADE, related_name='reverse_peers')
class Meta:
ordering = ("id",)
# add constraint for source and target to be unique together
constraints = [
models.UniqueConstraint(
fields=["source", "target"],
name="unique_source_target",
violation_error_message=_("Field source and target must be unique together."),
)
]
source = models.ForeignKey('Instance', on_delete=models.CASCADE, help_text=_("The source instance of this peer link."))
target = models.ForeignKey('ReceptorAddress', on_delete=models.CASCADE, help_text=_("The target receptor address of this peer link."))
class States(models.TextChoices):
ADDING = 'adding', _('Adding')
@@ -74,11 +88,6 @@ class InstanceLink(BaseModel):
choices=States.choices, default=States.ADDING, max_length=16, help_text=_("Indicates the current life cycle stage of this peer link.")
)
class Meta:
unique_together = ('source', 'target')
ordering = ("id",)
constraints = [models.CheckConstraint(check=~models.Q(source=models.F('target')), name='source_and_target_can_not_be_equal')]
class Instance(HasPolicyEditsMixin, BaseModel):
"""A model representing an AWX instance running against this database."""
@@ -108,6 +117,7 @@ class Instance(HasPolicyEditsMixin, BaseModel):
default="",
max_length=50,
)
# Auto-fields, implementation is different from BaseModel
created = models.DateTimeField(auto_now_add=True)
modified = models.DateTimeField(auto_now=True)
@@ -183,16 +193,9 @@ class Instance(HasPolicyEditsMixin, BaseModel):
node_state = models.CharField(
choices=States.choices, default=States.READY, max_length=16, help_text=_("Indicates the current life cycle stage of this instance.")
)
listener_port = models.PositiveIntegerField(
blank=True,
null=True,
default=None,
validators=[MinValueValidator(1024), MaxValueValidator(65535)],
help_text=_("Port that Receptor will listen for incoming connections on."),
)
peers = models.ManyToManyField('self', symmetrical=False, through=InstanceLink, through_fields=('source', 'target'), related_name='peers_from')
peers_from_control_nodes = models.BooleanField(default=False, help_text=_("If True, control plane cluster nodes should automatically peer to it."))
managed = models.BooleanField(help_text=_("If True, this instance is managed by the control plane."), default=False, editable=False)
peers = models.ManyToManyField('ReceptorAddress', through=InstanceLink, through_fields=('source', 'target'), related_name='peers_from')
POLICY_FIELDS = frozenset(('managed_by_policy', 'hostname', 'capacity_adjustment'))
@@ -239,6 +242,26 @@ class Instance(HasPolicyEditsMixin, BaseModel):
return True
return self.health_check_started > self.last_health_check
@property
def canonical_address(self):
return self.receptor_addresses.filter(canonical=True).first()
@property
def canonical_address_port(self):
# note: don't create a different query for receptor addresses, as this is prefetched on the View for optimization
for addr in self.receptor_addresses.all():
if addr.canonical:
return addr.port
return None
@property
def canonical_address_peers_from_control_nodes(self):
# note: don't create a different query for receptor addresses, as this is prefetched on the View for optimization
for addr in self.receptor_addresses.all():
if addr.canonical:
return addr.peers_from_control_nodes
return False
def get_cleanup_task_kwargs(self, **kwargs):
"""
Produce options to use for the command: ansible-runner worker cleanup
@@ -499,6 +522,35 @@ def schedule_write_receptor_config(broadcast=True):
write_receptor_config() # just run locally
@receiver(post_save, sender=ReceptorAddress)
def receptor_address_saved(sender, instance, **kwargs):
from awx.main.signals import disable_activity_stream
address = instance
control_instances = set(Instance.objects.filter(node_type__in=[Instance.Types.CONTROL, Instance.Types.HYBRID]))
if address.peers_from_control_nodes:
# if control_instances is not a subset of current peers of address, then
# that means we need to add some InstanceLinks
if not control_instances <= set(address.peers_from.all()):
with disable_activity_stream():
for control_instance in control_instances:
InstanceLink.objects.update_or_create(source=control_instance, target=address)
schedule_write_receptor_config()
else:
if address.peers_from.exists():
with disable_activity_stream():
address.peers_from.remove(*control_instances)
schedule_write_receptor_config()
@receiver(post_delete, sender=ReceptorAddress)
def receptor_address_deleted(sender, instance, **kwargs):
address = instance
if address.peers_from_control_nodes:
schedule_write_receptor_config()
@receiver(post_save, sender=Instance)
def on_instance_saved(sender, instance, created=False, raw=False, **kwargs):
'''
@@ -509,11 +561,14 @@ def on_instance_saved(sender, instance, created=False, raw=False, **kwargs):
2. a node changes its value of peers_from_control_nodes
3. a new control node comes online and has instances to peer to
'''
from awx.main.signals import disable_activity_stream
if created and settings.IS_K8S and instance.node_type in [Instance.Types.CONTROL, Instance.Types.HYBRID]:
inst = Instance.objects.filter(peers_from_control_nodes=True)
if set(instance.peers.all()) != set(inst):
instance.peers.set(inst)
schedule_write_receptor_config(broadcast=False)
peers_addresses = ReceptorAddress.objects.filter(peers_from_control_nodes=True)
if peers_addresses.exists():
with disable_activity_stream():
instance.peers.add(*peers_addresses)
schedule_write_receptor_config(broadcast=False)
if settings.IS_K8S and instance.node_type in [Instance.Types.HOP, Instance.Types.EXECUTION]:
if instance.node_state == Instance.States.DEPROVISIONING:
@@ -522,16 +577,6 @@ def on_instance_saved(sender, instance, created=False, raw=False, **kwargs):
# wait for jobs on the node to complete, then delete the
# node and kick off write_receptor_config
connection.on_commit(lambda: remove_deprovisioned_node.apply_async([instance.hostname]))
else:
control_instances = set(Instance.objects.filter(node_type__in=[Instance.Types.CONTROL, Instance.Types.HYBRID]))
if instance.peers_from_control_nodes:
if (control_instances & set(instance.peers_from.all())) != set(control_instances):
instance.peers_from.add(*control_instances)
schedule_write_receptor_config() # keep method separate to make pytest mocking easier
else:
if set(control_instances) & set(instance.peers_from.all()):
instance.peers_from.remove(*control_instances)
schedule_write_receptor_config()
if created or instance.has_policy_changes():
schedule_policy_task()
@@ -546,8 +591,6 @@ def on_instance_group_deleted(sender, instance, using, **kwargs):
@receiver(post_delete, sender=Instance)
def on_instance_deleted(sender, instance, using, **kwargs):
schedule_policy_task()
if settings.IS_K8S and instance.node_type in (Instance.Types.EXECUTION, Instance.Types.HOP) and instance.peers_from_control_nodes:
schedule_write_receptor_config()
class UnifiedJobTemplateInstanceGroupMembership(models.Model):

View File

@@ -25,6 +25,8 @@ from django.db.models import Q
# REST Framework
from rest_framework.exceptions import ParseError
from ansible_base.lib.utils.models import prevent_search
# AWX
from awx.api.versioning import reverse
from awx.main.constants import CLOUD_PROVIDERS
@@ -35,7 +37,7 @@ from awx.main.fields import (
OrderedManyToManyField,
)
from awx.main.managers import HostManager, HostMetricActiveManager
from awx.main.models.base import BaseModel, CommonModelNameNotUnique, VarsDictProperty, CLOUD_INVENTORY_SOURCES, prevent_search, accepts_json
from awx.main.models.base import BaseModel, CommonModelNameNotUnique, VarsDictProperty, CLOUD_INVENTORY_SOURCES, accepts_json
from awx.main.models.events import InventoryUpdateEvent, UnpartitionedInventoryUpdateEvent
from awx.main.models.unified_jobs import UnifiedJob, UnifiedJobTemplate
from awx.main.models.mixins import (
@@ -923,6 +925,7 @@ class InventorySourceOptions(BaseModel):
('rhv', _('Red Hat Virtualization')),
('controller', _('Red Hat Ansible Automation Platform')),
('insights', _('Red Hat Insights')),
('terraform', _('Terraform State')),
]
# From the options of the Django management base command
@@ -1628,6 +1631,20 @@ class satellite6(PluginFileInjector):
return ret
class terraform(PluginFileInjector):
plugin_name = 'terraform_state'
base_injector = 'managed'
namespace = 'cloud'
collection = 'terraform'
use_fqcn = True
def inventory_as_dict(self, inventory_update, private_data_dir):
env = super(terraform, self).get_plugin_env(inventory_update, private_data_dir, None)
ret = super().inventory_as_dict(inventory_update, private_data_dir)
ret['backend_config_files'] = env["TF_BACKEND_CONFIG_FILE"]
return ret
class controller(PluginFileInjector):
plugin_name = 'tower' # TODO: relying on routing for now, update after EEs pick up revised collection
base_injector = 'template'

View File

@@ -20,13 +20,14 @@ from django.core.exceptions import FieldDoesNotExist
# REST Framework
from rest_framework.exceptions import ParseError
from ansible_base.lib.utils.models import prevent_search
# AWX
from awx.api.versioning import reverse
from awx.main.constants import HOST_FACTS_FIELDS
from awx.main.models.base import (
BaseModel,
CreatedModifiedModel,
prevent_search,
accepts_json,
JOB_TYPE_CHOICES,
NEW_JOB_TYPE_CHOICES,

View File

@@ -9,7 +9,6 @@ import requests
# Django
from django.apps import apps
from django.conf import settings
from django.contrib.auth.models import User # noqa
from django.contrib.contenttypes.models import ContentType
from django.core.exceptions import ValidationError
from django.db import models
@@ -17,8 +16,9 @@ from django.db.models.query import QuerySet
from django.utils.crypto import get_random_string
from django.utils.translation import gettext_lazy as _
from ansible_base.lib.utils.models import prevent_search
# AWX
from awx.main.models.base import prevent_search
from awx.main.models.rbac import Role, RoleAncestorEntry
from awx.main.utils import parse_yaml_or_json, get_custom_venv_choices, get_licenser, polymorphic
from awx.main.utils.execution_environments import get_default_execution_environment
@@ -64,13 +64,12 @@ class ResourceMixin(models.Model):
@staticmethod
def _accessible_pk_qs(cls, accessor, role_field, content_types=None):
if type(accessor) == User:
if accessor._meta.model_name == 'user':
ancestor_roles = accessor.roles.all()
elif type(accessor) == Role:
ancestor_roles = [accessor]
else:
accessor_type = ContentType.objects.get_for_model(accessor)
ancestor_roles = Role.objects.filter(content_type__pk=accessor_type.id, object_id=accessor.id)
raise RuntimeError(f'Role filters only valid for users and ancestor role, received {accessor}')
if content_types is None:
ct_kwarg = dict(content_type_id=ContentType.objects.get_for_model(cls).id)
@@ -528,7 +527,6 @@ class CustomVirtualEnvMixin(models.Model):
class RelatedJobsMixin(object):
"""
This method is intended to be overwritten.
Called by get_active_jobs()
@@ -563,6 +561,7 @@ class WebhookTemplateMixin(models.Model):
SERVICES = [
('github', "GitHub"),
('gitlab', "GitLab"),
('bitbucket_dc', "BitBucket DataCenter"),
]
webhook_service = models.CharField(max_length=16, choices=SERVICES, blank=True, help_text=_('Service that webhook requests will be accepted from'))
@@ -623,6 +622,7 @@ class WebhookMixin(models.Model):
service_header = {
'github': ('Authorization', 'token {}'),
'gitlab': ('PRIVATE-TOKEN', '{}'),
'bitbucket_dc': ('Authorization', 'Bearer {}'),
}
service_statuses = {
'github': {
@@ -640,6 +640,14 @@ class WebhookMixin(models.Model):
'error': 'failed', # GitLab doesn't have an 'error' status distinct from 'failed' :(
'canceled': 'canceled',
},
'bitbucket_dc': {
'pending': 'INPROGRESS', # Bitbucket DC doesn't have any other statuses distinct from INPROGRESS, SUCCESSFUL, FAILED :(
'running': 'INPROGRESS',
'successful': 'SUCCESSFUL',
'failed': 'FAILED',
'error': 'FAILED',
'canceled': 'FAILED',
},
}
statuses = service_statuses[self.webhook_service]
@@ -648,11 +656,18 @@ class WebhookMixin(models.Model):
return
try:
license_type = get_licenser().validate().get('license_type')
data = {
'state': statuses[status],
'context': 'ansible/awx' if license_type == 'open' else 'ansible/tower',
'target_url': self.get_ui_url(),
}
if self.webhook_service == 'bitbucket_dc':
data = {
'state': statuses[status],
'key': 'ansible/awx' if license_type == 'open' else 'ansible/tower',
'url': self.get_ui_url(),
}
else:
data = {
'state': statuses[status],
'context': 'ansible/awx' if license_type == 'open' else 'ansible/tower',
'target_url': self.get_ui_url(),
}
k, v = service_header[self.webhook_service]
headers = {k: v.format(self.webhook_credential.get_input('token')), 'Content-Type': 'application/json'}
response = requests.post(status_api, data=json.dumps(data), headers=headers, timeout=30)

View File

@@ -5,6 +5,7 @@ from copy import deepcopy
import datetime
import logging
import json
import traceback
from django.db import models
from django.conf import settings
@@ -15,9 +16,11 @@ from django.utils.encoding import smart_str, force_str
from jinja2 import sandbox, ChainableUndefined
from jinja2.exceptions import TemplateSyntaxError, UndefinedError, SecurityError
from ansible_base.lib.utils.models import prevent_search
# AWX
from awx.api.versioning import reverse
from awx.main.models.base import CommonModelNameNotUnique, CreatedModifiedModel, prevent_search
from awx.main.models.base import CommonModelNameNotUnique, CreatedModifiedModel
from awx.main.utils import encrypt_field, decrypt_field, set_environ
from awx.main.notifications.email_backend import CustomEmailBackend
from awx.main.notifications.slack_backend import SlackBackend
@@ -482,14 +485,29 @@ class JobNotificationMixin(object):
if msg_template:
try:
msg = env.from_string(msg_template).render(**context)
except (TemplateSyntaxError, UndefinedError, SecurityError):
msg = ''
except (TemplateSyntaxError, UndefinedError, SecurityError) as e:
msg = '\r\n'.join([e.message, ''.join(traceback.format_exception(None, e, e.__traceback__).replace('\n', '\r\n'))])
if body_template:
try:
body = env.from_string(body_template).render(**context)
except (TemplateSyntaxError, UndefinedError, SecurityError):
body = ''
except (TemplateSyntaxError, UndefinedError, SecurityError) as e:
body = '\r\n'.join([e.message, ''.join(traceback.format_exception(None, e, e.__traceback__).replace('\n', '\r\n'))])
# https://datatracker.ietf.org/doc/html/rfc2822#section-2.2
# Body should have at least 2 CRLF, some clients will interpret
# the email incorrectly with blank body. So we will check that
if len(body.strip().splitlines()) <= 2:
# blank body
body = '\r\n'.join(
[
"The template rendering return a blank body.",
"Please check the template.",
"Refer to https://github.com/ansible/awx/issues/13983",
"for further information.",
]
)
return (msg, body)

View File

@@ -15,12 +15,10 @@ from django.utils.translation import gettext_lazy as _
# AWX
from awx.api.versioning import reverse
from django.contrib.auth.models import User # noqa
__all__ = [
'Role',
'batch_role_ancestor_rebuilding',
'get_roles_on_resource',
'ROLE_SINGLETON_SYSTEM_ADMINISTRATOR',
'ROLE_SINGLETON_SYSTEM_AUDITOR',
'role_summary_fields_generator',
@@ -170,16 +168,10 @@ class Role(models.Model):
return reverse('api:role_detail', kwargs={'pk': self.pk}, request=request)
def __contains__(self, accessor):
if type(accessor) == User:
if accessor._meta.model_name == 'user':
return self.ancestors.filter(members=accessor).exists()
elif accessor.__class__.__name__ == 'Team':
return self.ancestors.filter(pk=accessor.member_role.id).exists()
elif type(accessor) == Role:
return self.ancestors.filter(pk=accessor.pk).exists()
else:
accessor_type = ContentType.objects.get_for_model(accessor)
roles = Role.objects.filter(content_type__pk=accessor_type.id, object_id=accessor.id)
return self.ancestors.filter(pk__in=roles).exists()
raise RuntimeError(f'Role evaluations only valid for users, received {accessor}')
@property
def name(self):
@@ -460,31 +452,6 @@ class RoleAncestorEntry(models.Model):
object_id = models.PositiveIntegerField(null=False)
def get_roles_on_resource(resource, accessor):
"""
Returns a string list of the roles a accessor has for a given resource.
An accessor can be either a User, Role, or an arbitrary resource that
contains one or more Roles associated with it.
"""
if type(accessor) == User:
roles = accessor.roles.all()
elif type(accessor) == Role:
roles = [accessor]
else:
accessor_type = ContentType.objects.get_for_model(accessor)
roles = Role.objects.filter(content_type__pk=accessor_type.id, object_id=accessor.id)
return [
role_field
for role_field in RoleAncestorEntry.objects.filter(
ancestor__in=roles, content_type_id=ContentType.objects.get_for_model(resource).id, object_id=resource.id
)
.values_list('role_field', flat=True)
.distinct()
]
def role_summary_fields_generator(content_object, role_field):
global role_descriptions
global role_names

View File

@@ -0,0 +1,67 @@
from django.db import models
from django.core.validators import MinValueValidator, MaxValueValidator
from django.utils.translation import gettext_lazy as _
from awx.api.versioning import reverse
class Protocols(models.TextChoices):
TCP = 'tcp', 'TCP'
WS = 'ws', 'WS'
WSS = 'wss', 'WSS'
class ReceptorAddress(models.Model):
class Meta:
app_label = 'main'
constraints = [
models.UniqueConstraint(
fields=["address"],
name="unique_receptor_address",
violation_error_message=_("Receptor address must be unique."),
)
]
address = models.CharField(help_text=_("Routable address for this instance."), max_length=255)
port = models.IntegerField(help_text=_("Port for the address."), default=27199, validators=[MinValueValidator(0), MaxValueValidator(65535)])
websocket_path = models.CharField(help_text=_("Websocket path."), max_length=255, default="", blank=True)
protocol = models.CharField(
help_text=_("Protocol to use for the Receptor listener, 'tcp', 'wss', or 'ws'."), max_length=10, default=Protocols.TCP, choices=Protocols.choices
)
is_internal = models.BooleanField(help_text=_("If True, only routable within the Kubernetes cluster."), default=False)
canonical = models.BooleanField(help_text=_("If True, this address is the canonical address for the instance."), default=False)
peers_from_control_nodes = models.BooleanField(help_text=_("If True, control plane cluster nodes should automatically peer to it."), default=False)
instance = models.ForeignKey(
'Instance',
related_name='receptor_addresses',
on_delete=models.CASCADE,
null=False,
)
def __str__(self):
return self.get_full_address()
def get_full_address(self):
scheme = ""
path = ""
port = ""
if self.protocol == "ws":
scheme = "wss://"
if self.protocol == "ws" and self.websocket_path:
path = f"/{self.websocket_path}"
if self.port:
port = f":{self.port}"
return f"{scheme}{self.address}{port}{path}"
def get_peer_type(self):
if self.protocol == 'tcp':
return 'tcp-peer'
elif self.protocol in ['ws', 'wss']:
return 'ws-peer'
else:
return None
def get_absolute_url(self, request=None):
return reverse('api:receptor_address_detail', kwargs={'pk': self.pk}, request=request)

View File

@@ -30,8 +30,10 @@ from rest_framework.exceptions import ParseError
# Django-Polymorphic
from polymorphic.models import PolymorphicModel
from ansible_base.lib.utils.models import prevent_search, get_type_for_model
# AWX
from awx.main.models.base import CommonModelNameNotUnique, PasswordFieldsModel, NotificationFieldsModel, prevent_search
from awx.main.models.base import CommonModelNameNotUnique, PasswordFieldsModel, NotificationFieldsModel
from awx.main.dispatch import get_task_queuename
from awx.main.dispatch.control import Control as ControlDispatcher
from awx.main.registrar import activity_stream_registrar
@@ -42,7 +44,6 @@ from awx.main.utils.common import (
_inventory_updates,
copy_model_by_class,
copy_m2m_relationships,
get_type_for_model,
parse_yaml_or_json,
getattr_dne,
ScheduleDependencyManager,

View File

@@ -23,9 +23,11 @@ from crum import get_current_user
from jinja2 import sandbox
from jinja2.exceptions import TemplateSyntaxError, UndefinedError, SecurityError
from ansible_base.lib.utils.models import prevent_search
# AWX
from awx.api.versioning import reverse
from awx.main.models import prevent_search, accepts_json, UnifiedJobTemplate, UnifiedJob
from awx.main.models import accepts_json, UnifiedJobTemplate, UnifiedJob
from awx.main.models.notifications import NotificationTemplate, JobNotificationMixin
from awx.main.models.base import CreatedModifiedModel, VarsDictProperty
from awx.main.models.rbac import ROLE_SINGLETON_SYSTEM_ADMINISTRATOR, ROLE_SINGLETON_SYSTEM_AUDITOR

View File

@@ -1,5 +1,6 @@
# Copyright (c) 2019 Ansible, Inc.
# All Rights Reserved.
# -*-coding:utf-8-*-
class CustomNotificationBase(object):

View File

@@ -39,11 +39,15 @@ class TwilioBackend(AWXBaseEmailBackend, CustomNotificationBase):
logger.error(smart_str(_("Exception connecting to Twilio: {}").format(e)))
for m in messages:
try:
connection.messages.create(to=m.to, from_=m.from_email, body=m.subject)
sent_messages += 1
except Exception as e:
logger.error(smart_str(_("Exception sending messages: {}").format(e)))
if not self.fail_silently:
raise
failed = False
for dest in m.to:
try:
logger.debug(smart_str(_("FROM: {} / TO: {}").format(m.from_email, dest)))
connection.messages.create(to=dest, from_=m.from_email, body=m.subject)
sent_messages += 1
except Exception as e:
logger.error(smart_str(_("Exception sending messages: {}").format(e)))
failed = True
if not self.fail_silently and failed:
raise
return sent_messages

View File

@@ -4,13 +4,15 @@ import logging
from django.conf import settings
from django.urls import re_path
from channels.auth import AuthMiddlewareStack
from channels.routing import ProtocolTypeRouter, URLRouter
from ansible_base.lib.channels.middleware import DrfAuthMiddlewareStack
from . import consumers
logger = logging.getLogger('awx.main.routing')
_application = None
class AWXProtocolTypeRouter(ProtocolTypeRouter):
@@ -26,13 +28,91 @@ class AWXProtocolTypeRouter(ProtocolTypeRouter):
super().__init__(*args, **kwargs)
class MultipleURLRouterAdapter:
"""
Django channels doesn't nicely support Auth_1(urls_1), Auth_2(urls_2), ..., Auth_n(urls_n)
This class allows assocating a websocket url with an auth
Ordering matters. The first matching url will be used.
"""
def __init__(self, *auths):
self._auths = [a for a in auths]
async def __call__(self, scope, receive, send):
"""
Loop through the list of passed in URLRouter's (they may or may not be wrapped by auth).
We know we have exhausted the list of URLRouter patterns when we get a
ValueError('No route found for path %s'). When that happens, move onto the next
URLRouter.
If the final URLRouter raises an error, re-raise it in the end.
We know that we found a match when no error is raised, end the loop.
"""
last_index = len(self._auths) - 1
for i, auth in enumerate(self._auths):
try:
return await auth.__call__(scope, receive, send)
except ValueError as e:
if str(e).startswith('No route found for path'):
# Only surface the error if on the last URLRouter
if i == last_index:
raise
websocket_urlpatterns = [
re_path(r'api/websocket/$', consumers.EventConsumer.as_asgi()),
re_path(r'websocket/$', consumers.EventConsumer.as_asgi()),
]
websocket_relay_urlpatterns = [
re_path(r'websocket/relay/$', consumers.RelayConsumer.as_asgi()),
]
application = AWXProtocolTypeRouter(
{
'websocket': AuthMiddlewareStack(URLRouter(websocket_urlpatterns)),
}
)
def application_func(cls=AWXProtocolTypeRouter) -> ProtocolTypeRouter:
return cls(
{
'websocket': MultipleURLRouterAdapter(
URLRouter(websocket_relay_urlpatterns),
DrfAuthMiddlewareStack(URLRouter(websocket_urlpatterns)),
)
}
)
def __getattr__(name: str) -> ProtocolTypeRouter:
"""
Defer instantiating application.
For testing, we just need it to NOT run on import.
https://peps.python.org/pep-0562/#specification
Normally, someone would get application from this module via:
from awx.main.routing import application
and do something with the application:
application.do_something()
What does the callstack look like when the import runs?
...
awx.main.routing.__getattribute__(...) # <-- we don't define this so NOOP as far as we are concerned
if '__getattr__' in awx.main.routing.__dict__: # <-- this triggers the function we are in
return awx.main.routing.__dict__.__getattr__("application")
Why isn't this function simply implemented as:
def __getattr__(name):
if not _application:
_application = application_func()
return _application
It could. I manually tested it and it passes test_routing.py.
But my understanding after reading the PEP-0562 specification link above is that
performance would be a bit worse due to the extra __getattribute__ calls when
we reference non-global variables.
"""
if name == "application":
globs = globals()
if not globs['_application']:
globs['_application'] = application_func()
return globs['_application']
raise AttributeError(f"module {__name__!r} has no attribute {name!r}")

View File

@@ -17,6 +17,8 @@ from django.utils.timezone import now as tz_now
from django.conf import settings
from django.contrib.contenttypes.models import ContentType
from ansible_base.lib.utils.models import get_type_for_model
# AWX
from awx.main.dispatch.reaper import reap_job
from awx.main.models import (
@@ -34,7 +36,6 @@ from awx.main.models import (
from awx.main.scheduler.dag_workflow import WorkflowDAG
from awx.main.utils.pglock import advisory_lock
from awx.main.utils import (
get_type_for_model,
ScheduleTaskManager,
ScheduleWorkflowManager,
)
@@ -67,7 +68,7 @@ class TaskBase:
# initialize each metric to 0 and force metric_has_changed to true. This
# ensures each task manager metric will be overridden when pipe_execute
# is called later.
self.subsystem_metrics = s_metrics.Metrics(auto_pipe_execute=False)
self.subsystem_metrics = s_metrics.DispatcherMetrics(auto_pipe_execute=False)
self.start_time = time.time()
# We want to avoid calling settings in loops, so cache these settings at init time
@@ -104,7 +105,7 @@ class TaskBase:
try:
# increment task_manager_schedule_calls regardless if the other
# metrics are recorded
s_metrics.Metrics(auto_pipe_execute=True).inc(f"{self.prefix}__schedule_calls", 1)
s_metrics.DispatcherMetrics(auto_pipe_execute=True).inc(f"{self.prefix}__schedule_calls", 1)
# Only record metrics if the last time recording was more
# than SUBSYSTEM_METRICS_TASK_MANAGER_RECORD_INTERVAL ago.
# Prevents a short-duration task manager that runs directly after a

View File

@@ -29,7 +29,7 @@ class RunnerCallback:
self.safe_env = {}
self.event_ct = 0
self.model = model
self.update_attempts = int(settings.DISPATCHER_DB_DOWNTIME_TOLERANCE / 5)
self.update_attempts = int(getattr(settings, 'DISPATCHER_DB_DOWNTOWN_TOLLERANCE', settings.DISPATCHER_DB_DOWNTIME_TOLERANCE) / 5)
self.wrapup_event_dispatched = False
self.artifacts_processed = False
self.extra_update_fields = {}
@@ -95,17 +95,17 @@ class RunnerCallback:
if self.parent_workflow_job_id:
event_data['workflow_job_id'] = self.parent_workflow_job_id
event_data['job_created'] = self.job_created
if self.host_map:
host = event_data.get('event_data', {}).get('host', '').strip()
if host:
event_data['host_name'] = host
if host in self.host_map:
event_data['host_id'] = self.host_map[host]
else:
event_data['host_name'] = ''
event_data['host_id'] = ''
if event_data.get('event') == 'playbook_on_stats':
event_data['host_map'] = self.host_map
host = event_data.get('event_data', {}).get('host', '').strip()
if host:
event_data['host_name'] = host
if host in self.host_map:
event_data['host_id'] = self.host_map[host]
else:
event_data['host_name'] = ''
event_data['host_id'] = ''
if event_data.get('event') == 'playbook_on_stats':
event_data['host_map'] = self.host_map
if isinstance(self, RunnerCallbackForProjectUpdate):
# need a better way to have this check.

View File

@@ -114,7 +114,7 @@ class BaseTask(object):
def __init__(self):
self.cleanup_paths = []
self.update_attempts = int(settings.DISPATCHER_DB_DOWNTIME_TOLERANCE / 5)
self.update_attempts = int(getattr(settings, 'DISPATCHER_DB_DOWNTOWN_TOLLERANCE', settings.DISPATCHER_DB_DOWNTIME_TOLERANCE) / 5)
self.runner_callback = self.callback_class(model=self.model)
def update_model(self, pk, _attempt=0, **updates):

View File

@@ -27,7 +27,7 @@ from awx.main.utils.common import (
)
from awx.main.constants import MAX_ISOLATED_PATH_COLON_DELIMITER
from awx.main.tasks.signals import signal_state, signal_callback, SignalExit
from awx.main.models import Instance, InstanceLink, UnifiedJob
from awx.main.models import Instance, InstanceLink, UnifiedJob, ReceptorAddress
from awx.main.dispatch import get_task_queuename
from awx.main.dispatch.publish import task
from awx.main.utils.pglock import advisory_lock
@@ -676,36 +676,44 @@ RECEPTOR_CONFIG_STARTER = (
)
def should_update_config(instances):
def should_update_config(new_config):
'''
checks that the list of instances matches the list of
tcp-peers in the config
'''
current_config = read_receptor_config() # this gets receptor conf lock
current_peers = []
for config_entry in current_config:
for key, value in config_entry.items():
if key.endswith('-peer'):
current_peers.append(value['address'])
intended_peers = [f"{i.hostname}:{i.listener_port}" for i in instances]
logger.debug(f"Peers current {current_peers} intended {intended_peers}")
if set(current_peers) == set(intended_peers):
return False # config file is already update to date
return True
current_config = read_receptor_config() # this gets receptor conf lock
for config_entry in current_config:
if config_entry not in new_config:
logger.warning(f"{config_entry} should not be in receptor config. Updating.")
return True
for config_entry in new_config:
if config_entry not in current_config:
logger.warning(f"{config_entry} missing from receptor config. Updating.")
return True
return False
def generate_config_data():
# returns two values
# receptor config - based on current database peers
# should_update - If True, receptor_config differs from the receptor conf file on disk
instances = Instance.objects.filter(node_type__in=(Instance.Types.EXECUTION, Instance.Types.HOP), peers_from_control_nodes=True)
addresses = ReceptorAddress.objects.filter(peers_from_control_nodes=True)
receptor_config = list(RECEPTOR_CONFIG_STARTER)
for instance in instances:
peer = {'tcp-peer': {'address': f'{instance.hostname}:{instance.listener_port}', 'tls': 'tlsclient'}}
receptor_config.append(peer)
should_update = should_update_config(instances)
for address in addresses:
if address.get_peer_type():
peer = {
f'{address.get_peer_type()}': {
'address': f'{address.get_full_address()}',
'tls': 'tlsclient',
}
}
receptor_config.append(peer)
else:
logger.warning(f"Receptor address {address} has unsupported peer type, skipping.")
should_update = should_update_config(receptor_config)
return receptor_config, should_update
@@ -747,14 +755,13 @@ def write_receptor_config():
with lock:
with open(__RECEPTOR_CONF, 'w') as file:
yaml.dump(receptor_config, file, default_flow_style=False)
reload_receptor()
@task(queue=get_task_queuename)
def remove_deprovisioned_node(hostname):
InstanceLink.objects.filter(source__hostname=hostname).update(link_state=InstanceLink.States.REMOVING)
InstanceLink.objects.filter(target__hostname=hostname).update(link_state=InstanceLink.States.REMOVING)
InstanceLink.objects.filter(target__instance__hostname=hostname).update(link_state=InstanceLink.States.REMOVING)
node_jobs = UnifiedJob.objects.filter(
execution_node=hostname,

View File

@@ -6,6 +6,7 @@ import itertools
import json
import logging
import os
import psycopg
from io import StringIO
from contextlib import redirect_stdout
import shutil
@@ -62,7 +63,7 @@ from awx.main.tasks.receptor import get_receptor_ctl, worker_info, worker_cleanu
from awx.main.consumers import emit_channel_notification
from awx.main import analytics
from awx.conf import settings_registry
from awx.main.analytics.subsystem_metrics import Metrics
from awx.main.analytics.subsystem_metrics import DispatcherMetrics
from rest_framework.exceptions import PermissionDenied
@@ -113,7 +114,7 @@ def dispatch_startup():
cluster_node_heartbeat()
reaper.startup_reaping()
reaper.reap_waiting(grace_period=0)
m = Metrics()
m = DispatcherMetrics()
m.reset_values()
@@ -416,7 +417,7 @@ def handle_removed_image(remove_images=None):
@task(queue=get_task_queuename)
def cleanup_images_and_files():
_cleanup_images_and_files()
_cleanup_images_and_files(image_prune=True)
@task(queue=get_task_queuename)
@@ -495,7 +496,7 @@ def inspect_established_receptor_connections(mesh_status):
update_links = []
for link in all_links:
if link.link_state != InstanceLink.States.REMOVING:
if link.target.hostname in active_receptor_conns.get(link.source.hostname, {}):
if link.target.instance.hostname in active_receptor_conns.get(link.source.hostname, {}):
if link.link_state is not InstanceLink.States.ESTABLISHED:
link.link_state = InstanceLink.States.ESTABLISHED
update_links.append(link)
@@ -630,10 +631,18 @@ def cluster_node_heartbeat(dispatch_time=None, worker_tasks=None):
logger.error("Host {} last checked in at {}, marked as lost.".format(other_inst.hostname, other_inst.last_seen))
except DatabaseError as e:
if 'did not affect any rows' in str(e):
logger.debug('Another instance has marked {} as lost'.format(other_inst.hostname))
cause = e.__cause__
if cause and hasattr(cause, 'sqlstate'):
sqlstate = cause.sqlstate
sqlstate_str = psycopg.errors.lookup(sqlstate)
logger.debug('SQL Error state: {} - {}'.format(sqlstate, sqlstate_str))
if sqlstate == psycopg.errors.NoData:
logger.debug('Another instance has marked {} as lost'.format(other_inst.hostname))
else:
logger.exception("Error marking {} as lost.".format(other_inst.hostname))
else:
logger.exception('Error marking {} as lost'.format(other_inst.hostname))
logger.exception('No SQL state available. Error marking {} as lost'.format(other_inst.hostname))
# Run local reaper
if worker_tasks is not None:
@@ -788,10 +797,19 @@ def update_inventory_computed_fields(inventory_id):
try:
i.update_computed_fields()
except DatabaseError as e:
if 'did not affect any rows' in str(e):
logger.debug('Exiting duplicate update_inventory_computed_fields task.')
return
raise
# https://github.com/django/django/blob/eff21d8e7a1cb297aedf1c702668b590a1b618f3/django/db/models/base.py#L1105
# django raises DatabaseError("Forced update did not affect any rows.")
# if sqlstate is set then there was a database error and otherwise will re-raise that error
cause = e.__cause__
if cause and hasattr(cause, 'sqlstate'):
sqlstate = cause.sqlstate
sqlstate_str = psycopg.errors.lookup(sqlstate)
logger.error('SQL Error state: {} - {}'.format(sqlstate, sqlstate_str))
raise
# otherwise
logger.debug('Exiting duplicate update_inventory_computed_fields task.')
def update_smart_memberships_for_inventory(smart_inventory):

View File

@@ -0,0 +1,3 @@
{
"TF_BACKEND_CONFIG_FILE": "{{ file_reference }}"
}

View File

@@ -1,13 +1,8 @@
from awx.main.tests.functional.conftest import * # noqa
import os
import pytest
def pytest_addoption(parser):
parser.addoption("--release", action="store", help="a release version number, e.g., 3.3.0")
def pytest_generate_tests(metafunc):
# This is called for every test. Only get/set command line arguments
# if the argument is specified in the list of test "fixturenames".
option_value = metafunc.config.option.release
if 'release' in metafunc.fixturenames and option_value is not None:
metafunc.parametrize("release", [option_value])
@pytest.fixture()
def release():
return os.environ.get('VERSION_TARGET', '')

View File

@@ -1,19 +1,16 @@
import pytest
import yaml
import itertools
from unittest import mock
from django.db.utils import IntegrityError
from awx.api.versioning import reverse
from awx.main.models import Instance
from awx.main.models import Instance, ReceptorAddress
from awx.api.views.instance_install_bundle import generate_group_vars_all_yml
def has_peer(group_vars, peer):
peers = group_vars.get('receptor_peers', [])
for p in peers:
if f"{p['host']}:{p['port']}" == peer:
if p['address'] == peer:
return True
return False
@@ -24,119 +21,314 @@ class TestPeers:
def configure_settings(self, settings):
settings.IS_K8S = True
@pytest.mark.parametrize('node_type', ['control', 'hybrid'])
def test_prevent_peering_to_self(self, node_type):
@pytest.mark.parametrize('node_type', ['hop', 'execution'])
def test_peering_to_self(self, node_type, admin_user, patch):
"""
cannot peer to self
"""
control_instance = Instance.objects.create(hostname='abc', node_type=node_type)
with pytest.raises(IntegrityError):
control_instance.peers.add(control_instance)
instance = Instance.objects.create(hostname='abc', node_type=node_type)
addr = ReceptorAddress.objects.create(instance=instance, address='abc', canonical=True)
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': instance.pk}),
data={"hostname": "abc", "node_type": node_type, "peers": [addr.id]},
user=admin_user,
expect=400,
)
assert 'Instance cannot peer to its own address.' in str(resp.data)
@pytest.mark.parametrize('node_type', ['control', 'hybrid', 'hop', 'execution'])
def test_creating_node(self, node_type, admin_user, post):
"""
can only add hop and execution nodes via API
"""
post(
resp = post(
url=reverse('api:instance_list'),
data={"hostname": "abc", "node_type": node_type},
user=admin_user,
expect=400 if node_type in ['control', 'hybrid'] else 201,
)
if resp.status_code == 400:
assert 'Can only create execution or hop nodes.' in str(resp.data)
def test_changing_node_type(self, admin_user, patch):
"""
cannot change node type
"""
hop = Instance.objects.create(hostname='abc', node_type="hop")
patch(
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': hop.pk}),
data={"node_type": "execution"},
user=admin_user,
expect=400,
)
assert 'Cannot change node type.' in str(resp.data)
@pytest.mark.parametrize('node_type', ['hop', 'execution'])
def test_listener_port_null(self, node_type, admin_user, post):
"""
listener_port can be None
"""
post(
url=reverse('api:instance_list'),
data={"hostname": "abc", "node_type": node_type, "listener_port": None},
@pytest.mark.parametrize(
'payload_port, payload_peers_from, initial_port, initial_peers_from',
[
(-1, -1, None, None),
(-1, -1, 27199, False),
(-1, -1, 27199, True),
(None, -1, None, None),
(None, False, None, None),
(-1, False, None, None),
(27199, True, 27199, True),
(27199, False, 27199, False),
(27199, -1, 27199, True),
(27199, -1, 27199, False),
(-1, True, 27199, True),
(-1, False, 27199, False),
],
)
def test_no_op(self, payload_port, payload_peers_from, initial_port, initial_peers_from, admin_user, patch):
node = Instance.objects.create(hostname='abc', node_type='hop')
if initial_port is not None:
ReceptorAddress.objects.create(address=node.hostname, port=initial_port, canonical=True, peers_from_control_nodes=initial_peers_from, instance=node)
assert ReceptorAddress.objects.filter(instance=node).count() == 1
else:
assert ReceptorAddress.objects.filter(instance=node).count() == 0
data = {'enabled': True} # Just to have something to post.
if payload_port != -1:
data['listener_port'] = payload_port
if payload_peers_from != -1:
data['peers_from_control_nodes'] = payload_peers_from
patch(
url=reverse('api:instance_detail', kwargs={'pk': node.pk}),
data=data,
user=admin_user,
expect=201,
expect=200,
)
@pytest.mark.parametrize('node_type, allowed', [('control', False), ('hybrid', False), ('hop', True), ('execution', True)])
def test_peers_from_control_nodes_allowed(self, node_type, allowed, post, admin_user):
"""
only hop and execution nodes can have peers_from_control_nodes set to True
"""
post(
url=reverse('api:instance_list'),
data={"hostname": "abc", "peers_from_control_nodes": True, "node_type": node_type, "listener_port": 6789},
assert ReceptorAddress.objects.filter(instance=node).count() == (0 if initial_port is None else 1)
if initial_port is not None:
ra = ReceptorAddress.objects.get(instance=node, canonical=True)
assert ra.port == initial_port
assert ra.peers_from_control_nodes == initial_peers_from
@pytest.mark.parametrize(
'payload_port, payload_peers_from',
[
(27199, True),
(27199, False),
(27199, -1),
],
)
def test_creates_canonical_address(self, payload_port, payload_peers_from, admin_user, patch):
node = Instance.objects.create(hostname='abc', node_type='hop')
assert ReceptorAddress.objects.filter(instance=node).count() == 0
data = {'enabled': True} # Just to have something to post.
if payload_port != -1:
data['listener_port'] = payload_port
if payload_peers_from != -1:
data['peers_from_control_nodes'] = payload_peers_from
patch(
url=reverse('api:instance_detail', kwargs={'pk': node.pk}),
data=data,
user=admin_user,
expect=201 if allowed else 400,
expect=200,
)
def test_listener_port_is_required(self, admin_user, post):
"""
if adding instance to peers list, that instance must have listener_port set
"""
Instance.objects.create(hostname='abc', node_type="hop", listener_port=None)
post(
url=reverse('api:instance_list'),
data={"hostname": "ex", "peers_from_control_nodes": False, "node_type": "execution", "listener_port": None, "peers": ["abc"]},
assert ReceptorAddress.objects.filter(instance=node).count() == 1
ra = ReceptorAddress.objects.get(instance=node, canonical=True)
assert ra.port == payload_port
assert ra.peers_from_control_nodes == (payload_peers_from if payload_peers_from != -1 else False)
@pytest.mark.parametrize(
'payload_port, payload_peers_from, initial_port, initial_peers_from',
[
(None, False, 27199, True),
(None, -1, 27199, True),
(None, False, 27199, False),
(None, -1, 27199, False),
],
)
def test_deletes_canonical_address(self, payload_port, payload_peers_from, initial_port, initial_peers_from, admin_user, patch):
node = Instance.objects.create(hostname='abc', node_type='hop')
ReceptorAddress.objects.create(address=node.hostname, port=initial_port, canonical=True, peers_from_control_nodes=initial_peers_from, instance=node)
assert ReceptorAddress.objects.filter(instance=node).count() == 1
data = {'enabled': True} # Just to have something to post.
if payload_port != -1:
data['listener_port'] = payload_port
if payload_peers_from != -1:
data['peers_from_control_nodes'] = payload_peers_from
patch(
url=reverse('api:instance_detail', kwargs={'pk': node.pk}),
data=data,
user=admin_user,
expect=200,
)
assert ReceptorAddress.objects.filter(instance=node).count() == 0
@pytest.mark.parametrize(
'payload_port, payload_peers_from, initial_port, initial_peers_from',
[
(27199, True, 27199, False),
(27199, False, 27199, True),
(-1, True, 27199, False),
(-1, False, 27199, True),
],
)
def test_updates_canonical_address(self, payload_port, payload_peers_from, initial_port, initial_peers_from, admin_user, patch):
node = Instance.objects.create(hostname='abc', node_type='hop')
ReceptorAddress.objects.create(address=node.hostname, port=initial_port, canonical=True, peers_from_control_nodes=initial_peers_from, instance=node)
assert ReceptorAddress.objects.filter(instance=node).count() == 1
data = {'enabled': True} # Just to have something to post.
if payload_port != -1:
data['listener_port'] = payload_port
if payload_peers_from != -1:
data['peers_from_control_nodes'] = payload_peers_from
patch(
url=reverse('api:instance_detail', kwargs={'pk': node.pk}),
data=data,
user=admin_user,
expect=200,
)
assert ReceptorAddress.objects.filter(instance=node).count() == 1
ra = ReceptorAddress.objects.get(instance=node, canonical=True)
assert ra.port == initial_port # At the present time, changing ports is not allowed
assert ra.peers_from_control_nodes == payload_peers_from
@pytest.mark.parametrize(
'payload_port, payload_peers_from, initial_port, initial_peers_from, error_msg',
[
(-1, True, None, None, "Cannot enable peers_from_control_nodes"),
(None, True, None, None, "Cannot enable peers_from_control_nodes"),
(None, True, 21799, True, "Cannot enable peers_from_control_nodes"),
(None, True, 21799, False, "Cannot enable peers_from_control_nodes"),
(21800, -1, 21799, True, "Cannot change listener port"),
(21800, True, 21799, True, "Cannot change listener port"),
(21800, False, 21799, True, "Cannot change listener port"),
(21800, -1, 21799, False, "Cannot change listener port"),
(21800, True, 21799, False, "Cannot change listener port"),
(21800, False, 21799, False, "Cannot change listener port"),
],
)
def test_canonical_address_validation_error(self, payload_port, payload_peers_from, initial_port, initial_peers_from, error_msg, admin_user, patch):
node = Instance.objects.create(hostname='abc', node_type='hop')
if initial_port is not None:
ReceptorAddress.objects.create(address=node.hostname, port=initial_port, canonical=True, peers_from_control_nodes=initial_peers_from, instance=node)
assert ReceptorAddress.objects.filter(instance=node).count() == 1
else:
assert ReceptorAddress.objects.filter(instance=node).count() == 0
data = {'enabled': True} # Just to have something to post.
if payload_port != -1:
data['listener_port'] = payload_port
if payload_peers_from != -1:
data['peers_from_control_nodes'] = payload_peers_from
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': node.pk}),
data=data,
user=admin_user,
expect=400,
)
def test_peers_from_control_nodes_listener_port_enabled(self, admin_user, post):
assert error_msg in str(resp.data)
def test_changing_managed_listener_port(self, admin_user, patch):
"""
if peers_from_control_nodes is True, listener_port must an integer
Assert that all other combinations are allowed
if instance is managed, cannot change listener port at all
"""
for index, item in enumerate(itertools.product(['hop', 'execution'], [True, False], [None, 6789])):
node_type, peers_from, listener_port = item
# only disallowed case is when peers_from is True and listener port is None
disallowed = peers_from and not listener_port
post(
url=reverse('api:instance_list'),
data={"hostname": f"abc{index}", "peers_from_control_nodes": peers_from, "node_type": node_type, "listener_port": listener_port},
user=admin_user,
expect=400 if disallowed else 201,
)
hop = Instance.objects.create(hostname='abc', node_type="hop", managed=True)
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': hop.pk}),
data={"listener_port": 5678},
user=admin_user,
expect=400, # cannot set port
)
assert 'Cannot change listener port for managed nodes.' in str(resp.data)
ReceptorAddress.objects.create(instance=hop, address='hop', port=27199, canonical=True)
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': hop.pk}),
data={"listener_port": None},
user=admin_user,
expect=400, # cannot unset port
)
assert 'Cannot change listener port for managed nodes.' in str(resp.data)
def test_bidirectional_peering(self, admin_user, patch):
"""
cannot peer to node that is already to peered to it
if A -> B, then disallow B -> A
"""
hop1 = Instance.objects.create(hostname='hop1', node_type='hop')
hop1addr = ReceptorAddress.objects.create(instance=hop1, address='hop1', canonical=True)
hop2 = Instance.objects.create(hostname='hop2', node_type='hop')
hop2addr = ReceptorAddress.objects.create(instance=hop2, address='hop2', canonical=True)
hop1.peers.add(hop2addr)
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': hop2.pk}),
data={"peers": [hop1addr.id]},
user=admin_user,
expect=400,
)
assert 'Instance hop1 is already peered to this instance.' in str(resp.data)
def test_multiple_peers_same_instance(self, admin_user, patch):
"""
cannot peer to more than one address of the same instance
"""
hop1 = Instance.objects.create(hostname='hop1', node_type='hop')
hop1addr1 = ReceptorAddress.objects.create(instance=hop1, address='hop1', canonical=True)
hop1addr2 = ReceptorAddress.objects.create(instance=hop1, address='hop1alternate')
hop2 = Instance.objects.create(hostname='hop2', node_type='hop')
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': hop2.pk}),
data={"peers": [hop1addr1.id, hop1addr2.id]},
user=admin_user,
expect=400,
)
assert 'Cannot peer to the same instance more than once.' in str(resp.data)
@pytest.mark.parametrize('node_type', ['control', 'hybrid'])
def test_disallow_modifying_peers_control_nodes(self, node_type, admin_user, patch):
def test_changing_peers_control_nodes(self, node_type, admin_user, patch):
"""
for control nodes, peers field should not be
modified directly via patch.
"""
control = Instance.objects.create(hostname='abc', node_type=node_type)
hop1 = Instance.objects.create(hostname='hop1', node_type='hop', peers_from_control_nodes=True, listener_port=6789)
hop2 = Instance.objects.create(hostname='hop2', node_type='hop', peers_from_control_nodes=False, listener_port=6789)
assert [hop1] == list(control.peers.all()) # only hop1 should be peered
patch(
control = Instance.objects.create(hostname='abc', node_type=node_type, managed=True)
hop1 = Instance.objects.create(hostname='hop1', node_type='hop')
hop1addr = ReceptorAddress.objects.create(instance=hop1, address='hop1', peers_from_control_nodes=True, canonical=True)
hop2 = Instance.objects.create(hostname='hop2', node_type='hop')
hop2addr = ReceptorAddress.objects.create(instance=hop2, address='hop2', canonical=True)
assert [hop1addr] == list(control.peers.all()) # only hop1addr should be peered
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': control.pk}),
data={"peers": ["hop2"]},
data={"peers": [hop2addr.id]},
user=admin_user,
expect=400, # cannot add peers directly
expect=400, # cannot add peers manually
)
assert 'Setting peers manually for managed nodes is not allowed.' in str(resp.data)
patch(
url=reverse('api:instance_detail', kwargs={'pk': control.pk}),
data={"peers": ["hop1"]},
data={"peers": [hop1addr.id]},
user=admin_user,
expect=200, # patching with current peers list should be okay
)
patch(
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': control.pk}),
data={"peers": []},
user=admin_user,
expect=400, # cannot remove peers directly
)
assert 'Setting peers manually for managed nodes is not allowed.' in str(resp.data)
patch(
url=reverse('api:instance_detail', kwargs={'pk': control.pk}),
data={},
@@ -148,23 +340,25 @@ class TestPeers:
url=reverse('api:instance_detail', kwargs={'pk': hop2.pk}),
data={"peers_from_control_nodes": True},
user=admin_user,
expect=200, # patching without data should be fine too
expect=200,
)
assert {hop1, hop2} == set(control.peers.all()) # hop1 and hop2 should now be peered from control node
assert {hop1addr, hop2addr} == set(control.peers.all()) # hop1 and hop2 should now be peered from control node
def test_disallow_changing_hostname(self, admin_user, patch):
def test_changing_hostname(self, admin_user, patch):
"""
cannot change hostname
"""
hop = Instance.objects.create(hostname='hop', node_type='hop')
patch(
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': hop.pk}),
data={"hostname": "hop2"},
user=admin_user,
expect=400,
)
def test_disallow_changing_node_state(self, admin_user, patch):
assert 'Cannot change hostname.' in str(resp.data)
def test_changing_node_state(self, admin_user, patch):
"""
only allow setting to deprovisioning
"""
@@ -175,12 +369,54 @@ class TestPeers:
user=admin_user,
expect=200,
)
patch(
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': hop.pk}),
data={"node_state": "ready"},
user=admin_user,
expect=400,
)
assert "Can only change instances to the 'deprovisioning' state." in str(resp.data)
def test_changing_managed_node_state(self, admin_user, patch):
"""
cannot change node state of managed node
"""
hop = Instance.objects.create(hostname='hop', node_type='hop', managed=True)
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': hop.pk}),
data={"node_state": "deprovisioning"},
user=admin_user,
expect=400,
)
assert 'Cannot deprovision managed nodes.' in str(resp.data)
def test_changing_managed_peers_from_control_nodes(self, admin_user, patch):
"""
cannot change peers_from_control_nodes of managed node
"""
hop = Instance.objects.create(hostname='hop', node_type='hop', managed=True)
ReceptorAddress.objects.create(instance=hop, address='hop', peers_from_control_nodes=True, canonical=True)
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': hop.pk}),
data={"peers_from_control_nodes": False},
user=admin_user,
expect=400,
)
assert 'Cannot change peers_from_control_nodes for managed nodes.' in str(resp.data)
hop.peers_from_control_nodes = False
hop.save()
resp = patch(
url=reverse('api:instance_detail', kwargs={'pk': hop.pk}),
data={"peers_from_control_nodes": False},
user=admin_user,
expect=400,
)
assert 'Cannot change peers_from_control_nodes for managed nodes.' in str(resp.data)
@pytest.mark.parametrize('node_type', ['control', 'hybrid'])
def test_control_node_automatically_peers(self, node_type):
@@ -191,9 +427,10 @@ class TestPeers:
peer to hop should be removed if hop is deleted
"""
hop = Instance.objects.create(hostname='hop', node_type='hop', peers_from_control_nodes=True, listener_port=6789)
hop = Instance.objects.create(hostname='hop', node_type='hop')
hopaddr = ReceptorAddress.objects.create(instance=hop, address='hop', peers_from_control_nodes=True, canonical=True)
control = Instance.objects.create(hostname='abc', node_type=node_type)
assert hop in control.peers.all()
assert hopaddr in control.peers.all()
hop.delete()
assert not control.peers.exists()
@@ -203,26 +440,50 @@ class TestPeers:
if a new node comes online, other peer relationships should
remain intact
"""
hop1 = Instance.objects.create(hostname='hop1', node_type='hop', listener_port=6789, peers_from_control_nodes=True)
hop2 = Instance.objects.create(hostname='hop2', node_type='hop', listener_port=6789, peers_from_control_nodes=False)
hop1.peers.add(hop2)
hop1 = Instance.objects.create(hostname='hop1', node_type='hop')
hop2 = Instance.objects.create(hostname='hop2', node_type='hop')
hop2addr = ReceptorAddress.objects.create(instance=hop2, address='hop2', canonical=True)
hop1.peers.add(hop2addr)
# a control node is added
Instance.objects.create(hostname='control', node_type=node_type, listener_port=None)
Instance.objects.create(hostname='control', node_type=node_type)
assert hop1.peers.exists()
def test_group_vars(self, get, admin_user):
def test_reverse_peers(self, admin_user, get):
"""
if hop1 peers to hop2, hop1 should
be in hop2's reverse_peers list
"""
hop1 = Instance.objects.create(hostname='hop1', node_type='hop')
hop2 = Instance.objects.create(hostname='hop2', node_type='hop')
hop2addr = ReceptorAddress.objects.create(instance=hop2, address='hop2', canonical=True)
hop1.peers.add(hop2addr)
resp = get(
url=reverse('api:instance_detail', kwargs={'pk': hop2.pk}),
user=admin_user,
expect=200,
)
assert hop1.pk in resp.data['reverse_peers']
def test_group_vars(self):
"""
control > hop1 > hop2 < execution
"""
control = Instance.objects.create(hostname='control', node_type='control', listener_port=None)
hop1 = Instance.objects.create(hostname='hop1', node_type='hop', listener_port=6789, peers_from_control_nodes=True)
hop2 = Instance.objects.create(hostname='hop2', node_type='hop', listener_port=6789, peers_from_control_nodes=False)
execution = Instance.objects.create(hostname='execution', node_type='execution', listener_port=6789)
control = Instance.objects.create(hostname='control', node_type='control')
hop1 = Instance.objects.create(hostname='hop1', node_type='hop')
ReceptorAddress.objects.create(instance=hop1, address='hop1', peers_from_control_nodes=True, port=6789, canonical=True)
execution.peers.add(hop2)
hop1.peers.add(hop2)
hop2 = Instance.objects.create(hostname='hop2', node_type='hop')
hop2addr = ReceptorAddress.objects.create(instance=hop2, address='hop2', peers_from_control_nodes=False, port=6789, canonical=True)
execution = Instance.objects.create(hostname='execution', node_type='execution')
ReceptorAddress.objects.create(instance=execution, address='execution', peers_from_control_nodes=False, port=6789, canonical=True)
execution.peers.add(hop2addr)
hop1.peers.add(hop2addr)
control_vars = yaml.safe_load(generate_group_vars_all_yml(control))
hop1_vars = yaml.safe_load(generate_group_vars_all_yml(hop1))
@@ -265,13 +526,15 @@ class TestPeers:
control = Instance.objects.create(hostname='control1', node_type='control')
write_method.assert_not_called()
# new hop node with peers_from_control_nodes False (no)
hop1 = Instance.objects.create(hostname='hop1', node_type='hop', listener_port=6789, peers_from_control_nodes=False)
# new address with peers_from_control_nodes False (no)
hop1 = Instance.objects.create(hostname='hop1', node_type='hop')
hop1addr = ReceptorAddress.objects.create(instance=hop1, address='hop1', peers_from_control_nodes=False, canonical=True)
hop1.delete()
write_method.assert_not_called()
# new hop node with peers_from_control_nodes True (yes)
hop1 = Instance.objects.create(hostname='hop1', node_type='hop', listener_port=6789, peers_from_control_nodes=True)
# new address with peers_from_control_nodes True (yes)
hop1 = Instance.objects.create(hostname='hop1', node_type='hop')
hop1addr = ReceptorAddress.objects.create(instance=hop1, address='hop1', peers_from_control_nodes=True, canonical=True)
write_method.assert_called()
write_method.reset_mock()
@@ -280,20 +543,21 @@ class TestPeers:
write_method.assert_called()
write_method.reset_mock()
# new hop node with peers_from_control_nodes False and peered to another hop node (no)
hop2 = Instance.objects.create(hostname='hop2', node_type='hop', listener_port=6789, peers_from_control_nodes=False)
hop2.peers.add(hop1)
# new address with peers_from_control_nodes False and peered to another hop node (no)
hop2 = Instance.objects.create(hostname='hop2', node_type='hop')
ReceptorAddress.objects.create(instance=hop2, address='hop2', peers_from_control_nodes=False, canonical=True)
hop2.peers.add(hop1addr)
hop2.delete()
write_method.assert_not_called()
# changing peers_from_control_nodes to False (yes)
hop1.peers_from_control_nodes = False
hop1.save()
hop1addr.peers_from_control_nodes = False
hop1addr.save()
write_method.assert_called()
write_method.reset_mock()
# deleting hop node that has peers_from_control_nodes to False (no)
hop1.delete()
# deleting address that has peers_from_control_nodes to False (no)
hop1.delete() # cascade deletes to hop1addr
write_method.assert_not_called()
# deleting control nodes (no)
@@ -315,8 +579,8 @@ class TestPeers:
# not peered, so config file should not be updated
for i in range(3):
Instance.objects.create(hostname=f"exNo-{i}", node_type='execution', listener_port=6789, peers_from_control_nodes=False)
inst = Instance.objects.create(hostname=f"exNo-{i}", node_type='execution')
ReceptorAddress.objects.create(instance=inst, address=f"exNo-{i}", port=6789, peers_from_control_nodes=False, canonical=True)
_, should_update = generate_config_data()
assert not should_update
@@ -324,11 +588,13 @@ class TestPeers:
expected_peers = []
for i in range(3):
expected_peers.append(f"hop-{i}:6789")
Instance.objects.create(hostname=f"hop-{i}", node_type='hop', listener_port=6789, peers_from_control_nodes=True)
inst = Instance.objects.create(hostname=f"hop-{i}", node_type='hop')
ReceptorAddress.objects.create(instance=inst, address=f"hop-{i}", port=6789, peers_from_control_nodes=True, canonical=True)
for i in range(3):
expected_peers.append(f"exYes-{i}:6789")
Instance.objects.create(hostname=f"exYes-{i}", node_type='execution', listener_port=6789, peers_from_control_nodes=True)
inst = Instance.objects.create(hostname=f"exYes-{i}", node_type='execution')
ReceptorAddress.objects.create(instance=inst, address=f"exYes-{i}", port=6789, peers_from_control_nodes=True, canonical=True)
new_config, should_update = generate_config_data()
assert should_update

View File

@@ -2,12 +2,12 @@ from unittest import mock
import pytest
import json
from ansible_base.lib.utils.models import get_type_for_model
from awx.api.versioning import reverse
from awx.main.models.jobs import JobTemplate, Job
from awx.main.models.activity_stream import ActivityStream
from awx.main.access import JobTemplateAccess
from awx.main.utils.common import get_type_for_model
@pytest.fixture

View File

@@ -3,15 +3,19 @@ import pytest
from unittest import mock
import urllib.parse
from unittest.mock import PropertyMock
import importlib
# Django
from django.urls import resolve
from django.http import Http404
from django.apps import apps
from django.core.handlers.exception import response_for_exception
from django.contrib.auth.models import User
from django.core.serializers.json import DjangoJSONEncoder
from django.db.backends.sqlite3.base import SQLiteCursorWrapper
from django.db.models.signals import post_migrate
# AWX
from awx.main.models.projects import Project
from awx.main.models.ha import Instance
@@ -41,10 +45,19 @@ from awx.main.models.workflow import WorkflowJobTemplate
from awx.main.models.ad_hoc_commands import AdHocCommand
from awx.main.models.oauth import OAuth2Application as Application
from awx.main.models.execution_environments import ExecutionEnvironment
from awx.main.utils import is_testing
__SWAGGER_REQUESTS__ = {}
# HACK: the dab_resource_registry app required ServiceID in migrations which checks do not run
dab_rr_initial = importlib.import_module('ansible_base.resource_registry.migrations.0001_initial')
if is_testing():
post_migrate.connect(lambda **kwargs: dab_rr_initial.create_service_id(apps, None))
@pytest.fixture(scope="session")
def swagger_autogen(requests=__SWAGGER_REQUESTS__):
return requests

View File

@@ -193,6 +193,7 @@ class TestInventorySourceInjectors:
('satellite6', 'theforeman.foreman.foreman'),
('insights', 'redhatinsights.insights.insights'),
('controller', 'awx.awx.tower'),
('terraform', 'cloud.terraform.terraform_state'),
],
)
def test_plugin_proper_names(self, source, proper_name):

View File

@@ -309,3 +309,139 @@ def test_bulk_job_set_all_prompt(job_template, organization, inventory, project,
assert node[0].limit == 'kansas'
assert node[0].skip_tags == 'foobar'
assert node[0].job_tags == 'untagged'
@pytest.mark.django_db
@pytest.mark.parametrize('num_hosts, num_queries', [(1, 70), (10, 150), (25, 250)])
def test_bulk_host_delete_num_queries(organization, inventory, post, get, user, num_hosts, num_queries, django_assert_max_num_queries):
'''
If I am a...
org admin
inventory admin at org level
admin of a particular inventory
superuser
Bulk Host delete should take under a certain number of queries
'''
users_list = setup_admin_users_list(organization, inventory, user)
for u in users_list:
hosts = [{'name': str(uuid4())} for i in range(num_hosts)]
with django_assert_max_num_queries(num_queries):
bulk_host_create_response = post(reverse('api:bulk_host_create'), {'inventory': inventory.id, 'hosts': hosts}, u, expect=201).data
assert len(bulk_host_create_response['hosts']) == len(hosts), f"unexpected number of hosts created for user {u}"
hosts_ids_created = get_inventory_hosts(get, inventory.id, u)
bulk_host_delete_response = post(reverse('api:bulk_host_delete'), {'hosts': hosts_ids_created}, u, expect=201).data
assert len(bulk_host_delete_response['hosts'].keys()) == len(hosts), f"unexpected number of hosts deleted for user {u}"
@pytest.mark.django_db
def test_bulk_host_delete_rbac(organization, inventory, post, get, user):
'''
If I am a...
org admin
inventory admin at org level
admin of a particular invenotry
... I can bulk delete hosts
Everyone else cannot
'''
admin_users_list = setup_admin_users_list(organization, inventory, user)
users_list = setup_none_admin_uses_list(organization, inventory, user)
for indx, u in enumerate(admin_users_list):
bulk_host_create_response = post(
reverse('api:bulk_host_create'), {'inventory': inventory.id, 'hosts': [{'name': f'foobar-{indx}'}]}, u, expect=201
).data
assert len(bulk_host_create_response['hosts']) == 1, f"unexpected number of hosts created for user {u}"
assert Host.objects.filter(inventory__id=inventory.id)[0].name == f'foobar-{indx}'
hosts_ids_created = get_inventory_hosts(get, inventory.id, u)
bulk_host_delete_response = post(reverse('api:bulk_host_delete'), {'hosts': hosts_ids_created}, u, expect=201).data
assert len(bulk_host_delete_response['hosts'].keys()) == 1, f"unexpected number of hosts deleted by user {u}"
for indx, create_u in enumerate(admin_users_list):
bulk_host_create_response = post(
reverse('api:bulk_host_create'), {'inventory': inventory.id, 'hosts': [{'name': f'foobar2-{indx}'}]}, create_u, expect=201
).data
print(bulk_host_create_response)
assert bulk_host_create_response['hosts'][0]['name'] == f'foobar2-{indx}'
hosts_ids_created = get_inventory_hosts(get, inventory.id, create_u)
print(f"Try to delete {hosts_ids_created}")
for delete_u in users_list:
bulk_host_delete_response = post(reverse('api:bulk_host_delete'), {'hosts': hosts_ids_created}, delete_u, expect=403).data
assert "Lack permissions to delete hosts from this inventory." in bulk_host_delete_response['inventories'].values()
@pytest.mark.django_db
def test_bulk_host_delete_from_multiple_inv(organization, inventory, post, get, user):
'''
If I am inventory admin at org level
Bulk Host delete should be enabled only on my inventory
'''
num_hosts = 10
inventory.organization = organization
# Create second inventory
inv2 = organization.inventories.create(name="second-test-inv")
inv2.organization = organization
admin2_user = user('inventory2_admin', False)
inv2.admin_role.members.add(admin2_user)
admin_user = user('inventory_admin', False)
inventory.admin_role.members.add(admin_user)
organization.member_role.members.add(admin_user)
organization.member_role.members.add(admin2_user)
hosts = [{'name': str(uuid4())} for i in range(num_hosts)]
hosts2 = [{'name': str(uuid4())} for i in range(num_hosts)]
# create hosts in each of the inventories
bulk_host_create_response = post(reverse('api:bulk_host_create'), {'inventory': inventory.id, 'hosts': hosts}, admin_user, expect=201).data
assert len(bulk_host_create_response['hosts']) == len(hosts), f"unexpected number of hosts created for user {admin_user}"
bulk_host_create_response2 = post(reverse('api:bulk_host_create'), {'inventory': inv2.id, 'hosts': hosts2}, admin2_user, expect=201).data
assert len(bulk_host_create_response2['hosts']) == len(hosts), f"unexpected number of hosts created for user {admin2_user}"
# get all hosts ids - from both inventories
hosts_ids_created = get_inventory_hosts(get, inventory.id, admin_user)
hosts_ids_created += get_inventory_hosts(get, inv2.id, admin2_user)
expected_error = "Lack permissions to delete hosts from this inventory."
# try to delete ALL hosts with admin user of inventory 1.
for inv_name, invadmin in zip([inv2.name, inventory.name], [admin_user, admin2_user]):
bulk_host_delete_response = post(reverse('api:bulk_host_delete'), {'hosts': hosts_ids_created}, invadmin, expect=403).data
result_message = bulk_host_delete_response['inventories'][inv_name]
assert result_message == expected_error, f"deleted hosts without permission by user {invadmin}"
def setup_admin_users_list(organization, inventory, user):
inventory.organization = organization
inventory_admin = user('inventory_admin', False)
org_admin = user('org_admin', False)
org_inv_admin = user('org_admin', False)
superuser = user('admin', True)
for u in [org_admin, org_inv_admin, inventory_admin]:
organization.member_role.members.add(u)
organization.admin_role.members.add(org_admin)
organization.inventory_admin_role.members.add(org_inv_admin)
inventory.admin_role.members.add(inventory_admin)
return [inventory_admin, org_inv_admin, superuser, org_admin]
def setup_none_admin_uses_list(organization, inventory, user):
inventory.organization = organization
auditor = user('auditor', False)
member = user('member', False)
use_inv_member = user('member', False)
for u in [auditor, member, use_inv_member]:
organization.member_role.members.add(u)
inventory.use_role.members.add(use_inv_member)
organization.auditor_role.members.add(auditor)
return [auditor, member, use_inv_member]
def get_inventory_hosts(get, inv_id, use_user):
data = get(reverse('api:inventory_hosts_list', kwargs={'pk': inv_id}), use_user, expect=200).data
results = [host['id'] for host in data['results']]
return results

View File

@@ -81,6 +81,7 @@ def test_default_cred_types():
'aws_secretsmanager_credential',
'azure_kv',
'azure_rm',
'bitbucket_dc_token',
'centrify_vault_kv',
'conjur',
'controller',
@@ -100,6 +101,7 @@ def test_default_cred_types():
'satellite6',
'scm',
'ssh',
'terraform',
'thycotic_dsv',
'thycotic_tss',
'vault',

Some files were not shown because too many files have changed in this diff Show More