- use awx-python in shebang in dev env
- scl enable where needed for rhel7 & container installs
- use scram-sha-256 pg user hashing by default
- ensure psycopg2 is using the correct PG_CONFIG at build time for the right libpq version
I had 1.13 installed as part of Centos Extras and spent hours attempting to install AWX 4.0.0; the attempts all threw masses of permission denied errors.
Uninstalling that version and replacing with a current docker-ce then worked.
HTTPS is, by default, expected to be on port 443.
Also, with HSTS set, we need to be sure that users attempting to arrive
via HTTP are properly redirected to HTTPS.
This does so by:
* Setting up a 301 redirect for any URL to its HTTPS version
* Adjusting the internal port for HTTPS traffic to 8053
* Setting docker-compose to share port 443 → 8053
- This is configurable via an inventory variable
As it turns out, the docker-compose Python module is
required, and docker-py doesn’t cut it.
Even more confusing, docker-compose Python module installs
the docker Python module, which conflicts with docker-py. To
avoid this, there are additional docs to call this out.
When docker-compose become the sole method for using
Docker directly, some of this was shifted around in ways that
are inconsistent with other elements.
This adjusts it so that:
* The inventory variable default is set like the others, and
is less confusing
* We no longer mention the Standalone Docker in inventory
* We format our INSTALL docs w/r/t this var
this commit implements the bulk of `awx-manage run_dispatcher`, a new
command that binds to RabbitMQ via kombu and balances messages across
a pool of workers that are similar to celeryd workers in spirit.
Specifically, this includes:
- a new decorator, `awx.main.dispatch.task`, which can be used to
decorate functions or classes so that they can be designated as
"Tasks"
- support for fanout/broadcast tasks (at this point in time, only
`conf.Setting` memcached flushes use this functionality)
- support for job reaping
- support for success/failure hooks for job runs (i.e.,
`handle_work_success` and `handle_work_error`)
- support for auto scaling worker pool that scale processes up and down
on demand
- minimal support for RPC, such as status checks and pool recycle/reload
Shane McDonald
JensPfeifle
softwarefactory-project-zuul[bot]
Kim Ausloos
Evan Stork
Christian Adams
vrevelas
Simon Séhier
JP Mens
Kia Lam
softwarefactory-project-zuul[bot]
Nick Busey
Jeff Byrnes
Elijah DeLee
Markus Opahle
walkafwalka
Vladimir Pouzanov
Numblesix
jmferrer
Ryan Petrello
Taylor Smith
Lyosha Shchur
Raphaël B
Jared Tabor
Matthew Jones
therealmaxmouse
Cédric Levasseur
Paschalis Korosoglou
dovshap
Joachim Jablon
Jake McDermott
Matthew Jones