search

fititnt / ap-application-load-balancer

AP Application Load Balancer (AP-ALB). Sophisticated monolithic Ansible role to manage standalone and clusters of cross-platform and multicloud load balancers. Abstract HAProxy + OpenResty + On-the-fly auto HTTPS. Dedicated to Public Domain.
https://ap-application-load-balancer.etica.ai/
The Unlicense
1 stars 0 forks source link

AP-ALB on Ansible Galaxy #49

Open fititnt opened 4 years ago

fititnt commented 4 years ago

After all this time, let's publish to Ansible Galaxy https://galaxy.ansible.com/.

This will be my first role published there. And it's already more sofisticated than average ones.

fititnt commented 4 years ago

This is the first import. The E305: Use shell only when shell functionality is required and E301: Commands should not change things if nothing needs doing I was already aware that would show on the score. They are marked as HIGH severity, that's harsh, but ok. I will try to reduce this, but may be very complicated if supporting multiple OSs.

Here the print

Captura de tela de 2020-01-01 23-27-26

fititnt commented 4 years ago

Ansible Galaxy does not have the lastest versions of OpenSUSE for we target them on our role

https://galaxy.ansible.com/api/v1/platforms/?search=suse

fititnt commented 4 years ago

Beyond AP-ALB, we published under my username plus 2 packages, ansible-galaxy install fititnt.faker (https://github.com/fititnt/ansible-faker) and ansible-galaxy install fititnt.syntactic_sugar (https://github.com/fititnt/ansible-syntactic-sugar).

Captura de tela de 2020-01-12 22-24-03

The Quality Score from fititnt.ap_application_load_balancer improved from the first release, from 2.5 to 3.3:

Captura de tela de 2020-01-12 22-27-38

Note about the automatic quality score on AP-ALB

For who is not aware on how these scores works on Ansible Galaxy, actually several of they are actually possible for the developer force the linter to ignore.

Some issues are more aesthetic (like whitespace) and so easy to fix. Others do not. For example the new 2 Ansible roles where much, much much more simpler than AP-ALB and easy to make it score higher. The AP-ALB, in special because of the AP-ALB v0.8.x adaptations to support RHEL/CentOS 8 family and design changes to allow flexibility even for non-tested OSs #34 forced to take some shortcuts using shell, but even then I'm aware that is possible to improve this.

So, the issues that already where not removed from the lint are ones that are likely to need some refactoring later.