Refactor Reused Ansible Tasks into Roles (w/need-based delays) #15

New Issue

Closed

opened 2023-08-13 21:18:20 +02:00 by so-rose · 0 comments

so-rose commented

2023-08-13 21:18:20 +02:00

Owner

There's a lot of boilerplate, reuse, and just general mess in our playbooks right now.

The sustainable thing to do is to factored out into reusable roles, which can be used/reused both in this project and in other projects.

These roles should make sure to only utilize hard-coded delays (pause builtin module) when absolutely needed. Especially ex.:

Find a way to check if a volume needs to be unmounted and remounted due to a change during a stack deployment. This is the kind of optimization that makes the difference between needing to stop the stack or not.
Only if one is needed (rclone cleanup the old unmounted volume), should hard-coded pauses be used.

There's a lot of boilerplate, reuse, and just general mess in our playbooks right now. The sustainable thing to do is to factored out into reusable roles, which can be used/reused both in this project and in other projects. These roles should make sure to only utilize hard-coded delays (`pause` builtin module) when absolutely needed. Especially ex.: - Find a way to check if a volume needs to be unmounted and remounted due to a change during a stack deployment. This is the kind of optimization that makes the difference between needing to stop the stack or not. - **Only** if one is needed (rclone cleanup the old unmounted volume), should hard-coded pauses be used.