Add configuration option for enforcement models

We eventually want deployers to be able to choose which enforcement
model they want for their deployment. This commit adds a new
configuration options that exposes the enforcement model via config.

This will dictate how project limits are validated.

Change-Id: Ied398b92fef09f3a446960b0757c68a21ba50ed6
Related-Bug: 1765193
This commit is contained in:
Lance Bragstad 2018-04-19 15:13:47 +00:00
parent 000c071379
commit b8e8a8b550
1 changed files with 12 additions and 0 deletions

View File

@ -47,6 +47,17 @@ be useful to tune if you have a large number of unified limits in your
deployment.
"""))
enforcement_model = cfg.StrOpt(
'enforcement_model',
default='flat',
choices=['flat'],
help=utils.fmt("""
The enforcement model to use when validating limits associated to projects.
Enforcement models will behave differently depending on the existing limits,
which may result in backwards incompatible changes if a model is switched in a
running deployment.
"""))
GROUP_NAME = __name__.split('.')[-1]
ALL_OPTS = [
@ -54,6 +65,7 @@ ALL_OPTS = [
caching,
cache_time,
list_limit,
enforcement_model,
]