[bitnami/redis] replica anti-affinity #17222
Labels
feature-request
on-hold
Issues or Pull Requests with this label will never be considered stale
redis
Name and Version
bitnami/redis 17.11.6
What is the problem this feature will solve?
By default kubernetes may simply decide to put the redis master/slave hosts on a single worker. This can hide the fact that, while you've got multiple systems for HA, they are all on a single node.
What is the feature you are proposing to solve the problem?
Set a default pod anti-affinity for the redis master/slaves to spread out the pods that can be ignored if no other scheduling paths are viable.
What alternatives have you considered?
manually passing in the affinity to the chart as extra values.
The text was updated successfully, but these errors were encountered: