You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Nov 1, 2022. It is now read-only.
We had talked in #flux about some of the ways we could handle this
Proposed ideas:
1.) use git repo history for gc @hiddeco pointed out that this may be problematic
2.) use a combination of labels to identify that a hash belongs to a type, for example that hash above belongs to a Prometheus so fluxd won't garbage collect a StatefulSet
questions?
what if a Foo creates a Foo? maybe the labels would have a hash of some kind to identify the resource it belongs to?
The text was updated successfully, but these errors were encountered:
Hi friends!
I'm testing the garbage collection feature, and i'm noticing that some operators' children are being garbage collected!
example:
Prometheus operator creates a statefulset, and the statefulset gets cleaned up when gc runs
we've identified that this is because the statefulset is mirroring the labels from the
Prometheus
objectboth the
Prometheus
object and theStatefulSet
get thisWe had talked in #flux about some of the ways we could handle this
Proposed ideas:
1.) use git repo history for gc @hiddeco pointed out that this may be problematic
2.) use a combination of labels to identify that a hash belongs to a type, for example that hash above belongs to a
Prometheus
so fluxd won't garbage collect aStatefulSet
questions?
what if a
Foo
creates aFoo
? maybe the labels would have a hash of some kind to identify the resource it belongs to?The text was updated successfully, but these errors were encountered: