When TAS RF is updated (e.g. nodeTaints or tolerations), then the cache information is not updated.
What you expected to happen:
Update the cache or fail the update. I think updating cache would be ideal, but it might be hard to deliever before 0.10,
so we may just block updates for now (when topologyName is specified).
How to reproduce it (as minimally and precisely as possible):
Create a TAS setup, and add the .spec.tolerations.
issue: the tolerations are not added to the newly admitted workloads.
What happened:
When TAS RF is updated (e.g. nodeTaints or tolerations), then the cache information is not updated.
What you expected to happen:
Update the cache or fail the update. I think updating cache would be ideal, but it might be hard to deliever before 0.10, so we may just block updates for now (when topologyName is specified).
How to reproduce it (as minimally and precisely as possible):
Create a TAS setup, and add the .spec.tolerations. issue: the tolerations are not added to the newly admitted workloads.
Anything else we need to know?:
We had a similar decision about update to the levels and decided to block updates for now: https://github.com/kubernetes-sigs/kueue/issues/3614