Conversation
…olerating every taint Signed-off-by: Christopher Haar <christopher.haar@upbound.io>
Signed-off-by: Christopher Haar <christopher.haar@upbound.io>
Signed-off-by: Christopher Haar <christopher.haar@upbound.io>
Signed-off-by: Christopher Haar <christopher.haar@upbound.io>
4 tasks done
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description of your changes
Stacked on and a follow-up to #456, the first commit here is that PRs fix - will rebase when merged.
Nothing verified that a clouds serving stack actually lands where it should: the local e2e runs everything on one untainted kind node, so a toleration that was too broad (the wildcard tolerations #456 removes) or too narrow (an
Equaltoleration whose value drifted from the cloud's real taint) passed locally and only failed on a real cluster.nix run .#e2e -- --cloud eks|aks|gke|nebius|vultr|existingnow installs that cloud's serving stack on the local kind workload cluster, with KWOK faking two nodes labelled and tainted exactly like that cloud's GPU pool while the real kind node plays the untainted system pool. A single values-driven Chainsaw test applies the fake nodes and a ServingStack XR directly (source: Existinghard-wirescloud: Existing, so the XR is applied rather than composed through an InferenceCluster), waits for it to become Ready, then checks placement from both sides: the values file's expected DaemonSets must fan out to the fake GPU pool (a toleration too narrow fails here), and an audit of every pod's tolerations rejects wildcards and unexpected GPU tolerations wherever the pod landed (a toleration too broad fails here, approximating client-side what AKS rejects at admission). Everything cloud-specific lives ine2e/clouds/values-<cloud>.yaml.Validated with a local
--cloud aksFixes #
I have:
nix flake check(or./nix.sh flake check) and made sure it passes.git commit -s.