fix: stop using remove node label #7
Merged
Conversation
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
…rework chore: rework cluster-autoscaler Scaleway cloudprovider integration
…r-api-changes Adjust OWNERS so that only API changes need api review
* fix: ensure vpa_recommender_vpa_objects_count UpdateModeInPlaceOrRecreate is reset * move GetUpdateModes() to helpers.go * update copyright Co-authored-by: Adrian Moisey <[email protected]> --------- Co-authored-by: Adrian Moisey <[email protected]>
…t-in-pod-auto-scaler fix: deprecated import of cacheddiscovery in vertical-pod-autoscaler
…deStartupTime Make maxNodeStartupTime configurable
fix: deprecated import of cacheddiscovery in balancer
Add support for Intel Habana Gaudi GPUs in the cluster autoscaler by:
- Define ResourceIntelGPU resource name (habana.ai/gaudi)
- Add Intel GPU to GPUVendorResourceNames list
- Refactor GPU detection logic to iterate through all GPU vendor resource names
instead of checking vendors individually
This enables the autoscaler to properly detect and handle Intel GPU nodes
alongside existing NVIDIA, AMD, and DirectX GPU support.
Extract the GPU allocatable detection loop into a new NodeHasGpuAllocatable helper function in utils/gpu/gpu.go. This eliminates code duplication across gpu_processor.go and makes the logic more maintainable. The new function returns both the GPU allocatable value and whether it exists, allowing callers to get both pieces of information in a single call. Changes: - Add NodeHasGpuAllocatable() helper in utils/gpu/gpu.go - Update NodeHasGpu() to use the new helper - Simplify FilterOutNodesWithUnreadyResources() in gpu_processor.go - Simplify GetNodeGpuTarget() in gpu_processor.go
…support Add Intel GPU (Habana Gaudi) autoscaler support
…encytracker Node removal latency metrics added
nickstern2002
force-pushed
the
ns/drop-remove-node-label
branch
from
f152187 to
8ce8145
Compare
RixhersAjazi
previously approved these changes
Dec 2, 2025
RixhersAjazi
left a comment
RixhersAjazi
left a comment
There was a problem hiding this comment.
LGTM, just a quick question for my understanding. Would suggest waiting on others to review as well to ensure bigger picture context isn't missing.
cluster-autoscaler/cloudprovider/coreweave/coreweave_nodegroup.go
Outdated
Show resolved
Hide resolved
LanceEa
reviewed
Dec 3, 2025
LanceEa
left a comment
LanceEa
left a comment
There was a problem hiding this comment.
a couple of open questions but the gist looks right.
cluster-autoscaler/cloudprovider/coreweave/coreweave_nodegroup.go
Outdated
Show resolved
Hide resolved
LanceEa
previously approved these changes
Dec 3, 2025
Member
Author
|
Okay cool, I'm not going to merge this in as it will pollute our fork's commit history. Instead I am going to get this rebased on top of the upstream's master branch and open this PR in the upstream repo. I've already cut an image from this branch's latest commit so it is good to rollout now This will make this PR a little messy on our end but it will look the same in the upstream PR. Its better than having to ask CBS to manually hard reset our fork. Will eventually close this PR once the Upstream has been merged in |
nickstern2002
force-pushed
the
ns/drop-remove-node-label
branch
from
e761968 to
106ffbf
Compare
nickstern2002
force-pushed
the
ns/drop-remove-node-label
branch
from
106ffbf to
6d42554
Compare
Member
Author
|
Here is the upstream PR |
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.
What type of PR is this?
What this PR does / why we need it:
Two changes present in this PR
Dropping logic that adds the
remove-nodelabelremove-nodelabel is now returning to its original state of being controlled by a single internal controllerremove-nodelabel was causing conflicts with the internal controller responsible for reconciling node groups.Taking the absolute value of
deltain theDecreaseTargetSizeDecreaseTargetSizeis called delta is passed as a negative value. This causes the autoscaler to increase a nodePool's target size because the method subtracts delta.Which issue(s) this PR fixes:
Fixes #
Special notes for your reviewer:
Does this PR introduce a user-facing change?
Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.: