Skip to content

[Helm] Update ray-cluster default resource values#4588

Merged
rueian merged 14 commits intoray-project:masterfrom
Future-Outlier:1.4.2to1.6.0
Mar 17, 2026
Merged

[Helm] Update ray-cluster default resource values#4588
rueian merged 14 commits intoray-project:masterfrom
Future-Outlier:1.4.2to1.6.0

Conversation

@Future-Outlier
Copy link
Member

@Future-Outlier Future-Outlier commented Mar 11, 2026

Summary

  1. Update ray-cluster default resource values to follow Kubernetes best practices.
  2. Give head container more memory to avoid OOM.

Changes

ray-cluster Default Value Changes

The helm chart ray-cluster/values.yaml is updated to match
ray-cluster.sample.yaml.

Field Old New Reason
head.resources.limits.memory 2G 5Gi Increase head memory to avoid OOM
head.resources.requests.memory 2G 5Gi Match limits for Guaranteed QoS
worker.resources.limits.memory 1G 1Gi Use IEC binary unit (Gi)
worker.resources.requests.memory 1G 1Gi Use IEC binary unit (Gi)
worker.maxReplicas 3 5 Allow more room for autoscaling
additionalWorkerGroups.smallGroup.resources.limits.memory 1G 1Gi Use IEC binary unit (Gi)
additionalWorkerGroups.smallGroup.resources.requests.memory 1G 1Gi Use IEC binary unit (Gi)
additionalWorkerGroups.smallGroup.maxReplicas 3 5 Allow more room for autoscaling

Signed-off-by: Future-Outlier <eric901201@gmail.com>
…rce defaults

Signed-off-by: Future-Outlier <eric901201@gmail.com>
Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: bd30892654

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Signed-off-by: Future-Outlier <eric901201@gmail.com>
Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: b86b9310d6

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

@Future-Outlier Future-Outlier changed the title [Helm] Fix apiserver README chart version and align ray-cluster resource defaults [Helm] Bump helm chart versions to 1.6.0 and update RayCluster default values Mar 16, 2026
Signed-off-by: Future-Outlier <eric901201@gmail.com>
Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: c317388cad

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Signed-off-by: Future-Outlier <eric901201@gmail.com>
Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: ce3ecbd909

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

# to the chart and its templates, including the app version.
# Versions are expected to follow Semantic Versioning (https://semver.org/)
version: 1.1.0
version: 1.6.0
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

the chart version in master should actually always be nightly or something similar. Only release branches should have official versions

# -- Image repository.
repository: quay.io/kuberay/apiserver
# -- Image tag.
tag: nightly
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

keep nightly

description: A Helm chart for deploying the Kuberay operator on Kubernetes.

version: 1.1.0
version: 1.6.0
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

updated to nightly

repository: quay.io/kuberay/operator

# -- Image tag.
tag: nightly
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

keep as nightly

Signed-off-by: Future-Outlier <eric901201@gmail.com>
Signed-off-by: Future-Outlier <eric901201@gmail.com>
Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 299a4056cc

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: e90716cfdb

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

@Future-Outlier
Copy link
Member Author

Hi, @andrewsykim
For values.yaml, I've reverted the image tags back to nightly.

For Chart.yaml, I have to follow the SemVer's rule, so I will keep 1.1.0 as the same
https://helm.sh/docs/topics/charts/#charts-and-versioning

Signed-off-by: Future-Outlier <eric901201@gmail.com>
Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 1b14f85739

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Signed-off-by: Future-Outlier <eric901201@gmail.com>
@Future-Outlier
Copy link
Member Author

I've reverted the helm chart version and image tag changes. For values.yaml image tags, I've reverted them back to nightly. For Chart.yaml versions, I've kept them as 1.1.0 since the version field requires SemVer 2 format — non-SemVer names are explicitly disallowed by Helm, so we can't use nightly directly there.

I think it would be cleaner to make all helm chart version bumps and README updates in the kuberay-helm repo instead of in the kuberay repo, to keep things in one place. What do you think?

@Future-Outlier Future-Outlier changed the title [Helm] Bump helm chart versions to 1.6.0 and update RayCluster default values [Helm] Update ray-cluster default resource values Mar 16, 2026
@andrewsykim
Copy link
Member

For Chart.yaml, I have to follow the SemVer's rule, so I will keep 1.1.0 as the same
https://helm.sh/docs/topics/charts/#charts-and-versioning

Huh, I didn't know this. Perhaps we can use a placeholder version like v0.0.0 or similar

requests:
cpu: "1"
memory: "2Gi"
memory: "5Gi"
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Btw, I think the e2e test output is failing since you increased the memory here:


[2026-03-16T15:43:22Z] Running Suite: Kubectl Ray e2e Test Suite - /workdir/kubectl-plugin/test/e2e
--
[2026-03-16T15:43:22Z] ============================================================================
[2026-03-16T15:43:22Z] Random Seed: 1773674078
[2026-03-16T15:43:22Z]
[2026-03-16T15:43:22Z] Will run 30 of 30 specs
[2026-03-16T15:43:22Z] ••••••••••••
[2026-03-16T15:43:22Z] ##############################
[2026-03-16T15:43:22Z] • [FAILED] [36.174 seconds]
[2026-03-16T15:43:22Z] Calling ray plugin `get` command [It] succeed in getting ray cluster information
[2026-03-16T15:43:22Z] /workdir/kubectl-plugin/test/e2e/kubectl_ray_cluster_get_test.go:28
[2026-03-16T15:43:22Z]
[2026-03-16T15:43:22Z]   [FAILED] Expected
[2026-03-16T15:43:22Z]       <string>: NAME                 NAMESPACE       DESIRED WORKERS   AVAILABLE WORKERS   CPUS   GPUS   TPUS   MEMORY   CONDITION               STATUS   AGE
[2026-03-16T15:43:22Z]       raycluster-kuberay   test-ns-usf5j   1                 1                   2      0      0      6Gi      RayClusterProvisioned   ready    25s
[2026-03-16T15:43:22Z]   to contain substring
[2026-03-16T15:43:22Z]       <string>: NAME                 NAMESPACE       DESIRED WORKERS   AVAILABLE WORKERS   CPUS   GPUS   TPUS   MEMORY   CONDITION               STATUS   AGE
[2026-03-16T15:43:22Z]       raycluster-kuberay   test-ns-usf5j   1                 1                   2      0      0      3Gi      RayClusterProvisioned   ready
[2026-03-16T15:43:22Z]   In [It] at: /workdir/kubectl-plugin/test/e2e/kubectl_ray_cluster_get_test.go:69 @ 03/16/26 15:28:44.638
[2026-03-16T15:43:22Z] ##############################
[2026-03-16T15:43:22Z] •••••••••••••••••
[2026-03-16T15:43:22Z]
[2026-03-16T15:43:22Z] Summarizing 1 Failure:
[2026-03-16T15:43:22Z]   [FAIL] Calling ray plugin `get` command [It] succeed in getting ray cluster information
[2026-03-16T15:43:22Z]   /workdir/kubectl-plugin/test/e2e/kubectl_ray_cluster_get_test.go:69
[2026-03-16T15:43:22Z]
[2026-03-16T15:43:22Z] Ran 30 of 30 Specs in 1724.141 seconds
[2026-03-16T15:43:22Z] FAIL! -- 29 Passed \| 1 Failed \| 0 Pending \| 0 Skipped
[2026-03-16T15:43:22Z] ### FAIL: TestKubectlRayCommand (1724.14s)
[2026-03-16T15:43:22Z] FAIL
[2026-03-16T15:43:22Z] FAIL	github.com/ray-project/kuberay/kubectl-plugin/test/e2e	1724.148s
[2026-03-16T15:43:22Z] FAIL
[2026-03-16T15:43:24Z] 🚨 Error: The command exited with status 1
[2026-03-16T15:43:24Z] user command error: The plugin docker command hook exited with status 1


Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

fixed, tks!

@rueian
Copy link
Collaborator

rueian commented Mar 16, 2026

Seems like the test need to be fixed.

Signed-off-by: Future-Outlier <eric901201@gmail.com>
Copy link

@cursor cursor bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Fix All in Cursor

Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: dae0e0cbea

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Signed-off-by: Future-Outlier <eric901201@gmail.com>
@rueian rueian merged commit aca84af into ray-project:master Mar 17, 2026
31 checks passed
@github-project-automation github-project-automation bot moved this from can be merged to Done in @Future-Outlier's kuberay project Mar 17, 2026
rueian added a commit that referenced this pull request Mar 19, 2026
* Disable RayMultiHostIndexing feature for TestReconcile_Multihost_Replicas (#4583)

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* Disable the field alignment lint check (#4560)

* chore: disable govet fieldalignment lint

Signed-off-by: jinbum9958 <jinbum9958@gmail.com>

* chore: remove redundant //nolint:govet comments

Signed-off-by: jinbum9958 <jinbum9958@gmail.com>

* chore: restore RayCronJobSpec/RayCronJob comments

Signed-off-by: jinbum9958 <jinbum9958@gmail.com>

* chore: restore RayCronJob comment

Signed-off-by: jinbum9958 <jinbum9958@gmail.com>

* chore: regenerate artifacts for RayCronJob comment changes

Signed-off-by: jinbum9958 <jinbum9958@gmail.com>

---------

Signed-off-by: jinbum9958 <jinbum9958@gmail.com>

* Mark RC releases as pre-release in GoReleaser (#4590)

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* Add example YAML for manually enabling Ray k8s auth (#4582)

Signed-off-by: Andrew Sy Kim <andrewsy@google.com>

* [RayService] Rollback Support for Incremental Upgrades (#4109)

* Implement rollback support

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Fix unit test file

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Ensure upgrade in progress status is cleared

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Clarify rollback scenarios and clear pending apps during rollback

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Fix typo

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Add guard for rollback to prepare new cluster

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* fix e2e test checking pending cluster name

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Fix rollback calculation

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Fix httproute name in test

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Increase timeout for checking RayCluster deletion

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Fix rollback check and change httproute equal check so it doesn't always evaluate false

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Adjust logic to use timestamp of growing cluster

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Fix naming of vars

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

---------

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>
Signed-off-by: Ryan O'Leary <113500783+ryanaoleary@users.noreply.github.com>

* [RayService] Promote Incremental Upgrade Feature to Beta (#4599)

Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>

* Revert "[RayService] Promote Incremental Upgrade Feature to Beta (#4599)" (#4602)

This reverts commit 0c8aab9.

* [Github Action] Skip krew-index update for pre-release tags (#4587)

* [CI] Skip krew-index update for pre-release tags

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

---------

Signed-off-by: Future-Outlier <eric901201@gmail.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

* [Helm] update ray-cluster chart to apiVersion: v2 (#4593)

* Update kind configs and docs to v1.29.0 (#4595)

* Remove ingress template support for k8s < 1.19 (#4591)

* [Helm] Update ray-cluster default resource values (#4588)

* update kuberay 1.6.0 helm chart and readme

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* [Helm] Fix apiserver README chart version and align ray-cluster resource defaults

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* fix ctl test

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

---------

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* fix lint

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* chore: Use RayCluster name as SA name for RBAC auth (#4611)

Signed-off-by: JiangJiaWei1103 <waynechuang97@gmail.com>

* Add Google Artifact Registry image build/push guide (#4618)

* [History Server] Fix API response format to match Ray Dashboard frontend schema (#4615)

* [History Server] Fix API response format to match Ray Dashboard frontend schema

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* fix bug 1

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* add comments to explain frontend fields naming issue

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* fix tests

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* Add comments

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* update

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* Add frontend reference for this PR

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* fix codex and cursor bug

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* upadte

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* add actor.LabelSelector

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* better query for additional endpoints, request URI (path + query)

Signed-off-by: Future-Outlier <eric901201@gmail.com>

* Add comments

Signed-off-by: Future-Outlier <eric901201@gmail.com>

---------

Signed-off-by: Future-Outlier <eric901201@gmail.com>

---------

Signed-off-by: Future-Outlier <eric901201@gmail.com>
Signed-off-by: jinbum9958 <jinbum9958@gmail.com>
Signed-off-by: Andrew Sy Kim <andrewsy@google.com>
Signed-off-by: Ryan O'Leary <ryanaoleary@google.com>
Signed-off-by: Ryan O'Leary <113500783+ryanaoleary@users.noreply.github.com>
Signed-off-by: JiangJiaWei1103 <waynechuang97@gmail.com>
Co-authored-by: Han-Ju Chen (Future-Outlier) <eric901201@gmail.com>
Co-authored-by: Jinbum Kim <jinbum9958@gmail.com>
Co-authored-by: Andrew Sy Kim <andrewsy@google.com>
Co-authored-by: Ryan O'Leary <113500783+ryanaoleary@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Matt Boersma <Matt.Boersma@microsoft.com>
Co-authored-by: Jia-Wei Jiang <36886416+JiangJiaWei1103@users.noreply.github.com>
Co-authored-by: Chia-Yi Liang <aaronliang@google.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants