10 Pull requests merged by 6 people

• add node selector option for kubectl plugin create worker group

  #3235 merged Mar 27, 2025

• [Docs] update development md

  #3230 merged Mar 27, 2025

• kubectl-plugin: set global flags at the root cmd

  #3203 merged Mar 26, 2025

• [kubectl-plugin] Add head/worker node selector option

  #3228 merged Mar 26, 2025

• Integrate with rayci

  #3215 merged Mar 25, 2025

• [RayJob][Fix] Use --no-wait for job submission to avoid carrying the error return code to the log tailing

  #3216 merged Mar 25, 2025

• [operator] add // +optional to CRD fields with omitempty

  #3220 merged Mar 25, 2025

• [fix][proto] install missing unzip in Dockerfile

  #3221 merged Mar 24, 2025

• [operator] remove incorrect comment

  #3218 merged Mar 22, 2025

• [RayService][Test] make sure annotation populated to RayCluster

  #3210 merged Mar 21, 2025

14 Pull requests opened by 7 people

• [Prometheus] Add `ray_cluster_head_pod_ready_duration_seconds` metric

  #3222 opened Mar 24, 2025

• [RayJob][Test] refactor TestValidateRayJobSpec with table test

  #3223 opened Mar 25, 2025

• [Prometheus] Add `ray_services_created_total` metric

  #3224 opened Mar 25, 2025

• create cluster with config file

  #3225 opened Mar 25, 2025

• Update KubeRay release documentation

  #3226 opened Mar 25, 2025

• [plugin] warn if RayJob YAML has entrypoint

  #3229 opened Mar 26, 2025

• [operator] add `+optional` to CRD fields that are optional

  #3231 opened Mar 26, 2025

• Integrate with rayci (#3215)

  #3234 opened Mar 26, 2025

• [WIP] refactor metrics and add ray_jobs_created_total metric

  #3236 opened Mar 27, 2025

• [fix][plugin] add missing CLI flags for create cluster cmd

  #3237 opened Mar 27, 2025

• [refactor][plugin] RayClusterSpecObject

  #3238 opened Mar 27, 2025

• [feat] enforce DNS1035 validations on RayCluster, RayService, and RayJob names

  #3239 opened Mar 27, 2025

• [apiserver][feat] add pagination to ListClustersRequest

  #3240 opened Mar 27, 2025

• [kubectl-plugin] remove CPU limits by default

  #3243 opened Mar 28, 2025

30 Issues closed by 5 people

• [Bug] test issue for Exalate integration (please ignore)

  #3241 closed Mar 27, 2025

• [Bug] Head node autoscaler container fails to communicate to the kubernetes api with a 401 in Azure Kubernetes 1.30

  #2324 closed Mar 26, 2025

• What's the behavior for a Pod with restartPolicy Always when it is in PodFailed?

  #2293 closed Mar 26, 2025

• [Bug] Priority Class Name from worker group spec not forwarded to final templated yaml files

  #2086 closed Mar 26, 2025

• [Feat][kubectl-plugin] Support node selectors for creating clusters

  #3143 closed Mar 26, 2025

• [Autoscaler] Print the value of `WorkerGroupSpec.Replicas` when the number of worker Pods is not as expected in autoscaling-related tests.

  #2980 closed Mar 24, 2025

• [Refactor] Use constants for image tag, image repo, and versions in golang to avoid hard-coded strings

  #2939 closed Mar 22, 2025

• [release] Update TPU-related YAMLs to Ray 2.41

  #2944 closed Mar 22, 2025

• [release] Update RayService YAMLs to Ray 2.41

  #2948 closed Mar 22, 2025

• [release] Update the YAML files of the following examples to Ray 2.41.

  #2947 closed Mar 22, 2025

• [release] Update YuniKorn YAML files to Ray 2.41

  #2968 closed Mar 22, 2025

• [GCS FT] Use GcsFaultToleranceOptions instead of `ray.io/ft-enabled` in sample YAMLs

  #2957 closed Mar 22, 2025

• [release] Update Volcano YAML files to Ray 2.41

  #2967 closed Mar 22, 2025

• [Refactor] Make head Pod name deterministic

  #3013 closed Mar 22, 2025

• [release] Add doc for RayJob DeletionPolicy

  #3042 closed Mar 22, 2025

• [Doc] Update KubeRay upgrade guide for KubeRay v1.2.2 to v1.3.0

  #3054 closed Mar 22, 2025

• [Request] Test Ray#46861 does not break Kuberay redis cleanup

  #2319 closed Mar 22, 2025

• [release] Update sample YAMLs from Ray 2.9 to Ray 2.41

  #2920 closed Mar 22, 2025

• [Feature] Add timestamps for logs in e2e tests

  #2836 closed Mar 22, 2025

• [Feature] Update doc for GCS FT after API changes

  #2697 closed Mar 22, 2025

• [Umbrella] RayService Refactoring

  #2548 closed Mar 22, 2025

• [RayService][Refactor] RayService events

  #2547 closed Mar 22, 2025

• [Epic] Python tests cleanup

  #2509 closed Mar 22, 2025

• [Feature] Improve observability of e2e tests in Buildkite runner

  #2651 closed Mar 22, 2025

• [Bug] Re-enable flaky kubectl plugin e2e test "should reconnect after pod connection is lost"

  #2752 closed Mar 22, 2025

• [release] Update kubectl plugin doc to beta

  #3044 closed Mar 22, 2025

• [Refactor] Merge raycluster_gcs_ft_test.go and raycluster_gcsft_test.go

  #3002 closed Mar 22, 2025

• [Feature] Add shellcheck to pre-commit

  #2932 closed Mar 22, 2025

• [Bug] Test E2E (kubectl-plugin) Has Different Buildkite Display Format

  #3001 closed Mar 21, 2025

• [CI] Support upload artifacts in BuildKite

  #3106 closed Mar 21, 2025

6 Issues opened by 6 people

• [Bug] please ignore - one more test ticket for Exalate connection (to delete)

  #3242 opened Mar 27, 2025

• [Bug] Ray Dashboard breaks running k8s locally in kind w/ prometheus + grafana configured

  #3233 opened Mar 26, 2025

• [Feature] Improve error message from "kubectl ray session"

  #3232 opened Mar 26, 2025

• [Umbrella] Add Autoscaler e2e tests for partial placement groups

  #3227 opened Mar 26, 2025

• [Feature][metrics] ray_cluster_head_pod_ready_duration_seconds

  #3219 opened Mar 21, 2025

• [Bug] how to create jobs to use vgpu in hami with volcano and kuberay

  #3217 opened Mar 21, 2025

261 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• [refactor][operator]: make `RayStartParams` optional

  #3202 commented on Mar 27, 2025 • 10 new comments

• feat: add aggregated clusterrole

  #3193 commented on Mar 21, 2025 • 2 new comments

• [Prometheus] Add `ray_cluster_provisioned_duration_seconds` metric

  #3212 commented on Mar 25, 2025 • 1 new comment

• [Chore][CI] Limit the release-image-build github workflow to only take tag as input

  #3117 commented on Mar 26, 2025 • 1 new comment

• [Bug] [RayService] KubeRay does not handle voluntary disruptions well

  #1333 commented on Mar 22, 2025 • 0 new comments

• [Bug] Kuberay API Server Makefile target deploy-opeartor default image does not exist

  #1350 commented on Mar 22, 2025 • 0 new comments

• [Bug] Address Ray's suggestion. Claiming an entire K8s Node per Pod.

  #1378 commented on Mar 22, 2025 • 0 new comments

• [Core] Metric unintentional_worker_failures_total is not accurate

  #1918 commented on Mar 22, 2025 • 0 new comments

• [Bug] KubeRay switched traffic when Serve did not start

  #1559 commented on Mar 22, 2025 • 0 new comments

• [Bug] Distributed Inference Issues - GKE Autopilot

  #1568 commented on Mar 22, 2025 • 0 new comments

• [Bug] RayJob Volcano integration

  #1580 commented on Mar 22, 2025 • 0 new comments

• [Bug] KubeRay API Server ListAllComputeTemplates is intermittently failing.

  #1591 commented on Mar 22, 2025 • 0 new comments

• [Bug] KubeRay keep update RayJob CR when external Finalizer added to it.

  #1626 commented on Mar 22, 2025 • 0 new comments

• [Bug][GCS FT] Some Ray Serve configurations

  #1628 commented on Mar 22, 2025 • 0 new comments

• [autoscaler] Autoscaler can not know the overwritten port number.

  #1644 commented on Mar 22, 2025 • 0 new comments

• [Bug] Kustomize cluster-scope-resources creates a namespace.

  #1670 commented on Mar 22, 2025 • 0 new comments

• [Bug] The command `make deploy-with-webhooks` does not succeed in the first run

  #1692 commented on Mar 22, 2025 • 0 new comments

• [Bug] worker group cannot be removed from RayCluster

  #1739 commented on Mar 22, 2025 • 0 new comments

• RayCluster raylet health probe uses hardcoded default dashboard agent port

  #1760 commented on Mar 22, 2025 • 0 new comments

• [RayService] [CI] Some tests for pending/active clusters may spuriously pass because head pod is not manually set to ready

  #1768 commented on Mar 22, 2025 • 0 new comments

• [Bug] Autoscaler round up the amount of cpu resources a pod has.

  #1166 commented on Mar 22, 2025 • 0 new comments

• [Bug] RayService's Head Service does not inherit annotations from Head Pod's Service

  #1092 commented on Mar 22, 2025 • 0 new comments

• default K8s Service `appProtocol: tcp` causes Istio RequestAuthentication to not apply to that port

  #1025 commented on Mar 22, 2025 • 0 new comments

• [Bug] Can not create utils.py in rayservice directory

  #843 commented on Mar 22, 2025 • 0 new comments

• [Bug] change to request and limits does not trigger pod updates

  #787 commented on Mar 22, 2025 • 0 new comments

• [Bug] De-dupe RayCluster comments

  #765 commented on Mar 22, 2025 • 0 new comments

• [Bug] 409 conflicts when updating status

  #745 commented on Mar 22, 2025 • 0 new comments

• Verify the state of `ray pdb` and `ray memory` on KubeRay

  #710 commented on Mar 22, 2025 • 0 new comments

• [Bug] GPU multitenancy issues

  #687 commented on Mar 22, 2025 • 0 new comments

• [Bug] Investigate slow Ray pod termination

  #503 commented on Mar 22, 2025 • 0 new comments

• [RayService] Add envtests for RayService

  #2878 commented on Mar 22, 2025 • 0 new comments

• [Feature] Support `timeout` in Ray job

  #2743 commented on Mar 22, 2025 • 0 new comments

• [Doc] Best practice to reduce image pulling overhead

  #2742 commented on Mar 22, 2025 • 0 new comments

• [Feature][CLI] Support `working_dir` is bigger than 500MB for `ray job submit`

  #2717 commented on Mar 22, 2025 • 0 new comments

• [RayService] Make `ServiceStatus` to be equivalent to `Ready` condition

  #2851 commented on Mar 22, 2025 • 0 new comments

• Make sure RayService example still work

  #2307 commented on Mar 22, 2025 • 0 new comments

• [Feature] shell completion for `kubectl ray get [workergroups|nodes]` CLI commands

  #3051 commented on Mar 22, 2025 • 0 new comments

• [Doc] Rewrite RayCluster config doc to make it more organized

  #3055 commented on Mar 22, 2025 • 0 new comments

• Revisit "Using GPU" doc

  #3059 commented on Mar 22, 2025 • 0 new comments

• [Doc] Update kuberay's performance test documentation

  #3082 commented on Mar 22, 2025 • 0 new comments

• [Bug][kubectl-plugin] E2e test DeferCleanup in BeforeEach may cause namespace race condition

  #2758 commented on Mar 22, 2025 • 0 new comments

• [Bug] `wait-gcs-ready` init-container going out-of-memory indefinitely (OOMKilled)

  #2735 commented on Mar 22, 2025 • 0 new comments

• [Bug] Ray Job submit fails when GCS is enabled

  #2676 commented on Mar 22, 2025 • 0 new comments

• [Bug] Submit multiple RayJobs concurrently will cause ray-operator to slow down

  #2646 commented on Mar 22, 2025 • 0 new comments

• [Bug] apiserver API + Client do not support setting headServiceAnnotations

  #2626 commented on Mar 22, 2025 • 0 new comments

• [Bug] Incorrect labels on the Service for RayCluster head

  #2564 commented on Mar 22, 2025 • 0 new comments

• [RayService][Bug] Partial Removal of Deployments in ray-service.sample.yaml's ServeConfigV2 Causes WaitForServeDeploymentReady State

  #2557 commented on Mar 22, 2025 • 0 new comments

• [RayService][Bug] when doing multi-node serving with vLLM, non-primary worker pods reports unready.

  #2552 commented on Mar 22, 2025 • 0 new comments

• [Bug] ray.exceptions.RpcError: Timed out while waiting for GCS to become available.

  #2540 commented on Mar 22, 2025 • 0 new comments

• [Bug][High Availability] 502 Errors while Head Node in Recovery

  #1153 commented on Mar 22, 2025 • 0 new comments

• If the head node dies, the cluster is never restored [Bug]

  #1141 commented on Mar 22, 2025 • 0 new comments

• [Bug] lifecycle prestophook not available in helm chart values.yaml

  #1103 commented on Mar 22, 2025 • 0 new comments

• [Bug] Kuberay can not start when enableBatchScheduler=true

  #2430 commented on Mar 22, 2025 • 0 new comments

• [Bug] KubeRay Worker group pod keeps restarting on EKS - Fails to CrashLoopBackOff

  #2420 commented on Mar 22, 2025 • 0 new comments

• [Bug] Grafana DashBoard is too old

  #2400 commented on Mar 22, 2025 • 0 new comments

• [Bug] Bubble ImagePullErr and ImagePullBackoff to the Ray CRD

  #2387 commented on Mar 22, 2025 • 0 new comments

• [Bug] Autoscaler sideacr crashes, bringing down head pod, if request exceeds max pod replicas

  #2385 commented on Mar 22, 2025 • 0 new comments

• [Bug] Old RayServices not deleted after operator update to 1.2.1

  #2374 commented on Mar 22, 2025 • 0 new comments

• [Bug] RayJob does not shut down the submitter pod properly

  #2359 commented on Mar 22, 2025 • 0 new comments

• [Bug] RayJob can have JobStatus = Running while JobDeploymentStatus = Failed

  #2314 commented on Mar 22, 2025 • 0 new comments

• [Bug] Readiness and liveness probes failing when applying ray-service.sample.yaml file

  #2269 commented on Mar 22, 2025 • 0 new comments

• [Bug] What's minimum permission set for kuberay-operator?

  #2213 commented on Mar 22, 2025 • 0 new comments

• [Bug] "enable-batch-scheduler" bool flag is not working for schedulers other than Volcano

  #2185 commented on Mar 22, 2025 • 0 new comments

• [Bug] Readiness probe failed: timeout on minikube

  #2158 commented on Mar 22, 2025 • 0 new comments

• [Bug] [API Server] Can't specify cluster rayVersion in Ray Job

  #2109 commented on Mar 22, 2025 • 0 new comments

• [Bug] What's the relationship between watching `Endpoints` and RayService e2e tests?

  #2085 commented on Mar 22, 2025 • 0 new comments

• [Bug/RayJob] k8sjob doesn't contain ray worker logs with ray.shutdown() at the end

  #1975 commented on Mar 22, 2025 • 0 new comments

• [Bug] Ray cluster terminates more worker pods than the amount of replica scale down requested

  #1936 commented on Mar 22, 2025 • 0 new comments

• [Umbrella] Remove the RayCluster `.Status.State` field

  #2299 commented on Mar 22, 2025 • 0 new comments

• [Feature] Support get k8s pod log for interactive RayJob

  #2701 commented on Mar 22, 2025 • 0 new comments

• [Feature] Make `suspend` API for worker group atomic

  #2693 commented on Mar 22, 2025 • 0 new comments

• [Umbrella] `Suspend` API for worker groups

  #2692 commented on Mar 22, 2025 • 0 new comments

• [Feature] Emit metrics of cluster creation and other related metrics for observability

  #2681 commented on Mar 22, 2025 • 0 new comments

• [Feature] Don't submit scale requests if the worker group is suspended

  #2666 commented on Mar 22, 2025 • 0 new comments

• [Feature] Run Ginkgo tests with Ginkgo CLI

  #2658 commented on Mar 22, 2025 • 0 new comments

• Replace references of rayproject/ray-ml image

  #2292 commented on Mar 22, 2025 • 0 new comments

• [Feature] [API Server] Support `activeDeadlineSeconds` in API Server RayJob resource

  #2278 commented on Mar 22, 2025 • 0 new comments

• [Discuss][Feature] Authenticate requests sent to Ray Serve proxy actors

  #2276 commented on Mar 22, 2025 • 0 new comments

• [Feature] Remove wait-for-gcs init container

  #2275 commented on Mar 22, 2025 • 0 new comments

• [Feature] Support extended kube-scheduler as batch scheduler

  #2052 commented on Mar 22, 2025 • 0 new comments

• [Feature] [API Server] Allow autoscaling in python api server client

  #2029 commented on Mar 22, 2025 • 0 new comments

• [CI] RayJob `ray-job.shutdown.yaml` sample YAML test is flaky

  #2979 commented on Mar 24, 2025 • 0 new comments

• [CI] `TestAutoscalingRayService` is flaky

  #2981 commented on Mar 24, 2025 • 0 new comments

• [Doc] Explain the difference between head svc and serve of RayService

  #2995 commented on Mar 24, 2025 • 0 new comments

• [Bug] Minor inconsistency in RayJob submitter retries in v1.3

  #3211 commented on Mar 26, 2025 • 0 new comments

• [Feature][metrics] ray_services_ready_duration_seconds

  #3177 commented on Mar 26, 2025 • 0 new comments

• [Feature][metrics] ray_services_created_total

  #3176 commented on Mar 26, 2025 • 0 new comments

• [Bug] RayService applications do not load on v1.3.0

  #3133 commented on Mar 26, 2025 • 0 new comments

• [Discussion] KubeRay and Ray version compatibilities

  #187 commented on Mar 26, 2025 • 0 new comments

• [Epic][Feature] KubeRay v1.4.0 - Operator SLI Tracking

  #3171 commented on Mar 26, 2025 • 0 new comments

• [Umbrella] Autoscaler improvements

  #2600 commented on Mar 26, 2025 • 0 new comments

• [Bug] Compilation fails for apiserver

  #3061 commented on Mar 26, 2025 • 0 new comments

• [Bug] Minimum CPU and Memory requirements for KubeRay Head and worker pods

  #2186 commented on Mar 26, 2025 • 0 new comments

• [Feature] I want to be able to add annotations to the ServiceAccount that RayCluster creates

  #2322 commented on Mar 26, 2025 • 0 new comments

• [Bug] autoscaler doesn't launch larger nodes for pending tasks

  #3214 commented on Mar 26, 2025 • 0 new comments

• [Bug] [API Server] JobSubmission service does not work for cluster names >41 characters

  #2169 commented on Mar 27, 2025 • 0 new comments

• [release] Add doc tests for important KubeRay docs

  #3157 commented on Mar 28, 2025 • 0 new comments

• [RayService] Support Incremental Zero-Downtime Upgrades

  #3166 commented on Mar 25, 2025 • 0 new comments

• [Prometheus] Add `ray_clusters_created_total` metric

  #3204 commented on Mar 25, 2025 • 0 new comments

• [Feature] Add `DeleteWorkersOnFailure` deletion policy for RayJob

  #2765 commented on Mar 22, 2025 • 0 new comments

• [Feature] Check Redis cleanup in Golang e2e tests

  #2764 commented on Mar 22, 2025 • 0 new comments

• [Feature] Print out errors or upload logs as artifacts when kubectl plugin e2e test error occurs

  #2753 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add `force` to worker group `suspend` API

  #2744 commented on Mar 22, 2025 • 0 new comments

• [Feature] Allow setting `app.kubernetes.io/name` in head-svc created by RayService

  #2648 commented on Mar 22, 2025 • 0 new comments

• [Feature] [kuberay/apiserver] adding custom init container configuration to worker container specs

  #2623 commented on Mar 22, 2025 • 0 new comments

• [Feature] Support preserving only the Head pod with RayJob when shutdownAfterJobFinishes=true

  #2615 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add kubectl plugin to KubeRay repo README

  #2603 commented on Mar 22, 2025 • 0 new comments

• [auth] Support auth for RayJob / RayService

  #2586 commented on Mar 22, 2025 • 0 new comments

• [auth] RBAC to support both Autoscaler and Auth

  #2585 commented on Mar 22, 2025 • 0 new comments

• [Feature] Configure dashboard readiness probe based on dashboard-port in rayStartParams

  #2584 commented on Mar 22, 2025 • 0 new comments

• [Feature] How to propagate the labels from CR to K8s built-in resources?

  #2583 commented on Mar 22, 2025 • 0 new comments

• [Umbrella] KubeRay Authentication Support

  #2581 commented on Mar 22, 2025 • 0 new comments

• [Feature] Why KubeRay image becomes larger?

  #2580 commented on Mar 22, 2025 • 0 new comments

• [Feature] Integrate KubeRay with Koordinator

  #2573 commented on Mar 22, 2025 • 0 new comments

• [Feature] ENABLE_GCS_FT_REDIS_CLEANUP should be an field in RayService

  #2571 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add E2E Test for Autoscaler Nested Remote Functions

  #2568 commented on Mar 22, 2025 • 0 new comments

• [Chore] Golang import grouping by package

  #2567 commented on Mar 22, 2025 • 0 new comments

• [Umbrella] Controller Expectation

  #2566 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add e2e Ray v2 Autoscaler Tests with KubeRay

  #2561 commented on Mar 22, 2025 • 0 new comments

• [Feature] Light-weight job submitter

  #2537 commented on Mar 22, 2025 • 0 new comments

• [Feature] Identify and apply changes on ray-cluster

  #2534 commented on Mar 22, 2025 • 0 new comments

• [Umbrella] Multi-tenant observability

  #2526 commented on Mar 22, 2025 • 0 new comments

• [Feature] Deploy Ray Serve replicas across multiple K8s nodes / regions / zones to achieve HA

  #2500 commented on Mar 22, 2025 • 0 new comments

• Statistics of other types of gpu

  #2484 commented on Mar 22, 2025 • 0 new comments

• [Feature] Ray Cluster: Preserving Job State After Cluster Restart

  #2479 commented on Mar 22, 2025 • 0 new comments

• [Feature] Support cron scheduling for RayJob

  #2426 commented on Mar 22, 2025 • 0 new comments

• [Benchmark] Summarize the benchmark results and experiment setup

  #2402 commented on Mar 22, 2025 • 0 new comments

• [Feature] Consider setting AUTOSCALER_CONSERVE_GPU_NODES by default in Ray autoscaler

  #2381 commented on Mar 22, 2025 • 0 new comments

• [Feature] Remove `//nolint:gosec` to allow rule G115 after the false positive issue is solved

  #2369 commented on Mar 22, 2025 • 0 new comments

• [Doc] KubeRay configuration

  #2356 commented on Mar 22, 2025 • 0 new comments

• [RFC] Introduce new API-RayCluster Fleet and ReplicaSet in KubeRay

  #2323 commented on Mar 22, 2025 • 0 new comments

• [Feature] Api Server Support API: Run Ray Service on the cluster which is created already

  #1646 commented on Mar 22, 2025 • 0 new comments

• [Feature] Schedule RayJob

  #1662 commented on Mar 22, 2025 • 0 new comments

• [Feature] [ApiServer] Publish API Server Python client to PyPi

  #1708 commented on Mar 22, 2025 • 0 new comments

• [Feature] Finalizer to block deletion of RayCluster with running jobs

  #1740 commented on Mar 22, 2025 • 0 new comments

• [Feature] rayStartParams UX improvement

  #1758 commented on Mar 22, 2025 • 0 new comments

• [RayService] Refactor to unify cluster decision for active and pending RayClusters

  #1761 commented on Mar 22, 2025 • 0 new comments

• [RayService] [Enhancement] Avoid unnecessary pod deletion when updating RayCluster

  #1769 commented on Mar 22, 2025 • 0 new comments

• [Feature] Create `serve-svc` before all apps are running

  #1803 commented on Mar 22, 2025 • 0 new comments

• [Feature] [CLI] Add arm64 arch build for kuberay and output should ignore by git

  #1819 commented on Mar 22, 2025 • 0 new comments

• [Feature] Use ServiceAccount token when submitting RayJobs

  #1824 commented on Mar 22, 2025 • 0 new comments

• [Test] Add RayJob + Kueue e2e tests

  #1846 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add default volumes and volumeMounts for `/tmp/ray` to Ray Pods

  #1851 commented on Mar 22, 2025 • 0 new comments

• [Docs] Add docs for structured config and default sidecar containers

  #1869 commented on Mar 22, 2025 • 0 new comments

• [Feature] Support AWS IAM for Redis Auth

  #1888 commented on Mar 22, 2025 • 0 new comments

• [Doc] RayJob `suspend`

  #1900 commented on Mar 22, 2025 • 0 new comments

• [Doc] Create a doc for external Redis with TLS

  #1921 commented on Mar 22, 2025 • 0 new comments

• [Discussion][GCS FT] Delete Pods by the value of restartCount

  #1356 commented on Mar 22, 2025 • 0 new comments

• [Feature] publishNotReadyAddresses for services

  #1365 commented on Mar 22, 2025 • 0 new comments

• [Doc] Create a doc for #1386

  #1393 commented on Mar 22, 2025 • 0 new comments

• [Feature] Extract autoscaler as separate deployment per raycluster

  #1416 commented on Mar 22, 2025 • 0 new comments

• [Feature][GCS FT] Test the behavior of `cleanup_redis_storage`

  #1422 commented on Mar 22, 2025 • 0 new comments

• [Feature] Improve the Kustomization deployment

  #1434 commented on Mar 22, 2025 • 0 new comments

• [Feature] Allow for specifying options when creating ingresses

  #1436 commented on Mar 22, 2025 • 0 new comments

• Enable usage of Ray Serve when using RayCluster CRD

  #1451 commented on Mar 22, 2025 • 0 new comments

• [Feature] [api-server] API documentation revamp

  #1455 commented on Mar 22, 2025 • 0 new comments

• [Feature] Support networking.k8s.io.IngressSpec as ingress config for ray head

  #1475 commented on Mar 22, 2025 • 0 new comments

• [CI] Test the compatibility between KubeRay / Ray

  #1499 commented on Mar 22, 2025 • 0 new comments

• [Feature] Serve gRPC health check

  #1554 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add a doc and YAML for Ray Serve grpc support

  #1555 commented on Mar 22, 2025 • 0 new comments

• [Feature] Configurable RayCluster readiness definition

  #1631 commented on Mar 22, 2025 • 0 new comments

• [Feature] Move kuberay-helm to KubeRay repo

  #1635 commented on Mar 22, 2025 • 0 new comments

• [Feature] Aggregate roles on Kuberay resources to Kubernetes user-facing roles

  #1641 commented on Mar 22, 2025 • 0 new comments

• [Feature] Should we stop publishing images on DockerHub?

  #1924 commented on Mar 22, 2025 • 0 new comments

• [Feature][metrics] ray_cluster_provisioned_duration_seconds

  #3172 commented on Mar 21, 2025 • 0 new comments

• [release] Automate the Helm chart release process

  #3162 commented on Mar 21, 2025 • 0 new comments

• [release] Port KubeRay CI tests to Ray CI

  #3158 commented on Mar 21, 2025 • 0 new comments

• [Umbrella] Optimize release process

  #3098 commented on Mar 21, 2025 • 0 new comments

• [Bug] Unable to deploy vLLM with llama 3 8B on Red Hat Openshift

  #3164 commented on Mar 21, 2025 • 0 new comments

• [Bug] Ray can not schedule the workload when there is enough resource

  #3151 commented on Mar 21, 2025 • 0 new comments

• [Bug] kubectl plugin not detecting KubeRay on GKE RayOperator Add-On

  #3115 commented on Mar 21, 2025 • 0 new comments

• [Feature][metrics] ray_jobs_terminated_total

  #3180 commented on Mar 21, 2025 • 0 new comments

• [Feature][metrics] ray_clusters_created_total

  #3175 commented on Mar 21, 2025 • 0 new comments

• [Feature][metrics] ray_job_deployment_status

  #3179 commented on Mar 21, 2025 • 0 new comments

• [Feature][metrics] ray_job_execution_duration_seconds

  #3181 commented on Mar 21, 2025 • 0 new comments

• [Feature] Add ability to add a prestop hook

  #3185 commented on Mar 21, 2025 • 0 new comments

• [doc] Update kubectl plugin doc for `scale` command

  #3190 commented on Mar 21, 2025 • 0 new comments

• [Feature][metrics] ray_job_submitter_pod_startup_duration_seconds

  #3201 commented on Mar 21, 2025 • 0 new comments

• [Feature] RayService Incremental Upgrade Project Tracker

  #3209 commented on Mar 21, 2025 • 0 new comments

• [Feature] Revisit RBAC

  #3213 commented on Mar 21, 2025 • 0 new comments

• Improve logging configuration experience

  #1940 commented on Mar 22, 2025 • 0 new comments

• [Feature] Create a Script to Manually Clean up Redis

  #1941 commented on Mar 22, 2025 • 0 new comments

• [Feature] Set appProtocol correctly for head service

  #1946 commented on Mar 21, 2025 • 0 new comments

• [Feature] Support dynamic refresh of watched namespaces

  #2061 commented on Mar 21, 2025 • 0 new comments

• [Feature] KubeRay Scalability Benchmarking

  #2069 commented on Mar 21, 2025 • 0 new comments

• [Feature] [API Server] [RFC] Add persistence for job history using a SQL database

  #2114 commented on Mar 21, 2025 • 0 new comments

• [Feature] RayService CRD to have ImagePullSecret Reference

  #2137 commented on Mar 21, 2025 • 0 new comments

• FT GCS should handle draining of node where head pod is scheduled

  #2153 commented on Mar 21, 2025 • 0 new comments

• [Feature] Should we also set PublishNotReadyAddresses if the service is not headless?

  #2157 commented on Mar 21, 2025 • 0 new comments

• [Umbrella] Ray Autoscaling tests

  #2173 commented on Mar 21, 2025 • 0 new comments

• [Doc] Release schedule

  #2191 commented on Mar 21, 2025 • 0 new comments

• Use login shell for Submitter Pod command

  #2209 commented on Mar 21, 2025 • 0 new comments

• [Feature] Add API reference documentation for KubeRay custom resources

  #2247 commented on Mar 21, 2025 • 0 new comments

• [CI] CI doesn't detect whether a PR runs `make generate` or not

  #3091 commented on Mar 21, 2025 • 0 new comments

• [Bug] RayCluster K8s event for creating worker Pods

  #3056 commented on Mar 21, 2025 • 0 new comments

• [Feature][metrics] ray_jobs_created_total

  #3178 commented on Mar 21, 2025 • 0 new comments

• [Feature] raycluster created by rayservice controller may need some default labels?

  #412 commented on Mar 22, 2025 • 0 new comments

• [Feature] use kubernetes events to trigger autoscaling vs directly patching the CR

  #442 commented on Mar 22, 2025 • 0 new comments

• [Feature] add ut for job spec in api server

  #444 commented on Mar 22, 2025 • 0 new comments

• [Feature] [Documentation] Document state and status fields

  #454 commented on Mar 22, 2025 • 0 new comments

• [Feature] Enable replicated KubeRay operator deployments

  #474 commented on Mar 22, 2025 • 0 new comments

• [Feature] [Test] KubeRay vs. Kubernetes version compatibility testing

  #516 commented on Mar 22, 2025 • 0 new comments

• [Feature][RayService] Handle serve deployment delete during the cluster destroy.

  #647 commented on Mar 22, 2025 • 0 new comments

• integration test to stop ray job when delete job custom resource or timeout is triggered

  #664 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add a stale bot

  #681 commented on Mar 22, 2025 • 0 new comments

• [Feature] Python linting

  #682 commented on Mar 22, 2025 • 0 new comments

• [Feature] Streaming data from local machine to cluster

  #684 commented on Mar 22, 2025 • 0 new comments

• [Discussion][Feature] Clean up workersToDelete?

  #733 commented on Mar 22, 2025 • 0 new comments

• [Feature] [Test] E2E release tests on cloud K8s infra

  #737 commented on Mar 22, 2025 • 0 new comments

• [Feature] [Docs] Explain how to use Ray custom resources with examples

  #742 commented on Mar 22, 2025 • 0 new comments

• [Feature][Docs][Discussion] Provider consistent guidance on resource Request and Limits

  #744 commented on Mar 22, 2025 • 0 new comments

• [Feature] add ServiceAccountName

  #746 commented on Mar 22, 2025 • 0 new comments

• Create a table for YAML and who refers to it

  #3050 commented on Mar 22, 2025 • 0 new comments

• [Doc] Rewrite GPU training doc

  #3049 commented on Mar 22, 2025 • 0 new comments

• Use uv in sample YAML files

  #3039 commented on Mar 22, 2025 • 0 new comments

• [Feature] Please publish the api as distinct go module

  #3020 commented on Mar 22, 2025 • 0 new comments

• [Feature] Throw an error message on applying if redis password is set in GcsFaultToleranceOptions and rayStartParams both

  #2986 commented on Mar 22, 2025 • 0 new comments

• [Feature] Support to generate PersistentVolumeClaim for Pod

  #59 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add affinity and toleration in ComputeTemplate

  #177 commented on Mar 22, 2025 • 0 new comments

• [Feature] Enrich Github Issue Templates

  #186 commented on Mar 22, 2025 • 0 new comments

• [Feature] Revisit share memory setting for Ray Cluster

  #201 commented on Mar 22, 2025 • 0 new comments

• [Feature] [apiserver] support custom resource in ComputeTemplate

  #207 commented on Mar 22, 2025 • 0 new comments

• [Feature] Authn and Authz support for KubeRay API server and CLI

  #263 commented on Mar 22, 2025 • 0 new comments

• [Feature] [Docs] Document workflows for deploying KubeRay with GitOps systems (ArgoCD, Flux)

  #273 commented on Mar 22, 2025 • 0 new comments

• [Feature] CI test for single namespace

  #282 commented on Mar 22, 2025 • 0 new comments

• [Feature] Ray remote kernels

  #300 commented on Mar 22, 2025 • 0 new comments

• [Feature] Make a one click deployment of Kuberay by using EKS Blueprint or a eksctl example

  #301 commented on Mar 22, 2025 • 0 new comments

• [Feature] Pagination support for ListClusters

  #312 commented on Mar 22, 2025 • 0 new comments

• [Feature] Sort randomPodsToDelete in descending order of timestamp

  #747 commented on Mar 22, 2025 • 0 new comments

• [Feature] Publish Helm chart for RayService

  #1085 commented on Mar 22, 2025 • 0 new comments

• [Feature] Expose Ray --logging-level parameter

  #1098 commented on Mar 22, 2025 • 0 new comments

• [DOC] Add tips to use ray serve cli to check the kuberay cluster remotely.

  #1124 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add LLM examples using KubeRay

  #1131 commented on Mar 22, 2025 • 0 new comments

• How does ray cluster support third-party resources besides GPU and CPU

  #1135 commented on Mar 22, 2025 • 0 new comments

• [Feature][Discussion] RayCluster restart policy

  #1174 commented on Mar 22, 2025 • 0 new comments

• [RayJob] Add `RayCronJob` Custom Resource Definition

  #1206 commented on Mar 22, 2025 • 0 new comments

• [Feature] Use the observedGeneration to determine whether to update the status or not.

  #1217 commented on Mar 22, 2025 • 0 new comments

• [Refactor] Separate the logic for retrieving Serve application statuses and checking Serve application statuses

  #1234 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add warning messages for instances where CPUs aren't specified when using Autoscaler.

  #1254 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add kubeconform to CI pipeline

  #1267 commented on Mar 22, 2025 • 0 new comments

• Add To Kuberay Metrics

  #1272 commented on Mar 22, 2025 • 0 new comments

• [Doc] Add guidelines for setting up ALB ingress for RayService

  #1273 commented on Mar 22, 2025 • 0 new comments

• [Feature] Avoid manually importing Grafana dashboard JSON file

  #1285 commented on Mar 22, 2025 • 0 new comments

• [Enhancement] [Doc] Link shutdown sample YAML in RayJob doc

  #1302 commented on Mar 22, 2025 • 0 new comments

• Contribute Tutorial for model training/fine-tuning using KubeRay with CodeFlare and ODH stack.

  #1328 commented on Mar 22, 2025 • 0 new comments

• [CI][Release] Run CI for PRs against release branch

  #799 commented on Mar 22, 2025 • 0 new comments

• [Feature] Connect to RayCluster via GCS port rather than Ray client in compatibility test

  #848 commented on Mar 22, 2025 • 0 new comments

• [Feature] Autoscaler to support scaling from zero based on `custom_resource` or `accelerator_type`

  #863 commented on Mar 22, 2025 • 0 new comments

• [Feature] The raycluster-autoscaler serviceaccount inherit default serviceaccount imagePullSecrets

  #892 commented on Mar 22, 2025 • 0 new comments

• [Feature] Nightly helm releases

  #910 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add tutorial to explain how to collect metrics from operator for Prometheus

  #921 commented on Mar 22, 2025 • 0 new comments

• [Feature] Provide KubeRay backend implementation for SkyPilot

  #930 commented on Mar 22, 2025 • 0 new comments

• [Feature] Reconcile workers when volumeMount differs

  #944 commented on Mar 22, 2025 • 0 new comments

• [Feature][Doc] Troubleshooting for networking

  #955 commented on Mar 22, 2025 • 0 new comments

• [Feature] Add document for environment_variables and Audit all environment variables to identify which should not be modified by users.

  #957 commented on Mar 22, 2025 • 0 new comments

• [Feature] Automatically expire GCS FT Redis keys

  #959 commented on Mar 22, 2025 • 0 new comments

• [Feature] Doc about network policies

  #961 commented on Mar 22, 2025 • 0 new comments

• [Feature][Doc] Revisit the definitions for ports

  #964 commented on Mar 22, 2025 • 0 new comments

• [Feature] Ensure the number of healthy workers while keep the abnormal worker for troubleshooting

  #1022 commented on Mar 22, 2025 • 0 new comments

• [Umbrella] GCS fault tolerance on KubeRay

  #1033 commented on Mar 22, 2025 • 0 new comments

• Adding support for environment variables, labels, annotations, and toleraritions

  #1067 commented on Mar 22, 2025 • 0 new comments