ray-project/kuberay issues and pull requests

#3872 - [Bug] Restore random head pod name when Ray < 2.48 and Autoscaler v2 and GCS FT are enabled

Pull Request - State: open - Opened by machichima 17 days ago

#3837 - Get rid of wget dependency

Issue - State: open - Opened by jjyao about 1 month ago - 4 comments
Labels: 1.5.0

#3836 - Feature/cron scheduling rayjob 2426

Pull Request - State: open - Opened by DW-Han about 1 month ago

#3835 - Use Go 1.24.0 in go module

Pull Request - State: open - Opened by tenzen-y about 1 month ago

#3834 - Avoid requiring specific Go patch version in go module

Issue - State: open - Opened by tenzen-y about 1 month ago - 1 comment

#3833 - Add RayCluster YAML for verl example

Pull Request - State: closed - Opened by kevin85421 about 1 month ago

#3832 - [Feature] [kubectl-plugin] Improve support for autoscaling clusters

Issue - State: open - Opened by jleben about 1 month ago - 2 comments
Labels: enhancement, triage

#3831 - [feat][operator] validate Ray resource metadata in webhook

Pull Request - State: open - Opened by davidxia about 1 month ago

#3830 - [feat][python-client]: add rayjob support to kuberay python-client

Pull Request - State: open - Opened by kryanbeane about 1 month ago

#3829 - [Feature] Add RayJob support to Kuberay python-client

Issue - State: open - Opened by kryanbeane about 1 month ago - 1 comment
Labels: enhancement, triage

#3828 - [cherry-pick] Cherry-pick #3826 into release-1.4 branch

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago

#3827 - [Bug] Exiting because this node manager has mistakenly been marked as dead by the GCS

Issue - State: closed - Opened by hobin2017 about 1 month ago
Labels: bug, triage

#3826 - Fix ray nightly image env var setup

Pull Request - State: closed - Opened by dayshah about 1 month ago - 1 comment

#3825 - [Test][Release] Change upgrade test version to test upgrade from 1.3.2 to 1.4.0

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago - 1 comment

#3824 - [release] Update upgrade test during release process

Issue - State: closed - Opened by kevin85421 about 1 month ago
Labels: enhancement, release-blocker, 1.4.0

#3823 - [Fix][Release] Fix Krew release indenetation error

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago - 2 comments

#3822 - KubeRay v1.4.0 default non-login BASH shell issue tracking

Issue - State: open - Opened by MortalHappiness about 1 month ago

#3821 - [kubectl-plugin] Remove ephemeral storage check

Pull Request - State: open - Opened by win5923 about 1 month ago

#3820 - [Feature] Add prometheus metrics reset support

Issue - State: open - Opened by win5923 about 1 month ago
Labels: enhancement, triage

#3819 - [Chore] Remove CHANGELOG.md

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago

#3818 - [Fix] changelog-generator.py failed to parse some commit messages

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago - 3 comments

#3817 - [Doc] Update RayJob Quick Start Job When V1.5 Release

Issue - State: open - Opened by weizhaowz about 1 month ago - 1 comment
Labels: 1.5.0

#3816 - [Release] Update KubeRay version references for 1.4.0

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago

#3815 - [kubeclt-plugin] use solid value as default value in get and create

Pull Request - State: open - Opened by fscnick about 1 month ago - 1 comment

#3814 - [cherry-pick] Cherry-pick #3809 into release-1.4 branch

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago

#3813 - [cherry-pick] Cherry-pick #3804 into release-1.4 branch

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago

#3812 - [apiserver] Add migration doc from v1 to v2

Pull Request - State: open - Opened by nadongjun about 1 month ago

#3811 - [Fix] kubectl ray create cluster config file CPU overwrites the whole resource requests and limits

Pull Request - State: open - Opened by CheyuWu about 1 month ago

#3810 - [Cherry-pick][Helm Chart] Set honorLabel of serviceMonitor to true (#3805)

Pull Request - State: closed - Opened by owenowenisme about 1 month ago - 1 comment

#3809 - [kubeclt-plugin] fix get cluster all namespace

Pull Request - State: closed - Opened by fscnick about 1 month ago - 1 comment

#3808 - [Bug] Add default value for entrypoint flags in job_submit.go

Pull Request - State: open - Opened by 400Ping about 1 month ago - 1 comment

#3807 - [Bug] Wrong default value for entrypoint flags in job_submit.go

Issue - State: open - Opened by 400Ping about 1 month ago

#3806 - [Bug] kubectl-plugin get cluster without all-namespace still shows all-namespace.

Issue - State: closed - Opened by fscnick about 1 month ago
Labels: bug, release-blocker, 1.4.0

#3805 - [Helm Chart] Set honorLabel of serviceMonitor to `true`

Pull Request - State: closed - Opened by owenowenisme about 1 month ago - 1 comment
Labels: release-blocker, 1.4.0

#3804 - [Docs] Add kubectl plugin create cluster sample yaml config files

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago - 1 comment
Labels: release-blocker, 1.4.0

#3803 - [Bug] kubectl ray create cluster config file CPU overwrites the whole resource requests and limits

Issue - State: open - Opened by MortalHappiness about 1 month ago - 1 comment
Labels: bug, cli, 1.5.0

#3802 - [Doc] `ray job submit -- echo aaa; echo bbb` and `echo aaa` will be executed in the cluster but `echo bbb` will be executed in the submitter

Issue - State: open - Opened by kevin85421 about 1 month ago
Labels: enhancement, triage

#3801 - [Bug] Wrong default value for head and worker ray start params in create_cluster.go

Issue - State: open - Opened by MortalHappiness about 1 month ago - 1 comment
Labels: bug, good-first-issue, cli

#3800 - chore: reduce memory allocation on handling http response

Pull Request - State: open - Opened by fscnick about 1 month ago

#3799 - [Bug] RayCluster unavailable due to health probe failure

Issue - State: open - Opened by ChenYi015 about 1 month ago
Labels: bug, triage

#3798 - [cherry-pick] Cherry-pick #3795 into release-1.4 branch

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago

#3797 - [cherry-pick] Cherry-pick #3796 into release-1.4 branch

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago

#3796 - [Chore][Sample-yaml] Upgrade pytorch-lightning to 1.8.5 for `ray-job.pytorch-distributed-training.yaml`

Pull Request - State: closed - Opened by MortalHappiness about 1 month ago

#3795 - [Metrics] Remove serviceMonitor.yaml

Pull Request - State: closed - Opened by owenowenisme about 1 month ago - 1 comment

#3794 - Question: Autoscaler v1 vs v2 configuration and performance

Issue - State: open - Opened by testinfected about 2 months ago - 12 comments

#3793 - [Feature] Support `runtimeClassName` in values.yaml

Issue - State: open - Opened by unclebenel about 2 months ago
Labels: enhancement, triage

#3792 - [Test][Autoscaler] add fake single-host TPU tests

Pull Request - State: open - Opened by davidxia about 2 months ago

#3791 - [Bug] why readinessProbe port 8000 check

Issue - State: closed - Opened by Helion-z about 2 months ago - 8 comments
Labels: bug, triage

#3790 - Use ImplementationSpecific in ray-cluster.separate-ingress.yaml (#3781)

Pull Request - State: closed - Opened by MortalHappiness about 2 months ago

#3789 - [cherry-pick] Cherry-pick #3786 into release-1.4 branch

Pull Request - State: closed - Opened by MortalHappiness about 2 months ago

#3788 - [cherry-pick] Cherry-pick #3782 into release-1.4 branch

Pull Request - State: closed - Opened by MortalHappiness about 2 months ago

#3787 - [cherry-pick] Cherry-pick #3779 into release-1.4 branch

Pull Request - State: closed - Opened by MortalHappiness about 2 months ago

#3786 - Remove vLLM examples in favor of Ray Serve LLM

Pull Request - State: closed - Opened by kevin85421 about 2 months ago
Labels: release-blocker, 1.4.0

#3785 - pass client when call batchscheduler.New()

Pull Request - State: open - Opened by KunWuLuan about 2 months ago - 3 comments

#3784 - [Release] Update KubeRay version references for 1.4.0-rc.2

Pull Request - State: closed - Opened by MortalHappiness about 2 months ago

#3783 - [cherry-pick] Cherry-pick #3780 into release-1.4 branch

Pull Request - State: closed - Opened by MortalHappiness about 2 months ago

#3782 - Update update-ray-job.kueue-toy-sample.yaml

Pull Request - State: closed - Opened by troychiu about 2 months ago - 3 comments
Labels: release-blocker, 1.4.0

#3781 - Use ImplementationSpecific in ray-cluster.separate-ingress.yaml

Pull Request - State: closed - Opened by troychiu about 2 months ago - 3 comments
Labels: release-blocker, 1.4.0

#3780 - [Doc][Fix] correct the indention of storageClass in ray-cluster.persistent-redis.yaml

Pull Request - State: closed - Opened by rueian about 2 months ago - 3 comments

#3779 - [Feat] Add e2e test for applying `ray-job.interactive-mode.yaml`

Pull Request - State: closed - Opened by CheyuWu about 2 months ago - 4 comments

#3778 - [Feat] Add e2e test for applying `ray-job.interactive-mode.yaml`

Issue - State: closed - Opened by CheyuWu about 2 months ago
Labels: bug, triage

#3777 - [cherry-pick][doc] Improve APIServer v2 doc (#3773)

Pull Request - State: closed - Opened by kevin85421 about 2 months ago

#3776 - [Feature] gcsFaultToleranceOptions spec support in RayCluster Helm Chart

Issue - State: open - Opened by tplass-ias about 2 months ago
Labels: enhancement, triage

#3775 - [kubectl-plugin] Use a more Golang-native approach to retrieve the CR status for testing

Pull Request - State: open - Opened by machichima about 2 months ago

#3774 - [Release] Reset ray-operator version in root go.mod to v0.0.0

Pull Request - State: closed - Opened by MortalHappiness about 2 months ago - 1 comment

#3773 - [doc] Improve APIServer v2 doc

Pull Request - State: closed - Opened by kevin85421 about 2 months ago
Labels: release-blocker, 1.4.0

#3772 - [cherry-pick] Cherry-pick #3771 into release-1.4

Pull Request - State: closed - Opened by MortalHappiness about 2 months ago

#3771 - Revert "Fix issue where unescaped semicolons caused task execution failures. (#3691)"

Pull Request - State: closed - Opened by MortalHappiness about 2 months ago - 2 comments

#3769 - [scheduler-plugins] Support second scheduler mode

Issue - State: open - Opened by kevin85421 about 2 months ago - 1 comment
Labels: enhancement, raycluster, 1.5.0, scheduling

#3764 - [Bug] `ray-job.use-existing-raycluster.yaml` entrypoint error

Issue - State: closed - Opened by MortalHappiness about 2 months ago - 8 comments
Labels: bug, operator, P0, release-blocker, 1.4.0

#3763 - [Doc] Reference helm chart version in `helm-chart/kuberay-operator/README.md.gotmpl` with go template

Pull Request - State: closed - Opened by MortalHappiness about 2 months ago - 1 comment

#3746 - [scheduler-plugins] Kuberay should pass client when call batchscheduler.New()

Issue - State: open - Opened by KunWuLuan about 2 months ago - 2 comments
Labels: raycluster, 1.5.0, scheduling

#3741 - [Bug] ENABLE_RAY_HEAD_CLUSTER_IP_SERVICE set true does not create Ray Head with CLUSTER-IP set as expected

Issue - State: closed - Opened by amholler about 2 months ago - 1 comment
Labels: bug, triage

#3740 - [RayService][Test] create curl pod waiting until running

Pull Request - State: open - Opened by fscnick about 2 months ago

#3736 - test: enable upgrade to image built from source

Pull Request - State: open - Opened by pawelpaszki about 2 months ago - 2 comments

#3731 - [RayJob] Support deletion policies based on job status

Pull Request - State: closed - Opened by weizhaowz 2 months ago - 8 comments

#3714 - [Feature] Update RayJob DeletionPolicy API to differentiate success/failure scenarios

Issue - State: open - Opened by andrewsykim 2 months ago - 14 comments
Labels: enhancement, triage

#3703 - [Helm] Add priorityClassName for kuberay-operator chart

Pull Request - State: open - Opened by win5923 2 months ago - 1 comment

#3659 - [Feature] Provide a better experience to manage the KubeRay with ArgoCD and GitOps

Issue - State: open - Opened by cmontemuino 2 months ago - 12 comments
Labels: enhancement, triage, 1.5.0

#3643 - [Grafana] Allow auto-load dashboard jsons

Pull Request - State: open - Opened by owenowenisme 2 months ago - 1 comment

#3642 - [ray-operator][Bug] Rayjob is Failed or Succeed, but Raycluster status(jobDeploymentStatus) is still Running(#3553)

Pull Request - State: open - Opened by dushulin 2 months ago

#3627 - [Feature] [kubectl-plugin] Expose setting `shutdownAfterJobFinishes` and `ttlSecondsAfterFinished` in ray job submit

Pull Request - State: open - Opened by CheyuWu 3 months ago - 2 comments

GitHub / ray-project/kuberay issues and pull requests