1 Release published by 1 person

• ray-2.47.0 Ray-2.47.0

  published Jun 12, 2025

99 Pull requests merged by 47 people

• [serve] Set route_prefix and docs_path when re-deploying app

  #53753 merged Jun 14, 2025

• [cherry-pick][dashboard] Fix retrieving IP address from the GPUProfilingManager on the dashboard agent

  #53817 merged Jun 14, 2025

• Add tpu usage metrics to reporter_agent

  #53678 merged Jun 14, 2025

• [data] Refactor interface for actor_pool_map_operator

  #53752 merged Jun 13, 2025

• ray-llm container cu124 -> cu128 update

  #53730 merged Jun 13, 2025

• [dashboard] Fix retrieving IP address from the GPUProfilingManager on the dashboard agent

  #53807 merged Jun 13, 2025

• [ci/release] Trigger Ray release by running a Bazel binary

  #52962 merged Jun 13, 2025

• version change for 2.47.1

  #53813 merged Jun 13, 2025

• cherrypick #53671

  #53812 merged Jun 13, 2025

• [core] Move dependencies of NodeManger to main.cc for better testability

  #53782 merged Jun 13, 2025

• [core] Deflake test_object_spilling.py

  #53803 merged Jun 13, 2025

• [core] Deflake test_state_api.py

  #53804 merged Jun 13, 2025

• [tune] update BlockMetadata args in tests

  #53791 merged Jun 13, 2025

• [serve] Fix autoscaling metrics

  #53778 merged Jun 13, 2025

• pass route prefix to replica

  #53777 merged Jun 13, 2025

• [Serve] Call shared long poll client router registration in event loop

  #53613 merged Jun 13, 2025

• [core] Add timeout to ray.get call in test_update_object_location_batch_failure

  #53805 merged Jun 13, 2025

• [RLlib] Fix device check in Learner.

  #53706 merged Jun 13, 2025

• [core] Deflake test_client_builder.py

  #53774 merged Jun 13, 2025

• [core] Increase instance sizes for wheel / HA tests

  #53783 merged Jun 13, 2025

• [serve.llm] Organize spread out utils.py

  #53722 merged Jun 13, 2025

• [Doc] Added ray-serve llm doc

  #52832 merged Jun 12, 2025

• Remove Schema From BlockMetadata

  #53454 merged Jun 12, 2025

• [Core] Exit the Core Worker Early Error Received from Plasma Store

  #53679 merged Jun 12, 2025

• fix: WandbLogger crashing silently on a FileNotFoundError

  #50308 merged Jun 12, 2025

• [Serve] feat: make ray.serve.batch concurrent

  #53096 merged Jun 12, 2025

• [RLlib] Add ability to compute percentiles to MetricsLogger/Stats

  #52963 merged Jun 12, 2025

• [core][telemetry] move the open telemetry tests into a pytest module

  #53751 merged Jun 12, 2025

• [core] Speed up test_actor_advanced.py

  #53738 merged Jun 12, 2025

• Replace miniconda3 with miniforge

  #53436 merged Jun 12, 2025

• [serve] Improve test_metrics

  #53747 merged Jun 12, 2025

• update codeowners for ray serve

  #53717 merged Jun 11, 2025

• [Tune][Air] Fix MLflowLoggerCallback to enable its use with PBT (#27783)

  #42182 merged Jun 11, 2025

• Fix vLLM batch test by changing to Pixtral

  #53744 merged Jun 11, 2025

• Fix uv tests on macos x86_64

  #53741 merged Jun 11, 2025

• [core][telemetry/07] support counter metric on worker side

  #53418 merged Jun 11, 2025

• [docker] Update latest Docker dependencies for 2.47.0 release

  #53749 merged Jun 11, 2025

• [docker] Update latest Docker dependencies for 2.47.0 release

  #53748 merged Jun 11, 2025

• [train] Raise error when calling ray.train.report with a gpu tensor

  #53725 merged Jun 11, 2025

• [serve] Increase httpx timeout to 30s for backpressure test

  #53711 merged Jun 11, 2025

• [ci] Move simulate_storage from _private/ to _common/

  #53735 merged Jun 11, 2025

• [core] Give io context concurrency hint

  #53642 merged Jun 11, 2025

• [serve] Remove dependency on ray._private.ray_constants.py

  #53700 merged Jun 11, 2025

• [core] Warning when creating actor with restarts and arguments in plasma

  #53713 merged Jun 11, 2025

• [Docs] [istio mtls] Add warning on sidecar OOM for mTLS

  #53385 merged Jun 11, 2025

• [cherry-pick][Docs] Added user-guide for Joins (#52987)

  #53712 merged Jun 11, 2025

• Add observability for label-selectors

  #53423 merged Jun 11, 2025

• [Doc] Bind the version of kuberay to v1.3.0 in related docs

  #53661 merged Jun 11, 2025

• [docs] fix link in gcp-gke-tpu-cluster.md

  #53708 merged Jun 11, 2025

• Add perf metrics for 2.47.0

  #53668 merged Jun 11, 2025

• [train][template] pytorch + train + data template uses absolute links

  #53718 merged Jun 11, 2025

• [train] add trace to WorkerHealthCheckFailedError

  #53626 merged Jun 11, 2025

• fix CI, wrong import path

  #53715 merged Jun 10, 2025

• [core][gpu-objects] Fix the performance regression by clearing object_ref for small and non-GPU objects

  #53692 merged Jun 10, 2025

• [Docs] Finalize time-series tutorial, add lockfiles

  #53710 merged Jun 10, 2025

• [core] remove dead open telemetry code

  #53709 merged Jun 10, 2025

• E2e rag

  #53703 merged Jun 10, 2025

• [core] Add single-controller API for ray.util.collective and torch gloo backend

  #53319 merged Jun 10, 2025

• [core] Migrate ray.private.pydantic_compat from _private to _common

  #53686 merged Jun 10, 2025

• [core][3/N] Avoid unnecessary deserialization/serialization of ParentTaskId

  #53695 merged Jun 10, 2025

• [core] Remove deprecated storage parameter to ray.init

  #53669 merged Jun 10, 2025

• [serve.llm] delete dead code from prompt format days

  #53621 merged Jun 10, 2025

• [core] Fix test_multi_tenancy.py on Windows

  #53699 merged Jun 10, 2025

• [core] Remove unused object_ref_seed parameter

  #53698 merged Jun 10, 2025

• [core] early exit spill if spilling config is empty

  #53193 merged Jun 10, 2025

• [ci] Fix crane auth issue for nightly multi arch tagging

  #53483 merged Jun 10, 2025

• handle task cancellation error

  #53680 merged Jun 10, 2025

• Code refactoring in proxy

  #53644 merged Jun 10, 2025

• [core] Migrate wait_for_condition and async_wait_for_condition from _private to _common

  #53652 merged Jun 10, 2025

• Convert cluster compute config in release test to Kuberay compute config

  #53681 merged Jun 10, 2025

• [Core] Vendor setproctitle

  #53471 merged Jun 10, 2025

• add back run on anyscale button

  #53688 merged Jun 10, 2025

• [Compiled Graph] Enhance Compile Graph with Multi-Device Support

  #53395 merged Jun 10, 2025

• BLD: Remove redundant manylinux1 related flag in .bazelrc

  #53549 merged Jun 10, 2025

• [train][template] Add Anyscale template for pytorch + train + data

  #53220 merged Jun 10, 2025

• [core] Remove deprecated ray start CLI options

  #53675 merged Jun 10, 2025

• [core] Speed up & deflake test_multitenancy.py

  #53674 merged Jun 10, 2025

• [ci] change macos intel platform to 12_0

  #53671 merged Jun 10, 2025

• [Docs] Create lockfiles for various e2e tutorials

  #53672 merged Jun 10, 2025

• [Docs] Adds second notebook to timeseries tutorial

  #53561 merged Jun 10, 2025

• [ci] fix misconfig on byod scripts

  #53682 merged Jun 10, 2025

• Fix map_batches release test back_to_back option

  #53664 merged Jun 9, 2025

• [ci] Resize some runtime_env tests

  #53670 merged Jun 9, 2025

• [core] Creating an interface ObjectManager's for GrpcClientManager.

  #53656 merged Jun 9, 2025

• [core][1/N] Avoid unnecessary deserialization/serialization of TaskId

  #53577 merged Jun 9, 2025

• make constant for x-request-id

  #53667 merged Jun 9, 2025

• Remove ray.workflow package

  #53612 merged Jun 9, 2025

• add vale to pre-commit

  #53564 merged Jun 9, 2025

• Add DataContext + LogicalOp Args to Dataset Export

  #53554 merged Jun 9, 2025

• [core] Skip test on mac build

  #53662 merged Jun 9, 2025

• [serve.llm] Prefix-aware scheduler [2/N] Configure PrefixAwareReplicaScheduler as default scheduler in LLMServer

  #52725 merged Jun 9, 2025

• [core] Make preloading Jemalloc configurable for worker

  #47243 merged Jun 9, 2025

• [llm] bump vllm to 0.9.0.1

  #53443 merged Jun 9, 2025

• uint8_t* data ptr not used.

  #47565 merged Jun 9, 2025

• [core][2/N] Avoid unnecessary deserialization/serialization of ObjectId

  #53574 merged Jun 9, 2025

• [ci] add more docker groups to work with buildkite amis

  #53640 merged Jun 9, 2025

• [ci] use new docker account for releasing

  #53646 merged Jun 8, 2025

• [core] Correctly fail worker lease request if a task becomes infeasible after scheduling

  #52295 merged Jun 8, 2025

• pin flashinfer-python to 0.2.5

  #53637 merged Jun 7, 2025

62 Pull requests opened by 41 people

• [Air] Add Video FPS Support for `WandbLoggerCallback`

  #53638 opened Jun 7, 2025

• [Serve] Check multiple FastAPI ingress deployments in a single application

  #53647 opened Jun 8, 2025

• [core]: Correct podman output parsing for image uri in runtime env

  #53653 opened Jun 9, 2025

• [core] Use core worker client pool in GCS

  #53654 opened Jun 9, 2025

• [core] Adding a nightly benchmark for continuous, bidirectional object transfer on two nodes.

  #53657 opened Jun 9, 2025

• [core] Release resources only after tasks have stopped executing

  #53660 opened Jun 9, 2025

• Train Tests: Add wrapper to run tests in a loop

  #53683 opened Jun 10, 2025

• [train] Cleanups for training ingest benchmark

  #53684 opened Jun 10, 2025

• [refactor] Install uv from test-requirements.txt

  #53685 opened Jun 10, 2025

• [data] allow max_calls to be a static but not dynamic option

  #53687 opened Jun 10, 2025

• [WIP] Remove old uv runtime env plugin

  #53690 opened Jun 10, 2025

• Bump requests from 2.32.3 to 2.32.4 in /python

  #53691 opened Jun 10, 2025

• [RLlib] Examples folder do-over (vol 53): Learning 2-agent cartpole with global observation, 1 policy outputting all agents' actions, and individual rewards.

  #53697 opened Jun 10, 2025

• [Data] Add reading from Delta Lake tables and from Unity Catalog

  #53701 opened Jun 10, 2025

• [RLlib; Offline RL] Implement Offline Policy Evaluation (OPE) via Importance Sampling.

  #53702 opened Jun 10, 2025

• [Serve][LLM] Simplify _prepare_engine_config()

  #53704 opened Jun 10, 2025

• [HashShuffle] - Add warnings for when there are insufficient resources for Aggregators

  #53705 opened Jun 10, 2025

• (serve.llm): Refactor/Consolidate LoRA downloading

  #53714 opened Jun 10, 2025

• (serve.llm) Make _LLMServerBase.__init__ synchronous

  #53719 opened Jun 10, 2025

• [core][gpu objects] Integrate single-controller collective APIs with GPU objects

  #53720 opened Jun 10, 2025

• Bump scikit-learn from 1.3.2 to 1.5.1 in /doc/source/ray-overview/examples/e2e-timeseries

  #53721 opened Jun 10, 2025

• (serve.llm) Remove test leakage from placement bundle logic

  #53723 opened Jun 10, 2025

• [serve.llm] Add better logging verbosity controls

  #53728 opened Jun 11, 2025

• Minor Documentation Fixes in Protobuf Files

  #53731 opened Jun 11, 2025

• [RLlib; docs] Docs do-over (new API stack): `ConnectorV2` documentation.

  #53732 opened Jun 11, 2025

• [WIP] Remove test cases for `gcs_actor_based_scheduling`

  #53733 opened Jun 11, 2025

• [core][telemetry/10] support custom gauge+counter+sum metrics

  #53734 opened Jun 11, 2025

• [core][telemetry/11] support histogram metric on worker side

  #53740 opened Jun 11, 2025

• [ci] bazelize `get_contributors` script

  #53743 opened Jun 11, 2025

• [core][rfc] upgrade opentelemetry-sdk

  #53745 opened Jun 11, 2025

• Test

  #53746 opened Jun 11, 2025

• Add example gpt2 tuning script

  #53750 opened Jun 11, 2025

• [ci] add cibase tags for ci base envs

  #53755 opened Jun 12, 2025

• Fix ray import error when both ROCR_VISIBLE_DEVICES and HIP_VISIBLE_DEVICES are set

  #53757 opened Jun 12, 2025

• [core] Fix race condition in raylet graceful shutdown

  #53762 opened Jun 12, 2025

• [core] Control whether to construct a default concurrency group executor when max-concurrency=1 and there is no other concurrency group for an actor

  #53770 opened Jun 12, 2025

• [Core][Bug fix]Fix issue: the streaming generator will mark the inplasma object that is already ready as failed after the task fails.

  #53773 opened Jun 12, 2025

• [core] Add switch for the cache of runtime env

  #53775 opened Jun 12, 2025

• [serve] Add telemetry for users with Pydantic version < 2

  #53779 opened Jun 12, 2025

• [core] Fix GCS subscribers map race condition

  #53781 opened Jun 12, 2025

• Try parallelize tests within file

  #53784 opened Jun 12, 2025

• DepSets CLI tool

  #53785 opened Jun 13, 2025

• [core] shutdown grpc on worker force exit

  #53787 opened Jun 13, 2025

• [train] add proper filtering to metrics

  #53788 opened Jun 13, 2025

• [core] update to protbuf-28.2, absl-20240722, grpc-1.67 and patch for windows

  #53789 opened Jun 13, 2025

• Add `pin_memory` to `iter_torch_batches`

  #53792 opened Jun 13, 2025

• add new developer api to get application url

  #53796 opened Jun 13, 2025

• [core][gpu-objects] Support intra-process communication

  #53798 opened Jun 13, 2025

• Feature/sac discrete

  #53801 opened Jun 13, 2025

• [core] Deflake `test_multiprocessing.py`

  #53802 opened Jun 13, 2025

• [serve] Revert request timeout from serve instance fixtures

  #53809 opened Jun 13, 2025

• [Data] Avoid failing when no `batch_size` is specified while using GPUs as it's perfectly legitimate use-case

  #53810 opened Jun 13, 2025

• [core] deleting unused code from plasma client

  #53814 opened Jun 13, 2025

• Fix pickle error with remote code models in vLLM multiprocessing

  #53815 opened Jun 13, 2025

• [train] TrainStateActor periodically checks controller status and sets aborted

  #53818 opened Jun 13, 2025

• Bump gitpython from 3.1.40 to 3.1.41 in /python

  #53819 opened Jun 13, 2025

• Bump tqdm from 4.64.1 to 4.66.3 in /python

  #53820 opened Jun 13, 2025

• [Serve.llm][P/D] Support separate deployment config for PDProxy in Prefill disagg

  #53821 opened Jun 14, 2025

• [train][template] Remove ineffective post build script and pip install instead

  #53822 opened Jun 14, 2025

• Sharing progress with broader team

  #53823 opened Jun 14, 2025

• Disable TP=2 VLM batch test

  #53825 opened Jun 14, 2025

• [Doc][KubeRay] remove head pod trailing hash and adjust volcano output

  #53826 opened Jun 14, 2025

54 Issues closed by 15 people

• CI test windows://python/ray/tests:test_basic_client_mode is flaky

  #52117 closed Jun 13, 2025

• [Serve] check_health with custom exception does not enter failed state, infinite retries

  #53742 closed Jun 13, 2025

• CI test windows://python/ray/serve/tests:test_standalone is flaky

  #48420 closed Jun 13, 2025

• [core][gpu-objects] Object contains multiple tensors and/or mix of CPU data and GPU tensors

  #51274 closed Jun 13, 2025

• CI test windows://python/ray/serve/tests:test_standalone_with_comp_sche is flaky

  #48425 closed Jun 13, 2025

• CI test windows://python/ray/tests:test_actor_state_metrics is flaky

  #46303 closed Jun 13, 2025

• CI test linux://python/ray/tune:test_tuner is consistently_failing

  #53786 closed Jun 13, 2025

• CI test linux://rllib:examples/algorithms/appo_custom_algorithm_w_shared_data_actor is flaky

  #53176 closed Jun 13, 2025

• Release test serve_autoscaling_load_test.aws failed

  #53760 closed Jun 13, 2025

• Release test llm_serve_llama_3dot1_8B_quantized_tp1_2p6d failed

  #53769 closed Jun 13, 2025

• Release test llm_serve_llama_3dot1_8B_quantized_tp1_1p1d failed

  #53768 closed Jun 13, 2025

• Release test llm_serve_llama_3dot2_1B_no_accelerator failed

  #53765 closed Jun 13, 2025

• Release test llm_serve_llama_3dot2_1B_s3 failed

  #53767 closed Jun 13, 2025

• Release test llm_serve_llama_3dot1_8B_lora failed

  #53766 closed Jun 13, 2025

• Release test llm_serve_llama_3dot1_8B_quantized_tp_1 failed

  #53764 closed Jun 13, 2025

• Release test llm_serve_llama_3dot1_8B_tp_2 failed

  #53763 closed Jun 13, 2025

• Release test serve_scale_replicas.aws failed

  #53761 closed Jun 13, 2025

• CI test windows://python/ray/serve/tests:test_logging is flaky

  #46043 closed Jun 13, 2025

• [llm] vllm is throwing RuntimeError("Failed to infer device type")

  #51967 closed Jun 13, 2025

• CI test windows://python/ray/serve/tests:test_target_capacity is consistently_failing

  #48426 closed Jun 12, 2025

• CI test windows://python/ray/serve/tests:test_telemetry is flaky

  #48427 closed Jun 12, 2025

• CI test linux://python/ray/data:test_datasink is consistently_failing

  #52098 closed Jun 12, 2025

• CI test linux://rllib:examples/learners/ppo_with_torch_lr_schedulers is flaky

  #49181 closed Jun 12, 2025

• [Serve] concurrency in ray.serve.batch

  #53071 closed Jun 12, 2025

• Release test batch_inference_hetero failed

  #53601 closed Jun 12, 2025

• CI test linux://rllib:learning_tests_stateless_cartpole_appo_gpu is flaky

  #47295 closed Jun 12, 2025

• CI test darwin://python/ray/tests:test_runtime_env_uv_run_client_mode is consistently_failing

  #53650 closed Jun 11, 2025

• CI test windows://python/ray/serve/tests:test_backpressure is consistently_failing

  #50386 closed Jun 11, 2025

• [Kuberay] The reference of Kuberay code link should bind to a release version

  #53655 closed Jun 11, 2025

• [Data] add boundaries or sorted flag to GroupedData.map_groups

  #52577 closed Jun 11, 2025

• CI test windows://python/ray/tests:test_actor_failures is consistently_failing

  #52130 closed Jun 11, 2025

• [Dashboard/Core] In KubeRay, resource list in Cluster Dashboard tab

  #53641 closed Jun 11, 2025

• CI test windows://python/ray/tests:test_object_store_metrics is flaky

  #49514 closed Jun 11, 2025

• [core][gpu-objects] Performance regression caused by transferring object references for small objects

  #53623 closed Jun 10, 2025

• [Core] Ray worker fails to register with raylet when using grpcio>=1.71.0

  #53631 closed Jun 10, 2025

• CI test linux://rllib:learning_tests_multi_agent_cartpole_ppo_multi_gpu is flaky

  #46226 closed Jun 10, 2025

• [core] support .rayignore

  #53648 closed Jun 10, 2025

• [core] CUDA VISIBLE DEVICES is not being set for PlacementGroups

  #53643 closed Jun 10, 2025

• CI test linux://python/ray/tests:test_client_builder is flaky

  #43570 closed Jun 10, 2025

• Release test map_batches_fixed_size_actors_numpy_False failed

  #53658 closed Jun 10, 2025

• Release test map_batches_autoscaling_actors_numpy_False failed

  #53659 closed Jun 10, 2025

• CI test windows://python/ray/tests:test_multi_tenancy is consistently_failing

  #51506 closed Jun 10, 2025

• [Core] Ray doesn't respect object_store_memory when spilling is disabled

  #53086 closed Jun 10, 2025

• [Serve] quick request cancellation with model composition leads to unhandled `TaskCancelledError`s

  #53639 closed Jun 10, 2025

• Migrating from `manylinux1` to `manylinux2014`

  #53548 closed Jun 10, 2025

• CI test linux://rllib:examples/evaluation/evaluation_parallel_to_training_multi_agent_duration_auto is flaky

  #53255 closed Jun 10, 2025

• CI test linux://python/ray/data:test_map is consistently_failing

  #48164 closed Jun 9, 2025

• CI test windows://python/ray/tests:test_output is consistently_failing

  #51467 closed Jun 9, 2025

• CI test darwin://python/ray/tests:test_runtime_env_conda_and_pip_client_mode is consistently_failing

  #53649 closed Jun 9, 2025

• [Core] We should make preloading Jemalloc configurable for worker

  #47242 closed Jun 9, 2025

• slurmstepd: error: Detected 8 oom-kill event(s) in step 59137890.batch cgroup. Some of your processes may have been killed by the cgroup out-of-memory handler.

  #14394 closed Jun 9, 2025

• CI test darwin://python/ray/tests:test_job is consistently_failing

  #45537 closed Jun 9, 2025

• CI test windows://python/ray/tests:test_runtime_env_working_dir_remote_uri_client_mode is consistently_failing

  #53615 closed Jun 9, 2025

• Issue with TLS Authentication

  #53651 closed Jun 9, 2025

34 Issues opened by 31 people

• [bug][serve.llm] AssertionError: failed to get the hash of the compiled graph (VLM, batch, TP=2)

  #53824 opened Jun 14, 2025

• How to transfer tensors stored in GPU in actor with NCCL?

  #53816 opened Jun 13, 2025

• [flaky] test_scheduling_2.py::test_demand_report_when_scale_up

  #53811 opened Jun 13, 2025

• Release test random_shuffle_fixed_size failed

  #53806 opened Jun 13, 2025

• [Data] Custom Partitioner in Ray Data and Related Implementation Considerations

  #53800 opened Jun 13, 2025

• [Core] Ray causes a 25% slower GPU performance compared with manually written Multi-processing program on 8 Hopper GPUs

  #53799 opened Jun 13, 2025

• [Core] Transient network failure on RPC `WaitForActorRefDeleted` causes actor registration fail

  #53797 opened Jun 13, 2025

• How to enable tool calling in serve llm?

  #53795 opened Jun 13, 2025

• [Serve] Proxy issues: Request cancellation, intermittent 503 backpressure, and max_queued_requests configuration not applied

  #53794 opened Jun 13, 2025

• [RLlib] Checkpointing fails with CUDA GPU learner using the new API stack

  #53793 opened Jun 13, 2025

• [<Ray component: Core|RLlib|etc...>] Issue of port allocation

  #53790 opened Jun 13, 2025

• [RLlib][Unity] unity3d_env_local.py 'NoneType' for action spaces

  #53780 opened Jun 12, 2025

• Support gymnasium > 1.0.0

  #53776 opened Jun 12, 2025

• [Core] The streaming generator will mark the inplasma object that is already ready as failed after the task fails.

  #53772 opened Jun 12, 2025

• [core] Add a config to control whether to construct a default concurrency group executor when max-concurrency=1 and there is no other concurrency group for an actor

  #53771 opened Jun 12, 2025

• [Dashboard] Support ncu

  #53759 opened Jun 12, 2025

• [Core] Ray hangs with vllm0.8.5 v1 api for tp8+pp4

  #53758 opened Jun 12, 2025

• Core: Ray 2.45 causes Google's LIBTPU to be very spammy

  #53756 opened Jun 12, 2025

• [core] Race condition between raylet graceful shutdown and GCS health checks

  #53739 opened Jun 11, 2025

• Conflict between ROCR_VISIBLE_DEVICES and HIP_VISIBLE_DEVICES environment variables causes Ray import error

  #53737 opened Jun 11, 2025

• [Announcement] Ray Summit 2025 Call for Proposals Due June 30th

  #53729 opened Jun 11, 2025

• [core] Actor restarts don't work when an actor creation arg is evicted from plasma

  #53727 opened Jun 10, 2025

• Release test compiled_graphs failed

  #53716 opened Jun 10, 2025

• [Core] Custom docker image not scaling out

  #53696 opened Jun 10, 2025

• [<Ray component: Core|RLlib|etc...>] SAC config error about framework

  #53694 opened Jun 10, 2025

• [Data] Support for SQL/DataFrame capability

  #53693 opened Jun 10, 2025

• [Dashboard] Display gpu metrics for AMD/ROCm devices

  #53689 opened Jun 10, 2025

• [Serve] Serve-native CPU profiling in Replicas is broken

  #53677 opened Jun 9, 2025

• [Dask-on-Ray,core] Tasks not registering on the jobs and job is subsequently getting stuck

  #53666 opened Jun 9, 2025

• [Ray Core] Detached actor doesn't finish method after the client disconnects

  #53665 opened Jun 9, 2025

• [Ray serve] Unable to serve meta-llama/Llama-3.1-8B-Instruct

  #53663 opened Jun 9, 2025

• [Core] Transient network failure on RPC `MarkJobFinished` causes node crash

  #53645 opened Jun 8, 2025

• [core][autoscaler] Select different node types when a node type is unavailable

  #53636 opened Jun 7, 2025

• [core][dashboard]: Package already exists, skipping upload.

  #53635 opened Jun 7, 2025

260 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.

• [ci] First release test on GKE

  #53390 commented on Jun 13, 2025 • 33 new comments

• [train] Driver SIGINT calls controller abort

  #53600 commented on Jun 14, 2025 • 27 new comments

• [Docs] Add ServiceMonitor section and make some step optional in Grafana & Promethus page

  #53474 commented on Jun 14, 2025 • 19 new comments

• [core] fix detached actor being unexpectedly killed

  #53562 commented on Jun 13, 2025 • 18 new comments

• [core] Fix gcs register actor callback check

  #53634 commented on Jun 13, 2025 • 15 new comments

• [Docs][KubeRay] Add guide for writing KubeRay doctests

  #51708 commented on Jun 9, 2025 • 11 new comments

• [Serve] Set the docs path after app is initialized on the replica

  #53463 commented on Jun 11, 2025 • 5 new comments

• [Core] Add default Ray Node labels at Node init

  #53360 commented on Jun 13, 2025 • 5 new comments

• Update V2 Autoscaler to support scheduling using Node labels and LabelSelector API

  #53578 commented on Jun 13, 2025 • 4 new comments

• [Serve] Prioritize stopping most recently scaled-up replicas during downscaling

  #52929 commented on Jun 12, 2025 • 2 new comments

• [core][rocm] Allow CUDA_VISIBLE_DEVICS and HIP_VISIBLE_DEVICES

  #53531 commented on Jun 12, 2025 • 2 new comments

• [Core] Add Logic to Emit Task Events to Event Aggregator

  #53402 commented on Jun 13, 2025 • 2 new comments

• [core][telemetry/08] record counter metric e2e

  #53449 commented on Jun 11, 2025 • 1 new comment

• [Serve] Immediately terminate unscheduled replicas

  #52416 commented on Jun 8, 2025 • 0 new comments

• Use large instance for OOM test

  #52400 commented on Jun 8, 2025 • 0 new comments

• Debug multinode

  #52382 commented on Jun 12, 2025 • 0 new comments

• Bump http-proxy-middleware from 2.0.6 to 2.0.9 in /python/ray/dashboard/client

  #52370 commented on Jun 8, 2025 • 0 new comments

• [core] Avoid infinite recursion when an ActorHandle fails to unpickle

  #52363 commented on Jun 8, 2025 • 0 new comments

• Bump torch from 2.0.1 to 2.6.0 in /doc/source/templates/05_dreambooth_finetuning/dreambooth

  #52351 commented on Jun 8, 2025 • 0 new comments

• [core] Fix stuck of long-time running RayCluster

  #52303 commented on Jun 12, 2025 • 0 new comments

• [build] warning when username or homedir include @ character

  #52274 commented on Jun 8, 2025 • 0 new comments

• [autoscaler] Refactor locks behavior in local node provider

  #52269 commented on Jun 11, 2025 • 0 new comments

• [Do Not Merge][core/raylet] Split giant ray core C++ targets into small ones

  #52250 commented on Jun 8, 2025 • 0 new comments

• Added warning for serve config when no config exists

  #52224 commented on Jun 8, 2025 • 0 new comments

• [data] Add overwrite options for iceberg sink

  #52206 commented on Jun 8, 2025 • 0 new comments

• [core] Remove object store runner

  #51766 commented on Jun 7, 2025 • 0 new comments

• Adapt Dask on Ray to the new Dask Task class

  #52589 commented on Jun 10, 2025 • 0 new comments

• Bump h11 from 0.14.0 to 0.16.0 in /release

  #52582 commented on Jun 11, 2025 • 0 new comments

• [Dashboard] Add Worker ID column to Worker table in Node detail page

  #52581 commented on Jun 14, 2025 • 0 new comments

• [fix][Doc] Code snippet in `Ray Data Internal` document

  #52559 commented on Jun 9, 2025 • 0 new comments

• [data] Improve Ray Data progress bar summary string

  #52526 commented on Jun 9, 2025 • 0 new comments

• [core][autoscaler] make ready/infeasible/backlog counts clear in the autoscaler status verbose report

  #52520 commented on Jun 9, 2025 • 0 new comments

• fix bootstrap alert inheritence

  #52519 commented on Jun 9, 2025 • 0 new comments

• [core] Static Priority Scheduling (3/N)

  #52506 commented on Jun 9, 2025 • 0 new comments

• [core] Static Priority scheduling (4/N)

  #52489 commented on Jun 9, 2025 • 0 new comments

• [core] Support streaming arbitrary types to Status messages

  #52473 commented on Jun 9, 2025 • 0 new comments

• [core] Static Priority scheduling (2/N)

  #52465 commented on Jun 9, 2025 • 0 new comments

• feat(data): Add on_error and error_handler to map_batches

  #52457 commented on Jun 12, 2025 • 0 new comments

• [Data][doc] Update on ray.data.Dataset.map() type hints

  #52455 commented on Jun 9, 2025 • 0 new comments

• [Not for Merge] [Core] Add Task Event Aggregator Skeleton

  #52452 commented on Jun 12, 2025 • 0 new comments

• [TEST] test branch

  #52447 commented on Jun 9, 2025 • 0 new comments

• [core] Fix cgroup controller check

  #52443 commented on Jun 9, 2025 • 0 new comments

• [core] Static Priority Scheduling (1/N)

  #52439 commented on Jun 8, 2025 • 0 new comments

• [air/data][perf]: Hot-path check to reduce the number of calls of pa.concat_arrays

  #51792 commented on Jun 12, 2025 • 0 new comments

• [core] Get cloud provider with ray on kubernetes

  #51793 commented on Jun 7, 2025 • 0 new comments

• [core][wip] Trying bzlmod

  #51834 commented on Jun 7, 2025 • 0 new comments

• [core] Support `.options` chaining in `actor.options`

  #51836 commented on Jun 10, 2025 • 0 new comments

• [core][cgraph] Refactor get_devices() for compiled graph

  #51855 commented on Jun 8, 2025 • 0 new comments

• [DO NOT MERGE] Increase test timeout to repro in CI

  #51863 commented on Jun 8, 2025 • 0 new comments

• [Data] Fix bug where pandas blocks don't use tensor extension

  #51868 commented on Jun 8, 2025 • 0 new comments

• Solved warning system

  #51875 commented on Jun 11, 2025 • 0 new comments

• [DONOTMERGE] (mostly) Python PoC for GPU objects in Ray

  #51902 commented on Jun 11, 2025 • 0 new comments

• Add new autoscaling parameter `aggregation function`

  #51905 commented on Jun 8, 2025 • 0 new comments

• [WIP]

  #51939 commented on Jun 8, 2025 • 0 new comments

• [RLlib]fix a bug in the example

  #51944 commented on Jun 8, 2025 • 0 new comments

• [build] Fix typo & improve `build-docker.sh`

  #51950 commented on Jun 8, 2025 • 0 new comments

• [core][autoscaler] test with KubeRay 1.4.0

  #51958 commented on Jun 8, 2025 • 0 new comments

• [Fix][Core] Fail fast if the dashboard agent fails to launch the HTTP server

  #51960 commented on Jun 9, 2025 • 0 new comments

• Add return_future to ActorPool.get_next()

  #52004 commented on Jun 11, 2025 • 0 new comments

• test for raycirun

  #52012 commented on Jun 8, 2025 • 0 new comments

• [Chore][Dashboard] Move DataHead to python/ray/data/ folder

  #52013 commented on Jun 9, 2025 • 0 new comments

• [Chore][Dashboard] Move `TrainHead` to `python/ray/train` folder

  #52014 commented on Jun 9, 2025 • 0 new comments

• [Data] Implement proper limit pushdown #51966

  #52018 commented on Jun 8, 2025 • 0 new comments

• [Data] Make `from_items` lineage serializable

  #52026 commented on Jun 8, 2025 • 0 new comments

• Re-arrange integrations docs

  #52049 commented on Jun 8, 2025 • 0 new comments

• [WIP] Ray Data doc updates

  #52062 commented on Jun 8, 2025 • 0 new comments

• [Data] Preserving special types in mapping and filtering

  #52067 commented on Jun 8, 2025 • 0 new comments

• [Data,Train] Add helpful errors when running forbidden methods on sharded datasets

  #52079 commented on Jun 13, 2025 • 0 new comments

• [data] Add Dataset.write_datasink_lazy to support intermediate outputs.

  #52094 commented on Jun 8, 2025 • 0 new comments

• [Dashboard] Add GPU component usage

  #52102 commented on Jun 9, 2025 • 0 new comments

• Bump xgrammar from 0.1.16 to 0.1.18 in /python

  #52175 commented on Jun 8, 2025 • 0 new comments

• upgrade path to python protobuf 4

  #52194 commented on Jun 8, 2025 • 0 new comments

• [train] upgrade tensorflow-datasets

  #52195 commented on Jun 8, 2025 • 0 new comments

• [core][tests] Add chaos tests to verify the interaction between actor restarts, task retries, and lineage reconstruction

  #53021 commented on Jun 13, 2025 • 0 new comments

• [Data] Refactoring udf context sharing

  #53026 commented on Jun 13, 2025 • 0 new comments

• [data] fix lance dataset schema

  #53134 commented on Jun 9, 2025 • 0 new comments

• [docs] updating broken links on rllib torch doc

  #53161 commented on Jun 10, 2025 • 0 new comments

• Add progress bars to hash operators

  #53175 commented on Jun 13, 2025 • 0 new comments

• [WIP] Fix daft test

  #53338 commented on Jun 9, 2025 • 0 new comments

• [Serve] Allow HTTPs Options in Ray Serve

  #26814 commented on Jun 8, 2025 • 0 new comments

• [WIP] [core] Attempting a basic solution to streaming generator not adding errors to plasma

  #53393 commented on Jun 13, 2025 • 0 new comments

• try running things with protobuf 4

  #53442 commented on Jun 9, 2025 • 0 new comments

• feat: Add QPS-based autoscaling policy for Ray Serve

  #53445 commented on Jun 9, 2025 • 0 new comments

• Replace `python setup.py bdist_wheel` with `pip wheel`

  #53458 commented on Jun 13, 2025 • 0 new comments

• [Data] Added distinct function

  #53460 commented on Jun 11, 2025 • 0 new comments

• [core] Add as_completed and map_unordered APIs

  #53461 commented on Jun 12, 2025 • 0 new comments

• [Serve] make various default values of `AutoscalingConfig.max_replicas` consistent and >1

  #50222 commented on Jun 8, 2025 • 0 new comments

• [core][telemetry/09] record sum metric e2e

  #53512 commented on Jun 11, 2025 • 0 new comments

• [RLlib] MetricsLogger: Fix `get/set_state` to handle tensors in `self.values`.

  #53514 commented on Jun 13, 2025 • 0 new comments

• [core] Turn executed task inserted into a RAY_CHECK

  #53522 commented on Jun 8, 2025 • 0 new comments

• [core][telemetry/10] record histogram metric e2e

  #53523 commented on Jun 11, 2025 • 0 new comments

• [serve.llm] Update ray-llm docker

  #53532 commented on Jun 7, 2025 • 0 new comments

• [RLlib] Upgrade RLlink protocol for external env/simulator training.

  #53550 commented on Jun 13, 2025 • 0 new comments

• Script to generate test coverage for doc files

  #53556 commented on Jun 13, 2025 • 0 new comments

• [Serve] Multiple FastAPI ingress deployments in a single application are not disallowed

  #53024 commented on Jun 8, 2025 • 0 new comments

• [Not for Merge] Event Aggregator Perf

  #53576 commented on Jun 13, 2025 • 0 new comments

• [Doc] vale ignores anchors of headers

  #53580 commented on Jun 9, 2025 • 0 new comments

• [core][telemetry/11] record legacy-legacy metrics e2e

  #53596 commented on Jun 11, 2025 • 0 new comments

• [Train] Allow customization of FPS for wandb logger; instead of slow 4 FPS

  #50186 commented on Jun 7, 2025 • 0 new comments

• Fix 53605

  #53607 commented on Jun 9, 2025 • 0 new comments

• [core] Remove experimental `max_cpu_frac_per_node`

  #53610 commented on Jun 10, 2025 • 0 new comments

• [core] Support broadcast collective for compiled graphs

  #53625 commented on Jun 10, 2025 • 0 new comments

• [Serve] Unable to load meta-llama/Llama-3.3-70B-Instruct

  #53571 commented on Jun 7, 2025 • 0 new comments

• [core] [easy] readability improvements for IO Workers

  #52590 commented on Jun 10, 2025 • 0 new comments

• [Dashboard] Allow getting dashboard URL via RuntimeContext

  #52676 commented on Jun 10, 2025 • 0 new comments

• check if ray is installed when using conda env

  #52677 commented on Jun 10, 2025 • 0 new comments

• Bump transformers from 4.30.1 to 4.50.0 in /doc/source/templates/testing/docker/03_serving_stable_diffusion

  #52681 commented on Jun 14, 2025 • 0 new comments

• [core] Minor pull manager cleanup

  #52724 commented on Jun 9, 2025 • 0 new comments

• [Core][Refactor] Create separate RPCs for cancelling prepared PG bundle and removing PG

  #52751 commented on Jun 9, 2025 • 0 new comments

• [core] Remove copy when receiving small object returns

  #52777 commented on Jun 9, 2025 • 0 new comments

• [core] Remove small task output copy on task execution path

  #52778 commented on Jun 9, 2025 • 0 new comments

• [core][refactor] Move `to_resubmit_` from CoreWorker to TaskManager to avoid an abstraction leak

  #52779 commented on Jun 10, 2025 • 0 new comments

• [WIP][Data] replace tensor array with naive pyarrow tensor

  #52784 commented on Jun 10, 2025 • 0 new comments

• [Core] Increase timeout of start_api_server and make it configurable

  #52789 commented on Jun 14, 2025 • 0 new comments

• [ci] try running cicd unit tests in forge env

  #52792 commented on Jun 12, 2025 • 0 new comments

• Bump minimum pyarrow version to 17

  #52820 commented on Jun 14, 2025 • 0 new comments

• [DONT-MERGE] Adding a sleep to start debugging windows tests.

  #52822 commented on Jun 10, 2025 • 0 new comments

• [Data] remove empty lance read tasks

  #52831 commented on Jun 14, 2025 • 0 new comments

• Train Tests: Use map_batches for image_classification

  #52837 commented on Jun 9, 2025 • 0 new comments

• [DNR] remove ensure_liveness

  #52913 commented on Jun 10, 2025 • 0 new comments

• [data] ResourceManager: always reserve min resource requirement

  #52914 commented on Jun 10, 2025 • 0 new comments

• [core] Synchronize locations with pinned_at_raylet_id

  #52920 commented on Jun 10, 2025 • 0 new comments

• [core] Add sync get node info to NodeInfoAccessor

  #52928 commented on Jun 10, 2025 • 0 new comments

• Ray Data Enhancement: Percentiles and Statistical Aggregations (PR #52588)

  #52937 commented on Jun 10, 2025 • 0 new comments

• Add pinned_memory and non_blocking transfer for default collate_fn

  #52948 commented on Jun 10, 2025 • 0 new comments

• [Data] Replace `_MapWorker` name with operator names

  #52949 commented on Jun 10, 2025 • 0 new comments

• [Data] fix write_iceberg error

  #52956 commented on Jun 10, 2025 • 0 new comments

• [deps] upgrade pandas to always use 2+

  #52961 commented on Jun 11, 2025 • 0 new comments

• [Core] Propogate InvalidArgument Status from LabelSelector Data Type

  #52964 commented on Jun 13, 2025 • 0 new comments

• [RLlib; Offline RL] - Use `iter_torch_batches` in learner

  #52968 commented on Jun 13, 2025 • 0 new comments

• [core] Use GetResourceLoadRequest as a substitute liveness check

  #52971 commented on Jun 10, 2025 • 0 new comments

• [Data] Fixing null-safety when converting to `TensorArray`

  #52977 commented on Jun 10, 2025 • 0 new comments

• [Not for Merge] Increate Timeout for asan tests to Repro in CI

  #52993 commented on Jun 10, 2025 • 0 new comments

• [Core] Identify Mac M1/M2 GPUs as valid GPUs

  #39136 commented on Jun 9, 2025 • 0 new comments

• [Data] Adding streaming capability for `ray.data.Dataset.unique`

  #51207 commented on Jun 9, 2025 • 0 new comments

• [distributed debugger] vscode extension does not accept windows path when configuring cluster

  #53088 commented on Jun 9, 2025 • 0 new comments

• [container] Publish multi-architecture container images

  #41727 commented on Jun 9, 2025 • 0 new comments

• [Data] Significant Memory Leak / OOM When Reading Large Parquet Files with RayData

  #49158 commented on Jun 9, 2025 • 0 new comments

• Uv sync with project using Ray fails installing on Python 3.13

  #52819 commented on Jun 9, 2025 • 0 new comments

• [data] ObjectRefs passed to map UDF are not automatically deref'ed

  #49207 commented on Jun 9, 2025 • 0 new comments

• [Serve] Make replica scheduler backoff configurable

  #52871 commented on Jun 10, 2025 • 0 new comments

• [Conda] Ray should raise exception when ray is not installed in conda environment

  #52672 commented on Jun 10, 2025 • 0 new comments

• [Core] Submitted containerized job is stuck in pending mode

  #37293 commented on Jun 10, 2025 • 0 new comments

• [ray.serve.llm] serve.llm with streaming has overhead compared to vllm-v0 for a single replica when concurrency > 32

  #52746 commented on Jun 10, 2025 • 0 new comments

• Ray serve + core steaming is slow at high concurrency

  #52745 commented on Jun 10, 2025 • 0 new comments

• [core] ray.init does not work if run in a node with external ip while the cluster is started internally

  #8244 commented on Jun 10, 2025 • 0 new comments

• ray azure does not work out of the box

  #52511 commented on Jun 10, 2025 • 0 new comments

• [Azure] Ray up for Azure fails

  #48976 commented on Jun 10, 2025 • 0 new comments

• Support Availability Zone Deployment in Azure

  #39966 commented on Jun 10, 2025 • 0 new comments

• [Serve] Autoscaling not working correctly when `max_replica_per_node` is set in Ray Serve

  #53582 commented on Jun 10, 2025 • 0 new comments

• [Serve] `fastapi_app` is still mutable in the deployment constructor after being passed to `@serve.ingress`

  #52775 commented on Jun 10, 2025 • 0 new comments

• [core] ray.init() not possible even while on same network as Ray Cluster.

  #53520 commented on Jun 10, 2025 • 0 new comments

• [core] TPU Visible Chips not set correctly

  #53569 commented on Jun 10, 2025 • 0 new comments

• [Bug] Dashboard can't start with TLS on

  #22466 commented on Jun 10, 2025 • 0 new comments

• [Train] Support for `lightning.pytorch` on the `mps` backend

  #49858 commented on Jun 10, 2025 • 0 new comments

• [Ray Core/Dashboard] - Installing Ray via UV breaks dashboard.

  #53608 commented on Jun 10, 2025 • 0 new comments

• [Serve.llm] Clean up output logs and give option to opt out of different verbosity levels

  #53492 commented on Jun 10, 2025 • 0 new comments

• [llm] Roadmap for Data and Serve LLM APIs

  #51313 commented on Jun 11, 2025 • 0 new comments

• [Data] [optimizer] map/map_batches should output the same number of rows as the input

  #36295 commented on Jun 11, 2025 • 0 new comments

• [Core] Read-only buffer error in some scikit-learn models

  #52571 commented on Jun 11, 2025 • 0 new comments

• [Core] ASSERTION FAILED: queue.num_items() == 0

  #53510 commented on Jun 11, 2025 • 0 new comments

• [Core] Ray dashboard agent high memory usage

  #52639 commented on Jun 11, 2025 • 0 new comments

• [Core][ROCm] Setting CUDA_VISIBLE_DEVICES leads to an assertion

  #52701 commented on Jun 11, 2025 • 0 new comments

• [Core] `ray job submit` doesn't always catch the last lines of the job logs

  #48701 commented on Jun 11, 2025 • 0 new comments

• [doc] Add documentation guide for MPI on Ray.

  #41626 commented on Jun 11, 2025 • 0 new comments

• [Dashboard] Decoupling dashboard and dashboard lifetime from Ray Cluster

  #46444 commented on Jun 9, 2025 • 0 new comments

• [Core] Ray_tasks and ray_memory_manager_worker_eviction_total metrics should emit 0 instead of null for each state at start

  #47616 commented on Jun 9, 2025 • 0 new comments

• [Ray debugger] Unable to use debugger on Ray Cluster on k8s

  #45541 commented on Jun 9, 2025 • 0 new comments

• [Dashboard] Should specify the time range in job detail page for load the cluster status and scale metrics

  #41781 commented on Jun 9, 2025 • 0 new comments

• [Dashboard] Add job retention mechanism

  #35700 commented on Jun 9, 2025 • 0 new comments

• [Ray Dashboard] Set Route Prefix/Base Address

  #35269 commented on Jun 9, 2025 • 0 new comments

• [Core] Expose logs for runtime environment installation process on worker nodes for remote Ray clusters

  #34310 commented on Jun 9, 2025 • 0 new comments

• [State API] ray log truncation message improvements

  #32392 commented on Jun 9, 2025 • 0 new comments

• [Core] Core metrics observed from worker nodes do not propagate to Prometheus

  #31675 commented on Jun 9, 2025 • 0 new comments

• [core][state] list objects show objects if spilled.

  #31374 commented on Jun 9, 2025 • 0 new comments

• Troubleshooting the root cause that cluster_status is undefined

  #40076 commented on Jun 9, 2025 • 0 new comments

• [Event] Support rotation for the event log files

  #39591 commented on Jun 9, 2025 • 0 new comments

• [Dashboard] Support row in the dashboard.

  #38024 commented on Jun 9, 2025 • 0 new comments

• [dashboard][UI] [TaskTable] Rendering task table has a 8s delay

  #36656 commented on Jun 9, 2025 • 0 new comments

• [UI] [Log viewer] Prevent the infinite loading for logs fetch

  #36486 commented on Jun 9, 2025 • 0 new comments

• [Dashboard] Kill a job

  #30182 commented on Jun 9, 2025 • 0 new comments

• [Dashboard] [CI] Add tests for uncaught exceptions

  #29809 commented on Jun 9, 2025 • 0 new comments

• [Core] Setting python log level for ray processes

  #29758 commented on Jun 9, 2025 • 0 new comments

• [Core] timeline doesn't show all infos.

  #28320 commented on Jun 9, 2025 • 0 new comments

• [State Observability] Improve ray list job implementation.

  #26832 commented on Jun 9, 2025 • 0 new comments

• [autoscaler][logs] Improve status logging

  #26670 commented on Jun 9, 2025 • 0 new comments

• [Autoscaler/Core][Code quality] Handle autoscaler event logging through RPC, not logs.

  #26186 commented on Jun 9, 2025 • 0 new comments

• [core] Node IDs not consistent across APIs

  #25090 commented on Jun 9, 2025 • 0 new comments

• [Jobs] Setting `RAY_LOG_TO_STDERR` results in empty job logs

  #24886 commented on Jun 9, 2025 • 0 new comments

• [Core Observability] Include name to actor log prefix + process name

  #24876 commented on Jun 9, 2025 • 0 new comments

• [jobs] [Feature] Support streaming job logs to stdout/stderr

  #23564 commented on Jun 9, 2025 • 0 new comments

• [dashboard] Wonky GPU display

  #14664 commented on Jun 9, 2025 • 0 new comments

• [metrics] ray.util.metrics API should closely mirror the prometheus python API

  #14496 commented on Jun 9, 2025 • 0 new comments

• [metrics] Report general metrics for gRPC

  #14368 commented on Jun 9, 2025 • 0 new comments

• [core] Getting node IP address by object ref

  #13630 commented on Jun 9, 2025 • 0 new comments

• [Logs] Spdlog doesn't rotate raylet.out and gcs_server.out

  #13466 commented on Jun 9, 2025 • 0 new comments

• Ray dashboard_url and prom_discovery.json files not scoped to session dir

  #12662 commented on Jun 9, 2025 • 0 new comments

• verify windows wheels.

  #43442 commented on Jun 8, 2025 • 0 new comments

• [core] Fix max_calls option when used on a worker that is part of a workflow

  #43700 commented on Jun 11, 2025 • 0 new comments

• remove flaky marker from test

  #44033 commented on Jun 12, 2025 • 0 new comments

• RuntimeContext support get actor namespace

  #45025 commented on Jun 11, 2025 • 0 new comments

• [Core] Improve logging during accelerator auto-detection

  #45240 commented on Jun 14, 2025 • 0 new comments

• Enable setting OS disk size in Azure

  #45867 commented on Jun 10, 2025 • 0 new comments

• Fix malformed `temp_dir` path when connecting Windows workers to cluster with Linux head

  #45930 commented on Jun 9, 2025 • 0 new comments

• [URL] Change the absolute path to a relative path to solve the ingres…

  #45933 commented on Jun 10, 2025 • 0 new comments

• python/ray/autoscaler/gcp/*.yaml: change scheduling from dict to list

  #46500 commented on Jun 14, 2025 • 0 new comments

• Fix mlflow artifact logging

  #46570 commented on Jun 10, 2025 • 0 new comments

• [Core] If possible, force flush the trace when the worker ends.

  #46654 commented on Jun 14, 2025 • 0 new comments

• [Core]Support Merge code search path from env variable

  #46771 commented on Jun 11, 2025 • 0 new comments

• [bazel] move python rules up

  #47260 commented on Jun 12, 2025 • 0 new comments

• [RayCluster] Introduce how to run ray remote job with ray client （#47…

  #47771 commented on Jun 11, 2025 • 0 new comments

• Add kuberay operator addon to cmd in gke-gcs-bucket.md

  #48268 commented on Jun 11, 2025 • 0 new comments

• (WIP) [core][compiled graphs] Unify code paths for NCCL P2P and collectives scheduling

  #48649 commented on Jun 12, 2025 • 0 new comments

• [Core]: Fix ConnectionError on Autoscaler CR lookups in K8s clusters …

  #48675 commented on Jun 7, 2025 • 0 new comments

• Update azure.md - Missing azure dependency

  #49104 commented on Jun 11, 2025 • 0 new comments

• [RLlib] Add NPU and HPU support to RLlib

  #49535 commented on Jun 13, 2025 • 0 new comments

• [Core] Add virtual cluster

  #49717 commented on Jun 12, 2025 • 0 new comments

• [DATA]Add custom resources in data autoscaling

  #49756 commented on Jun 9, 2025 • 0 new comments

• [core] Thread-safe gcs node manager

  #50024 commented on Jun 7, 2025 • 0 new comments

• [core] Always create a default executor

  #51058 commented on Jun 8, 2025 • 0 new comments

• [CI] Replace `black` with `ruff format`

  #51332 commented on Jun 8, 2025 • 0 new comments

• [Dashboard] Support reporting AMD GPU usage

  #51345 commented on Jun 10, 2025 • 0 new comments

• [Core] Cover cpplint for ray/src/ray/common

  #51551 commented on Jun 8, 2025 • 0 new comments

• [py_modules] Don't install the wheel package if it's already installed

  #51629 commented on Jun 11, 2025 • 0 new comments

• windows dev setup

  #51678 commented on Jun 8, 2025 • 0 new comments

• pyarrow.lib.ArrowInvalid: Struct child array #5 does not match type field: null vs double

  #53529 commented on Jun 8, 2025 • 0 new comments

• [core] Lazily subscribe to node changes from workers

  #51718 commented on Jun 7, 2025 • 0 new comments

• [Core] Native CPU affinity support for accelerators

  #51719 commented on Jun 10, 2025 • 0 new comments

• [Core] Turn off RayTaskError cause wrapping functionality

  #48320 commented on Jun 11, 2025 • 0 new comments

• [Core] `ray.init()` and `ray start` fails on Windows 11 in ray 2.45+

  #52739 commented on Jun 12, 2025 • 0 new comments

• [Core] ray._raylet.ObjectRef and ray.types.ObjectRef type compabtibility

  #53591 commented on Jun 12, 2025 • 0 new comments

• [Serve] Proxy actor not started on worker node when using kuberay

  #50349 commented on Jun 12, 2025 • 0 new comments

• [Serve] Specify different images for each deployment

  #52994 commented on Jun 12, 2025 • 0 new comments

• [RLlib] gym.spaces.Sequence unbatching error

  #53293 commented on Jun 12, 2025 • 0 new comments

• [RLlib] ActionMaskingTorchRLModule can't set up `conv_filters`

  #53325 commented on Jun 12, 2025 • 0 new comments

• [RLlib] Type of `AlgorithmConfig.training(learner_connector` is wrong

  #53368 commented on Jun 12, 2025 • 0 new comments

• [RLlib] Env runners error out when interacting with Repeated observation spaces

  #53327 commented on Jun 12, 2025 • 0 new comments

• AttributeError: 'NoneType' object has no attribute 'enable_rl_module_and_learner' with highway-env

  #53398 commented on Jun 12, 2025 • 0 new comments

• [RLLIB] EnvContext.vector_index is always 0

  #53419 commented on Jun 12, 2025 • 0 new comments

• [RLlib][PPO new-API] Large discrepancy between Algorithm.evaluate() and manual inference via restored EnvToModule/ModuleToEnv pipelines on CarRacing-v3

  #53588 commented on Jun 12, 2025 • 0 new comments

• [Core | Serve] Compatibility issue with pydantic>=2.10

  #52211 commented on Jun 12, 2025 • 0 new comments

• [Core] Prevent schedulling non-GPU tasks to GPU nodes

  #47866 commented on Jun 12, 2025 • 0 new comments

• [core][compiled graphs] Slow NCCL init on H200 server

  #53619 commented on Jun 12, 2025 • 0 new comments

• [Data] `dataset.write_iceberg` error

  #52967 commented on Jun 12, 2025 • 0 new comments

• [Data] PyArrow 20.0.0 Backward Incompatability (`unexpected keyword argument 'maps_as_pydicts'`)

  #52685 commented on Jun 12, 2025 • 0 new comments

• [LLM] We need to create a more robust way of handling actor shutdown

  #53179 commented on Jun 13, 2025 • 0 new comments

• CI test windows://python/ray/serve/tests:test_batching is consistently_failing

  #46016 commented on Jun 13, 2025 • 0 new comments

• Global Per-Epoch Shuffling with TorchTrainer

  #47460 commented on Jun 13, 2025 • 0 new comments

• Release test sort_autoscaling failed

  #53546 commented on Jun 13, 2025 • 0 new comments

• CI test windows://python/ray/serve/tests:test_request_timeout is flaky

  #48417 commented on Jun 13, 2025 • 0 new comments

• [Core|Dataset] Ray job stuck with idle actors with no tasks

  #45822 commented on Jun 13, 2025 • 0 new comments

• [core|serve] Migrate shared utilities from `ray._private` to `ray._common`

  #53478 commented on Jun 13, 2025 • 0 new comments

• CI test windows://python/ray/tests:test_object_spilling is consistently_failing

  #45961 commented on Jun 13, 2025 • 0 new comments

• CI test windows://python/ray/tests:test_object_spilling_asan is consistently_failing

  #45962 commented on Jun 13, 2025 • 0 new comments

• CI test windows://python/ray/tests:test_object_spilling_debug_mode is consistently_failing

  #43796 commented on Jun 13, 2025 • 0 new comments

• CI test linux://python/ray/tests:test_runtime_env_container is consistently_failing

  #45223 commented on Jun 14, 2025 • 0 new comments

• [Dashboard] Provide a job dashboard URL link instead of the dashboard link when ray.init is called.

  #35427 commented on Jun 11, 2025 • 0 new comments

• Add Apple silicon GPU(mps) support to ray

  #38464 commented on Jun 12, 2025 • 0 new comments

• updates to setup-dev.py to work around the types.py import issues

  #38948 commented on Jun 14, 2025 • 0 new comments

• [core] Fix a corner case where the RPC never return

  #39801 commented on Jun 11, 2025 • 0 new comments