Owner

ray-project

3 indexed repositories · View on GitHub

kuberay
A toolkit to run Ray applications on Kubernetes

Go · 2694 stars
raydp
RayDP provides simple APIs for running Spark on Ray and integrating Spark with AI libraries.

Scala · 377 stars
travis-tracker-v2

Python · 2 stars

Beginner friendly Everything

Open All

[Data] Shuffle v2 progress bars: misleading totals for aggregations, redundant sub-bars Open

data

ray-project/ray#66314 · 1 comment · 1 assignee · Opened Sep 19, 2026
[Data] Plan to deprecate built-in ClickHouse datasource in favor of community-maintained ray-clickhouse Open

data docs tech-debt

Difficulty 4/5 3-5 days Newbie friendliness 25/100

ray-project/ray#66309 · Opened Sep 18, 2026
[Core][runtime_env] Support Kerberos-authenticated HTTPS code package downloads Open

community-backlog core security

Difficulty 5/5 Over a week Newbie friendliness 20/100

ray-project/ray#66278 · Opened Sep 18, 2026
[Serve] Handle freshness depends on clocks from different machines Open

bug community-backlog serve stability triage

Difficulty 3/5 1-2 days Newbie friendliness 74/100

ray-project/ray#66271 · 1 comment · Opened Sep 18, 2026
[Serve][Docs] In-place updates guide doesn't mention that code-update replica restarts are rolling Open

community-backlog docs serve usability

Difficulty 2/5 1-3 hours Newbie friendliness 88/100

ray-project/ray#66265 · Opened Sep 18, 2026
[Core] [RDT] nccl transport will hang to timeout while actor without "ray.method(tensor_transport="nccl")" decorator Open

bug community-backlog core gpu stability triage usability

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ray-project/ray#66264 · 1 comment · Opened Sep 18, 2026
[RFC][Data] Add Apache Gravitino catalog provider for Ray Data Open

community-backlog data question usability

Difficulty 5/5 Over a week Newbie friendliness 25/100

ray-project/ray#66263 · Opened Sep 18, 2026
[Bug] `TestZeroDowntimeUpgradeAfterOperatorUpgrade` fails at ready-endpoints check Open

bug ci P0

ray-project/kuberay#5317 · 4 comments · 1 assignee · Opened Sep 18, 2026
[Bug][Prometheus] `rayClusterConditionProvisioned` is missing from `Describe()` Open

bug good-first-issue

ray-project/kuberay#5310 · 1 comment · 1 assignee · Opened Sep 17, 2026
[Bug] logStdoutEncoder and logFileEncoder setting from `--config` are ignored Open

bug good-first-issue

ray-project/kuberay#5309 · 1 comment · 1 assignee · Opened Sep 17, 2026
[Core] Worker hangs at exit and leaks as ray::IDLE when log rotation is on and a child process holds the redirected stream Open

bug community-backlog core observability stability triage

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ray-project/ray#66231 · Opened Sep 17, 2026
[Feature] Forward Kubernetes Pod events for autoscaler-driven remediation Open

enhancement

ray-project/kuberay#5304 · 2 comments · 2 assignees · Opened Sep 16, 2026
[Flaky] Test Python Client setup times out waiting for kuberay-operator on master Open

bug ci flaky P0

ray-project/kuberay#5303 · 1 comment · 1 assignee · Opened Sep 16, 2026
[Flaky] TestCollector times out waiting for the MinIO API endpoint on master Open

bug ci flaky P0

ray-project/kuberay#5302 · 4 comments · 1 assignee · Opened Sep 16, 2026
[Bug][RayCluster][YuniKorn] Hard-gang placeholder timeout causes unbounded Pod recreation against a terminated YuniKorn application Open

Difficulty 5/5 Over a week Newbie friendliness 35/100

ray-project/kuberay#5301 · 2 comments · Opened Sep 16, 2026
[Core] After GCS failover, raylet grants lease requests queued by the previous GCS; worker stays ray::IDLE and leaks the node's GPU Open

community-backlog core stability

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ray-project/ray#66199 · 1 comment · Opened Sep 16, 2026
[Data] Actor pools cannot scale down to zero in multi-step pipelines with downstream bottleneck Open

community-backlog data performance

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ray-project/ray#66196 · Opened Sep 16, 2026
[Epic][History Server] Remaining work towards GA Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

ray-project/kuberay#5297 · Opened Sep 16, 2026
[Data] from_pandas with override_num_blocks argument is broken on pandas version 3 Open

bug community-backlog data stability triage

Difficulty 3/5 1-2 days Newbie friendliness 76/100

ray-project/ray#66185 · 1 comment · Opened Sep 15, 2026
[Data] read_iceberg migration to Datasource V2 (tracking issue) Open

data performance

Difficulty 5/5 Over a week Newbie friendliness 25/100

ray-project/ray#66181 · Opened Sep 15, 2026
[Serve] Replica scheduling hangs/cancelled on worker reachable only via routed subnet (plain actors work fine) Open

community-backlog serve stability

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ray-project/ray#66164 · Opened Sep 15, 2026
[CI] gVisor sandbox GPU tests fail: NVIDIA driver (550.90.07) unsupported by gVisor nvproxy Open

community-backlog core stability

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ray-project/ray#66161 · 1 comment · Opened Sep 14, 2026
Runtime environment setup fails with gateway HTTP 403 on IPv6; raw IPv6 Host header reproduces the failure Open

bug community-backlog core stability triage

Difficulty 4/5 3-5 days Newbie friendliness 68/100

ray-project/ray#66159 · 1 comment · Opened Sep 14, 2026
[Data] read_clickhouse: validate and quote table and column identifiers Open

community-backlog data security usability

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ray-project/ray#66156 · 4 comments · Opened Sep 14, 2026
[Data] read_clickhouse: support server-side typed parameters for filter values Open

community-backlog data usability

Difficulty 4/5 3-5 days Newbie friendliness 54/100

ray-project/ray#66155 · 4 comments · Opened Sep 14, 2026
[core] Raylet contacts the runtime env agent before checking maximum_startup_concurrency, exhausting ephemeral ports and self-terminating Open

core stability

ray-project/ray#66153 · 1 assignee · Opened Sep 14, 2026
[Data] read_iceberg re-reads a shared delete file once per data file instead of once per read task Open

community-backlog data performance

Difficulty 4/5 3-5 days Newbie friendliness 58/100

ray-project/ray#66147 · Opened Sep 13, 2026
[Data] write_iceberg UPSERT collects every upserted key on the driver, even when no file needs a rewrite Open

community-backlog data performance question stability

Difficulty 5/5 Over a week Newbie friendliness 48/100

ray-project/ray#66145 · 1 comment · Opened Sep 13, 2026
[Serve] A batch_size_fn exception stops batching and hangs subsequent valid requests Open

community-backlog serve stability

Difficulty 4/5 3-5 days Newbie friendliness 52/100

ray-project/ray#66140 · Opened Sep 13, 2026
[Data] Upgrade pyiceberg to 0.12.x and adopt its commit retry and concurrency validation Open

community-backlog data stability

Difficulty 5/5 Over a week Newbie friendliness 42/100

ray-project/ray#66134 · Opened Sep 12, 2026
[Data] write_iceberg OVERWRITE with an overwrite_filter rewrites every candidate file on the driver Open

community-backlog data performance question

Difficulty 4/5 3-5 days Newbie friendliness 65/100

ray-project/ray#66133 · Opened Sep 12, 2026
[RLLib] Change connector configuration callbacks to transform ConnectorPipelineV2 Open

community-backlog enhancement rllib triage usability

ray-project/ray#66128 · 3 comments · 1 assignee · Opened Sep 12, 2026
[CI][History Server] Move the remaining historyserver gosec exclusions into the permanent section with a reason Open

enhancement good-first-issue

ray-project/kuberay#5279 · 2 comments · 1 assignee · Opened Sep 12, 2026
[CI][History Server] Remove the gosec G110 lint exclusion by deleting the unused DecompressStream helper Open

enhancement good-first-issue

Difficulty 2/5 1-3 hours Newbie friendliness 84/100

ray-project/kuberay#5278 · Opened Sep 12, 2026
[CI][History Server] Remove the nilerr lint exclusion by fixing the os.Stat error handling in MoveSessionLogsToPrevLogs Open

enhancement good-first-issue

ray-project/kuberay#5277 · 2 comments · 1 assignee · Opened Sep 12, 2026
[CI][History Server] Remove the errcheck lint exclusion by handling go-restful response write errors Open

enhancement good-first-issue

ray-project/kuberay#5276 · 2 comments · 1 assignee · Opened Sep 12, 2026
[CI][History Server] Remove the noctx lint exclusion for historyserver/test by using context-aware HTTP and exec calls Open

enhancement good-first-issue

ray-project/kuberay#5275 · 2 comments · 1 assignee · Opened Sep 12, 2026
[CI][History Server] Remove the unused lint exclusion for historyserver/test by restoring the /api/cluster_status e2e tests Open

enhancement good-first-issue

ray-project/kuberay#5274 · 1 comment · 1 assignee · Opened Sep 12, 2026
[CI][History Server] Remove the gosec G114 lint exclusion by giving the event collector timeouts Open

enhancement good-first-issue

ray-project/kuberay#5273 · 3 comments · 1 assignee · Opened Sep 12, 2026
[CI][History Server] Remove the SA1019 aws-sdk-go lint exclusion by migrating the S3 backend to aws-sdk-go-v2 Open

enhancement

ray-project/kuberay#5272 · 3 comments · 1 assignee · Opened Sep 12, 2026
[Umbrella][CI][History Server] Burn down the historyserver golangci-lint exclusion baseline Open

enhancement good-first-issue

Difficulty 4/5 3-5 days Newbie friendliness 45/100

ray-project/kuberay#5271 · 1 reaction · Opened Sep 12, 2026
[Data] Potential unnecessary copy converting flat Arrow tensor columns to pandas with zero_copy_batch=True Open

data performance

Difficulty 4/5 3-5 days Newbie friendliness 52/100

ray-project/ray#66118 · 1 comment · Opened Sep 11, 2026
[Core] Default actors to only placing on cluster minimum capacity, when autoscaling enabled Open

community-backlog core enhancement performance stability triage

Difficulty 5/5 Over a week Newbie friendliness 38/100

ray-project/ray#66115 · Opened Sep 11, 2026
[Core] `_internal_kv_put(..., overwrite=False)` return value is inverted relative to caller expectation Open

community-backlog core docs stability usability

Difficulty 2/5 1-3 hours Newbie friendliness 72/100

ray-project/ray#66114 · 1 comment · Opened Sep 11, 2026
[Core] Document the `internal_kv` durability contract, and add per-namespace quota and TTL Open

community-backlog core docs stability

Difficulty 5/5 Over a week Newbie friendliness 35/100

ray-project/ray#66113 · Opened Sep 11, 2026
[Feature] Add support for allowing batch scheduler plugins to emit events and set status conditions Open

enhancement

ray-project/kuberay#5270 · 9 comments · 1 assignee · Opened Sep 11, 2026
[Data] Improve ClickHouse read error observability Open

community-backlog data observability usability

Difficulty 4/5 3-5 days Newbie friendliness 52/100

ray-project/ray#66098 · 2 comments · Opened Sep 11, 2026
[core] Nsight runtime_env replaces the selected Python command with bare `python`, so workers fail silently when `python` is not on PATH Open

community-backlog core stability

Difficulty 2/5 1-3 hours Newbie friendliness 86/100

ray-project/ray#66093 · Opened Sep 11, 2026
[RFC][Data] Fit preprocessor Chains in one dataset scan by deferring statistics to the first transform() Open

data performance question

Difficulty 5/5 Over a week Newbie friendliness 25/100

ray-project/ray#66090 · Opened Sep 11, 2026
[Data] iter_batches(local_shuffle_buffer_size=...) yields short batches mid-stream, breaking DDP step parity Open

bug community-backlog data stability triage

Difficulty 4/5 3-5 days Newbie friendliness 48/100

ray-project/ray#66085 · Opened Sep 11, 2026

Showing the newest 100

This page lists what was indexed most recently. Advanced filter has the whole inventory, narrowed by language, difficulty and how long a task takes.

Open advanced filter

ray-project

Showing the newest 100

Get new issues in your inbox