Merge branch 'master' into hpguo/train_structured_log

README.rst

@@ -137,4 +137,4 @@ Getting Involved
 .. _`StackOverflow`: https://stackoverflow.com/questions/tagged/ray
 .. _`Meetup Group`: https://www.meetup.com/Bay-Area-Ray-Meetup/
 .. _`Twitter`: https://twitter.com/raydistributed
-.. _`Slack`: https://forms.gle/9TSdDYUgxYs8SA9e8
+.. _`Slack`: https://www.ray.io/join-slack?utm_source=github&utm_medium=ray_readme&utm_campaign=getting_involved

ci/env/install-dependencies.sh

@@ -244,9 +244,9 @@ install_pip_packages() {
 
     # For DAG visualization
     requirements_packages+=("pydot")
-    requirements_packages+=("pytesseract")
-    requirements_packages+=("spacy>=3")
-    requirements_packages+=("spacy_langdetect")
+    requirements_packages+=("pytesseract==0.3.13")
+    requirements_packages+=("spacy==3.7.5")
+    requirements_packages+=("spacy_langdetect==0.1.2")
   fi
 
   # Additional RLlib test dependencies.

ci/ray_ci/docker_container.py

@@ -20,8 +20,8 @@
 ]
 GPU_PLATFORM = "cu12.1.1-cudnn8"
 
-PYTHON_VERSIONS_RAY = ["3.9", "3.10", "3.11"]
-PYTHON_VERSIONS_RAY_ML = ["3.9", "3.10"]
+PYTHON_VERSIONS_RAY = ["3.9", "3.10", "3.11", "3.12"]
+PYTHON_VERSIONS_RAY_ML = ["3.9", "3.10", "3.11"]
 ARCHITECTURES_RAY = ["x86_64", "aarch64"]
 ARCHITECTURES_RAY_ML = ["x86_64"]
 
@@ -46,6 +46,15 @@ def __init__(
         upload: bool = False,
     ) -> None:
         assert "RAYCI_CHECKOUT_DIR" in os.environ, "RAYCI_CHECKOUT_DIR not set"
+
+        assert python_version in PYTHON_VERSIONS_RAY
+        assert platform in PLATFORMS_RAY
+        assert architecture in ARCHITECTURES_RAY
+        if image_type == RayType.RAY_ML:
+            assert python_version in PYTHON_VERSIONS_RAY_ML
+            assert platform in PLATFORMS_RAY_ML
+            assert architecture in ARCHITECTURES_RAY_ML
+
         rayci_checkout_dir = os.environ["RAYCI_CHECKOUT_DIR"]
         self.python_version = python_version
         self.platform = platform
@@ -122,13 +131,13 @@ def _get_image_tags(self, external: bool = False) -> List[str]:
         versions = self._get_image_version_tags(external)
 
         platforms = [self.get_platform_tag()]
-        if self.platform == "cpu" and self.image_type == "ray":
+        if self.platform == "cpu" and self.image_type == RayType.RAY:
             # no tag is alias to cpu for ray image
             platforms.append("")
         elif self.platform == GPU_PLATFORM:
             # gpu is alias to cu118 for ray image
             platforms.append("-gpu")
-            if self.image_type == "ray-ml":
+            if self.image_type == RayType.RAY_ML:
                 # no tag is alias to gpu for ray-ml image
                 platforms.append("")
 

ci/ray_ci/test_ray_docker_container.py

@@ -7,6 +7,7 @@
 
 from ci.ray_ci.builder_container import DEFAULT_PYTHON_VERSION
 from ci.ray_ci.container import _DOCKER_ECR_REPO
+from ci.ray_ci.docker_container import GPU_PLATFORM
 from ci.ray_ci.ray_docker_container import RayDockerContainer
 from ci.ray_ci.test_base import RayCITestBase
 from ci.ray_ci.utils import RAY_VERSION
@@ -203,8 +204,8 @@ def test_canonical_tag(self) -> None:
         container = RayDockerContainer(v, "cpu", "ray", "aarch64")
         assert container._get_canonical_tag() == f"{sha}-{pv}-cpu-aarch64"
 
-        container = RayDockerContainer(v, "cu11.8.0-cudnn8", "ray-ml")
-        assert container._get_canonical_tag() == f"{sha}-{pv}-cu118"
+        container = RayDockerContainer(v, GPU_PLATFORM, "ray-ml")
+        assert container._get_canonical_tag() == f"{sha}-{pv}-cu121"
 
         with mock.patch.dict(os.environ, {"BUILDKITE_BRANCH": "releases/1.0.0"}):
             container = RayDockerContainer(v, "cpu", "ray")

doc/BUILD

@@ -467,6 +467,7 @@ doctest(
             "source/rllib/rllib-sample-collection.rst",
         ],
     ),
+    data = ["//rllib:cartpole-v1_large"],
     tags = ["team:rllib"],
 )
 

doc/source/_includes/rllib/new_api_stack.rst

@@ -4,8 +4,4 @@
     The Ray Team plans to transition algorithms, example scripts, and documentation to the new code base
     thereby incrementally replacing the "old API stack" (e.g., ModelV2, Policy, RolloutWorker) throughout the subsequent minor releases leading up to Ray 3.0.
 
-    Note, however, that so far only PPO (single- and multi-agent) and SAC (single-agent only)
-    support the "new API stack" and continue to run by default with the old APIs.
-    You can continue to use the existing custom (old stack) classes.
-
     :doc:`See here </rllib/rllib-new-api-stack>` for more details on how to use the new API stack.

doc/source/_templates/navbar-anyscale.html

@@ -1,4 +1,4 @@
-<a id="try-anyscale-href" href="https://console.anyscale.com">
+<a id="try-anyscale-href" href="https://console.anyscale.com/?utm_source=ray_docs&utm_medium=docs&utm_campaign=navbar">
   <div id="try-anyscale-text">
     <span>Try Ray on Anyscale</span>
   </div>

doc/source/cluster/kubernetes/examples/gpu-training-example.md

@@ -40,10 +40,10 @@ kubectl apply -f https://raw.githubusercontent.com/GoogleCloudPlatform/container
 #   (Method 2) "gcloud container clusters get-credentials <your-cluster-name> --region <your-region> --project <your-project>"
 #   (Method 3) "kubectl config use-context ..."
 
-# Install both CRDs and KubeRay operator v1.0.0.
+# Install both CRDs and KubeRay operator.
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 helm repo update
-helm install kuberay-operator kuberay/kuberay-operator --version 1.0.0
+helm install kuberay-operator kuberay/kuberay-operator --version 1.2.2
 
 # Create a Ray cluster
 kubectl apply -f https://raw.githubusercontent.com/ray-project/ray/master/doc/source/cluster/kubernetes/configs/ray-cluster.gpu.yaml
@@ -116,7 +116,7 @@ It is optional.
 # Create the KubeRay operator
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 helm repo update
-helm install kuberay-operator kuberay/kuberay-operator --version 1.0.0
+helm install kuberay-operator kuberay/kuberay-operator --version 1.2.2
 
 # Create a Ray cluster
 kubectl apply -f https://raw.githubusercontent.com/ray-project/ray/master/doc/source/cluster/kubernetes/configs/ray-cluster.gpu.yaml

doc/source/cluster/kubernetes/examples/mobilenet-rayservice.md

@@ -19,7 +19,7 @@ Note that the YAML file in this example uses `serveConfigV2`, which is supported
 
 ```sh
 # Download `ray-service.mobilenet.yaml`
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.0.0/ray-operator/config/samples/ray-service.mobilenet.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-service.mobilenet.yaml
 
 # Create a RayService
 kubectl apply -f ray-service.mobilenet.yaml

doc/source/cluster/kubernetes/examples/rayjob-batch-inference-example.md

@@ -37,12 +37,12 @@ The KubeRay operator Pod must be on the CPU node if you have set up the taint fo
 
 ## Step 2: Submit the RayJob
 
-Create the RayJob custom resource with [ray-job.batch-inference.yaml](https://github.com/ray-project/kuberay/blob/v1.0.0/ray-operator/config/samples/ray-job.batch-inference.yaml).
+Create the RayJob custom resource with [ray-job.batch-inference.yaml](https://github.com/ray-project/kuberay/blob/v1.2.2/ray-operator/config/samples/ray-job.batch-inference.yaml).
 
 Download the file with `curl`:
 
 ```bash
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.0.0/ray-operator/config/samples/ray-job.batch-inference.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-job.batch-inference.yaml
 ```
 
 Note that the `RayJob` spec contains a spec for the `RayCluster`. This tutorial uses a single-node cluster with 4 GPUs. For production use cases, use a multi-node cluster where the head node doesn't have GPUs, so that Ray can automatically schedule GPU workloads on worker nodes which won't interfere with critical Ray processes on the head node.

doc/source/cluster/kubernetes/examples/stable-diffusion-rayservice.md

@@ -18,7 +18,7 @@ Please note that the YAML file in this example uses `serveConfigV2`, which is su
 
 ```sh
 # Step 3.1: Download `ray-service.stable-diffusion.yaml`
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.0.0/ray-operator/config/samples/ray-service.stable-diffusion.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-service.stable-diffusion.yaml
 
 # Step 3.2: Create a RayService
 kubectl apply -f ray-service.stable-diffusion.yaml

doc/source/cluster/kubernetes/examples/text-summarizer-rayservice.md

@@ -17,7 +17,7 @@ Please note that the YAML file in this example uses `serveConfigV2`, which is su
 
 ```sh
 # Step 3.1: Download `ray-service.text-summarizer.yaml`
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.0.0/ray-operator/config/samples/ray-service.text-summarizer.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-service.text-summarizer.yaml
 
 # Step 3.2: Create a RayService
 kubectl apply -f ray-service.text-summarizer.yaml

doc/source/cluster/kubernetes/getting-started/raycluster-quick-start.md

@@ -26,8 +26,8 @@ Deploy the KubeRay operator with the [Helm chart repository](https://github.com/
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 helm repo update
 
-# Install both CRDs and KubeRay operator v1.1.1.
-helm install kuberay-operator kuberay/kuberay-operator --version 1.1.1
+# Install both CRDs and KubeRay operator v1.2.2.
+helm install kuberay-operator kuberay/kuberay-operator --version 1.2.2
 
 # Confirm that the operator is running in the namespace `default`.
 kubectl get pods
@@ -46,14 +46,14 @@ Once the KubeRay operator is running, we are ready to deploy a RayCluster. To do
   :::{tab-item} ARM64 (Apple Silicon)
   ```sh
   # Deploy a sample RayCluster CR from the KubeRay Helm chart repo:
-  helm install raycluster kuberay/ray-cluster --version 1.1.1 --set 'image.tag=2.9.0-aarch64'
+  helm install raycluster kuberay/ray-cluster --version 1.2.2 --set 'image.tag=2.9.0-aarch64'
   ```
   :::
 
   :::{tab-item} x86-64 (Intel/Linux)
   ```sh
   # Deploy a sample RayCluster CR from the KubeRay Helm chart repo:
-  helm install raycluster kuberay/ray-cluster --version 1.1.1
+  helm install raycluster kuberay/ray-cluster --version 1.2.2
   ```
   :::
 

doc/source/cluster/kubernetes/getting-started/rayjob-quick-start.md

@@ -6,7 +6,7 @@
 
 * KubeRay v0.6.0 or higher
   * KubeRay v0.6.0 or v1.0.0: Ray 1.10 or higher.
-  * KubeRay v1.1.1 is highly recommended: Ray 2.8.0 or higher. This document is mainly for KubeRay v1.1.1.
+  * KubeRay v1.1.1 or newer is highly recommended: Ray 2.8.0 or higher.
 
 ## What's a RayJob?
 
@@ -78,7 +78,7 @@ Follow the [RayCluster Quickstart](kuberay-operator-deploy) to install the lates
 ## Step 3: Install a RayJob
 
 ```sh
-kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.1.1/ray-operator/config/samples/ray-job.sample.yaml
+kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-job.sample.yaml
 ```
 
 ## Step 4: Verify the Kubernetes cluster status
@@ -158,13 +158,13 @@ The Python script `sample_code.py` used by `entrypoint` is a simple Ray script t
 ## Step 6: Delete the RayJob
 
 ```sh
-kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.1.1/ray-operator/config/samples/ray-job.sample.yaml
+kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-job.sample.yaml
 ```
 
 ## Step 7: Create a RayJob with `shutdownAfterJobFinishes` set to true
 
 ```sh
-kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.1.1/ray-operator/config/samples/ray-job.shutdown.yaml
+kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-job.shutdown.yaml
 ```
 
 The `ray-job.shutdown.yaml` defines a RayJob custom resource with `shutdownAfterJobFinishes: true` and `ttlSecondsAfterFinished: 10`.
@@ -192,7 +192,7 @@ kubectl get raycluster
 
 ```sh
 # Step 10.1: Delete the RayJob
-kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.1.1/ray-operator/config/samples/ray-job.shutdown.yaml
+kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-job.shutdown.yaml
 
 # Step 10.2: Delete the KubeRay operator
 helm uninstall kuberay-operator

doc/source/cluster/kubernetes/getting-started/rayservice-quick-start.md

@@ -40,7 +40,7 @@ Please note that the YAML file in this example uses `serveConfigV2` to specify a
 ## Step 3: Install a RayService
 
 ```sh
-kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.1.1/ray-operator/config/samples/ray-service.sample.yaml
+kubectl apply -f https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-service.sample.yaml
 ```
 
 ## Step 4: Verify the Kubernetes cluster status
@@ -110,7 +110,7 @@ curl -X POST -H 'Content-Type: application/json' rayservice-sample-serve-svc:800
 
 ```sh
 # Delete the RayService.
-kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.1.1/ray-operator/config/samples/ray-service.sample.yaml
+kubectl delete -f https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-service.sample.yaml
 
 # Uninstall the KubeRay operator.
 helm uninstall kuberay-operator

doc/source/cluster/kubernetes/index.md

@@ -17,10 +17,10 @@ references
 
 In this section we cover how to execute your distributed Ray programs on a Kubernetes cluster.
 
-Using the [KubeRay Operator](https://github.com/ray-project/kuberay) is the
+Using the [KubeRay operator](https://github.com/ray-project/kuberay) is the
 recommended way to do so. The operator provides a Kubernetes-native way to manage Ray clusters.
 Each Ray cluster consists of a head node pod and a collection of worker node pods. Optional
-autoscaling support allows the KubeRay Operator to size your Ray clusters according to the
+autoscaling support allows the KubeRay operator to size your Ray clusters according to the
 requirements of your Ray workload, adding and removing Ray pods as needed. KubeRay supports
 heterogenous compute nodes (including GPUs) as well as running multiple Ray clusters with
 different Ray versions in the same Kubernetes cluster.

doc/source/cluster/kubernetes/k8s-ecosystem/ingress.md

@@ -32,10 +32,10 @@ Three examples show how to use ingress to access your Ray cluster:
 # Step 1: Install KubeRay operator and CRD
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 helm repo update
-helm install kuberay-operator kuberay/kuberay-operator --version 1.0.0
+helm install kuberay-operator kuberay/kuberay-operator --version 1.2.2
 
 # Step 2: Install a RayCluster
-helm install raycluster kuberay/ray-cluster --version 1.0.0
+helm install raycluster kuberay/ray-cluster --version 1.2.2
 
 # Step 3: Edit the `ray-operator/config/samples/ray-cluster-alb-ingress.yaml`
 #
@@ -122,10 +122,10 @@ Now run the following commands:
 # Step 1: Install KubeRay operator and CRD
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 helm repo update
-helm install kuberay-operator kuberay/kuberay-operator --version 1.0.0
+helm install kuberay-operator kuberay/kuberay-operator --version 1.2.2
 
 # Step 2: Install a RayCluster
-helm install raycluster kuberay/ray-cluster --version 1.0.0
+helm install raycluster kuberay/ray-cluster --version 1.2.2
 
 # Step 3: Edit ray-cluster-gclb-ingress.yaml to replace the service name with the name of the head service from the RayCluster. (Output of `kubectl get svc`)
 
@@ -185,12 +185,12 @@ kubectl wait --namespace ingress-nginx \
 # Step 3: Install KubeRay operator and CRD
 helm repo add kuberay https://ray-project.github.io/kuberay-helm/
 helm repo update
-helm install kuberay-operator kuberay/kuberay-operator --version 1.1.1
+helm install kuberay-operator kuberay/kuberay-operator --version 1.2.2
 
 # Step 4: Install RayCluster and create an ingress separately.
 # More information about change of setting was documented in https://github.com/ray-project/kuberay/pull/699
 # and `ray-operator/config/samples/ray-cluster.separate-ingress.yaml`
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.1.1/ray-operator/config/samples/ray-cluster.separate-ingress.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-cluster.separate-ingress.yaml
 kubectl apply -f ray-operator/config/samples/ray-cluster.separate-ingress.yaml
 
 # Step 5: Check the ingress created in Step 4.

doc/source/cluster/kubernetes/k8s-ecosystem/istio.md

@@ -66,7 +66,7 @@ In this mode, you _must_ disable the KubeRay init container injection by setting
 
 ```bash
 # Set ENABLE_INIT_CONTAINER_INJECTION=false on the KubeRay operator.
-helm upgrade kuberay-operator kuberay/kuberay-operator --version 1.1.1 \
+helm upgrade kuberay-operator kuberay/kuberay-operator --version 1.2.2 \
   --set env\[0\].name=ENABLE_INIT_CONTAINER_INJECTION \
   --set-string env\[0\].value=false
 

doc/source/cluster/kubernetes/k8s-ecosystem/kubeflow.md

@@ -42,7 +42,7 @@ kustomize version --short
 ```sh
 # Create a RayCluster CR, and the KubeRay operator will reconcile a Ray cluster
 # with 1 head Pod and 1 worker Pod.
-helm install raycluster kuberay/ray-cluster --version 1.0.0 --set image.tag=2.2.0-py38-cpu
+helm install raycluster kuberay/ray-cluster --version 1.2.2 --set image.tag=2.2.0-py38-cpu
 
 # Check RayCluster
 kubectl get pod -l ray.io/cluster=raycluster-kuberay

doc/source/cluster/kubernetes/k8s-ecosystem/prometheus-grafana.md

@@ -87,7 +87,7 @@ kubectl get service
   * `# HELP`: Describe the meaning of this metric.
   * `# TYPE`: See [this document](https://prometheus.io/docs/concepts/metric_types/) for more details.
 
-* Three required environment variables are defined in [ray-cluster.embed-grafana.yaml](https://github.com/ray-project/kuberay/blob/v1.0.0/ray-operator/config/samples/ray-cluster.embed-grafana.yaml). See [Configuring and Managing Ray Dashboard](https://docs.ray.io/en/latest/cluster/configure-manage-dashboard.html) for more details about these environment variables.
+* Three required environment variables are defined in [ray-cluster.embed-grafana.yaml](https://github.com/ray-project/kuberay/blob/v1.2.2/ray-operator/config/samples/ray-cluster.embed-grafana.yaml). See [Configuring and Managing Ray Dashboard](https://docs.ray.io/en/latest/cluster/configure-manage-dashboard.html) for more details about these environment variables.
   ```yaml
   env:
     - name: RAY_GRAFANA_IFRAME_HOST

doc/source/cluster/kubernetes/k8s-ecosystem/pyspy.md

@@ -33,7 +33,7 @@ Follow [this document](kuberay-operator-deploy) to install the latest stable Kub
 
 ```bash
 # Download `ray-cluster.py-spy.yaml`
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.0.0/ray-operator/config/samples/ray-cluster.py-spy.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-cluster.py-spy.yaml
 
 # Create a RayCluster
 kubectl apply -f ray-cluster.py-spy.yaml

doc/source/cluster/kubernetes/k8s-ecosystem/volcano.md

@@ -35,7 +35,7 @@ batchScheduler:
 * Pass the `--set batchScheduler.enabled=true` flag when running on the command line:
 ```shell
 # Install the Helm chart with --enable-batch-scheduler flag set to true
-helm install kuberay-operator kuberay/kuberay-operator --version 1.0.0 --set batchScheduler.enabled=true
+helm install kuberay-operator kuberay/kuberay-operator --version 1.2.2 --set batchScheduler.enabled=true
 ```
 
 ### Step 4: Install a RayCluster with the Volcano scheduler
@@ -45,7 +45,7 @@ The RayCluster custom resource must include the `ray.io/scheduler-name: volcano`
 ```shell
 # Path: kuberay/ray-operator/config/samples
 # Includes label `ray.io/scheduler-name: volcano` in the metadata.labels
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.0.0/ray-operator/config/samples/ray-cluster.volcano-scheduler.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-cluster.volcano-scheduler.yaml
 kubectl apply -f ray-cluster.volcano-scheduler.yaml
 
 # Check the RayCluster
@@ -113,7 +113,7 @@ Next, create a RayCluster with a head node (1 CPU + 2Gi of RAM) and two workers
 ```shell
 # Path: kuberay/ray-operator/config/samples
 # Includes  the `ray.io/scheduler-name: volcano` and `volcano.sh/queue-name: kuberay-test-queue` labels in the metadata.labels
-curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.0.0/ray-operator/config/samples/ray-cluster.volcano-scheduler-queue.yaml
+curl -LO https://raw.githubusercontent.com/ray-project/kuberay/v1.2.2/ray-operator/config/samples/ray-cluster.volcano-scheduler-queue.yaml
 kubectl apply -f ray-cluster.volcano-scheduler-queue.yaml
 ```