[Cluster] Fix Kuberay capitalization (#37791)

replace Kuberay with KubeRay, and add Kuberay to the list of banned words in the linter

---------

Signed-off-by: Archit Kulkarni <[email protected]>

.gitignore

@@ -16,7 +16,7 @@
 .jar
 /dashboard/client/build
 
-# Kuberay config lives in a separate repository
+# KubeRay config lives in a separate repository
 python/ray/autoscaler/kuberay/config
 
 # Files generated by flatc should be ignored

.vale/styles/Google/WordList.yml

@@ -62,7 +62,7 @@ swap:
  in order to: to
  ingest: import|load
  k8s: Kubernetes
- "[Kk]uberay": KubeRay
+ "[Kk]ube[Rr]ay": KubeRay
  long press: touch & hold
  network IP address: internal IP address
  omnibox: address bar

ci/lint/check-banned-words.sh

@@ -2,7 +2,7 @@
 
 # Checks Python and doc files for common mispellings.
 
-BANNED_WORDS="RLLib Rllib"
+BANNED_WORDS="RLLib Rllib Kuberay"
 
 echo "Checking for common mis-spellings..."
 for word in $BANNED_WORDS; do

doc/source/cluster/vms/user-guides/community/index.rst

@@ -22,7 +22,7 @@ The following is a list of community supported cluster managers.
 Using a custom cloud or cluster manager
 =======================================
 
-The Ray cluster launcher currently supports AWS, Azure, GCP, Aliyun, vSphere and Kuberay out of the box. To use the Ray cluster launcher and Autoscaler on other cloud providers or cluster managers, you can implement the `node_provider.py <https://github.com/ray-project/ray/tree/master/python/ray/autoscaler/node_provider.py>`_ interface (100 LOC).
+The Ray cluster launcher currently supports AWS, Azure, GCP, Aliyun, vSphere and KubeRay out of the box. To use the Ray cluster launcher and Autoscaler on other cloud providers or cluster managers, you can implement the `node_provider.py <https://github.com/ray-project/ray/tree/master/python/ray/autoscaler/node_provider.py>`_ interface (100 LOC).
 Once the node provider is implemented, you can register it in the `provider section <https://github.com/ray-project/ray/blob/master/python/ray/autoscaler/local/example-full.yaml#L18>`_ of the cluster launcher config.
 
 .. code-block:: yaml
@@ -31,5 +31,5 @@ Once the node provider is implemented, you can register it in the `provider sect
  type: "external"
  module: "my.module.MyCustomNodeProvider"
 
-You can refer to `AWSNodeProvider <https://github.com/ray-project/ray/blob/master/python/ray/autoscaler/_private/aws/node_provider.py#L95>`_, `KuberayNodeProvider <https://github.com/ray-project/ray/blob/master/python/ray/autoscaler/_private/kuberay/node_provider.py#L148>`_ and
+You can refer to `AWSNodeProvider <https://github.com/ray-project/ray/blob/master/python/ray/autoscaler/_private/aws/node_provider.py#L95>`_, `KubeRayNodeProvider <https://github.com/ray-project/ray/blob/master/python/ray/autoscaler/_private/kuberay/node_provider.py#L148>`_ and
  `LocalNodeProvider <https://github.com/ray-project/ray/blob/master/python/ray/autoscaler/_private/local/node_provider.py#L166>`_ for more examples.

doc/source/ray-observability/user-guides/debug-apps/debug-memory.rst

@@ -136,7 +136,7 @@ Then check the memory usage from the head node from the node memory usage view i
 The Ray head node has more memory-demanding system components such as GCS or the dashboard. 
 Also, the driver runs from a head node by default. If the head node has the same memory capacity as worker nodes
 and if you execute the same number of Tasks and Actors from a head node, it can easily have out-of-memory problems.
-In this case, do not run any Tasks and Actors on the head node by specifying ``--num-cpus=0`` when starting a head node by ``ray start --head``. If you use Kuberay, view `here <kuberay-num-cpus>`.
+In this case, do not run any Tasks and Actors on the head node by specifying ``--num-cpus=0`` when starting a head node by ``ray start --head``. If you use KubeRay, view `here <kuberay-num-cpus>`.
 
 .. _troubleshooting-out-of-memory-reduce-parallelism:
 

doc/source/serve/production-guide/kubernetes.md

@@ -246,7 +246,7 @@ Monitor your Serve application using the Ray Dashboard.
 - Learn about the [Ray Serve logs](serve-logging) and how to [persistent logs](kuberay-logging) on Kubernetes.
 
 :::{note}
-- To troubleshoot application deployment failures in Serve, you can check the Kuberay operator logs by running `kubectl logs -f <kuberay-operator-pod-name>` (e.g., `kubectl logs -f kuberay-operator-7447d85d58-lv7pf`). The Kuberay operator logs contain information about the Serve application deployment event and Serve application health checks.
+- To troubleshoot application deployment failures in Serve, you can check the KubeRay operator logs by running `kubectl logs -f <kuberay-operator-pod-name>` (e.g., `kubectl logs -f kuberay-operator-7447d85d58-lv7pf`). The KubeRay operator logs contain information about the Serve application deployment event and Serve application health checks.
 - You can also check the controller log and deployment log, which are located under `/tmp/ray/session_latest/logs/serve/` in both the head node pod and worker node pod. These logs contain information about specific deployment failure reasons and autoscaling events.
 :::
 

python/ray/autoscaler/_private/kuberay/node_provider.py

@@ -46,16 +46,16 @@
 # it decreases the number of replicas and adds the exact pods that should be
 # terminated to the scaleStrategy).
 
-# KuberayNodeProvider inherits from BatchingNodeProvider.
+# KubeRayNodeProvider inherits from BatchingNodeProvider.
 # Thus, the autoscaler's create and terminate requests are batched into a single
 # Scale Request object which is submitted at the end of autoscaler update.
 # KubeRay node provider converts the ScaleRequest into a RayCluster CR patch
 # and applies the patch in the submit_scale_request method.
 
-# To reduce potential for race conditions, KuberayNodeProvider
+# To reduce potential for race conditions, KubeRayNodeProvider
 # aborts the autoscaler update if the operator has not yet processed workersToDelete -
-# see KuberayNodeProvider.safe_to_scale().
-# Once it is confirmed that workersToDelete have been cleaned up, KuberayNodeProvider
+# see KubeRayNodeProvider.safe_to_scale().
+# Once it is confirmed that workersToDelete have been cleaned up, KubeRayNodeProvider
 # clears the workersToDelete list.
 
 
@@ -202,25 +202,25 @@ def _worker_group_replicas(raycluster: Dict[str, Any], group_index: int):
  return raycluster["spec"]["workerGroupSpecs"][group_index].get("replicas", 1)
 
 
-class KuberayNodeProvider(BatchingNodeProvider): # type: ignore
+class KubeRayNodeProvider(BatchingNodeProvider): # type: ignore
  def __init__(
  self,
  provider_config: Dict[str, Any],
  cluster_name: str,
  _allow_multiple: bool = False,
  ):
- logger.info("Creating KuberayNodeProvider.")
+ logger.info("Creating KubeRayNodeProvider.")
  self.namespace = provider_config["namespace"]
  self.cluster_name = cluster_name
 
  self.headers, self.verify = load_k8s_secrets()
 
  assert (
  provider_config.get(WORKER_LIVENESS_CHECK_KEY, True) is False
- ), f"To use KuberayNodeProvider, must set `{WORKER_LIVENESS_CHECK_KEY}:False`."
+ ), f"To use KubeRayNodeProvider, must set `{WORKER_LIVENESS_CHECK_KEY}:False`."
  assert (
  provider_config.get(WORKER_RPC_DRAIN_KEY, False) is True
- ), f"To use KuberayNodeProvider, must set `{WORKER_RPC_DRAIN_KEY}:True`."
+ ), f"To use KubeRayNodeProvider, must set `{WORKER_RPC_DRAIN_KEY}:True`."
  BatchingNodeProvider.__init__(
  self, provider_config, cluster_name, _allow_multiple
  )

python/ray/autoscaler/_private/providers.py

@@ -91,9 +91,9 @@ def _import_kubernetes(provider_config):
 
 
 def _import_kuberay(provider_config):
- from ray.autoscaler._private.kuberay.node_provider import KuberayNodeProvider
+ from ray.autoscaler._private.kuberay.node_provider import KubeRayNodeProvider
 
- return KuberayNodeProvider
+ return KubeRayNodeProvider
 
 
 def _import_aliyun(provider_config):
@@ -188,7 +188,7 @@ def _import_external(provider_config):
  "gcp": "GCP",
  "azure": "Azure",
  "kubernetes": "Kubernetes",
- "kuberay": "Kuberay",
+ "kuberay": "KubeRay",
  "aliyun": "Aliyun",
  "external": "External",
  "vsphere": "vSphere",

python/ray/autoscaler/kuberay/init-config.sh

@@ -1,6 +1,6 @@
 #!/bin/bash
 
-# Clone pinned Kuberay commit to temporary directory, copy the CRD definitions
+# Clone pinned KubeRay commit to temporary directory, copy the CRD definitions
 # into the autoscaler folder.
 KUBERAY_BRANCH="v1.0.0"
 OPERATOR_TAG="v1.0.0"

python/ray/autoscaler/kuberay/ray-cluster.complete.yaml

@@ -1,5 +1,5 @@
 # This is adapted from https://github.com/ray-project/kuberay/blob/master/ray-operator/config/samples/ray-cluster.complete.yaml
-# It is a general RayCluster that has most fields in it for maximum flexibility in the Ray/Kuberay integration MVP.
+# It is a general RayCluster that has most fields in it for maximum flexibility in the Ray/KubeRay integration MVP.
 apiVersion: ray.io/v1alpha1
 kind: RayCluster
 metadata:

python/ray/tests/kuberay/test_kuberay_node_provider.py

@@ -10,7 +10,7 @@
  _worker_group_index,
  _worker_group_max_replicas,
  _worker_group_replicas,
- KuberayNodeProvider,
+ KubeRayNodeProvider,
  ScaleRequest,
 )
 from ray.autoscaler._private.util import NodeID
@@ -88,7 +88,7 @@ def test_worker_group_replicas(group_index, expected_max_replicas, expected_repl
 def test_create_node_cap_at_max(
  attempted_target_replica_count, expected_target_replica_count
 ):
- """Validates that KuberayNodeProvider does not attempt to create more nodes than
+ """Validates that KubeRayNodeProvider does not attempt to create more nodes than
  allowed by maxReplicas. For the config in this test, maxReplicas is fixed at 300.
 
  Args:
@@ -98,8 +98,8 @@ def test_create_node_cap_at_max(
  capped at maxReplicas (300, for the config in this test.)
  """
  raycluster = get_basic_ray_cr()
- with mock.patch.object(KuberayNodeProvider, "__init__", return_value=None):
- kr_node_provider = KuberayNodeProvider(provider_config={}, cluster_name="fake")
+ with mock.patch.object(KubeRayNodeProvider, "__init__", return_value=None):
+ kr_node_provider = KubeRayNodeProvider(provider_config={}, cluster_name="fake")
  scale_request = ScaleRequest(
  workers_to_delete=set(),
  desired_num_workers={"small-group": attempted_target_replica_count},
@@ -171,9 +171,9 @@ def mock_get(node_provider, path):
  raise ValueError("Invalid path.")
 
  with mock.patch.object(
- KuberayNodeProvider, "__init__", return_value=None
- ), mock.patch.object(KuberayNodeProvider, "_get", mock_get):
- kr_node_provider = KuberayNodeProvider(provider_config={}, cluster_name="fake")
+ KubeRayNodeProvider, "__init__", return_value=None
+ ), mock.patch.object(KubeRayNodeProvider, "_get", mock_get):
+ kr_node_provider = KubeRayNodeProvider(provider_config={}, cluster_name="fake")
  kr_node_provider.cluster_name = "fake"
  nodes = kr_node_provider.non_terminated_nodes({})
  assert kr_node_provider.node_data_dict == expected_node_data
@@ -238,16 +238,16 @@ def mock_get(node_provider, path):
  ],
 )
 def test_submit_scale_request(node_data_dict, scale_request, expected_patch_payload):
- """Test the KuberayNodeProvider's RayCluster patch payload given a dict
+ """Test the KubeRayNodeProvider's RayCluster patch payload given a dict
  of current node counts and a scale request.
  """
  raycluster = get_basic_ray_cr()
  # Add another worker group for the sake of this test.
  blah_group = copy.deepcopy(raycluster["spec"]["workerGroupSpecs"][1])
  blah_group["groupName"] = "blah-group"
  raycluster["spec"]["workerGroupSpecs"].append(blah_group)
- with mock.patch.object(KuberayNodeProvider, "__init__", return_value=None):
- kr_node_provider = KuberayNodeProvider(provider_config={}, cluster_name="fake")
+ with mock.patch.object(KubeRayNodeProvider, "__init__", return_value=None):
+ kr_node_provider = KubeRayNodeProvider(provider_config={}, cluster_name="fake")
  kr_node_provider.node_data_dict = node_data_dict
  patch_payload = kr_node_provider._scale_request_to_patch_payload(
  scale_request=scale_request, raycluster=raycluster
@@ -277,9 +277,9 @@ def mock_patch(kuberay_provider, path, patch_payload):
  kuberay_provider._patched_raycluster = patch.apply(kuberay_provider._raycluster)
 
  with mock.patch.object(
- KuberayNodeProvider, "__init__", return_value=None
- ), mock.patch.object(KuberayNodeProvider, "_patch", mock_patch):
- kr_node_provider = KuberayNodeProvider(provider_config={}, cluster_name="fake")
+ KubeRayNodeProvider, "__init__", return_value=None
+ ), mock.patch.object(KubeRayNodeProvider, "_patch", mock_patch):
+ kr_node_provider = KubeRayNodeProvider(provider_config={}, cluster_name="fake")
  kr_node_provider.cluster_name = "fake"
  kr_node_provider._patched_raycluster = raycluster
  kr_node_provider._raycluster = raycluster