Add a note about the need to restart the kueue pod (kubernetes-sigs#2685

)
ChristianZaccaria · Jul 24, 2024 · 014d4ec · 014d4ec
1 parent f679118
commit 014d4ec
Show file tree

Hide file tree

Showing 9 changed files with 46 additions and 1 deletion.
diff --git a/site/content/en/docs/tasks/run/jobsets.md b/site/content/en/docs/tasks/run/jobsets.md
@@ -17,6 +17,11 @@ This guide is for [batch users](/docs/tasks#batch-user) that have a basic unders
 
 2. See [JobSet Installation](https://jobset.sigs.k8s.io/docs/installation/) for installation and configuration details of JobSet Operator.
 
+{{% alert title="Note" color="note" %}}
+In order to use JobSet you need to restart Kueue after the installation.
+You can do it by running: `kubectl delete pods -lcontrol-plane=controller-manager -nkueue-system`.
+{{% /alert %}}
+
 ## JobSet definition
 
 When running [JobSets](https://jobset.sigs.k8s.io/docs/concepts/) on

diff --git a/site/content/en/docs/tasks/run/kubeflow/mpijobs.md b/site/content/en/docs/tasks/run/kubeflow/mpijobs.md
@@ -16,7 +16,12 @@ Check [administer cluster quotas](/docs/tasks/manage/administer_cluster_quotas)
 
 Check [the MPI Operator installation guide](https://github.com/kubeflow/mpi-operator#installation).
 
-You can [modify kueue configurations from installed releases](/docs/installation#install-a-custom-configured-released-version) to include MPIJobs as an allowed workload.  
+You can [modify kueue configurations from installed releases](/docs/installation#install-a-custom-configured-released-version) to include MPIJobs as an allowed workload.
+
+{{% alert title="Note" color="note" %}}
+In order to use MPIJob you need to restart Kueue after the installation.
+You can do it by running: `kubectl delete pods -lcontrol-plane=controller-manager -nkueue-system`.
+{{% /alert %}}
 
 ## MPI Operator definition
 

diff --git a/site/content/en/docs/tasks/run/kubeflow/mxjobs.md b/site/content/en/docs/tasks/run/kubeflow/mxjobs.md
@@ -20,6 +20,11 @@ Note that the minimum requirement training-operator version is v1.7.0.
 
 You can [modify kueue configurations from installed releases](/docs/installation#install-a-custom-configured-released-version) to include MXJobs as an allowed workload.
 
+{{% alert title="Note" color="note" %}}
+In order to use Training Operator you need to restart Kueue after the installation.
+You can do it by running: `kubectl delete pods -lcontrol-plane=controller-manager -nkueue-system`.
+{{% /alert %}}
+
 ## MXJob definition
 
 ### a. Queue selection

diff --git a/site/content/en/docs/tasks/run/kubeflow/paddlejobs.md b/site/content/en/docs/tasks/run/kubeflow/paddlejobs.md
@@ -20,6 +20,11 @@ Note that the minimum requirement training-operator version is v1.7.0.
 
 You can [modify kueue configurations from installed releases](/docs/installation#install-a-custom-configured-released-version) to include PaddleJobs as an allowed workload.
 
+{{% alert title="Note" color="note" %}}
+In order to use Training Operator you need to restart Kueue after the installation.
+You can do it by running: `kubectl delete pods -lcontrol-plane=controller-manager -nkueue-system`.
+{{% /alert %}}
+
 ## PaddleJob definition
 
 ### a. Queue selection

diff --git a/site/content/en/docs/tasks/run/kubeflow/pytorchjobs.md b/site/content/en/docs/tasks/run/kubeflow/pytorchjobs.md
@@ -20,6 +20,11 @@ Note that the minimum requirement training-operator version is v1.7.0.
 
 You can [modify kueue configurations from installed releases](/docs/installation#install-a-custom-configured-released-version) to include PyTorchJobs as an allowed workload.
 
+{{% alert title="Note" color="note" %}}
+In order to use Training Operator you need to restart Kueue after the installation.
+You can do it by running: `kubectl delete pods -lcontrol-plane=controller-manager -nkueue-system`.
+{{% /alert %}}
+
 ## PyTorchJob definition
 
 ### a. Queue selection

diff --git a/site/content/en/docs/tasks/run/kubeflow/tfjobs.md b/site/content/en/docs/tasks/run/kubeflow/tfjobs.md
@@ -20,6 +20,11 @@ Note that the minimum requirement training-operator version is v1.7.0.
 
 You can [modify kueue configurations from installed releases](/docs/installation#install-a-custom-configured-released-version) to include TFJobs as an allowed workload.
 
+{{% alert title="Note" color="note" %}}
+In order to use Training Operator you need to restart Kueue after the installation.
+You can do it by running: `kubectl delete pods -lcontrol-plane=controller-manager -nkueue-system`.
+{{% /alert %}}
+
 ## TFJob definition
 
 ### a. Queue selection

diff --git a/site/content/en/docs/tasks/run/kubeflow/xgboostjobs.md b/site/content/en/docs/tasks/run/kubeflow/xgboostjobs.md
@@ -20,6 +20,11 @@ Note that the minimum requirement training-operator version is v1.7.0.
 
 You can [modify kueue configurations from installed releases](/docs/installation#install-a-custom-configured-released-version) to include XGBoostJobs as an allowed workload.
 
+{{% alert title="Note" color="note" %}}
+In order to use Training Operator you need to restart Kueue after the installation.
+You can do it by running: `kubectl delete pods -lcontrol-plane=controller-manager -nkueue-system`.
+{{% /alert %}}
+
 ## XGBoostJob definition
 
 ### a. Queue selection

diff --git a/site/content/en/docs/tasks/run/rayclusters.md b/site/content/en/docs/tasks/run/rayclusters.md
@@ -19,6 +19,11 @@ This guide is for [batch users](/docs/tasks#batch-user) that have a basic unders
 
 3. See [KubeRay Installation](https://ray-project.github.io/kuberay/deploy/installation/) for installation and configuration details of KubeRay.
 
+{{% alert title="Note" color="note" %}}
+In order to use RayCluster you need to restart Kueue after the installation.
+You can do it by running: `kubectl delete pods -lcontrol-plane=controller-manager -nkueue-system`.
+{{% /alert %}}
+
 ## RayCluster definition
 
 When running [RayClusters](https://docs.ray.io/en/latest/cluster/getting-started.html) on

diff --git a/site/content/en/docs/tasks/run/rayjobs.md b/site/content/en/docs/tasks/run/rayjobs.md
@@ -18,6 +18,11 @@ This guide is for [batch users](/docs/tasks#batch-user) that have a basic unders
 
 2. See [KubeRay Installation](https://ray-project.github.io/kuberay/deploy/installation/) for installation and configuration details of KubeRay.
 
+{{% alert title="Note" color="note" %}}
+In order to use RayJob you need to restart Kueue after the installation.
+You can do it by running: `kubectl delete pods -lcontrol-plane=controller-manager -nkueue-system`.
+{{% /alert %}}
+
 ## RayJob definition
 
 When running [RayJobs](https://ray-project.github.io/kuberay/guidance/rayjob/) on