diff --git a/content/docs/components/training/tftraining.md b/content/docs/components/training/tftraining.md index 9d68e0e5a9..b703ea292f 100644 --- a/content/docs/components/training/tftraining.md +++ b/content/docs/components/training/tftraining.md @@ -211,7 +211,7 @@ kubectl apply -f tfevent-volume kubectl apply -f tf_job_mnist.yaml ``` -Monitor the job (see the [TFJob docs](/docs/components/tftraining/#monitoring-your-job)): +Monitor the job (see the [detailed guide below](#monitoring-your-job)): ``` kubectl -n kubeflow get tfjob mnist -o yaml @@ -238,12 +238,8 @@ Typically you can change the following values in the TFJob yaml file: ### Accessing the TFJob dashboard -The TFJob dashboard is available at `/tfjobs/ui/`. Specifically: - -* If you're using the central Kubeflow UI, you can access the TFJob dashboard - by clicking **TFJOB DASHBOARD**: - - ![Central UI](/docs/images/central-ui.png) +The TFJob dashboard has the title **kubeflow/tf-operator**. +You can access it at `/tfjobs/ui/`. Specifically: * If you followed the guide to [deploying Kubeflow on GCP](/docs/gke/deploy/), you can diff --git a/content/docs/gke/pipelines-tutorial.md b/content/docs/gke/pipelines-tutorial.md index fc5125dac8..4ab5d3b87e 100644 --- a/content/docs/gke/pipelines-tutorial.md +++ b/content/docs/gke/pipelines-tutorial.md @@ -87,6 +87,12 @@ Notes: [clean up your GCP resources](#cleanup) when you've finished with them. * This guide uses [Cloud Shell][cloud-shell] to manage your GCP environment, to save you the steps of installing [Cloud SDK][cloud-sdk] and [kubectl][kubectl]. +### Start your Cloud Shell + +Follow the link to activate a +[Cloud Shell environment](https://console.cloud.google.com/cloudshell) in your +browser. + ### Set up some handy environment variables Set up the following environment variables for use throughout the tutorial: @@ -190,7 +196,7 @@ Deploy Kubeflow on GCP: alt="Prediction UI" class="mt-3 mb-3 p-3 border border-info rounded"> -1. Click **Pipeline Dashboard** to access the pipelines UI. The pipelines UI +1. Click **Pipelines** to access the pipelines UI. The pipelines UI looks like this: Pipelines UI @@ -69,13 +70,24 @@ getting-started guide for your chosen environment. For example: to your Google Account you may not need to log in again.) * On all other platforms, sign in using any username and password. +1. Select a namespace: + * Click the namespace dropdown to see the list of available namespaces. + * Choose the namespace that corresponds to your Kubeflow profile. (See + the page on [multi-user isolation](/docs/other-guides/multi-user-overview/) + for more information about namespaces.) + + Selecting a Kubeflow namespace + 1. Click **NEW SERVER** on the **Notebook Servers** page: The Kubeflow notebook servers page - You should see the **New Notebook Server** page: + You should see a page for entering details of your new server. Here is a + partial screenshot of the page: Form for adding a Kubeflow notebook server +## URL pattern with Google Cloud Platform (GCP) -## Google Cloud Platform (Kubernetes Engine) - -If you followed the guide to [deploying Kubeflow on Google Cloud Platform -(GCP)](/docs/gke/deploy/), Kubeflow -is deployed with Cloud Identity-Aware Proxy (Cloud IAP) or basic authentication, -and the Kubeflow landing page is accessible at a URL of the following pattern: +If you followed the guide to [deploying Kubeflow on GCP](/docs/gke/deploy/), +the Kubeflow central UI is accessible at a URL of the following pattern: ``` -https://.endpoints..cloud.goog/ +https://.endpoints..cloud.goog/ ``` -This URL brings up the landing page illustrated above. +The URL brings up the dashboard illustrated above. -When deployed with Cloud IAP, Kubeflow uses the +If you deploy Kubeflow with Cloud Identity-Aware Proxy (IAP), Kubeflow uses the [Let's Encrypt](https://letsencrypt.org/) service to provide an SSL certificate for the Kubeflow UI. For troubleshooting issues with your certificate, see the guide to [monitoring your Cloud IAP setup](/docs/gke/deploy/monitor-iap-setup/). -## Using Kubectl and port-forwarding +## Using kubectl and port-forwarding -If you're not using the Cloud IAP option or if you haven't yet set up your -Kubeflow endpoint, you can access Kubeflow via `kubectl` and port-forwarding. +You can access Kubeflow via `kubectl` and port-forwarding as follows: 1. Install `kubectl` if you haven't already done so: @@ -72,18 +67,21 @@ Kubeflow endpoint, you can access Kubeflow via `kubectl` and port-forwarding. http://localhost:8080/ ``` - * This will only work if you haven't enabled basic auth or Cloud IAP. If - authentication is enabled requests will be rejected - because you are not connecting over HTTPS and attaching proper credentials. + * Port-forwarding will not work if you're using basic authentication with GCP. - * Depending on how you've configured Kubeflow, not all UIs will work behind port-forwarding to the reverse proxy. + * Depending on how you've configured Kubeflow, not all UIs work behind + port-forwarding to the reverse proxy. - * Some web applications need to be configured to know the base URL they are serving on. - * So if you deployed Kubeflow with an ingress serving at `https://acme.mydomain.com` and configured an application - to be served at the URL `https://acme.mydomain.com/myapp` then the app may not work when served on - `https://localhost:8080/myapp` because the paths do not match. + For some web applications, you need to configure the base URL on which + the app is serving. + + For example, if you deployed Kubeflow with an ingress serving at + `https://example.mydomain.com` and configured an application + to be served at the URL `https://example.mydomain.com/myapp`, then the + app may not work when served on + `https://localhost:8080/myapp` because the paths do not match. ## Next steps -See how to [set up your Jupyter notebooks](/docs/notebooks/setup/) in -Kubeflow. +* See how to [access the TFJob dashboard](/docs/components/training/tftraining/). +* [Set up your Jupyter notebooks](/docs/notebooks/setup/) in Kubeflow. diff --git a/content/docs/pipelines/pipelines-quickstart.md b/content/docs/pipelines/pipelines-quickstart.md index 6e2be2a440..ebf91e8aa5 100644 --- a/content/docs/pipelines/pipelines-quickstart.md +++ b/content/docs/pipelines/pipelines-quickstart.md @@ -19,9 +19,7 @@ Kubeflow Pipelines. If you need a more in-depth guide, see the Follow these steps to deploy Kubeflow and open the pipelines dashboard: -1. Follow the guide to [deploying Kubeflow on GCP](/docs/gke/deploy/), - including the step to deploy Kubeflow using the - [Kubeflow deployment UI](https://deploy.kubeflow.cloud/). +1. Follow the guide to [deploying Kubeflow on GCP](/docs/gke/deploy/). {{% pipelines-compatibility %}} @@ -44,7 +42,7 @@ Follow these steps to deploy Kubeflow and open the pipelines dashboard: 1. Run ```kubectl port-forward -n kubeflow svc/ml-pipeline-ui 8080:80``` and go to `http://localhost:8080/`. -1. Click **Pipeline Dashboard** to access the pipelines UI. The pipelines UI looks like +1. Click **Pipelines** to access the pipelines UI. The pipelines UI looks like this: Pipelines UI -1. Click **Create an experiment**: +1. Click **Create experiment**: Starting an experiment on the pipelines UI @@ -90,7 +88,7 @@ workload: You can find the source code for the basic parallel join sample in the [Kubeflow Pipelines -repo](https://github.com/kubeflow/pipelines/blob/master/samples/basic/parallel_join.py). +repo](https://github.com/kubeflow/pipelines/blob/master/samples/core/parallel_join/parallel_join.py). ## Run an ML pipeline @@ -124,7 +122,7 @@ Follow these steps to set up the necessary GCP services and run the sample: alt="XGBoost sample on the pipelines UI" class="mt-3 mb-3 border border-info rounded"> -1. Click **Create an experiment**. +1. Click **Create experiment**. 1. Follow the prompts to create an **experiment** and then create a **run**. Supply the following **run parameters**: @@ -163,7 +161,7 @@ Follow these steps to set up the necessary GCP services and run the sample: You can find the source code for the XGBoost training sample in the [Kubeflow Pipelines -repo](https://github.com/kubeflow/pipelines/tree/master/samples/xgboost-spark). +repo](https://github.com/kubeflow/pipelines/tree/master/samples/core/xgboost-spark). ## Clean up your GCP environment diff --git a/layouts/shortcodes/code-webui-port-forward.html b/layouts/shortcodes/code-webui-port-forward.html index bc9c2808ce..1df8dcf12a 100644 --- a/layouts/shortcodes/code-webui-port-forward.html +++ b/layouts/shortcodes/code-webui-port-forward.html @@ -1,3 +1,3 @@ -
export NAMESPACE=kubeflow
+
export NAMESPACE=istio-system
 kubectl port-forward svc/ambassador -n ${NAMESPACE} 8080:80
 

\ No newline at end of file