<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Danny Thomson on Medium]]></title>
        <description><![CDATA[Stories by Danny Thomson on Medium]]></description>
        <link>https://medium.com/@dthomson25?source=rss-7b34a52228e2------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*6huGrWhsStAmvtaTdyydlg.jpeg</url>
            <title>Stories by Danny Thomson on Medium</title>
            <link>https://medium.com/@dthomson25?source=rss-7b34a52228e2------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 12:08:30 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@dthomson25/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Argo Rollouts v0.8 Release!]]></title>
            <link>https://blog.argoproj.io/argo-rollouts-v0-8-release-f14d2329cbd9?source=rss-7b34a52228e2------2</link>
            <guid isPermaLink="false">https://medium.com/p/f14d2329cbd9</guid>
            <category><![CDATA[argoproj]]></category>
            <category><![CDATA[kubernetes]]></category>
            <category><![CDATA[progressive-delivery]]></category>
            <category><![CDATA[canary-deployments]]></category>
            <category><![CDATA[blue-green-deployment]]></category>
            <dc:creator><![CDATA[Danny Thomson]]></dc:creator>
            <pubDate>Wed, 15 Apr 2020 23:51:43 GMT</pubDate>
            <atom:updated>2020-04-15T23:51:43.588Z</atom:updated>
            <content:encoded><![CDATA[<p>The ArgoProj team is excited to announce that Argo Rollout v0.8 has been released. It is one of our biggest releases yet and has received contributions from 5 separate companies! Below is a list of some of the big exciting new features with this release:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/458/0*9SrKNF28eA19yNXZ" /></figure><h4>Nginx Ingress Controller Support</h4><p>One of the most popular Ingress controllers for Kubernetes is the Nginx Ingress controller. It offers support for fine-grain traffic shaping to allow users to send a percentage of traffic to a canary Ingress that mirrors the primary Ingress. The Argo Rollout controller now understands how to create and update a canary Ingress and uses that understanding to send the desired percentage of traffic to the canary ReplicaSet.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/512/0*PhCpoVWYoApqnAmx.png" /></figure><p>Check out the <a href="https://argoproj.github.io/argo-rollouts/features/traffic-management/nginx/">documentation here</a>!</p><h4>AWS ALB Ingress Controller Support</h4><p>The Application Load Balancer (ALB) is a popular way of connecting external traffic to resources within an AWS account. An ALB is highly configurable and the ALB Ingress controller allows users to automate the creation of ALBs for Kubernetes clusters. ALBs have a concept of weighted target groups to allow users to route traffic to different targets with specific weights, and the ALB Ingress controller added support for the weighted target groups in <a href="https://github.com/kubernetes-sigs/aws-alb-ingress-controller/releases/tag/v1.1.5">v1.1.5</a>. In Argo Rollouts v0.8, a Rollout can be configured to use a specific Ingress for traffic shaping. The controller then modifies the AWS ALB Ingress to set the desired weight for traffic to the canary and stable ReplicaSets.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/270/1*HTYmQwU1OEGB-t2mTME72w.png" /></figure><p>Check out the <a href="https://argoproj.github.io/argo-rollouts/features/traffic-management/alb/">documentation here</a>!</p><h4>Blue Green Analysis</h4><p>Previously, only Rollouts using the canary strategy could create AnalysisRuns to verify a new ReplicaSet. Now, users can configure their BlueGreen Rollouts to create an AnalysisRun before or after promoting the new ReplicaSet to the active service. This allows users to run a smoke test before they promote the Rollout or revert a deployed Rollout if the analysis fails!</p><p>Check out the <a href="https://argoproj.github.io/argo-rollouts/features/analysis/#bluegreen-pre-promotion-analysis">documentation here</a>!</p><h4>Scheduling Pods Restarts</h4><p>Similar to a deployment, Argo Rollouts supports restarting the pods owned by the Rollout. Users can restart all the pods of a Rollout by running kubectl argo rollouts restart ROLLOUT_NAME or setting the .spec.RestartAt field to the time they want the pods to restart. One added benefit of this approach is that users can set the time in the future which allows them to schedule a restart.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*nS4nTsbL80r-Cg1pS7uEvg.png" /></figure><p>Check out the <a href="https://argoproj.github.io/argo-rollouts/features/restart/">documentation here</a>!</p><h4>Anti-Affinity</h4><p>After running BlueGreen Rollouts at Intuit, we noticed that Pods would occasionally by restarted after a deployment. This occurred because the cluster auto scaler would scale up the cluster for the deployment and then scale down after the deployment. To address that issue, Argo Rollout can add anti-affinity to new ReplicaSets to prevent the pod it creates from running on nodes with the previous version.</p><p>Check out the <a href="https://argoproj.github.io/argo-rollouts/features/anti-affinity/anti-affinity/">documentation here</a>!</p><h3>Where Can I Get This?</h3><p>For more details and installation instructions, check the v0.8 release notes</p><h3>Thanks to the Community!</h3><p>Thank you to the following contributors for their work in this release!</p><ul><li>cronik</li><li>dthomson25</li><li>duboisf</li><li>jessesuen</li><li>khhirani</li><li>moensch</li><li>nghialv</li></ul><h3>What’s next</h3><p>For anyone who is curious what’s coming in the v0.9 release, the ArgoProj is focusing on the following big features:</p><ul><li><a href="https://smi-spec.io/">Service Mesh Interface (SMI)</a> support</li><li><a href="https://github.com/argoproj/argo-rollouts/issues/358">Validation for the Argo Rollouts resources</a></li></ul><p>Additionally, the ArgoProj team is considering how we can integrate Header-based routing into Argo Rollouts. We created the <a href="https://github.com/argoproj/argo-rollouts/issues/474">following issue</a> to track our progress and welcome any feedback!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f14d2329cbd9" width="1" height="1" alt=""><hr><p><a href="https://blog.argoproj.io/argo-rollouts-v0-8-release-f14d2329cbd9">Argo Rollouts v0.8 Release!</a> was originally published in <a href="https://blog.argoproj.io/">Argo Project</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introducing Argo Rollouts]]></title>
            <link>https://blog.argoproj.io/introducing-argo-rollouts-59dd0fad476c?source=rss-7b34a52228e2------2</link>
            <guid isPermaLink="false">https://medium.com/p/59dd0fad476c</guid>
            <category><![CDATA[kubernetes]]></category>
            <category><![CDATA[canary-deployments]]></category>
            <category><![CDATA[argoproj]]></category>
            <category><![CDATA[blue-green-deployment]]></category>
            <category><![CDATA[cicd]]></category>
            <dc:creator><![CDATA[Danny Thomson]]></dc:creator>
            <pubDate>Mon, 20 May 2019 21:29:04 GMT</pubDate>
            <atom:updated>2019-05-27T03:53:30.143Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/850/0*Bdurdo54dEl5AWF_.jpg" /></figure><p>Less than a year ago, we started building <a href="https://argoproj.github.io/argo-cd/"><strong>Argo CD</strong></a>, a <strong>declarative, GitOps continuous delivery</strong> tool for Kubernetes. The project has been growing in popularity, with many developers and DevOps engineers adopting it. A huge thanks to our awesome community of contributors and users!</p><p>Today, we are excited to announce <a href="https://github.com/argoproj/argo-rollouts"><strong>Argo Rollouts</strong></a><strong> </strong>a new, Kubernetes-native, open source project brought to you by the Argo community and Intuit to provide <strong>missing deployment strategies like Blue/Green and Canary in a Kubernetes native and GitOps friendly manner</strong>. Argo Rollouts augments the functionality of the Kubernetes Deployment resource with additional deployment strategies.</p><p>You can watch the introduction and demo of Argo Rollouts at KubeCon Barcelona 2019 <a href="https://www.youtube.com/watch?v=yeVkTTO9nOA&amp;feature=youtu.be&amp;t=589">here</a>!</p><h4>Why Argo Rollouts?</h4><p>Argo Rollouts fills the need for the missing industry standard deployment strategies like <strong>Canary</strong> and <strong>Blue Green</strong> in Kubernetes. If a user wanted to use these strategies previously, they would have to orchestrate the deployment strategies themselves. Users were forced to craft fragile Jenkins scripts and overly repetitive Kubernetes manifests.</p><p>Using the learnings from those attempts, the Argoproj team tried to offer Pre-Sync, Sync, and Post-Sync hooks in Argo CD as a potential solution. However, we found that solution falls short of the ideal user experience we wanted to provide. It did not provide the visibility we would like to have during a deployment and still put the onus on the user to orchestrate the deployments.</p><p>We went back to the drawing board and realized that <strong>the logic can be encapsulated into a Custom Resource Definition</strong>. From there, we came up with the following fundamental requirements for Argo Rollouts to guide us during development:</p><ul><li>The deployment orchestration is abstracted from the users</li><li>Reuse as much of the Kubernetes deployment object spec as possible to make adoption easy</li><li>Provide a GitOps friendly model that seamlessly plugs into Argo CD</li><li>Application deployments should be reliable and idempotent</li></ul><h4>How does Argo Rollouts controller work?</h4><p>Argo Rollouts introduces a controller into a Kubernetes cluster to manage a new object type called a <strong>Rollout</strong>. A Rollout object is identical to a Deployment object except for a couple of keys fields. Below is an example of a Kubernetes Deployment spec converted to use an Argo Rollout using the BlueGreen deployment strategy.</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/d0750ee2de52e3ff75d8f10460e76fce/href">https://medium.com/media/d0750ee2de52e3ff75d8f10460e76fce/href</a></iframe><p>Three changes were required to convert a Deployment into a Rollout:</p><ol><li>Changing kind from Deployment to Rollout</li><li>Changing apiVersion to argoproj.io/v1alpha1</li><li>Adding a specific deployment strategy in the rollout (Blue Green in the example above)</li></ol><p>For the specific deployment strategy, Argo Rollouts currently provides Blue Green and ReplicaSet Based Canary.</p><p><strong>Blue Green Deployments</strong></p><p>For context, Blue Green deployments allow users deploy and run both the new and the old versions of their applications at the same time. The old version will receive the regular traffic, and developers can run tests on the new version until they feel comfortable with rolling out the new version. Then, the traffic is cut over from the old version to the new version, and the old version is eventually scaled down.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n6qJ4Ottpore9nZ2yUfYrQ.png" /><figcaption>Network View in Argo CD showing what pods are being selected by the Service’s Selector</figcaption></figure><p>The Argo Rollouts controller achieves this strategy by managing the ReplicaSets and filtering the traffic by modifying Service selectors. The ReplicaSets are created from the spec.template field of the Rollout and services are specified in the spec.strategy.blueGreen field. Each ReplicaSet created by the rollout has a unique hash (in the rollouts-pod-template-hash in the label) that the controller will add to the service’s selector to limit traffic to that one ReplicaSet.</p><p>When the spec.template field changes, the Rollout will create a new ReplicaSet and wait for it to become available. Once that occurs, the controller will modify the preview Service’s selector to send traffic to the new Replicaset and enter a paused state by setting the spec.paused field to true. During this time, the Rollout will pause and wait until an external operator (like a CD system or a user) changes the value to false. After the Rollout is unpaused, the Rollout will modify the active service’s selector to point it at the new ReplicaSet and scale down the old one.</p><p><strong>ReplicaSet Based Canary Deployments</strong></p><p>A canary rollout is a deployment strategy where the operator releases a new version of their application to a small percentage of the real traffic. Canaries are often used to verify the new release is functional and performant without having the risk of exposing the new version to all traffic.</p><p>In a ReplicaSet-based Canary rollout, the traffic is split evenly among all running pods. To achieve the canary rollout, users specify a list of steps to execute to transition to the new version using the spec.Strategy.Canary field. The available steps are setWeight, which indicates the percentage of the overall replica count that should be in the new replicaset and pause, which freezes the Rollout until a specified event occurs. When a new version is introduced, the Argo Rollout controller will execute each of these steps. Once it does, the controller will mark the current template within the rollout to be the new stable resource. Below is an example of a Canary Rollout:</p><iframe src="" width="0" height="0" frameborder="0" scrolling="no"><a href="https://medium.com/media/6724643c6088bf4d11fcd4550d7b057a/href">https://medium.com/media/6724643c6088bf4d11fcd4550d7b057a/href</a></iframe><p>When a new template is introduced in the above example, the rollout will first create the new replicaset with a replicas count of 1 and scale down the stable replicaset to 9 to achieve a weight of 10%. Once the pods are marked as available, the rollout will enter a paused state for 30 seconds before resuming. The controller will incrementally progress to a weight of 50% (5 new and 5 stable). At that point, the rollout will enter a paused state until an external operator unpauses it because the pause does not have a duration. After unpausing, the controller will slowly roll all the replicas to the new ReplicaSet before marking it as the new stable ReplicaSet as it executed all the steps.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MKVPKGa_f2Qn6ejw_qWHOQ.png" /><figcaption>A rollout with 5 replicas using the canary strategy to splitting 20% of traffic to a new version</figcaption></figure><p><strong>Trying out Argo Rollouts</strong></p><p>You can get started in 2 easy steps</p><ul><li>Install the Argo Rollouts controller into your namespace with the following commands:</li></ul><pre>kubectl create namespace argo-rollouts<br>kubectl apply -n argo-rollouts -f https://raw.githubusercontent.com/argoproj/argo-rollouts/stable/manifests/install.yaml</pre><ul><li>Switch your deployment to a Rollout by switching the kind and apiVersion and adding a deployment strategy like the examples above.</li></ul><h4>Argo Rollouts features</h4><ul><li>Blue Green Deployments</li><li>ReplicaSet-based Canary Deployments with customizable steps</li><li>Integration with Argo CD</li><li>HPA support</li><li>Controller level Prometheus metrics</li></ul><h4>Going forward</h4><p>We plan to continue growing Argo Rollouts functionality in the upcoming months. We want to incorporate Istio and other service mesh tools to support smarter traffic shaping, implement support for running multiple concurrent versions of software for experimentation and build a promotion-automation CRD. If any of this sounds exciting to you, please reach out in the <a href="https://join.slack.com/t/argoproj/shared_invite/enQtMzExODU3MzIyNjYzLTA5MTFjNjI0Nzg3NzNiMDZiNmRiODM4Y2M1NWQxOGYzMzZkNTc1YWVkYTZkNzdlNmYyZjMxNWI3NjY2MDc1MzI">Argoproj slack channel</a>!</p><p>Thank You!</p><p>The Argoproj Community</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=59dd0fad476c" width="1" height="1" alt=""><hr><p><a href="https://blog.argoproj.io/introducing-argo-rollouts-59dd0fad476c">Introducing Argo Rollouts</a> was originally published in <a href="https://blog.argoproj.io/">Argo Project</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Using GitOps to Deploy Kubeflow with Argo CD]]></title>
            <link>https://blog.argoproj.io/using-gitops-to-deploy-kubeflow-with-argo-cd-76f6b27807c?source=rss-7b34a52228e2------2</link>
            <guid isPermaLink="false">https://medium.com/p/76f6b27807c</guid>
            <category><![CDATA[kubeflow]]></category>
            <category><![CDATA[kubernetes]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[gitops]]></category>
            <category><![CDATA[argo]]></category>
            <dc:creator><![CDATA[Danny Thomson]]></dc:creator>
            <pubDate>Fri, 26 Oct 2018 17:10:41 GMT</pubDate>
            <atom:updated>2018-10-26T17:45:22.753Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*dxy2UUUP7jtJxfNj.jpg" /></figure><p><a href="http://kubeflow.org/">Kubeflow</a> is an open source project for making machine learning (ML) on Kubernetes simple, portable, and scalable. <a href="https://github.com/argoproj/argo-cd">Argo CD</a> is a GitOps-based Continuous Delivery tool for Kubernetes. With Argo CD, you specify the desired state of your applications on Kubernetes using declarative specifications and Argo CD will reconcile the differences between the desired state and the actual live state in your Kubernetes cluster.</p><p>This blog post shows how you can get the benefits of GitOps for Kubeflow by using Argo CD as a deployment tool. In a future blog post, we will show you how you can extend these benefits by using Kubeflow with Argo CD to train and deploy your machine learning models. Essentially, GitOps for ML!</p><p>The rest of the article will focus on walking through the steps to bring up a GKE cluster and using Argo CD to deploy Kubeflow from a Github repo. It mostly follows the regular <a href="https://www.kubeflow.org/docs/started/getting-started-gke">Kubeflow GKE Getting Started Guide</a> with slight variations to install Argo CD and setup your Github repo.</p><p>Please check out the <a href="https://drive.google.com/a/kubeflow.org/file/d/1jXri3ye6EweRMDEs_P5KRFUeJjEGF4D8/view?usp=sharing">demo</a> at the recent Kubeflow Community Meeting.</p><h3>Creating OAuth Credentials</h3><p>The first step in setting up your GKE cluster with Kubeflow is creating your OAuth client credentials by following the steps in the <a href="https://www.kubeflow.org/docs/started/getting-started-gke/#Create-oauth-client-credentials">GKE Getting Started Guide</a>. You should now have two environmental variables.</p><pre>export CLIENT_ID=&lt;CLIENT_ID from OAuth page&gt;<br>export CLIENT_SECRET=&lt;CLIENT_SECRET from OAuth page&gt;</pre><h3>Creating the GKE Cluster</h3><p>Kubeflow has a CLI tool called kfctl.sh to simplify the GKE cluster process. The CLI tool uses the <a href="https://cloud.google.com/deployment-manager/docs/">Deployment Manager</a> to declaratively manage all non K8s resources. Since the Deployment Manager has a declarative specification, you can to customize your GKE cluster as per their requirements.</p><p>In order for the kfctl.sh tool to deploy the configs to the Deployment Manager, the Google Cloud SDK and kubectl binary need to be installed. You can install the Google Cloud SDK with the instructions <a href="https://cloud.google.com/sdk/downloads">here</a> and install kubectl by running gcloud components install kubectl. For this demo, I decided to deploy my GKE cluster to us-central-1c, and I ran gcloud config set compute/zone us-central1-c to set the zone.</p><p>To download the kfctl.sh tool, you can run</p><pre>mkdir ${KUBEFLOW_REPO}<br>cd ${KUBEFLOW_REPO}<br>export KUBEFLOW_TAG=<a href="https://github.com/kubeflow/kubeflow/releases/tag/v0.3.1">v0.3.1</a> <br>curl https://raw.githubusercontent.com/kubeflow/kubeflow/${KUBEFLOW_TAG}/scripts/download.sh | bash</pre><ul><li><strong>KUBEFLOW_REPO</strong> path to a directory where you want to download the source to. (i.e. /tmp/kubeflow/)</li><li><strong>KUBEFLOW_TAG</strong> a tag corresponding to the version to checkout such as master for latest code. You can find the latest version <a href="https://github.com/kubeflow/kubeflow/releases">here</a>.</li><li>Alternatively, you can also just clone the Kubeflow repository using git.</li></ul><p>Now that you have the kftcl.sh downloaded, you can use it to create the config to your GKE cluster by running:</p><pre>./scripts/kfctl.sh init ${KFAPP} --platform gcp --project ${PROJECT}<br>cd ${KFAPP}<br>${KUBEFLOW_REPO}/scripts/kfctl.sh generate platform gcp<br>${KUBEFLOW_REPO}/scripts/kfctl.sh apply platform gcp</pre><ul><li><strong>KFAPP</strong> The name of a directory to store your configs. This directory will be created when you run init and will become your Github repo.</li><li><strong>PROJECT</strong> Your Google Cloud project name</li></ul><p>Once the kfctl.sh apply platform gcp command finishes, you will have a GKE cluster provisioned for Kubeflow. If your deployment fails, I recommend checking this <a href="https://www.kubeflow.org/docs/guides/gke/troubleshooting-gke/">GKE troubleshooting guide</a>.</p><h3>Commit your Kubeflow <strong>Ksonnet</strong> to Github</h3><p>In the <strong>KUBEFLOW_REPO</strong>, you will need to run the following command to generate the Ksonnet components that define your Kubeflow application.</p><pre>${KUBEFLOW_REPO}/scripts/kfctl.sh generate k8s</pre><p>If you would like to learn more on Ksonnet components, I recommend checking out the <a href="https://www.kubeflow.org/docs/guides/components/ksonnet/#why-kubeflow-uses-ksonnet">Kubeflow documentation on ksonnet</a>.</p><p>Next, we need to add a ksonnet environment to our directory. This command requires the ksonnet binary ks to be installed on your machine. You can follow the directions <a href="https://www.kubeflow.org/docs/guides/components/ksonnet/#installing-ksonnet">here</a> to install ksonnet. Once you have ksonnet setup, you can run:</p><pre>cd ${KUBEFLOW_REPO}/${KFAPP}/ks_app/<br>ks env add default --server https://kubernetes.default.svc</pre><p>Since I am deploying Argo CD to the same Kubernetes cluster as Kubeflow, I added the --server <a href="https://kubernetes.default.svc/">https://kubernetes.default.svc</a> flag to set the API server endpoint for this environment to the internal Kubernetes API server endpoint. Two nice side effects with setting our ksonnet environment like this are that Argo CD can deploy Kubeflow without having to go to the external internet and developers can not run the ksonnet apply command without adding extra flags to specify the true API endpoint.</p><p>At this point, your directory is ready to be committed to Github. Please follow your regular methods to push up your changes. For example, if you were using SSH based access, you could run within the <strong>KFAPP</strong> directory:</p><pre>cd ${KFAPP}<br>git init<br>git add --all<br>git commit -m &quot;Initial Kubeflow ksonnet files&quot;<br>git remote add origin git@github.com:argoproj/kubeflow-ks.git<br>git push --set-upstream origin master</pre><p><strong>NOTE:</strong> This will also store your GKE configs in the github repo.</p><p>You can take a look at <a href="https://github.com/argoproj/kubeflow-ks.git">https://github.com/argoproj/kubeflow-ks.git</a> to see an example of the ksonnet files that you should have in your repo with the caveat that the repo does not have the gcp_config folder.</p><h3><strong>Install Argo CD</strong></h3><p>With the Kubeflow ksonnet Github repo and GKE cluster ready, we now need to install Argo CD. For reference, most of the following steps follow the <a href="https://github.com/argoproj/argo-cd/blob/master/docs/getting_started.md">Argo CD getting started guide</a>.</p><h4>Install Argo CD to the Kubernetes Cluster</h4><pre>ARGO_CD_LATEST=$(curl --silent &quot;https://api.github.com/repos/argoproj/argo-cd/releases/latest&quot; | grep &#39;&quot;tag_name&quot;&#39; | sed -E &#39;s/.*&quot;([^&quot;]+)&quot;.*/\1/&#39;)</pre><pre>kubectl create namespace argocd</pre><pre>kubectl apply -n argocd -f https://raw.githubusercontent.com/argoproj/argo-cd/${ARGO_CD_LATEST}/manifests/install.yaml</pre><p>These three lines download the latest released Argo CD manifest files and install them into the argocd namespace</p><h4>Install Argo CD CLI</h4><p>If you use macOS, you can run:</p><pre>brew install argoproj/tap/argocd</pre><p>On Linux:</p><pre>curl -sSL -o /usr/local/bin/argocd https://github.com/argoproj/argo-cd/releases/download/${ARGO_CD_LATEST}/argocd-linux-amd64<br>chmod +x /usr/local/bin/argocd</pre><p><strong>Set Extra Argo CD Permissions</strong></p><p>Since the GKE cluster has RBAC enabled, you will need to grant your account the ability to create new cluster roles by running:</p><pre>kubectl create clusterrolebinding YOURNAME-cluster-admin-binding --clusterrole=cluster-admin --user=YOUREMAIL@gmail.com</pre><h4>Expose the Argo CD API server</h4><p>By default, the default manifest files for Argo CD does not expose a public endpoint. Here we will connect to Argo CD using the kubectl port-forwarding and expose Argo CD over port 8080</p><pre>kubectl port-forward service/argocd-server 8080:443</pre><p>You can read about other options to connect to your Argo CD instance <a href="https://github.com/argoproj/argo-cd/blob/master/docs/getting_started.md#3-access-the-argocd-api-server">here</a>.</p><h4>Login via the CLI as admin user</h4><p>The initial password for the admin user is autogenerated to be the pod name of the Argo CD API server. This can be retrieved with the command:</p><pre>kubectl get pods -n argocd -l app=argocd-server -o name | cut -d&#39;/&#39; -f 2</pre><p>Using the above password, login to Argo CD as admin over localhost:8080:</p><pre>argocd login localhost:8080</pre><p>After logging in, change the password using the command:</p><pre>argocd account update-password<br>argocd relogin</pre><h3><strong>Deploying Kubeflow</strong></h3><p>We’re finally ready to deploy Kubeflow using Argo CD!</p><pre>export KUBEFLOW_REPO_URL=&#39;Replace with a ssh or https git endpoint&#39;<br>argocd app create kubeflow --name kubeflow --repo $KUBEFLOW_REPO_URL --path ks_app --env default<br>argocd app sync kubeflow</pre><p>You can now view the Kubeflow application by running:</p><pre>argocd app get kubeflow</pre><p>or take a look at the application using the Argo CD GUI by pointing your browser to the IP address for the Argo CD service running on your cluster:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2IoCQS9lUFkpv1wuLgU6Cg.png" /></figure><p>NOTE: There is a <a href="https://github.com/kubeflow/kubeflow/issues/1145">known issue</a> with the IAP component that prevents the envoy service from becoming synced and causes all subsequent syncs to fail. As a workaround for this issue, we recommend that you sync individual resources by adding the resource flag to your sync command.</p><p>You can then access your Kubeflow UI by going to https://&lt;KFAPP&gt;.endpoint.&lt;PROJECT&gt;.cloud.googl/</p><ul><li>It can take 10–15 minutes for the endpoint to become available. Kubeflow needs to provision a signed SSL certificate and register a DNS name.</li></ul><h3><strong>Further Kubeflow Customizations</strong></h3><p>When you commit a change that modifies the Ksonnet application directory of your Kubeflow repository (the ks_app directory if you used the kfctl.sh script), Argo CD will detect that your application is out of sync with your git repo. To sync the new resource, you can run</p><pre>argocd app sync kubeflow --resource GROUP:KIND:NAME</pre><p>or from the UI:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/672/1*skniX7ckWDLxMLqrX3kdkQ.png" /></figure><h3><strong>Further Argo CD configurations</strong></h3><p>Check out the <a href="https://github.com/argoproj/argo-cd/tree/master/docs#argocd-documentation">Argo CD documentation</a> to read more about how to configure other features like auto-sync, SSO, RBAC, service tokens, and more.</p><h3>Next Steps</h3><p>While we can benefit from all the advantages GitOps offers by deploying Kubeflow with Argo CD, this integration still has a lot of room for improvement. Just as an example, the installation of Argo CD does not quite follow a GitOps model. The instructions tell users to install the Argo CD manifests from a remote URL instead of pulling them from a Git repo. To be more in line with a GitOps solution, users should have a more declarative installation of Argo CD (this issue is being tracked <a href="https://github.com/argoproj/argo-cd/issues/536">here</a>). The Argo CD team is excited to work with Kubeflow team to create a deeper integration that betters both tools.</p><p>Please check out the <a href="https://github.com/argoproj/argo-cd/">Argo CD code base</a>, and send any feedback or questions to the#argo-cd channel in the <a href="https://join.slack.com/t/argoproj/shared_invite/enQtMzExODU3MzIyNjYzLTA5MTFjNjI0Nzg3NzNiMDZiNmRiODM4Y2M1NWQxOGYzMzZkNTc1YWVkYTZkNzdlNmYyZjMxNWI3NjY2MDc1MzI">argoproj slack channel</a>!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=76f6b27807c" width="1" height="1" alt=""><hr><p><a href="https://blog.argoproj.io/using-gitops-to-deploy-kubeflow-with-argo-cd-76f6b27807c">Using GitOps to Deploy Kubeflow with Argo CD</a> was originally published in <a href="https://blog.argoproj.io/">Argo Project</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>