Further instructions for deploying the TeraFlow OS and an automation script.

.gitignore

@@ -146,3 +146,5 @@ dmypy.json
 # Cython debug symbols
 cython_debug/
 
+# Other
+/tmp

INSTALL.md

+# TeraFlow OS SDN Controller Installation Instructions 
+Assuming you have a running Kubernetes deployment installed following the instructions provided in [Wiki: Installing Kubernetes on your Linux machine](../../wikis/Installing-Kubernetes-on-your-Linux-machine), the following instructions will let you deploy TeraFlow OS SDN Controller in your local Kubernetes environment.
+
+Among others, it is assumed that you can run the command <code> kubectl cluster-info</code> in the machine you use for running the deployment script below.
+
+
+## 1. Clone the repository
+Create a folder (e.g., ~/deploy-test/teraflow/controller) to place the source code, move to that folder, clone the repository, and checkout the target branch you want to deploy (e.g., master or develop):
+
+<code>mkdir -p ~/deploy-test/teraflow/controller; cd ~/deploy-test/teraflow/controller</code>
+
+<code>git clone https://gitlab.com/teraflow-h2020/controller.git .</code>
+
+<code>git checkout develop</code>
+
+
+## 2. Adapt deployment script
+Find the script <code>deploy_to_kubernetes.sh</code> and modify the configuration settings at the top of the script according to your local Kubernetes environment.
+
+
+## 3. Run the deployment script
+Execute the script <code>deploy_to_kubernetes.sh</code> to build and deploy the components at your local Kubernetes environment.

README.md

-# Teraflow SDN controller
+# TeraFlow OS SDN Controller
 
 [Teraflow H2020 project](https://teraflow-h2020.eu/) - Secured autonomic traffic management for a Tera of SDN Flows
 
 Branch "master" : [![pipeline status](https://gitlab.com/teraflow-h2020/controller/badges/master/pipeline.svg)](https://gitlab.com/teraflow-h2020/controller/-/commits/master) [![coverage report](https://gitlab.com/teraflow-h2020/controller/badges/master/coverage.svg)](https://gitlab.com/teraflow-h2020/controller/-/commits/master)
 
 Branch "develop" : [![pipeline status](https://gitlab.com/teraflow-h2020/controller/badges/develop/pipeline.svg)](https://gitlab.com/teraflow-h2020/controller/-/commits/develop) [![coverage report](https://gitlab.com/teraflow-h2020/controller/badges/develop/coverage.svg)](https://gitlab.com/teraflow-h2020/controller/-/commits/develop)
+
+# Installation Instructions
+To install TeraFlow OS SDN Controller in your local Kubernetes deployment, we assume you deployed Kubernetes following the instructions provided in [Wiki: Installing Kubernetes on your Linux machine](../../wikis/Installing-Kubernetes-on-your-Linux-machine). Then, follow instructions in [INSTALL.md](./INSTALL.md).

deploy_in_kubernetes.sh

+#!/bin/bash
+
+########################################################################################################################
+# Define your deployment settings here
+########################################################################################################################
+
+# Set the URL of your local Docker registry where the images will be uploaded to. Leave it blank if you do not want to
+# use any Docker registry.
+REGISTRY_IMAGE=""
+#REGISTRY_IMAGE="http://my-container-registry.local/"
+
+# Set the list of components you want to build images for, and deploy.
+COMPONENTS="context device service compute monitoring centralizedattackdetector"
+
+# Set the tag you want to use for your images.
+IMAGE_TAG="tf-dev"
+
+# Set the name of the Kubernetes namespace to deploy to.
+K8S_NAMESPACE="tf-dev"
+
+
+########################################################################################################################
+# Automated steps start here
+########################################################################################################################
+
+# Constants
+CLONE_URL="https://gitlab.com/teraflow-h2020/controller.git"
+GITLAB_REPO_URL="registry.gitlab.com/teraflow-h2020/controller"
+TMP_FOLDER="./tmp"
+
+# Create a tmp folder for files modified during the deployment
+TMP_MANIFESTS_FOLDER="$TMP_FOLDER/manifests"
+mkdir -p $TMP_MANIFESTS_FOLDER
+TMP_LOGS_FOLDER="$TMP_FOLDER/logs"
+mkdir -p $TMP_LOGS_FOLDER
+
+
+# Re-create the namespace to prevent being affected by garbage on it
+kubectl delete namespace $K8S_NAMESPACE
+kubectl create namespace $K8S_NAMESPACE
+
+for COMPONENT in $COMPONENTS; do
+    echo "Processing '$COMPONENT' component..."
+    IMAGE_NAME="$COMPONENT:$IMAGE_TAG"
+    IMAGE_URL="$REGISTRY_IMAGE/$IMAGE_NAME"
+
+    echo "  Building Docker image..."
+    BUILD_LOG="$TMP_LOGS_FOLDER/build_${COMPONENT}.log"
+    docker build -t "$IMAGE_NAME" -f ./src/$COMPONENT/Dockerfile ./src/ > $BUILD_LOG
+
+    if [ -n "$REGISTRY_IMAGE" ]; then
+        echo "Pushing Docker image to '$REGISTRY_IMAGE'..."
+
+        TAG_LOG="$TMP_LOGS_FOLDER/tag_${COMPONENT}.log"
+        docker tag "$IMAGE_NAME" "$IMAGE_URL" > $TAG_LOG
+
+        PUSH_LOG="$TMP_LOGS_FOLDER/push_${COMPONENT}.log"
+        docker push "$IMAGE_URL" > "$PUSH_LOG"
+    fi
+
+    echo "  Adapting manifest file..."
+    MANIFEST="$TMP_MANIFESTS_FOLDER/${COMPONENT}service.yaml"
+    cp ./manifests/${COMPONENT}service.yaml $MANIFEST
+    if [ -n "$REGISTRY_IMAGE" ]; then
+        sed -E -i "s#image: $GITLAB_REPO_URL/$COMPONENT:latest#image: $IMAGE_URL#g" $MANIFEST
+        sed -E -i "s#imagePullPolicy: .*#imagePullPolicy: Always#g" $MANIFEST
+    else
+        sed -E -i "s#image: $GITLAB_REPO_URL/$COMPONENT:latest#image: $IMAGE_NAME#g" $MANIFEST
+        sed -E -i "s#imagePullPolicy: .*#imagePullPolicy: Never#g" $MANIFEST
+    fi
+
+    echo "  Deploying to Kubernetes..."
+    DEPLOY_LOG="$TMP_LOGS_FOLDER/push_${COMPONENT}.log"
+    kubectl --namespace $K8S_NAMESPACE apply -f $MANIFEST > $DEPLOY_LOG
+done
+
+kubectl --namespace $K8S_NAMESPACE get all
+
+echo "Done!"