Merge aac0b98168 into 66ff3a82b3

Kubernetes/Kubernetes-Lite/k3s.sh

@@ -30,44 +30,74 @@ KVVERSION="v0.6.3"
 # K3S Version
 k3sVersion="v1.28.7+k3s1"
 
-# Set the IP addresses of the master and work nodes
+# Define node types and their properties
-master1=192.168.3.21
+declare -A nodes=(
-master2=192.168.3.22
+    ["master1"]="ip=192.168.2.132,user=laneone,interface=eth0,type=master"
-master3=192.168.3.23
+    ["master2"]="ip=192.168.2.133,user=laneone,interface=eth0,type=master"
-worker1=192.168.3.24
+    ["master3"]="ip=192.168.2.134,user=laneone,interface=eth0,type=master"
-worker2=192.168.3.25
+    ["worker1"]="ip=192.168.2.129,user=laneone,interface=eth0,type=worker,labels=longhorn=true,worker=true"
-worker3=192.168.3.26
+    ["worker2"]="ip=192.168.2.130,user=laneone,interface=eth0,type=worker,labels=longhorn=true,worker=true"
-
+    ["worker3"]="ip=192.168.2.131,user=laneone,interface=eth0,type=worker,labels=longhorn=true,worker=true"
-# User of remote machines
+    ["worker4"]="ip=192.168.2.125,user=laneone,interface=enp34s0,type=worker,labels=worker=true,auth=password,password=l"
-user=ubuntu
+    ["worker5"]="ip=192.168.2.104,user=laneone,interface=enp104s0,type=worker,labels=worker=true,auth=password,password=l"
-
+)
-# Interface used on remotes
-interface=eth0
 
 # Set the virtual IP address (VIP)
-vip=192.168.3.50
+vip=192.168.2.50
-
-# Array of master nodes
-masters=($master2 $master3)
-
-# Array of worker nodes
-workers=($worker1 $worker2 $worker3)
-
-# Array of all
-all=($master1 $master2 $master3 $worker1 $worker2 $worker3)
-
-# Array of all minus master
-allnomaster1=($master2 $master3 $worker1 $worker2 $worker3)
 
 #Loadbalancer IP range
-lbrange=192.168.3.60-192.168.3.80
+lbrange=192.168.2.60-192.168.2.100
 
 #ssh certificate name variable
 certName=id_rsa
 
+# Additional k3s flags for metrics
+common_extra_args="--kubelet-arg containerd=/run/k3s/containerd/containerd.sock"
+server_extra_args="--no-deploy servicelb --no-deploy traefik --kube-controller-manager-arg bind-address=0.0.0.0 --kube-proxy-arg metrics-bind-address=0.0.0.0 --kube-scheduler-arg bind-address=0.0.0.0 --etcd-expose-metrics true"
+agent_extra_args="--node-label worker=true"
+
+# Create Grafana admin credentials
+
+grafana_user="adminuser"  # desired grafana username
+grafana_password="adminpassword"  # Generates a random 12-character password
+
+
+#############################################
+#            HELPER FUNCTIONS               #
+#############################################
+
+get_node_ip() {
+    echo "${nodes[$1]}" | grep -oP 'ip=\K[^,]+'
+}
+
+get_node_user() {
+    echo "${nodes[$1]}" | grep -oP 'user=\K[^,]+'
+}
+
+get_node_interface() {
+    echo "${nodes[$1]}" | grep -oP 'interface=\K[^,]+'
+}
+
+get_node_type() {
+    echo "${nodes[$1]}" | grep -oP 'type=\K[^,]+'
+}
+
+get_node_labels() {
+    echo "${nodes[$1]}" | grep -oP 'labels=\K[^,]*' | tr ',' ' '
+}
+
+get_node_auth() {
+    echo "${nodes[$1]}" | grep -oP 'auth=\K[^,]*'
+}
+
+get_node_password() {
+    echo "${nodes[$1]}" | grep -oP 'password=\K[^,]*'
+}
+
 #############################################
 #            DO NOT EDIT BELOW              #
 #############################################
+
 # For testing purposes - in case time is wrong due to VM snapshots
 sudo timedatectl set-ntp off
 sudo timedatectl set-ntp on
@@ -115,19 +145,21 @@ EOF
 done
 
 # Step 1: Bootstrap First k3s Node
-mkdir ~/.kube
+mkdir -p ~/.kube
+first_master=$(echo "${!nodes[@]}" | tr ' ' '\n' | grep "master" | head -n1)
 k3sup install \
-  --ip $master1 \
+  --ip $(get_node_ip $first_master) \
-  --user $user \
+  --user $(get_node_user $first_master) \
   --tls-san $vip \
   --cluster \
   --k3s-version $k3sVersion \
-  --k3s-extra-args "--disable traefik --disable servicelb --flannel-iface=$interface --node-ip=$master1 --node-taint node-role.kubernetes.io/master=true:NoSchedule" \
+  --k3s-extra-args "--disable traefik --disable servicelb --flannel-iface=$(get_node_interface $first_master) --node-ip=$(get_node_ip $first_master) --node-taint node-role.kubernetes.io/master=true:NoSchedule $common_extra_args $server_extra_args" \
   --merge \
   --sudo \
   --local-path $HOME/.kube/config \
   --ssh-key $HOME/.ssh/$certName \
   --context k3s-ha
+
 echo -e " \033[32;5mFirst Node bootstrapped successfully!\033[0m"
 
 # Step 2: Install Kube-VIP for HA
@@ -140,39 +172,74 @@ cat kube-vip | sed 's/$interface/'$interface'/g; s/$vip/'$vip'/g' > $HOME/kube-v
 # Step 4: Copy kube-vip.yaml to master1
 scp -i ~/.ssh/$certName $HOME/kube-vip.yaml $user@$master1:~/kube-vip.yaml
 
-
 # Step 5: Connect to Master1 and move kube-vip.yaml
 ssh $user@$master1 -i ~/.ssh/$certName <<- EOF
   sudo mkdir -p /var/lib/rancher/k3s/server/manifests
   sudo mv kube-vip.yaml /var/lib/rancher/k3s/server/manifests/kube-vip.yaml
 EOF
 
+# Function to set up passwordless sudo
+setup_passwordless_sudo() {
+    local node=$1
+    local user=$(get_node_user $node)
+    local ip=$(get_node_ip $node)
+    local auth_method=$(get_node_auth $node)
+    local password=$(get_node_password $node)
+
+    echo "Setting up passwordless sudo for $user on $ip"
+    
+    if [ "$auth_method" == "password" ]; then
+        sshpass -p "$password" ssh -o StrictHostKeyChecking=no $user@$ip "echo '$password' | sudo -S sh -c 'echo \"$user ALL=(ALL) NOPASSWD:ALL\" > /etc/sudoers.d/$user && chmod 0440 /etc/sudoers.d/$user'"
+    else
+        ssh -i $HOME/.ssh/$certName -o StrictHostKeyChecking=no $user@$ip "sudo sh -c 'echo \"$user ALL=(ALL) NOPASSWD:ALL\" > /etc/sudoers.d/$user && chmod 0440 /etc/sudoers.d/$user'"
+    fi
+}
+
 # Step 6: Add new master nodes (servers) & workers
-for newnode in "${masters[@]}"; do
+for node in "${!nodes[@]}"; do
+    setup_passwordless_sudo $node
+
+    if [ "$(get_node_type $node)" == "master" ] && [ "$node" != "$first_master" ]; then
         k3sup join \
-    --ip $newnode \
+          --ip $(get_node_ip $node) \
-    --user $user \
+          --user $(get_node_user $node) \
           --sudo \
           --k3s-version $k3sVersion \
           --server \
-    --server-ip $master1 \
+          --server-ip $(get_node_ip $first_master) \
           --ssh-key $HOME/.ssh/$certName \
-    --k3s-extra-args "--disable traefik --disable servicelb --flannel-iface=$interface --node-ip=$newnode --node-taint node-role.kubernetes.io/master=true:NoSchedule" \
+          --k3s-extra-args "--disable traefik --disable servicelb --flannel-iface=$(get_node_interface $node) --node-ip=$(get_node_ip $node) --node-taint node-role.kubernetes.io/master=true:NoSchedule $common_extra_args $server_extra_args" \
-    --server-user $user
+          --server-user $(get_node_user $first_master)
-  echo -e " \033[32;5mMaster node joined successfully!\033[0m"
+        echo -e " \033[32;5mMaster node $node joined successfully!\033[0m"
-done
+    elif [ "$(get_node_type $node)" == "worker" ]; then
-
+        labels=$(get_node_labels $node)
-# add workers
+        label_args=""
-for newagent in "${workers[@]}"; do
+        if [ ! -z "$labels" ]; then
-  k3sup join \
+            label_args="--node-label \"$labels\""
-    --ip $newagent \
+        fi
-    --user $user \
+        auth_method=$(get_node_auth $node)
+        if [ "$auth_method" == "password" ]; then
+            password=$(get_node_password $node)
+            sshpass -p "$password" k3sup join \
+              --ip $(get_node_ip $node) \
+              --user $(get_node_user $node) \
               --sudo \
               --k3s-version $k3sVersion \
-    --server-ip $master1 \
+              --server-ip $(get_node_ip $first_master) \
+              --k3s-extra-args "$label_args $common_extra_args $agent_extra_args" \
+              --ssh-key $HOME/.ssh/$certName
+        else
+            k3sup join \
+              --ip $(get_node_ip $node) \
+              --user $(get_node_user $node) \
+              --sudo \
+              --k3s-version $k3sVersion \
+              --server-ip $(get_node_ip $first_master) \
               --ssh-key $HOME/.ssh/$certName \
-    --k3s-extra-args "--node-label \"longhorn=true\" --node-label \"worker=true\""
+              --k3s-extra-args "$label_args $common_extra_args $agent_extra_args"
-  echo -e " \033[32;5mAgent node joined successfully!\033[0m"
+        fi
+        echo -e " \033[32;5mWorker node $node joined successfully!\033[0m"
+    fi
 done
 
 # Step 7: Install kube-vip as network LoadBalancer - Install the kube-vip Cloud Provider
@@ -183,7 +250,7 @@ kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.12.1/manif
 kubectl apply -f https://raw.githubusercontent.com/metallb/metallb/v0.13.12/config/manifests/metallb-native.yaml
 # Download ipAddressPool and configure using lbrange above
 curl -sO https://raw.githubusercontent.com/JamesTurland/JimsGarage/main/Kubernetes/K3S-Deploy/ipAddressPool
-cat ipAddressPool | sed 's/$lbrange/'$lbrange'/g' > $HOME/ipAddressPool.yaml
+cat ipAddressPool | sed 's/$lbrange/'$lbrange'/g' > ipAddressPool.yaml
 
 # Step 9: Test with Nginx
 kubectl apply -f https://raw.githubusercontent.com/inlets/inlets-operator/master/contrib/nginx-sample-deployment.yaml -n default
@@ -249,8 +316,18 @@ echo -e " \033[32;5mBe patient as it downloads and configures a number of pods i
 echo -e " \033[32;5mInstalling Longhorn - It can take a while for all pods to deploy...\033[0m"
 kubectl apply -f https://raw.githubusercontent.com/JamesTurland/JimsGarage/main/Kubernetes/Longhorn/longhorn.yaml
 kubectl get pods \
---namespace longhorn-system \
+--namespace longhorn-system
---watch
+
+echo "Waiting for Longhorn UI deployment to be fully ready..."
+while ! (kubectl wait --for=condition=available deployment/longhorn-driver-deployer -n longhorn-system --timeout=600s && \
+         kubectl wait --for=condition=available deployment/longhorn-ui -n longhorn-system --timeout=600s && \
+         kubectl wait --for=condition=available deployment/csi-attacher -n longhorn-system --timeout=600s && \
+         kubectl wait --for=condition=available deployment/csi-provisioner -n longhorn-system --timeout=600s && \
+         kubectl wait --for=condition=available deployment/csi-resizer -n longhorn-system --timeout=600s && \
+         kubectl wait --for=condition=available deployment/csi-snapshotter -n longhorn-system --timeout=600s); do
+    echo "Waiting for Longhorn UI deployment to be fully ready..."
+    sleep 1
+done
 
 # Step 17: Print out confirmation
 
@@ -258,3 +335,124 @@ kubectl get nodes
 kubectl get svc -n longhorn-system
 
 echo -e " \033[32;5mHappy Kubing! Access Longhorn through Rancher UI\033[0m"
+
+# Step 18: Download and modify values.yaml for Prometheus
+
+# Ensure yq is installed
+if ! command -v yq &> /dev/null; then
+    echo "yq is not installed. Installing yq..."
+    wget https://github.com/mikefarah/yq/releases/latest/download/yq_linux_amd64 -O /usr/bin/yq && chmod +x /usr/bin/yq
+fi
+
+echo -e " \033[32;5mSetting up Prometheus...\033[0m"
+
+# Download values.yaml
+wget https://raw.githubusercontent.com/techno-tim/launchpad/master/kubernetes/kube-prometheus-stack/values.yml -O values.yaml
+
+# Get master node IPs
+master_ips=$(for node in "${!nodes[@]}"; do
+    if [ "$(get_node_type $node)" == "master" ]; then
+        echo "$(get_node_ip $node)"
+    fi
+done | sort -u)
+
+echo '------'
+echo 'Master IPs: ' 
+echo $master_ips
+echo '------'
+
+# Function to update endpoints in values.yaml
+update_endpoints() {
+    local component=$1
+    echo "Updating endpoints for $component"
+    
+    # Create the new endpoints content
+    local new_endpoints=""
+    for ip in $master_ips; do
+        new_endpoints+="    - $ip\n"
+    done
+    
+    # Use awk to replace the endpoints section
+    awk -v component="$component" -v new_endpoints="$new_endpoints" '
+    $0 ~ "^" component ":" { 
+        print $0
+        in_component = 1
+        next
+    }
+    in_component && /^[a-z]/ { 
+        in_component = 0 
+    }
+    in_component && /^ *endpoints:/ { 
+        print "  endpoints:"
+        print new_endpoints
+        skip = 1
+        next
+    }
+    skip && /^[^ ]/ { 
+        skip = 0 
+    }
+    !skip { print }
+    ' values.yaml > values.yaml.tmp && mv values.yaml.tmp values.yaml
+    
+    echo "Updated $component endpoints"
+}
+
+# Update endpoints for different components
+components=("kubeControllerManager" "kubeEtcd" "kubeScheduler" "kubeProxy")
+for component in "${components[@]}"; do
+    update_endpoints "$component"
+done
+
+# Create Grafana admin credentials
+echo -e " \033[32;5mCreating Grafana admin credentials...\033[0m"
+
+# Create Kubernetes secret for Grafana
+kubectl create namespace monitoring --dry-run=client -o yaml | kubectl apply -f -
+kubectl create secret generic grafana-admin-credentials \
+    --from-literal=admin-user=$grafana_user \
+    --from-literal=admin-password=$grafana_password \
+    -n monitoring \
+    --dry-run=client -o yaml | kubectl apply -f -
+
+echo -e " \033[32;5mGrafana admin credentials created. Username: $grafana_user, Password: $grafana_password\033[0m"
+echo -e " \033[32;5mPlease make note of these credentials and store them securely.\033[0m"
+
+# Update Grafana admin credentials in values.yaml
+yq eval '.grafana.admin.existingSecret = "grafana-admin-credentials"' -i values.yaml
+yq eval '.grafana.admin.userKey = "admin-user"' -i values.yaml
+yq eval '.grafana.admin.passwordKey = "admin-password"' -i values.yaml
+
+# Verify the changes
+for component in "${components[@]}"; do
+    echo "Endpoints for ${component}:"
+    yq eval ".${component}.endpoints" values.yaml
+done
+
+echo -e " \033[32;5mvalues.yaml has been updated with master node IPs\033[0m"
+
+# Step 19: Install Prometheus using Helm
+echo -e " \033[32;5mInstalling Prometheus...\033[0m"
+
+# Add prometheus-community helm repo
+helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
+helm repo update
+
+# Install Prometheus stack
+helm install prometheus prometheus-community/kube-prometheus-stack \
+  -f values.yaml \
+  --namespace monitoring \
+  --create-namespace
+
+# Wait for the Grafana deployment to be ready
+kubectl -n monitoring rollout status deploy/grafana
+
+echo "Changing Grafana service to LoadBalancer type..."
+    kubectl patch svc grafana -n monitoring -p '{"spec": {"type": "LoadBalancer"}}'
+
+echo -e " \033[32;5mPrometheus has been installed!\033[0m"
+
+
+# Show external ip on which to access grafana
+kubectl get svc/grafana -n monitoring
+
+echo -e " \033[32;5m Happy Charting! \033[0m"