Add snapshot restore e2e test (#6396)

tests/e2e/snapshotrestore/Vagrantfile

+ENV['VAGRANT_NO_PARALLEL'] = 'no'
+NODE_ROLES = (ENV['E2E_NODE_ROLES'] ||
+  ["server-0", "server-1", "server-2", "agent-0", "agent-1"])
+NODE_BOXES = (ENV['E2E_NODE_BOXES'] ||
+  ['generic/ubuntu2004', 'generic/ubuntu2004', 'generic/ubuntu2004', 'generic/ubuntu2004', 'generic/ubuntu2004'])
+GITHUB_BRANCH = (ENV['E2E_GITHUB_BRANCH'] || "master")
+RELEASE_VERSION = (ENV['E2E_RELEASE_VERSION'] || "")
+EXTERNAL_DB = (ENV['E2E_EXTERNAL_DB'] || "etcd")
+NODE_CPUS = (ENV['E2E_NODE_CPUS'] || 2).to_i
+NODE_MEMORY = (ENV['E2E_NODE_MEMORY'] || 1024).to_i
+# Virtualbox >= 6.1.28 require `/etc/vbox/network.conf` for expanded private networks 
+NETWORK_PREFIX = "10.10.10"
+install_type = ""
+
+def provision(vm, role, role_num, node_num)
+  vm.box = NODE_BOXES[node_num]
+  vm.hostname = role
+  # An expanded netmask is required to allow VM<-->VM communication, virtualbox defaults to /32
+  node_ip = "#{NETWORK_PREFIX}.#{100+node_num}"
+  vm.network "private_network", ip: node_ip, netmask: "255.255.255.0"
+
+  scripts_location = Dir.exists?("./scripts") ? "./scripts" : "../scripts"
+  vagrant_defaults = File.exists?("./vagrantdefaults.rb") ? "./vagrantdefaults.rb" : "../vagrantdefaults.rb"
+  load vagrant_defaults
+
+  defaultOSConfigure(vm)
+  install_type = getInstallType(vm, RELEASE_VERSION, GITHUB_BRANCH)
+  
+  vm.provision "shell", inline: "ping -c 2 k3s.io"
+  
+  db_type = getDBType(role, role_num, vm)
+
+  if role.include?("server") && role_num == 0
+    vm.provision 'k3s-primary-server', type: 'k3s', run: 'once' do |k3s|
+      k3s.args = "server "
+      k3s.config = <<~YAML
+        token: vagrant
+        node-external-ip: #{NETWORK_PREFIX}.100
+        flannel-iface: eth1
+        tls-san: #{NETWORK_PREFIX}.100.nip.io
+        #{db_type}
+      YAML
+      k3s.env = %W[K3S_KUBECONFIG_MODE=0644 #{install_type}]
+      k3s.config_mode = '0644' # side-step https://github.com/k3s-io/k3s/issues/4321
+    end
+  
+  elsif role.include?("server") && role_num != 0
+    vm.provision 'k3s-secondary-server', type: 'k3s', run: 'once' do |k3s|
+      k3s.args = "server"
+      k3s.config = <<~YAML
+        server: "https://#{NETWORK_PREFIX}.100:6443"
+        token: vagrant
+        node-external-ip: #{node_ip}
+        flannel-iface: eth1
+        #{db_type}
+      YAML
+      k3s.env = %W[K3S_KUBECONFIG_MODE=0644 K3S_TOKEN=vagrant #{install_type}]
+      k3s.config_mode = '0644' # side-step https://github.com/k3s-io/k3s/issues/4321
+    end
+  end
+  
+  if role.include?("agent")
+    vm.provision 'k3s-agent', type: 'k3s', run: 'once' do |k3s|
+      k3s.args = "agent"
+      k3s.config = <<~YAML
+        server: "https://#{NETWORK_PREFIX}.100:6443"
+        token: vagrant
+        node-external-ip: #{node_ip}
+        flannel-iface: eth1
+        #{db_type}
+      YAML
+      k3s.env = %W[K3S_KUBECONFIG_MODE=0644 #{install_type}]
+      k3s.config_mode = '0644' # side-step https://github.com/k3s-io/k3s/issues/4321
+    end
+  end
+  if vm.box.to_s.include?("microos")
+    vm.provision 'k3s-reload', type: 'reload', run: 'once'
+    if !EXTERNAL_DB.empty?
+      vm.provision "shell", inline: "docker start #{EXTERNAL_DB}"
+    end
+  end
+  # This step does not run by default and is designed to be called by higher level tools
+end
+
+def getDBType(role, role_num, vm)
+  if ( EXTERNAL_DB == "" || EXTERNAL_DB == "etcd" )
+    if role.include?("server") && role_num == 0
+      return "cluster-init: true"
+    end
+  else
+    puts "Unknown EXTERNAL_DB: " + EXTERNAL_DB
+    abort
+  end
+  return ""
+end
+
+Vagrant.configure("2") do |config|
+  config.vagrant.plugins = ["vagrant-k3s", "vagrant-reload"]
+  # Default provider is libvirt, virtualbox is only provided as a backup
+  config.vm.provider "libvirt" do |v|
+    v.cpus = NODE_CPUS
+    v.memory = NODE_MEMORY
+  end
+  config.vm.provider "virtualbox" do |v|
+    v.cpus = NODE_CPUS
+    v.memory = NODE_MEMORY
+  end
+  
+  if NODE_ROLES.kind_of?(String)
+    NODE_ROLES = NODE_ROLES.split(" ", -1)
+  end
+  if NODE_BOXES.kind_of?(String)
+    NODE_BOXES = NODE_BOXES.split(" ", -1)
+  end
+
+  # Must iterate on the index, vagrant does not understand iterating 
+  # over the node roles themselves
+  NODE_ROLES.length.times do |i|
+    name = NODE_ROLES[i]
+    role_num = name.split("-", -1).pop.to_i
+    config.vm.define name do |node|
+      provision(node.vm, name, role_num, i)
+    end
+  end
+end

tests/e2e/snapshotrestore/snapshotrestore_test.go

+package snapshotrestore
+
+import (
+	"flag"
+	"fmt"
+	"os"
+	"strings"
+	"testing"
+
+	"github.com/k3s-io/k3s/tests/e2e"
+	. "github.com/onsi/ginkgo/v2"
+	. "github.com/onsi/gomega"
+)
+
+// Valid nodeOS:
+// generic/ubuntu2004, generic/centos7, generic/rocky8,
+// opensuse/Leap-15.3.x86_64, dweomer/microos.amd64
+
+var nodeOS = flag.String("nodeOS", "generic/ubuntu2004", "VM operating system")
+var serverCount = flag.Int("serverCount", 3, "number of server nodes")
+var agentCount = flag.Int("agentCount", 1, "number of agent nodes")
+var hardened = flag.Bool("hardened", false, "true or false")
+var ci = flag.Bool("ci", false, "running on CI")
+var local = flag.Bool("local", false, "deploy a locally built K3s binary")
+
+// Environment Variables Info:
+// E2E_EXTERNAL_DB: mysql, postgres, etcd (default: etcd)
+// E2E_RELEASE_VERSION=v1.23.1+k3s2 (default: latest commit from master)
+
+func Test_E2ESnapshotRestore(t *testing.T) {
+	RegisterFailHandler(Fail)
+	flag.Parse()
+	RunSpecs(t, "SnapshotRestore Test Suite")
+}
+
+var (
+	kubeConfigFile  string
+	serverNodeNames []string
+	agentNodeNames  []string
+	snapshotname    string
+)
+
+var _ = Describe("Verify Create", Ordered, func() {
+	Context("Cluster :", func() {
+		It("Starts up with no issues", func() {
+			var err error
+			if *local {
+				serverNodeNames, agentNodeNames, err = e2e.CreateLocalCluster(*nodeOS, *serverCount, *agentCount)
+			} else {
+				serverNodeNames, agentNodeNames, err = e2e.CreateCluster(*nodeOS, *serverCount, *agentCount)
+			}
+			Expect(err).NotTo(HaveOccurred(), e2e.GetVagrantLog(err))
+			fmt.Println("CLUSTER CONFIG")
+			fmt.Println("OS:", *nodeOS)
+			fmt.Println("Server Nodes:", serverNodeNames)
+			fmt.Println("Agent Nodes:", agentNodeNames)
+			kubeConfigFile, err = e2e.GenKubeConfigFile(serverNodeNames[0])
+			Expect(err).NotTo(HaveOccurred())
+		})
+
+		It("Checks Node and Pod Status", func() {
+			fmt.Printf("\nFetching node status\n")
+			Eventually(func(g Gomega) {
+				nodes, err := e2e.ParseNodes(kubeConfigFile, false)
+				g.Expect(err).NotTo(HaveOccurred())
+				for _, node := range nodes {
+					g.Expect(node.Status).Should(Equal("Ready"))
+				}
+			}, "420s", "5s").Should(Succeed())
+			_, _ = e2e.ParseNodes(kubeConfigFile, true)
+
+			fmt.Printf("\nFetching Pods status\n")
+			Eventually(func(g Gomega) {
+				pods, err := e2e.ParsePods(kubeConfigFile, false)
+				g.Expect(err).NotTo(HaveOccurred())
+				for _, pod := range pods {
+					if strings.Contains(pod.Name, "helm-install") {
+						g.Expect(pod.Status).Should(Equal("Completed"), pod.Name)
+					} else {
+						g.Expect(pod.Status).Should(Equal("Running"), pod.Name)
+					}
+				}
+			}, "620s", "5s").Should(Succeed())
+			_, _ = e2e.ParsePods(kubeConfigFile, true)
+		})
+
+		It("Verifies test workload before snapshot is created", func() {
+			res, err := e2e.DeployWorkload("clusterip.yaml", kubeConfigFile, *hardened)
+			Expect(err).NotTo(HaveOccurred(), "Cluster IP manifest not deployed: "+res)
+
+			Eventually(func(g Gomega) {
+				cmd := "kubectl get pods -o=name -l k8s-app=nginx-app-clusterip --field-selector=status.phase=Running --kubeconfig=" + kubeConfigFile
+				res, err := e2e.RunCommand(cmd)
+				g.Expect(err).NotTo(HaveOccurred())
+				g.Expect(res).Should((ContainSubstring("test-clusterip")), "failed cmd: "+cmd+" result: "+res)
+			}, "240s", "5s").Should(Succeed())
+		})
+
+		It("Verifies Snapshot is created", func() {
+			Eventually(func(g Gomega) {
+				cmd := "sudo k3s etcd-snapshot"
+				_, err := e2e.RunCmdOnNode(cmd, "server-0")
+				g.Expect(err).NotTo(HaveOccurred())
+				cmd = "sudo ls /var/lib/rancher/k3s/server/db/snapshots/"
+				snapshotname, err = e2e.RunCmdOnNode(cmd, "server-0")
+				fmt.Println("Snapshot Name", snapshotname)
+				g.Expect(snapshotname).Should(ContainSubstring("on-demand-server-0"))
+			}, "420s", "10s").Should(Succeed())
+		})
+
+		It("Verifies another test workload after snapshot is created", func() {
+			_, err := e2e.DeployWorkload("nodeport.yaml", kubeConfigFile, *hardened)
+			Expect(err).NotTo(HaveOccurred(), "NodePort manifest not deployed")
+			Eventually(func(g Gomega) {
+				cmd := "kubectl get pods -o=name -l k8s-app=nginx-app-nodeport --field-selector=status.phase=Running --kubeconfig=" + kubeConfigFile
+				res, err := e2e.RunCommand(cmd)
+				g.Expect(err).NotTo(HaveOccurred())
+				g.Expect(res).Should(ContainSubstring("test-nodeport"), "nodeport pod was not created")
+			}, "240s", "5s").Should(Succeed())
+		})
+
+		It("Verifies snapshot is restored successfully and validates only test workload1 is present", func() {
+			//Stop k3s on all nodes
+			for _, nodeName := range serverNodeNames {
+				cmd := "sudo systemctl stop k3s"
+				_, err := e2e.RunCmdOnNode(cmd, nodeName)
+				Expect(err).NotTo(HaveOccurred())
+			}
+			//Restores from snapshot on server-0
+			for _, nodeName := range serverNodeNames {
+				if nodeName == "server-0" {
+					cmd := "sudo k3s server --cluster-init --cluster-reset --cluster-reset-restore-path=/var/lib/rancher/k3s/server/db/snapshots/" + snapshotname
+					res, err := e2e.RunCmdOnNode(cmd, nodeName)
+					Expect(err).NotTo(HaveOccurred())
+					Expect(res).Should(ContainSubstring("Managed etcd cluster membership has been reset, restart without --cluster-reset flag now"))
+
+					cmd = "sudo systemctl start k3s"
+					_, err = e2e.RunCmdOnNode(cmd, nodeName)
+					Expect(err).NotTo(HaveOccurred())
+				}
+			}
+			//Verifies node is up and pods running
+			fmt.Printf("\nFetching node status\n")
+			Eventually(func(g Gomega) {
+				nodes, err := e2e.ParseNodes(kubeConfigFile, false)
+				g.Expect(err).NotTo(HaveOccurred())
+				for _, node := range nodes {
+					g.Expect(node.Status).Should(Equal("Ready"))
+				}
+			}, "420s", "5s").Should(Succeed())
+			_, _ = e2e.ParseNodes(kubeConfigFile, true)
+
+			fmt.Printf("\nFetching Pods status\n")
+			Eventually(func(g Gomega) {
+				pods, err := e2e.ParsePods(kubeConfigFile, false)
+				g.Expect(err).NotTo(HaveOccurred())
+				for _, pod := range pods {
+					if strings.Contains(pod.Name, "helm-install") {
+						g.Expect(pod.Status).Should(Equal("Completed"), pod.Name)
+					} else {
+						g.Expect(pod.Status).Should(Equal("Running"), pod.Name)
+					}
+				}
+			}, "620s", "5s").Should(Succeed())
+			_, _ = e2e.ParsePods(kubeConfigFile, true)
+			//Verifies test workload1 is present
+			//Verifies test workload2 is not present
+			cmd := "kubectl get pods --kubeconfig=" + kubeConfigFile
+			res, err := e2e.RunCommand(cmd)
+			Expect(err).NotTo(HaveOccurred())
+			Expect(res).Should(ContainSubstring("test-clusterip"))
+			Expect(res).ShouldNot(ContainSubstring("test-nodeport"))
+		})
+	})
+})
+
+var failed bool
+var _ = AfterEach(func() {
+	failed = failed || CurrentSpecReport().Failed()
+})
+
+var _ = AfterSuite(func() {
+	if failed && !*ci {
+		fmt.Println("FAILED!")
+	} else {
+		Expect(e2e.DestroyCluster()).To(Succeed())
+		Expect(os.Remove(kubeConfigFile)).To(Succeed())
+	}
+})