Added file and container image caching (#4828)
* File and container image downloads are now cached localy, so that repeated vagrant up/down runs do not trigger downloading of those files. This is especially useful on laptops with kubernetes runnig locally on vm's. The total size of the cache, after an ansible run, is currently around 800MB, so bandwidth (=time) savings can be quite significant.
* When download_run_once is false, the default is still not to cache, but setting download_force_cache will still enable caching.
* The local cache location can be set with download_cache_dir and defaults to /tmp/kubernetes_cache
* A local docker instance is no longer required to cache docker images; Images are cached to file. A local docker instance is still required, though, if you wish to download images on localhost.
* Fixed a FIXME, wher the argument was that delegate_to doesn't play nice with omit. That is a correct observation and the fix is to use default(inventory_host) instead of default(omit). See ansible/ansible#26009
* Removed "Register docker images info" task from download_container and set_docker_image_facts because it was faulty and unused.
* Removed redundant when:download.{container,enabled,run_once} conditions from {sync,download}_container.yml
* All features of commit d6fd0d2acaec9f53e75d82db30411f96a5bf2cc9 by Timoses <timosesu@gmail.com>, merged May 1st 2019, are included in this patch. Not all code was included verbatim, but each feature of that commit was checked to be working in this patch. One notable change: The actual downloading of the kubeadm images was moved to {download,sync)_container, to enable caching.
Note 1: I considered splitting this patch, but most changes that are not directly related to caching, are a pleasant by-product of implementing the caching code, so splitting would be impractical.
Note 2: I have my doubts about the usefulness of the upload, download and upgrade tags in the download role. Must they remain or can they be removed? If anybody knows, then please speak up. 5 years ago Added file and container image caching (#4828)
* File and container image downloads are now cached localy, so that repeated vagrant up/down runs do not trigger downloading of those files. This is especially useful on laptops with kubernetes runnig locally on vm's. The total size of the cache, after an ansible run, is currently around 800MB, so bandwidth (=time) savings can be quite significant.
* When download_run_once is false, the default is still not to cache, but setting download_force_cache will still enable caching.
* The local cache location can be set with download_cache_dir and defaults to /tmp/kubernetes_cache
* A local docker instance is no longer required to cache docker images; Images are cached to file. A local docker instance is still required, though, if you wish to download images on localhost.
* Fixed a FIXME, wher the argument was that delegate_to doesn't play nice with omit. That is a correct observation and the fix is to use default(inventory_host) instead of default(omit). See ansible/ansible#26009
* Removed "Register docker images info" task from download_container and set_docker_image_facts because it was faulty and unused.
* Removed redundant when:download.{container,enabled,run_once} conditions from {sync,download}_container.yml
* All features of commit d6fd0d2acaec9f53e75d82db30411f96a5bf2cc9 by Timoses <timosesu@gmail.com>, merged May 1st 2019, are included in this patch. Not all code was included verbatim, but each feature of that commit was checked to be working in this patch. One notable change: The actual downloading of the kubeadm images was moved to {download,sync)_container, to enable caching.
Note 1: I considered splitting this patch, but most changes that are not directly related to caching, are a pleasant by-product of implementing the caching code, so splitting would be impractical.
Note 2: I have my doubts about the usefulness of the upload, download and upgrade tags in the download role. Must they remain or can they be removed? If anybody knows, then please speak up. 5 years ago Added file and container image caching (#4828)
* File and container image downloads are now cached localy, so that repeated vagrant up/down runs do not trigger downloading of those files. This is especially useful on laptops with kubernetes runnig locally on vm's. The total size of the cache, after an ansible run, is currently around 800MB, so bandwidth (=time) savings can be quite significant.
* When download_run_once is false, the default is still not to cache, but setting download_force_cache will still enable caching.
* The local cache location can be set with download_cache_dir and defaults to /tmp/kubernetes_cache
* A local docker instance is no longer required to cache docker images; Images are cached to file. A local docker instance is still required, though, if you wish to download images on localhost.
* Fixed a FIXME, wher the argument was that delegate_to doesn't play nice with omit. That is a correct observation and the fix is to use default(inventory_host) instead of default(omit). See ansible/ansible#26009
* Removed "Register docker images info" task from download_container and set_docker_image_facts because it was faulty and unused.
* Removed redundant when:download.{container,enabled,run_once} conditions from {sync,download}_container.yml
* All features of commit d6fd0d2acaec9f53e75d82db30411f96a5bf2cc9 by Timoses <timosesu@gmail.com>, merged May 1st 2019, are included in this patch. Not all code was included verbatim, but each feature of that commit was checked to be working in this patch. One notable change: The actual downloading of the kubeadm images was moved to {download,sync)_container, to enable caching.
Note 1: I considered splitting this patch, but most changes that are not directly related to caching, are a pleasant by-product of implementing the caching code, so splitting would be impractical.
Note 2: I have my doubts about the usefulness of the upload, download and upgrade tags in the download role. Must they remain or can they be removed? If anybody knows, then please speak up. 5 years ago |
|
# -*- mode: ruby -*- # # vi: set ft=ruby :
# For help on using kubespray with vagrant, check out docs/vagrant.md
require 'fileutils'
Vagrant.require_version ">= 2.0.0"
CONFIG = File.join(File.dirname(__FILE__), ENV['KUBESPRAY_VAGRANT_CONFIG'] || 'vagrant/config.rb')
FLATCAR_URL_TEMPLATE = "https://%s.release.flatcar-linux.net/amd64-usr/current/flatcar_production_vagrant.json"
# Uniq disk UUID for libvirt DISK_UUID = Time.now.utc.to_i
SUPPORTED_OS = { "flatcar-stable" => {box: "flatcar-stable", user: "core", box_url: FLATCAR_URL_TEMPLATE % ["stable"]}, "flatcar-beta" => {box: "flatcar-beta", user: "core", box_url: FLATCAR_URL_TEMPLATE % ["beta"]}, "flatcar-alpha" => {box: "flatcar-alpha", user: "core", box_url: FLATCAR_URL_TEMPLATE % ["alpha"]}, "flatcar-edge" => {box: "flatcar-edge", user: "core", box_url: FLATCAR_URL_TEMPLATE % ["edge"]}, "ubuntu1604" => {box: "generic/ubuntu1604", user: "vagrant"}, "ubuntu1804" => {box: "generic/ubuntu1804", user: "vagrant"}, "ubuntu2004" => {box: "generic/ubuntu2004", user: "vagrant"}, "centos" => {box: "centos/7", user: "vagrant"}, "centos-bento" => {box: "bento/centos-7.6", user: "vagrant"}, "centos8" => {box: "centos/8", user: "vagrant"}, "centos8-bento" => {box: "bento/centos-8", user: "vagrant"}, "fedora33" => {box: "fedora/33-cloud-base", user: "vagrant"}, "fedora34" => {box: "fedora/34-cloud-base", user: "vagrant"}, "opensuse" => {box: "bento/opensuse-leap-15.2", user: "vagrant"}, "opensuse-tumbleweed" => {box: "opensuse/Tumbleweed.x86_64", user: "vagrant"}, "oraclelinux" => {box: "generic/oracle7", user: "vagrant"}, "oraclelinux8" => {box: "generic/oracle8", user: "vagrant"}, "rhel7" => {box: "generic/rhel7", user: "vagrant"}, "rhel8" => {box: "generic/rhel8", user: "vagrant"}, }
if File.exist?(CONFIG) require CONFIG end
# Defaults for config options defined in CONFIG $num_instances ||= 3 $instance_name_prefix ||= "k8s" $vm_gui ||= false $vm_memory ||= 2048 $vm_cpus ||= 2 $shared_folders ||= {} $forwarded_ports ||= {} $subnet ||= "172.18.8" $subnet_ipv6 ||= "fd3c:b398:0698:0756" $os ||= "ubuntu1804" $network_plugin ||= "flannel" # Setting multi_networking to true will install Multus: https://github.com/intel/multus-cni $multi_networking ||= false $download_run_once ||= "True" $download_force_cache ||= "True" # The first three nodes are etcd servers $etcd_instances ||= $num_instances # The first two nodes are kube masters $kube_master_instances ||= $num_instances == 1 ? $num_instances : ($num_instances - 1) # All nodes are kube nodes $kube_node_instances ||= $num_instances # The following only works when using the libvirt provider $kube_node_instances_with_disks ||= false $kube_node_instances_with_disks_size ||= "20G" $kube_node_instances_with_disks_number ||= 2 $override_disk_size ||= false $disk_size ||= "20GB" $local_path_provisioner_enabled ||= false $local_path_provisioner_claim_root ||= "/opt/local-path-provisioner/" $libvirt_nested ||= false
$playbook ||= "cluster.yml"
host_vars = {}
$box = SUPPORTED_OS[$os][:box] # if $inventory is not set, try to use example $inventory = "inventory/sample" if ! $inventory $inventory = File.absolute_path($inventory, File.dirname(__FILE__))
# if $inventory has a hosts.ini file use it, otherwise copy over # vars etc to where vagrant expects dynamic inventory to be if ! File.exist?(File.join(File.dirname($inventory), "hosts.ini")) $vagrant_ansible = File.join(File.dirname(__FILE__), ".vagrant", "provisioners", "ansible") FileUtils.mkdir_p($vagrant_ansible) if ! File.exist?($vagrant_ansible) $vagrant_inventory = File.join($vagrant_ansible,"inventory") FileUtils.rm_f($vagrant_inventory) FileUtils.ln_s($inventory, $vagrant_inventory) end
if Vagrant.has_plugin?("vagrant-proxyconf") $no_proxy = ENV['NO_PROXY'] || ENV['no_proxy'] || "127.0.0.1,localhost" (1..$num_instances).each do |i| $no_proxy += ",#{$subnet}.#{i+100}" end end
Vagrant.configure("2") do |config|
config.vm.box = $box if SUPPORTED_OS[$os].has_key? :box_url config.vm.box_url = SUPPORTED_OS[$os][:box_url] end config.ssh.username = SUPPORTED_OS[$os][:user]
# plugin conflict if Vagrant.has_plugin?("vagrant-vbguest") then config.vbguest.auto_update = false end
# always use Vagrants insecure key config.ssh.insert_key = false
if ($override_disk_size) unless Vagrant.has_plugin?("vagrant-disksize") system "vagrant plugin install vagrant-disksize" end config.disksize.size = $disk_size end
(1..$num_instances).each do |i| config.vm.define vm_name = "%s-%01d" % [$instance_name_prefix, i] do |node|
node.vm.hostname = vm_name
if Vagrant.has_plugin?("vagrant-proxyconf") node.proxy.http = ENV['HTTP_PROXY'] || ENV['http_proxy'] || "" node.proxy.https = ENV['HTTPS_PROXY'] || ENV['https_proxy'] || "" node.proxy.no_proxy = $no_proxy end
["vmware_fusion", "vmware_workstation"].each do |vmware| node.vm.provider vmware do |v| v.vmx['memsize'] = $vm_memory v.vmx['numvcpus'] = $vm_cpus end end
node.vm.provider :virtualbox do |vb| vb.memory = $vm_memory vb.cpus = $vm_cpus vb.gui = $vm_gui vb.linked_clone = true vb.customize ["modifyvm", :id, "--vram", "8"] # ubuntu defaults to 256 MB which is a waste of precious RAM vb.customize ["modifyvm", :id, "--audio", "none"] end
node.vm.provider :libvirt do |lv| lv.nested = $libvirt_nested lv.cpu_mode = "host-model" lv.memory = $vm_memory lv.cpus = $vm_cpus lv.default_prefix = 'kubespray' # Fix kernel panic on fedora 28 if $os == "fedora" lv.cpu_mode = "host-passthrough" end end
if $kube_node_instances_with_disks # Libvirt driverletters = ('a'..'z').to_a node.vm.provider :libvirt do |lv| # always make /dev/sd{a/b/c} so that CI can ensure that # virtualbox and libvirt will have the same devices to use for OSDs (1..$kube_node_instances_with_disks_number).each do |d| lv.storage :file, :device => "hd#{driverletters[d]}", :path => "disk-#{i}-#{d}-#{DISK_UUID}.disk", :size => $kube_node_instances_with_disks_size, :bus => "ide" end end end
if $expose_docker_tcp node.vm.network "forwarded_port", guest: 2375, host: ($expose_docker_tcp + i - 1), auto_correct: true end
$forwarded_ports.each do |guest, host| node.vm.network "forwarded_port", guest: guest, host: host, auto_correct: true end
if ["rhel7","rhel8"].include? $os # Vagrant synced_folder rsync options cannot be used for RHEL boxes as Rsync package cannot # be installed until the host is registered with a valid Red Hat support subscription node.vm.synced_folder ".", "/vagrant", disabled: false $shared_folders.each do |src, dst| node.vm.synced_folder src, dst end else node.vm.synced_folder ".", "/vagrant", disabled: false, type: "rsync", rsync__args: ['--verbose', '--archive', '--delete', '-z'] , rsync__exclude: ['.git','venv'] $shared_folders.each do |src, dst| node.vm.synced_folder src, dst, type: "rsync", rsync__args: ['--verbose', '--archive', '--delete', '-z'] end end
ip = "#{$subnet}.#{i+100}" node.vm.network :private_network, ip: ip, :libvirt__guest_ipv6 => 'yes', :libvirt__ipv6_address => "#{$subnet_ipv6}::#{i+100}", :libvirt__ipv6_prefix => "64", :libvirt__forward_mode => "none", :libvirt__dhcp_enabled => false
# Disable swap for each vm node.vm.provision "shell", inline: "swapoff -a"
# ubuntu1804 and ubuntu2004 have IPv6 explicitly disabled. This undoes that. if ["ubuntu1804", "ubuntu2004"].include? $os node.vm.provision "shell", inline: "rm -f /etc/modprobe.d/local.conf" node.vm.provision "shell", inline: "sed -i '/net.ipv6.conf.all.disable_ipv6/d' /etc/sysctl.d/99-sysctl.conf /etc/sysctl.conf" end
# Disable firewalld on oraclelinux/redhat vms if ["oraclelinux","oraclelinux8","rhel7","rhel8"].include? $os node.vm.provision "shell", inline: "systemctl stop firewalld; systemctl disable firewalld" end
host_vars[vm_name] = { "ip": ip, "flannel_interface": "eth1", "kube_network_plugin": $network_plugin, "kube_network_plugin_multus": $multi_networking, "download_run_once": $download_run_once, "download_localhost": "False", "download_cache_dir": ENV['HOME'] + "/kubespray_cache", # Make kubespray cache even when download_run_once is false "download_force_cache": $download_force_cache, # Keeping the cache on the nodes can improve provisioning speed while debugging kubespray "download_keep_remote_cache": "False", "docker_rpm_keepcache": "1", # These two settings will put kubectl and admin.config in $inventory/artifacts "kubeconfig_localhost": "True", "kubectl_localhost": "True", "local_path_provisioner_enabled": "#{$local_path_provisioner_enabled}", "local_path_provisioner_claim_root": "#{$local_path_provisioner_claim_root}", "ansible_ssh_user": SUPPORTED_OS[$os][:user] }
# Only execute the Ansible provisioner once, when all the machines are up and ready. if i == $num_instances node.vm.provision "ansible" do |ansible| ansible.playbook = $playbook $ansible_inventory_path = File.join( $inventory, "hosts.ini") if File.exist?($ansible_inventory_path) ansible.inventory_path = $ansible_inventory_path end ansible.become = true ansible.limit = "all,localhost" ansible.host_key_checking = false ansible.raw_arguments = ["--forks=#{$num_instances}", "--flush-cache", "-e ansible_become_pass=vagrant"] ansible.host_vars = host_vars #ansible.tags = ['download'] ansible.groups = { "etcd" => ["#{$instance_name_prefix}-[1:#{$etcd_instances}]"], "kube_control_plane" => ["#{$instance_name_prefix}-[1:#{$kube_master_instances}]"], "kube_node" => ["#{$instance_name_prefix}-[1:#{$kube_node_instances}]"], "k8s_cluster:children" => ["kube_control_plane", "kube_node"], } end end
end end end
|