X-Git-Url: https://git.openstreetmap.org./chef.git/blobdiff_plain/0f8970107f610c90fd9a8377eebb065893569a27..166cf2983434a0a2509387466df3461b18a39625:/cookbooks/hardware/recipes/default.rb?ds=sidebyside diff --git a/cookbooks/hardware/recipes/default.rb b/cookbooks/hardware/recipes/default.rb index b1d7caf0b..b6ac62291 100644 --- a/cookbooks/hardware/recipes/default.rb +++ b/cookbooks/hardware/recipes/default.rb @@ -17,7 +17,6 @@ # limitations under the License. # -include_recipe "apt" include_recipe "git" include_recipe "munin" include_recipe "prometheus" @@ -57,16 +56,33 @@ end case manufacturer when "HP" + include_recipe "apt::management-component-pack" + package "hponcfg" + execute "update-ilo" do + action :nothing + command "/usr/sbin/hponcfg -f /etc/ilo-defaults.xml" + end + + template "/etc/ilo-defaults.xml" do + source "ilo-defaults.xml.erb" + owner "root" + group "root" + mode "644" + notifies :run, "execute[update-ilo]" + end + package "hp-health" do action :install notifies :restart, "service[hp-health]" + only_if { node[:lsb][:release].to_f < 22.04 } end service "hp-health" do action [:enable, :start] supports :status => true, :restart => true + only_if { node[:lsb][:release].to_f < 22.04 } end if product.end_with?("Gen8", "Gen9") @@ -141,7 +157,7 @@ if File.exist?("/etc/default/grub") execute "update-grub" do action :nothing command "/usr/sbin/update-grub" - not_if { ENV["TEST_KITCHEN"] } + not_if { kitchen? } end template "/etc/default/grub" do @@ -178,8 +194,17 @@ if node[:kernel][:modules].include?("ipmi_si") package "ipmitool" package "freeipmi-tools" + template "/etc/prometheus/ipmi_local.yml" do + source "ipmi_local.yml.erb" + owner "root" + group "root" + mode "644" + end + prometheus_exporter "ipmi" do port 9290 + options "--config.file=/etc/prometheus/ipmi_local.yml" + subscribes :restart, "template[/etc/prometheus/ipmi_local.yml]" end end @@ -190,13 +215,30 @@ service "irqbalance" do supports :status => false, :restart => true, :reload => false end -# Link Layer Discovery Protocol Daemon package "lldpd" + service "lldpd" do action [:start, :enable] supports :status => true, :restart => true, :reload => true end +ohai_plugin "lldp" do + template "lldp.rb.erb" +end + +package %w[ + rasdaemon + ruby-sqlite3 +] + +service "rasdaemon" do + action [:enable, :start] +end + +prometheus_exporter "rasdaemon" do + port 9797 +end + tools_packages = [] status_packages = {} @@ -219,9 +261,6 @@ if node[:virtualization][:role] != "guest" || when "mpt2sas", "mpt3sas" tools_packages << "sas2ircu" status_packages["sas2ircu-status"] ||= [] - when "megaraid_mm" - tools_packages << "megactl" - status_packages["megaraid-status"] ||= [] when "megaraid_sas" tools_packages << "megacli" status_packages["megaclisas-status"] ||= [] @@ -265,7 +304,7 @@ if tools_packages.include?("areca") depth 1 user "root" group "root" - not_if { ENV["TEST_KITCHEN"] } + not_if { kitchen? } end else directory "/opt/areca" do @@ -274,6 +313,8 @@ else end end +include_recipe "apt::hwraid" unless status_packages.empty? + if status_packages.include?("cciss-vol-status") template "/usr/local/bin/cciss-vol-statusd" do source "cciss-vol-statusd.erb" @@ -286,6 +327,7 @@ if status_packages.include?("cciss-vol-status") systemd_service "cciss-vol-statusd" do description "Check cciss_vol_status values in the background" exec_start "/usr/local/bin/cciss-vol-statusd" + nice 10 private_tmp true protect_system "full" protect_home true @@ -302,7 +344,7 @@ else end end -%w[cciss-vol-status mpt-status sas2ircu-status megaraid-status megaclisas-status aacraid-status].each do |status_package| +%w[cciss-vol-status mpt-status sas2ircu-status megaclisas-status aacraid-status].each do |status_package| if status_packages.include?(status_package) package status_package @@ -344,32 +386,36 @@ nvmes = if node[:hardware][:pci] [] end +unless nvmes.empty? + package "nvme-cli" +end + intel_nvmes = nvmes.select { |pci| pci[:vendor_name] == "Intel Corporation" } if !intel_ssds.empty? || !intel_nvmes.empty? package "unzip" - intel_mas_tool_version = "1.4" - intel_mas_package_version = "#{intel_mas_tool_version}.102-0" + sst_tool_version = "1.3" + sst_package_version = "#{sst_tool_version}.208-0" - remote_file "#{Chef::Config[:file_cache_path]}/Intel_MAS_CLI_Tool_#{intel_mas_tool_version}_Linux.zip" do - source "https://downloadmirror.intel.com/30059/eng/Intel%C2%AE_MAS_CLI_Tool_Linux.zip" + remote_file "#{Chef::Config[:file_cache_path]}/SST_CLI_Linux_#{sst_tool_version}.zip" do + source "https://downloadmirror.intel.com/743764/SST_CLI_Linux_#{sst_tool_version}.zip" end - execute "#{Chef::Config[:file_cache_path]}/Intel_MAS_CLI_Tool_#{intel_mas_tool_version}_Linux.zip" do - command "unzip Intel_MAS_CLI_Tool_#{intel_mas_tool_version}_Linux.zip intelmas_#{intel_mas_package_version}_amd64.deb" + execute "#{Chef::Config[:file_cache_path]}/SST_CLI_Linux_#{sst_tool_version}.zip" do + command "unzip SST_CLI_Linux_#{sst_tool_version}.zip sst_#{sst_package_version}_amd64.deb" cwd Chef::Config[:file_cache_path] user "root" group "root" - not_if { ::File.exist?("#{Chef::Config[:file_cache_path]}/intelmas_#{intel_mas_package_version}_amd64.deb") } + not_if { ::File.exist?("#{Chef::Config[:file_cache_path]}/sst_#{sst_package_version}_amd64.deb") } end - dpkg_package "intelmas" do - version "#{intel_mas_package_version}" - source "#{Chef::Config[:file_cache_path]}/intelmas_#{intel_mas_package_version}_amd64.deb" + dpkg_package "sst" do + version "#{sst_package_version}" + source "#{Chef::Config[:file_cache_path]}/sst_#{sst_package_version}_amd64.deb" end - dpkg_package "isdct" do + dpkg_package "intelmas" do action :purge end end @@ -531,16 +577,21 @@ if File.exist?("/etc/mdadm/mdadm.conf") end end -template "/etc/modules" do - source "modules.erb" - owner "root" - group "root" - mode "644" +file "/etc/modules" do + action :delete end -service "kmod" do - action :nothing - subscribes :start, "template[/etc/modules]" +node[:hardware][:modules].each do |module_name| + kernel_module module_name do + action :install + not_if { kitchen? } + end +end + +node[:hardware][:blacklisted_modules].each do |module_name| + kernel_module module_name do + action :blacklist + end end if node[:hardware][:watchdog] @@ -618,3 +669,7 @@ if node[:hardware][:shm_size] notifies :run, "execute[remount-dev-shm]" end end + +prometheus_collector "ohai" do + interval "15m" +end