戰(zhàn) 4—— Neutron 私有網(wǎng)絡(luò)、Cinder 塊存儲與鏡像封裝部署)
1 私有網(wǎng)絡(luò)配置1.1 控制節(jié)點(diǎn)配置配置服務(wù)組件[rootcontroller ~]# vim /etc/neutron/neutron.conf[DEFAULT] service_plugins router allow_overlapping_ips True配置 Modular Layer 2 (ML2) 插件[rootcontroller ~]# vim /etc/neutron/plugins/ml2/ml2_conf.ini[ml2] type_drivers flat,vlan,vxlan tenant_network_types vxlan mechanism_drivers linuxbridge,l2population [ml2_type_vxlan] vni_ranges 1:1000配置 Linuxbridge 代理[rootcontroller ~]# vim /etc/neutron/plugins/ml2/linuxbridge_agent.ini[vxlan] enable_vxlan True local_ip 192.168.40.161 l2_population True配置layer3代理[rootcontroller ~]# vim /etc/neutron/l3_agent.ini[DEFAULT] interface_driver neutron.agent.linux.interface.BridgeInterfaceDriver external_network_bridge 重啟服務(wù)并驗(yàn)證[rootcontroller ~]# systemctl restart neutron-server.service neutron-linuxbridge-agent.service[rootcontroller ~]# systemctl enable --now neutron-l3-agent.service[rootcontroller ~]# neutron agent-list1.2 計(jì)算節(jié)點(diǎn)配置配置 Linuxbridge 代理[rootcompute1 ~]# vim /etc/neutron/plugins/ml2/linuxbridge_agent.ini[vxlan] enable_vxlan True local_ip 192.168.40.162 l2_population True重啟服務(wù)[rootcompute1 ~]# systemctl restart neutron-linuxbridge-agent.service1.3 dashboard 配置[rootcontroller ~]# vim /etc/openstack-dashboard/local_settingsOPENSTACK_NEUTRON_NETWORK { enable_router: True, enable_quotas: True, enable_ipv6: True, enable_distributed_router: True, enable_ha_router: True, enable_lb: True, enable_firewall: True, enable_vpn: True, enable_fip_topology_check: True,[rootcontroller ~]# systemctl restart httpd.service memcached.service1.4 前端頁面測試http://192.168.40.161/dashboard登錄demo用戶創(chuàng)建私有網(wǎng)絡(luò)查看網(wǎng)絡(luò)拓?fù)涞卿沘dmin將 public 網(wǎng)絡(luò)設(shè)為外部網(wǎng)絡(luò)登錄demo添加路由器添加接口與私有網(wǎng)絡(luò)連接刪除一臺公有網(wǎng)絡(luò)的云主機(jī)在私有網(wǎng)絡(luò)添加一臺云主機(jī)查看網(wǎng)絡(luò)拓?fù)鋵?controller 的解析加入 Windows 的 hosts 文件中打開控制臺已分配IP綁定浮動IP遠(yuǎn)程登錄測試2 塊存儲服務(wù)配置存儲節(jié)點(diǎn)添加一塊 50G 的新硬盤給 block1 節(jié)點(diǎn)同步部署資源和軟件源保證多節(jié)點(diǎn)環(huán)境配置一致[rootcompute1 ~]# scp -r /opt/mitaka/ block1:/opt/[rootcompute1 ~]# scp -r /etc/yum.repos.d/openstack.repo block1:/etc/yum.repos.d/2.1 控制節(jié)點(diǎn)配置創(chuàng)建數(shù)據(jù)庫[rootcontroller ~]# mysql -p123456MariaDB[(none)]CREATE DATABASE cinder;MariaDB[(none)]GRANT ALL PRIVILEGES ON cinder.* TOcinderlocalhostIDENTIFIED BYcinder;MariaDB[(none)]GRANT ALL PRIVILEGES ON cinder.* TOcinder%IDENTIFIED BYcinder;MariaDB[(none)]quit創(chuàng)建 cinder 用戶[rootcontroller ~]# . admin-openrc[rootcontroller ~]# openstack user create --domain default --password cinder cinder添加 admin 角色到 cinder 用戶上[rootcontroller ~]# openstack role add --project service --user cinder admin創(chuàng)建 cinder 和 cinderv2 服務(wù)實(shí)體[rootcontroller ~]# openstack service create --name cinder --description OpenStack Block Storage volume[rootcontroller ~]# openstack service create --name cinderv2 --description OpenStack Block Storage volumev2創(chuàng)建塊設(shè)備存儲服務(wù)的 API 入口點(diǎn)[rootcontroller ~]# openstack endpoint create --region RegionOne volume public http://controller:8776/v1/%\(tenant_id\)s[rootcontroller ~]# openstack endpoint create --region RegionOne volume internal http://controller:8776/v1/%\(tenant_id\)s[rootcontroller ~]# openstack endpoint create --region RegionOne volume admin http://controller:8776/v1/%\(tenant_id\)s[rootcontroller ~]# openstack endpoint create --region RegionOne volumev2 public http://controller:8776/v2/%\(tenant_id\)s[rootcontroller ~]# openstack endpoint create --region RegionOne volumev2 internal http://controller:8776/v2/%\(tenant_id\)s[rootcontroller ~]# openstack endpoint create --region RegionOne volumev2 admin http://controller:8776/v2/%\(tenant_id\)s安裝并配置組件[rootcontroller ~]# yum install openstack-cinder[rootcontroller ~]# vim /etc/cinder/cinder.conf[database] connection mysqlpymysql://cinder:cindercontroller/cinder [DEFAULT] rpc_backend rabbit auth_strategy keystone my_ip 192.168.40.161 [oslo_messaging_rabbit] rabbit_host controller rabbit_userid openstack rabbit_password openstack [keystone_authtoken] auth_uri http://controller:5000 auth_url http://controller:35357 memcached_servers controller:11211 auth_type password project_domain_name default user_domain_name default project_name service username cinder password cinder [oslo_concurrency] lock_path /var/lib/cinder/tmp初始化塊設(shè)備服務(wù)的數(shù)據(jù)庫[rootcontroller ~]# su -s /bin/sh -c cinder-manage db sync cinder配置計(jì)算節(jié)點(diǎn)[rootcontroller ~]# vim /etc/nova/nova.conf[cinder] os_region_name RegionOne重啟計(jì)算服務(wù)啟動塊設(shè)備存儲服務(wù)并將其配置為開機(jī)自啟[rootcontroller ~]# systemctl restart openstack-nova-api.service[rootcontroller ~]# systemctl enable --now openstack-cinder-api.service openstack-cinder-scheduler.service驗(yàn)證[rootcontroller ~]# cinder service-list2.2 存儲節(jié)點(diǎn)配置安裝 LVM 包啟動 LVM 的 metadata 服務(wù)并且設(shè)置該服務(wù)隨系統(tǒng)啟動[rootblock1 ~]# yum install lvm2[rootblock1 ~]# systemctl enable --now lvm2-lvmetad.service創(chuàng)建LVM 物理卷 /dev/sdb 和 LVM 卷組 cinder-volumes[rootblock1 ~]# pvcreate /dev/sdb[rootblock1 ~]# vgcreate cinder-volumes /dev/sdbLVM 過濾器只允許 LVM 識別系統(tǒng)盤 sda 和 Cinder 數(shù)據(jù)盤 sdb屏蔽其他所有磁盤防止宿主機(jī)誤識別虛擬機(jī)內(nèi)部 LVM避免數(shù)據(jù)損壞[rootblock1 ~]# vim /etc/lvm/lvm.conf安裝并配置組件[rootblock1 ~]# yum install openstack-cinder targetcli python-keystone[rootblock1 ~]# vim /etc/cinder/cinder.conf[database] connection mysqlpymysql://cinder:cindercontroller/cinder [DEFAULT] rpc_backend rabbit auth_strategy keystone my_ip 192.168.30.163 enabled_backends lvm glance_api_servers http://controller:9292 [oslo_messaging_rabbit] rabbit_host controller rabbit_userid openstack rabbit_password openstack [keystone_authtoken] auth_uri http://controller:5000 auth_url http://controller:35357 memcached_servers controller:11211 auth_type password project_domain_name default user_domain_name default project_name service username cinder password cinder [lvm] volume_driver cinder.volume.drivers.lvm.LVMVolumeDriver volume_group cinder-volumes iscsi_protocol iscsi iscsi_helper lioadm # 此部分在原文件中沒有加在文件最后 [oslo_concurrency] lock_path /var/lib/cinder/tmp啟動塊存儲卷服務(wù)及其依賴的服務(wù)并將其配置為隨系統(tǒng)啟動[rootblock1 ~]# systemctl enable --now openstack-cinder-volume.service target.service驗(yàn)證[rootcontroller ~]# cinder service-list2.3 前端頁面測試在 demo 用戶下可以看見卷創(chuàng)建云硬盤連接到云主機(jī) esc-2遠(yuǎn)程連接查看不分區(qū)直接格式化$sudomkfs.ext4 /dev/vdbumount 卸載之后分離云硬盤擴(kuò)容到10G重新連接到 ecs-2在遠(yuǎn)程查看云硬盤后臺擴(kuò)大容量后由于未分區(qū)虛擬機(jī)內(nèi)部可直接用resize2fs在線擴(kuò)容 ext 文件系統(tǒng)才能讓 df 識別到擴(kuò)容后的全部磁盤空間卸載、分離后為云硬盤創(chuàng)建快照用快照創(chuàng)建硬盤3 封裝及使用鏡像參考文檔https://docs.openstack.org/image-guide/create-images-manually-example-centos-image.html#download-a-centos-install-isoKVMKernel?based Virtual Machine基于內(nèi)核的虛擬機(jī)是Linux 內(nèi)核自帶的虛擬化技術(shù)屬于 Type?1 裸金屬虛擬化內(nèi)核變成虛擬機(jī)監(jiān)控器免費(fèi)開源Rocky Linux、CentOS、Ubuntu 都原生自帶。簡單理解讓一臺物理 Linux 機(jī)器可以同時跑很多臺獨(dú)立虛擬機(jī)就像 VMware Workstation但跑在 Linux 服務(wù)器上服務(wù)器虛擬化行業(yè)大量使用。新建虛擬機(jī)封裝流程裝好系統(tǒng)→預(yù)配置軟件→清理所有唯一標(biāo)識機(jī)器 ID、ssh 密鑰、網(wǎng)卡規(guī)則、日志、歷史→關(guān)機(jī)→做成 qcow2 模板鏡像封裝完成后絕對不能再開機(jī)安裝完成后開啟按照文檔要求進(jìn)行配置[rootbogon ~]# cd /etc/yum.repos.d/[rootbogon yum.repos.d]# rm -fr *[rootbogon yum.repos.d]# lsCentOS-Base.repo# 從控制節(jié)點(diǎn)拷貝[rootbogon ~]# yum install -y vim bash-completion chrony[rootbogon ~]# vim /etc/chrony.conf[rootbogon ~]# systemctl enable --now chronyd[rootbogon ~]# systemctl disable --now firewalld[rootbogon ~]# yum install acpid[rootbogon ~]# systemctl enable acpid[rootbogon ~]# yum install cloud-init[rootbogon ~]# yum install cloud-utils-growpart[rootbogon ~]# cd /etc/ssh/[rootbogon ssh]# rm -f ssh_host_*[rootbogon ~]# vim /boot/grub2/grub.cfg[rootbogon ssh]# grub2-mkconfig -o /boot/grub2/grub.cfgpoweroff 關(guān)機(jī)導(dǎo)出一定要導(dǎo)出為 ova 格式OVF是一套文件不是單個文件.ovf(虛擬機(jī)配置) .vmdk(磁盤) .mf(校驗(yàn)文件)好幾個散文件拷貝的時候必須把 3 個文件全部一起復(fù)制少一個就壞掉。OVA本質(zhì)就是把 ovfvmdkmf 全部打包壓縮成單個 tar 壓縮包。新開虛擬機(jī) server2以 rocky 9.6 為鏡像上傳打包好的.ova文件把 VMware 的 ova 模板轉(zhuǎn)換成 KVM (qemu?qcow2) 虛擬機(jī)鏡像輸出到本地目錄[rootserver2 ~]# yum install -y guestfs-tools.x86_64 virt-v2v.x86_64[rootserver2 ~]# mkdir /opt/v2v[rootserver2 ~]# virt-v2v -i ova demo.ova -o local -of qcow2 -os /opt/v2v/libvirt 虛擬化服務(wù)未啟動Rocky 9 新版本 libvirt 是拆分多守護(hù)進(jìn)程模式不再是傳統(tǒng)單一libvirtd拆分成virtqemud、virtnetworkd、virtstoraged、virtsecretd分別負(fù)責(zé) QEMU 虛擬機(jī)、網(wǎng)絡(luò)、存儲、密鑰管理開啟對應(yīng)服務(wù)即可[rootserver2 ~]# systemctl list-unit-files | grep virt[rootserver2 ~]# systemctl start virtqemud virtnetworkd virtstoraged virtsecretd[rootserver2 ~]# virt-v2v -i ova demo.ova -o local -of qcow2 -os /opt/v2v/輸出 qcow2 格式的鏡像成功清理qcow2鏡像里主機(jī)唯一標(biāo)識、日志和密鑰把它處理成可批量克隆的虛擬機(jī)模板母盤[rootserver2 ~]# cd /opt/v2v/[rootserver2 v2v]# virt-sysprep -a demo-sda[rootserver2 v2v]# virt-sparsify --compress demo-sda demo.qcow2導(dǎo)出創(chuàng)建鏡像創(chuàng)建云主機(jī)類型啟動云主機(jī)實(shí)例啟動實(shí)例可看到引導(dǎo)記錄打開控制臺系統(tǒng)會自動創(chuàng)建一個用戶 centos綁定浮動 VIP遠(yuǎn)程連接查看磁盤將 my_disk 連接到 ecs-3分區(qū)格式化后掛載如若要給硬盤擴(kuò)容需先取消掛載分離云硬盤再擴(kuò)容擴(kuò)容成功后重新連接到云主機(jī)與 2.3 小節(jié)中的擴(kuò)容不同由于磁盤已經(jīng)分區(qū)需要刪除分區(qū)重新分區(qū)并掛載[rootecs-3 ~]# fdisk /dev/vdb[rootecs-3 ~]# partprobe[rootecs-3 ~]# mount /dev/vdb1 /data/由于已經(jīng)格式化需要在線擴(kuò)容文件系統(tǒng)[rootecs-3 ~]# xfs_growfs /dev/vdb1 OpenStack 私有云實(shí)戰(zhàn) 4—— Neutron 私有網(wǎng)絡(luò)、Cinder 塊存儲與鏡像封裝部署
1 私有網(wǎng)絡(luò)配置1.1 控制節(jié)點(diǎn)配置配置服務(wù)組件[rootcontroller ~]# vim /etc/neutron/neutron.conf[DEFAULT] service_plugins router allow_overlapping_ips True配置 Modular Layer 2 (ML2) 插件[rootcontroller ~]# vim /etc/neutron/plugins/ml2/ml2_conf.ini[ml2] type_drivers flat,vlan,vxlan tenant_network_types vxlan mechanism_drivers linuxbridge,l2population [ml2_type_vxlan] vni_ranges 1:1000配置 Linuxbridge 代理[rootcontroller ~]# vim /etc/neutron/plugins/ml2/linuxbridge_agent.ini[vxlan] enable_vxlan True local_ip 192.168.40.161 l2_population True配置layer3代理[rootcontroller ~]# vim /etc/neutron/l3_agent.ini[DEFAULT] interface_driver neutron.agent.linux.interface.BridgeInterfaceDriver external_network_bridge 重啟服務(wù)并驗(yàn)證[rootcontroller ~]# systemctl restart neutron-server.service neutron-linuxbridge-agent.service[rootcontroller ~]# systemctl enable --now neutron-l3-agent.service[rootcontroller ~]# neutron agent-list1.2 計(jì)算節(jié)點(diǎn)配置配置 Linuxbridge 代理[rootcompute1 ~]# vim /etc/neutron/plugins/ml2/linuxbridge_agent.ini[vxlan] enable_vxlan True local_ip 192.168.40.162 l2_population True重啟服務(wù)[rootcompute1 ~]# systemctl restart neutron-linuxbridge-agent.service1.3 dashboard 配置[rootcontroller ~]# vim /etc/openstack-dashboard/local_settingsOPENSTACK_NEUTRON_NETWORK { enable_router: True, enable_quotas: True, enable_ipv6: True, enable_distributed_router: True, enable_ha_router: True, enable_lb: True, enable_firewall: True, enable_vpn: True, enable_fip_topology_check: True,[rootcontroller ~]# systemctl restart httpd.service memcached.service1.4 前端頁面測試http://192.168.40.161/dashboard登錄demo用戶創(chuàng)建私有網(wǎng)絡(luò)查看網(wǎng)絡(luò)拓?fù)涞卿沘dmin將 public 網(wǎng)絡(luò)設(shè)為外部網(wǎng)絡(luò)登錄demo添加路由器添加接口與私有網(wǎng)絡(luò)連接刪除一臺公有網(wǎng)絡(luò)的云主機(jī)在私有網(wǎng)絡(luò)添加一臺云主機(jī)查看網(wǎng)絡(luò)拓?fù)鋵?controller 的解析加入 Windows 的 hosts 文件中打開控制臺已分配IP綁定浮動IP遠(yuǎn)程登錄測試2 塊存儲服務(wù)配置存儲節(jié)點(diǎn)添加一塊 50G 的新硬盤給 block1 節(jié)點(diǎn)同步部署資源和軟件源保證多節(jié)點(diǎn)環(huán)境配置一致[rootcompute1 ~]# scp -r /opt/mitaka/ block1:/opt/[rootcompute1 ~]# scp -r /etc/yum.repos.d/openstack.repo block1:/etc/yum.repos.d/2.1 控制節(jié)點(diǎn)配置創(chuàng)建數(shù)據(jù)庫[rootcontroller ~]# mysql -p123456MariaDB[(none)]CREATE DATABASE cinder;MariaDB[(none)]GRANT ALL PRIVILEGES ON cinder.* TOcinderlocalhostIDENTIFIED BYcinder;MariaDB[(none)]GRANT ALL PRIVILEGES ON cinder.* TOcinder%IDENTIFIED BYcinder;MariaDB[(none)]quit創(chuàng)建 cinder 用戶[rootcontroller ~]# . admin-openrc[rootcontroller ~]# openstack user create --domain default --password cinder cinder添加 admin 角色到 cinder 用戶上[rootcontroller ~]# openstack role add --project service --user cinder admin創(chuàng)建 cinder 和 cinderv2 服務(wù)實(shí)體[rootcontroller ~]# openstack service create --name cinder --description OpenStack Block Storage volume[rootcontroller ~]# openstack service create --name cinderv2 --description OpenStack Block Storage volumev2創(chuàng)建塊設(shè)備存儲服務(wù)的 API 入口點(diǎn)[rootcontroller ~]# openstack endpoint create --region RegionOne volume public http://controller:8776/v1/%\(tenant_id\)s[rootcontroller ~]# openstack endpoint create --region RegionOne volume internal http://controller:8776/v1/%\(tenant_id\)s[rootcontroller ~]# openstack endpoint create --region RegionOne volume admin http://controller:8776/v1/%\(tenant_id\)s[rootcontroller ~]# openstack endpoint create --region RegionOne volumev2 public http://controller:8776/v2/%\(tenant_id\)s[rootcontroller ~]# openstack endpoint create --region RegionOne volumev2 internal http://controller:8776/v2/%\(tenant_id\)s[rootcontroller ~]# openstack endpoint create --region RegionOne volumev2 admin http://controller:8776/v2/%\(tenant_id\)s安裝并配置組件[rootcontroller ~]# yum install openstack-cinder[rootcontroller ~]# vim /etc/cinder/cinder.conf[database] connection mysqlpymysql://cinder:cindercontroller/cinder [DEFAULT] rpc_backend rabbit auth_strategy keystone my_ip 192.168.40.161 [oslo_messaging_rabbit] rabbit_host controller rabbit_userid openstack rabbit_password openstack [keystone_authtoken] auth_uri http://controller:5000 auth_url http://controller:35357 memcached_servers controller:11211 auth_type password project_domain_name default user_domain_name default project_name service username cinder password cinder [oslo_concurrency] lock_path /var/lib/cinder/tmp初始化塊設(shè)備服務(wù)的數(shù)據(jù)庫[rootcontroller ~]# su -s /bin/sh -c cinder-manage db sync cinder配置計(jì)算節(jié)點(diǎn)[rootcontroller ~]# vim /etc/nova/nova.conf[cinder] os_region_name RegionOne重啟計(jì)算服務(wù)啟動塊設(shè)備存儲服務(wù)并將其配置為開機(jī)自啟[rootcontroller ~]# systemctl restart openstack-nova-api.service[rootcontroller ~]# systemctl enable --now openstack-cinder-api.service openstack-cinder-scheduler.service驗(yàn)證[rootcontroller ~]# cinder service-list2.2 存儲節(jié)點(diǎn)配置安裝 LVM 包啟動 LVM 的 metadata 服務(wù)并且設(shè)置該服務(wù)隨系統(tǒng)啟動[rootblock1 ~]# yum install lvm2[rootblock1 ~]# systemctl enable --now lvm2-lvmetad.service創(chuàng)建LVM 物理卷 /dev/sdb 和 LVM 卷組 cinder-volumes[rootblock1 ~]# pvcreate /dev/sdb[rootblock1 ~]# vgcreate cinder-volumes /dev/sdbLVM 過濾器只允許 LVM 識別系統(tǒng)盤 sda 和 Cinder 數(shù)據(jù)盤 sdb屏蔽其他所有磁盤防止宿主機(jī)誤識別虛擬機(jī)內(nèi)部 LVM避免數(shù)據(jù)損壞[rootblock1 ~]# vim /etc/lvm/lvm.conf安裝并配置組件[rootblock1 ~]# yum install openstack-cinder targetcli python-keystone[rootblock1 ~]# vim /etc/cinder/cinder.conf[database] connection mysqlpymysql://cinder:cindercontroller/cinder [DEFAULT] rpc_backend rabbit auth_strategy keystone my_ip 192.168.30.163 enabled_backends lvm glance_api_servers http://controller:9292 [oslo_messaging_rabbit] rabbit_host controller rabbit_userid openstack rabbit_password openstack [keystone_authtoken] auth_uri http://controller:5000 auth_url http://controller:35357 memcached_servers controller:11211 auth_type password project_domain_name default user_domain_name default project_name service username cinder password cinder [lvm] volume_driver cinder.volume.drivers.lvm.LVMVolumeDriver volume_group cinder-volumes iscsi_protocol iscsi iscsi_helper lioadm # 此部分在原文件中沒有加在文件最后 [oslo_concurrency] lock_path /var/lib/cinder/tmp啟動塊存儲卷服務(wù)及其依賴的服務(wù)并將其配置為隨系統(tǒng)啟動[rootblock1 ~]# systemctl enable --now openstack-cinder-volume.service target.service驗(yàn)證[rootcontroller ~]# cinder service-list2.3 前端頁面測試在 demo 用戶下可以看見卷創(chuàng)建云硬盤連接到云主機(jī) esc-2遠(yuǎn)程連接查看不分區(qū)直接格式化$sudomkfs.ext4 /dev/vdbumount 卸載之后分離云硬盤擴(kuò)容到10G重新連接到 ecs-2在遠(yuǎn)程查看云硬盤后臺擴(kuò)大容量后由于未分區(qū)虛擬機(jī)內(nèi)部可直接用resize2fs在線擴(kuò)容 ext 文件系統(tǒng)才能讓 df 識別到擴(kuò)容后的全部磁盤空間卸載、分離后為云硬盤創(chuàng)建快照用快照創(chuàng)建硬盤3 封裝及使用鏡像參考文檔https://docs.openstack.org/image-guide/create-images-manually-example-centos-image.html#download-a-centos-install-isoKVMKernel?based Virtual Machine基于內(nèi)核的虛擬機(jī)是Linux 內(nèi)核自帶的虛擬化技術(shù)屬于 Type?1 裸金屬虛擬化內(nèi)核變成虛擬機(jī)監(jiān)控器免費(fèi)開源Rocky Linux、CentOS、Ubuntu 都原生自帶。簡單理解讓一臺物理 Linux 機(jī)器可以同時跑很多臺獨(dú)立虛擬機(jī)就像 VMware Workstation但跑在 Linux 服務(wù)器上服務(wù)器虛擬化行業(yè)大量使用。新建虛擬機(jī)封裝流程裝好系統(tǒng)→預(yù)配置軟件→清理所有唯一標(biāo)識機(jī)器 ID、ssh 密鑰、網(wǎng)卡規(guī)則、日志、歷史→關(guān)機(jī)→做成 qcow2 模板鏡像封裝完成后絕對不能再開機(jī)安裝完成后開啟按照文檔要求進(jìn)行配置[rootbogon ~]# cd /etc/yum.repos.d/[rootbogon yum.repos.d]# rm -fr *[rootbogon yum.repos.d]# lsCentOS-Base.repo# 從控制節(jié)點(diǎn)拷貝[rootbogon ~]# yum install -y vim bash-completion chrony[rootbogon ~]# vim /etc/chrony.conf[rootbogon ~]# systemctl enable --now chronyd[rootbogon ~]# systemctl disable --now firewalld[rootbogon ~]# yum install acpid[rootbogon ~]# systemctl enable acpid[rootbogon ~]# yum install cloud-init[rootbogon ~]# yum install cloud-utils-growpart[rootbogon ~]# cd /etc/ssh/[rootbogon ssh]# rm -f ssh_host_*[rootbogon ~]# vim /boot/grub2/grub.cfg[rootbogon ssh]# grub2-mkconfig -o /boot/grub2/grub.cfgpoweroff 關(guān)機(jī)導(dǎo)出一定要導(dǎo)出為 ova 格式OVF是一套文件不是單個文件.ovf(虛擬機(jī)配置) .vmdk(磁盤) .mf(校驗(yàn)文件)好幾個散文件拷貝的時候必須把 3 個文件全部一起復(fù)制少一個就壞掉。OVA本質(zhì)就是把 ovfvmdkmf 全部打包壓縮成單個 tar 壓縮包。新開虛擬機(jī) server2以 rocky 9.6 為鏡像上傳打包好的.ova文件把 VMware 的 ova 模板轉(zhuǎn)換成 KVM (qemu?qcow2) 虛擬機(jī)鏡像輸出到本地目錄[rootserver2 ~]# yum install -y guestfs-tools.x86_64 virt-v2v.x86_64[rootserver2 ~]# mkdir /opt/v2v[rootserver2 ~]# virt-v2v -i ova demo.ova -o local -of qcow2 -os /opt/v2v/libvirt 虛擬化服務(wù)未啟動Rocky 9 新版本 libvirt 是拆分多守護(hù)進(jìn)程模式不再是傳統(tǒng)單一libvirtd拆分成virtqemud、virtnetworkd、virtstoraged、virtsecretd分別負(fù)責(zé) QEMU 虛擬機(jī)、網(wǎng)絡(luò)、存儲、密鑰管理開啟對應(yīng)服務(wù)即可[rootserver2 ~]# systemctl list-unit-files | grep virt[rootserver2 ~]# systemctl start virtqemud virtnetworkd virtstoraged virtsecretd[rootserver2 ~]# virt-v2v -i ova demo.ova -o local -of qcow2 -os /opt/v2v/輸出 qcow2 格式的鏡像成功清理qcow2鏡像里主機(jī)唯一標(biāo)識、日志和密鑰把它處理成可批量克隆的虛擬機(jī)模板母盤[rootserver2 ~]# cd /opt/v2v/[rootserver2 v2v]# virt-sysprep -a demo-sda[rootserver2 v2v]# virt-sparsify --compress demo-sda demo.qcow2導(dǎo)出創(chuàng)建鏡像創(chuàng)建云主機(jī)類型啟動云主機(jī)實(shí)例啟動實(shí)例可看到引導(dǎo)記錄打開控制臺系統(tǒng)會自動創(chuàng)建一個用戶 centos綁定浮動 VIP遠(yuǎn)程連接查看磁盤將 my_disk 連接到 ecs-3分區(qū)格式化后掛載如若要給硬盤擴(kuò)容需先取消掛載分離云硬盤再擴(kuò)容擴(kuò)容成功后重新連接到云主機(jī)與 2.3 小節(jié)中的擴(kuò)容不同由于磁盤已經(jīng)分區(qū)需要刪除分區(qū)重新分區(qū)并掛載[rootecs-3 ~]# fdisk /dev/vdb[rootecs-3 ~]# partprobe[rootecs-3 ~]# mount /dev/vdb1 /data/由于已經(jīng)格式化需要在線擴(kuò)容文件系統(tǒng)[rootecs-3 ~]# xfs_growfs /dev/vdb1