35.director安装

• • 红帽基于tripleO---openstack on openstack开发了openstack director
  • RHOSP版本不同，安装过程会有所不同，请参照红帽官方文档

 

• • undercloud环境配置截图：

两个网卡

分区：

yum源配置：

hostname设置：

hosts文件配置：

网卡显示： 【部署成功后的截图】

 

• • 部署undercloud：

部署undercloud的时候，不能使用root用户来操作，需要使用普通用户stack

 

[root@undercloud ~]# useradd stack

[root@undercloud ~]# echo redhat | passwd --stdin stack

Changing password for user stack.

passwd: all authentication tokens updated successfully.

 

 

[root@undercloud ~]# su - stack

[stack@undercloud ~]$ mkdir images----之后要用

[root@undercloud ~]$ yum install python-tripleoclient.noarch -y

 

• • 根据undercloud的配置文件进行部署undercloud

[stack@undercloud ~]$ cp /usr/share/instack-undercloud/undercloud.conf.sample undercloud.conf

 

• • [stack@undercloud ~]$ openstack undercloud install

--------------------- END PROFILING ---------------------

[2018-02-09 18:24:53,314] (os-refresh-config) [INFO] Completed phase post-configure

os-refresh-config completed successfully

Generated new ssh key in ~/.ssh/id_rsa

Created flavor "baremetal" with profile "None"

Created flavor "control" with profile "control"

Created flavor "compute" with profile "compute"

Created flavor "ceph-storage" with profile "ceph-storage"

Created flavor "block-storage" with profile "block-storage"

Created flavor "swift-storage" with profile "swift-storage"

#############################################################################

Undercloud install complete.

 

The file containing this installation's passwords is at

/home/stack/undercloud-passwords.conf.

 

There is also a stackrc file at /home/stack/stackrc.

 

These files are needed to interact with the OpenStack services, and should be

secured.

#############################################################################

 

• • 部署overcloud

安装好了undercloud之后，会通过PXE boot启动overcloud的两台裸机来部署openstack

【overcloud里涉及的两台机器是裸机】

 

裸机的配置：

 

 

• • 准备部署overcloud的环境

[root@undercloud ~]# yum list rhosp-director*

[root@undercloud ~]# yum install rhosp-director* -y

 

 

export NOVA_VERSION=1.1

export OS_PASSWORD=$(sudo hiera admin_password)

export OS_AUTH_URL=http://192.0.2.1:5000/v2.0

export OS_USERNAME=admin

export OS_TENANT_NAME=admin

export COMPUTE_API_VERSION=1.1

export OS_NO_CACHE=True

export OS_CLOUDNAME=undercloud

export OS_IMAGE_API_VERSION=1

• • 解压镜像文件：

 

• • 开始导入镜像：

[stack@undercloud images]$ openstack overcloud image upload --image-path ~/images

 

• • 更新subnet的DNS server：

[stack@undercloud images]$ neutron subnet-list

[stack@undercloud images]$ neutron subnet-show 7afd290f-76da-4970-87fc-79e5b97fbdf7

[stack@undercloud images]$ neutron subnet-update 7afd290f-76da-4970-87fc-79e5b97fbdf7 --dns_nameservers

 

 

• • 通过PXE boot启动overcloud的两台裸机来部署openstack， pxe boot所需要的文件放在/httpboot下，用来引导系统的：

 

• • 如果想要undercloud远程PXE boot管理两台裸机，必须保证其支持IPMI-智能平台管理界面，因此需要在undercloud上做些配置。

由于vmware 是不支持IPMI的， 因此单独作别的设置来满足条件，红帽官方文档给提供了一个fake_pxe的方式启动。

 

• • 修改配置文件提供对fake_pxe启动的支持：

vim /etc/ironic/ironic.conf

添加fake_pxe到enabled_drivers 来启动这个驱动

 

• • 重启ironic服务：

# systemctl restart openstack-ironic-api.service openstack-ironic-conductor.service openstack-ironic-inspector-dnsmasq.service openstack-ironic-inspector.service

 

• • 采用导入json文件让undercloud识别两台裸机：

[stack@undercloud ~]$ vim instackenv-twonodes.json

 

{ "nodes": [         {                 "arch": "x86_64",                 "name": "controller",                 "pm_type": "fake_pxe",                 "mac": [                         "00:50:56:20:8F:97"                 ],                 "cpu": "4",                 "memory": "4096",                 "disk": "200"         },         {                 "arch": "x86_64",                 "name": "compute1",                 "pm_type": "fake_pxe",                 "mac": [                         "00:50:56:25:60:6D"                 ],                 "cpu": "4",                 "memory": "4096",                 "disk": "200"         }        ] }

• • undercloud 目前不知道要配置的node list，节点属性

$ ironic node-list

• • 导入json文件：

[stack@undercloud ~]$ openstack baremetal import --json instackenv-twonodes.json

再次查看ironic node：

• • 为overcloud里的节点分配内核：

# openstack baremetal configure boot

 

• • 检查所有节点的硬件属性

# openstack baremetal introspection bulk start

 

通过ironic node-list查看节点状态，当要求开机power on的时候，手动开机即可。

 

 

当需要关机power off，手动关机即可：

 

• • 指定节点哪台是控制节点，哪台是计算节点，设置标签命令：

# ironic node-update e36186b9-f504-48e6-91c6-8a7fd8b0c5f4 add properties/capabilities='profile:control,boot_option:local'

# ironic node-update 475b7dd3-2fa9-42a1-a57e-f6354dd615a1 add properties/capabilities='profile:compute,boot_option:local'

 

 

• • 查看node profile：

# openstack overcloud profiles list

 

 

• • 裸机introspection 完成：

• • 部署overcloud node命令：

# openstack overcloud deploy --templates -e /usr/share/openstack-tripleo-heat-templates/environments/network-isolation.yaml

--netron-network-type vxlan --neutron-tunnel-types vxlan

• • 在进行overcloud部署时候，使用ironic node-list 命令检查节点状态，当节点状态由deploying变成deploy wait-callback后，手动启动这个节点

 

 

• • active之后，手动重启两台裸机节点

 

• • 最后登录两台nodes： 使用用户heat-admin登录：