Well, why install 6.0. The previous post mentioned that 6.1 to the public network connection requirements are relatively high, my experimental environment only a NAT and a bridge network can be used, and I do not have to be able to route the network (temporarily)
2015.09.04 Update Yangjiang
Environment: Vmware Workstation
PC:32GB memory, SSD hard drive
Network preparation:
Setting up a VMware NAT network: 10.20.2.0/255.255.255.0, saving
Start the Windows XP virtual machine and get IP,10.20.2.3,GW 10.20.2.2,dns automatically 10.20.2.2
Ping www.baidu.com
Stop Windows services VMware DHCP Service
(Why should I stop the VMware DHCP service?) Because I want to use the VMware NAT network to do the admin network for OpenStack, the OpenStack Admin network does not allow other DHCP services, in the future OpenStack compute node to take the IP from the controll node's DHCP server, Do PXE boot)
Insert the CD ISO file to start a clean fuel virtual machine
On the CD-ROM installation splash screen, press the TAB key to modify the IP address
My NAT network Environment 10.20.0.0/24, it is strange that the Mirantis default is not to use the NAT network gateway 10.20.0.2 This IP, modify it.
Modify IP address:10.20.2.4, Default gateway:10.20.2.2, DHCP gatway:10.20.2.2
Enter, start the installation process, the virtual machines restart, the middle prompts whether to enter fuel setup
(Sadly, when I saw it, it had gone slightly)
Start extracting and loading Docker images.
Show a bunch of debug:executing ' docker ... ' info
Installation for about 20-3 minutes
(Note: If a good IP is not modified now, the IP address is modified by modifying/etc/sysconfig/network-scripts/ifcfg-eth0 to 10.20.0.4,fuel 6.0 still with the old IP address 10.20.0.2.)
This time causes fuel boot failure, access to Http://xxxx:8000/website always shows funnel
Reference log file/var/log/docker will see error log messages)
After installation, the Linux login interface prompts to access the http://10.20.2.4:8000/into the Fuel Management interface
Operating system Account Root/r00tme,fuel account Admin/admin
Now, log in to Http://10.20.2.4:8000/, click the button to create my first OpenStack environment demo01, the wizard modifies two places
1. Installing OpenStack with CentOS
2. Instead of using the HA installation controller, use a multi-node
Other: Cinder default (without celth), network with neutron net default
In the fuel interface, demo01 the OpenStack admin interface, click on the Second tab network,
Modify the public and floating networks to 172.16.0.x to 192.168.2.x, and the corresponding gateway, click Save Settings at the bottom right of the screen
In VMware,
Create 3 empty virtual machines, configure two network Nat/bridge, HDD 40GB or above.
Start the virtual machine, and the virtual machines are automatically booted with the PXE Mount OpenStack Mirror (because 3 empty virtual machines and fuel virtual machines are connected through the NAT network)
After startup, show Bootstrap login
In the fuel interface, demo01 the OpenStack admin interface, click on the Second tab network,
Click Verify network at the bottom right of the screen
After about a few minutes, the verification is complete, prompting the validation through
Click on the first tab node, add nodes, set 3 empty virtual machines as controller, Compute, Storage-cinder
Click Network Configuration to view
Click Deploy change, about 1 hours, deployment complete, prompt
Deployment of Environment ' demo01 ' are done. Access The OpenStack Dashboard (Horizon) at http://192.168.2.20/or via internal network at http://10.20.2.7/
Recommendation: Click on the Fifth tab of the health check to allow testing
Login to the OpenStack Admin interface http://10.20.2.7/
Strangely, the Admin user cannot launch instance!!!
Take a closer look, the left navigation bar, project, Admin, Identity, in project can be launch instance, in the Admin-and system launch instance
Sweat. 2015.09.04 Yangjiang
A lot of details are omitted, please refer to the online games of the Illustrated article:
Http://blog.vsharing.com/wpskl/A1738986.html