LXC On OpenSuse

Sep 1, 2014
Technology

LXC 相关操作

列出本机已有的容器：

# lxc-ls
xxxxxyyySimulator1  xxxxxyyySimulator2

开启虚拟机：

# lxc-start -n xxxxxyyySimulator1

本机开启终端连接到开启后的虚拟机：

# lxc-console --name xxxxxyyySimulator1

Type <Ctrl+a q> to exit the console, <Ctrl+a Ctrl+a> to enter Ctrl+a itself

Welcome to openSUSE 12.3 "Dartmouth" - Kernel 3.11.10-21-default (tty1).


xxxxxyyySimulator1 login: root
Password: 
Last login: Fri Aug 29 15:40:54 from xxx.xxx.xx.59
Have a lot of fun...
xxxxxyyySimulator1:~ #

用户名和密码都是"root”. ctrl+a后按q即可退出该终端。

销毁容器:

lxc-destroy -n XXXXXXXXXX

克隆容器:

bash /usr/bin/lxc-clone -o xxxxxyyySimulator1 -n xxxxxyyySimulator2

其中-o 是源容器， -n 后接的是目的容器名，目的容器会自动创建。

LXC 容器修改

比如，网络配置在下列文件里：

# cat /var/lib/lxc/xxxxxyyySimulator2/config | more
# Template used to create this container: /usr/share/lxc/templates/lxc-opensuse
# Template script checksum (SHA-1): xxxxxxxxxxxxxxxxxxxxxxxxxxxx

#lxc.network.type = empty
lxc.network.type = veth
lxc.network.flags = up
lxc.network.link = br0
lxc.network.name = eth0
lxc.network.ipv4 = xxx.xxx.xx.67/24
lxc.network.ipv4.gateway = xxx.xxx.xx.1

network.ipv4代表其IP地址，而Gateway则代表其默认路由。

虚拟机位置：

linux:/var/lib/lxc # du -hs *
483M	xxxxxyyySimulator1
483M	xxxxxyyySimulator2
linux:/var/lib/lxc # pwd
/var/lib/lxc

克隆后的虚拟机也在同一目录下。

启动后的虚拟机，都可以被视为真实的物理机，可以通过ssh直接连上去操作。

LXC Computer

Aug 27, 2014
Technology

Network Configuration

Add the rules in udevd:

linux-:~ # cd /etc/udev/rules.d/
linux-:/etc/udev/rules.d # cat 10-network.rules
SUBSYSTEM=="net", ACTION=="add", ATTR{address}=="00:22:22:22:22", NAME="eth1"
SUBSYSTEM=="net", ACTION=="add", ATTR{address}=="00:22:22:22:22", NAME="eth0"
SUBSYSTEM=="net", ACTION=="add", ATTR{address}=="00:22:22:22:22", NAME="eth2"
SUBSYSTEM=="net", ACTION=="add", ATTR{address}=="00:22:22:22:22", NAME="eth3"

Add following network configuration:

linux-:/etc/sysconfig # cd network/
linux-:/etc/sysconfig/network # cat ifcfg-eth0
# Loopback (lo) configuration
IPADDR=1xx.xx.xx.xxx
NETMASK=255.255.255.0
BROADCAST=1xx.xx.xx.xxx
STARTMODE=auto
USERCONTROL=yes
FIREWALL=no

Default Gateway Setup:

linux-:~ # cat /etc/sysconfig/network/ifroute-br0
# Destination     Dummy/Gateway     Netmask            Device
#
default		xxx.xxx.xx.1	    255.255.255.255	br0

Restart computer then you got the fixed ip address in eth0.
Add the default route so we could get outside.

vim routes
xxx.xxx.xx.1 -eth0 
or manually: 
route add default gw xxx.xxx.xx.1 eth0

Bridge Networking Configuration:

linux-:/etc/sysconfig/network # cat ifcfg-br0
STARTMODE='auto'
BOOTPROTO='static'
DNS1=xxx.xxx.xx.1
GATEWAY=xxx.xxx.xx.1
IPADDR=xxx.xxx.xx.59
NETMASK=255.255.255.0
ONBOOT=yes
USERCONTROL='no'
BRIDGE='yes'
BRIDGE_PORTS='eth0'
BRIDGE_AGEINGTIME='20'
BRIDGE_FORWARDDELAY='0'
BRIDGE_HELLOTIME='2'
BRIDGE_MAXAGE='20'
BRIDGE_PATHCOSTS='3'
BRIDGE_STP='on'
linux-:/etc/sysconfig/network # cat ifcfg-eth0
BOOTPROTO='static'
STARTMODE='ifplugd'
IFPLUGD_PRIORITY='1'
NAME = '1000 mBIT ETHERNET'
USERCTL=no

The route should changed to:

route add default gw xxx.xxx.xx.1 br0

LXC Install

Enable the free ways:

ssh -C -L 127.0.0.1:9001:1xx.xxx.2xx:2xxxx root@1xx.xx.1xx.xxx

Use zypper to install the container.

zypper search lxc
# zypper install lxc lxc-devel yast2-lxc libvirt-daemon-lxc libvirt-daemon-driver-lxc
# lxc-checkconfig 
# ls /usr/share/lxc/templates/

Yes we have the opensuse specified configuraton.

Create The first Container:

lxc-create -n ixxxxxSimulator1 -t /usr/share/lxc/templates/lxc-opensuse

List the installed Container:

linux-:~ # lxc-ls
xxxxhxxSimulator1

Username and password are root.

Start the lxc machine via:

lxc-start -n xxxxxSimulator1

LXC Configuration

No Network, Add it!
First we remove the desktop kernel. and use the default kernel

# uname -a
Linux XXXXSimulator1 3.11.6-4-desktop 
# zypper in kernel-default
# zypper rm kernel-desktop
# uname -a
Linux linux- 3.11.10-21-default

Enable the xfce4 for the default vnc server desktop:

zypper in -t pattern xfce

Change the default lxc configuraiton of network:

$ vim /var/lib/lxc/XXXXSimulator1/config
lxc.network.type = veth
lxc.network.flags = up
lxc.network.link = br0
lxc.network.name = eth0

Now if you start the lxc container, the network eth0 will be automatically started.

LXC Expand

Duplicate LXC Machine.
This is strange when we directly call lxc-clone will cause failed silently:

 # lxc-clone -o XXXXSimulator1 -n XXXXSimulator2
linux-:~ # echo $?
1

Then we use this:

# bash /usr/bin/lxc-clone -o XXXXSimulator1 -n XXXXSimulator2
Tweaking configuration
Copying rootfs...
Updating rootfs...
'XXXXSimulator2' created
linux-:~ # lxc-ls
XXXXSimulator1  XXXXSimulator2

Change the XXXXSimulator2’s configuration:

$ vim /var/lib/lxc/XXXXSimulator2/config
lxc.network.ipv4 = xxx.xxx.xx.67

Now start the two LXC via:

# lxc-start  -n XXXXSimulator2
# lxc-start  -n XXXXSimulator1
[Trusty@Linux01 ~]$ ping -c 1 xxx.xxx.xx.66
PING xxx.xxx.xx.66 (xxx.xxx.xx.66) 56(84) bytes of data.
64 bytes from xxx.xxx.xx.66: icmp_seq=1 ttl=64 time=1.50 ms

--- xxx.xxx.xx.66 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 1ms
rtt min/avg/max/mdev = 1.506/1.506/1.506/0.000 ms
[Trusty@Linux01 ~]$ ping -c 1 xxx.xxx.xx.67
PING xxx.xxx.xx.67 (xxx.xxx.xx.67) 56(84) bytes of data.
64 bytes from xxx.xxx.xx.67: icmp_seq=1 ttl=64 time=1.56 ms

--- xxx.xxx.xx.67 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 1ms
rtt min/avg/max/mdev = 1.567/1.567/1.567/0.000 ms

Later we could configure the LXC, to let the container start at bootup, Or control its behavior.

LXC Computer Configuration

The IP address and Default Gateway Configuration:

lxc.network.type = veth
lxc.network.flags = up
lxc.network.link = br0
lxc.network.name = eth0
lxc.network.ipv4 = xxx.xxx.xx.59/24
lxc.network.ipv4.gateway = xxx.xxx.xx.1

Then start the LXC Container, you will see the ip address/netmask already configured.

LXC Destroy

Destroyed the unused Container2:

linux-:~ # lxc-ls
XXXXSimulator1  XXXXSimulator2
linux-:~ # lxc-destroy -n XXXXSimulator2
linux-:~ # lxc-ls
XXXXSimulator1

Puppet on ArchLinux(2)

Aug 13, 2014
Technology

Add Arch

step1, add the hosts into /etc/hosts:

# Puppet 
10.0.0.88	puppet
10.0.0.89	client

step2, edit the /etc/puppet/puppet.conf:

[agent]
    # add server
    [agent]
    server = puppet

Restart the puppet.service:

systemctl restart puppet.service
systemctl enable puppet.service

step3, in 10.0.0.88(server), add the ssl certification of archlinux:

root@Ubuntu88:/home/Trusty# !44
puppet cert --list
Warning: Setting templatedir is deprecated. See http://links.puppetlabs.com/env-settings-deprecations
   (at /usr/lib/ruby/vendor_ruby/puppet/settings.rb:1095:in `block in issue_deprecations')
  "XXXyyy.lan" (SHA256) 8XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
root@Ubuntu88:/home/Trusty# puppet cert --sign XXXyyy.lan
Warning: Setting templatedir is deprecated. See http://links.puppetlabs.com/env-settings-deprecations
   (at /usr/lib/ruby/vendor_ruby/puppet/settings.rb:1095:in `block in issue_deprecations')
Notice: Signed certificate request for XXXyyy.lan
Notice: Removing file Puppet::SSL::CertificateRequest XXXyyy.lan at '/var/lib/puppet/ssl/ca/requests/XXXyyy.lan.pem'

Now check the /tmp, we will see our test file in last chapter.

For test pupose, we will disable archLinux’s puppet via:

[root@TrustyArch tmp]# systemctl stop puppet.service
[root@TrustyArch tmp]# systemctl disable puppet.service
Removed symlink /etc/systemd/system/multi-user.target.wants/puppet.service.

Install package

Add following lines into 10.0.0.88, /etc/puppet/manifests/site.pp:

package {
    'xplot':
        ensure => installed
}

Then restart the puppetmaster, in 10.0.0.89, the package xplot will be installed.

Virtualbox Modprobe Problem

Aug 13, 2014
Technology

After upgrading the Linux Kernel, my virtualbox cannot automatically load the kernel modules for virtualbox. Following is the steps for finding out the problems and solving them.

Locating Problem

I could manually modprobe the virtualbox driver, but failed to load at boot, so I first check the status of the systemd’s output.
Checking the systemd’s modules load service status:

# systemctl status systemd-modules-load.service
● systemd-modules-load.service - Load Kernel Modules
   Loaded: loaded (/usr/lib/systemd/system/systemd-modules-load.service; static)
   Active: failed (Result: exit-code) since Wed 2014-08-13 13:32:34 CST; 1h 24min ago
     Docs: man:systemd-modules-load.service(8)
           man:modules-load.d(5)
  Process: 142 ExecStart=/usr/lib/systemd/systemd-modules-load (code=exited, status=1/FAILURE)
 Main PID: 142 (code=exited, status=1/FAILURE)

Aug 13 13:32:34 XXXyyy systemd[1]: systemd-modules-load.service: main process exited, code=exited, status=1/FAILURE
Aug 13 13:32:34 XXXyyy systemd[1]: Failed to start Load Kernel Modules.
Aug 13 13:32:34 XXXyyy systemd[1]: Unit systemd-modules-load.service entered failed state.
Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable.

Manually reload this service and check the status:

[root@XXXyyy Trusty]# systemctl restart systemd-modules-load
Job for systemd-modules-load.service failed. See 'systemctl status systemd-modules-load.service' and 'journalctl -xn' for details.
[root@XXXyyy Trusty]# systemctl status systemd-modules-load
● systemd-modules-load.service - Load Kernel Modules
   Loaded: loaded (/usr/lib/systemd/system/systemd-modules-load.service; static)
   Active: failed (Result: exit-code) since Wed 2014-08-13 14:59:31 CST; 13s ago
     Docs: man:systemd-modules-load.service(8)
           man:modules-load.d(5)
  Process: 21364 ExecStart=/usr/lib/systemd/systemd-modules-load (code=exited, status=1/FAILURE)
 Main PID: 21364 (code=exited, status=1/FAILURE)

Aug 13 14:59:31 XXXyyy systemd[1]: systemd-modules-load.service: main process exited, code=exited, status=1/FAILURE
Aug 13 14:59:31 XXXyyy systemd[1]: Failed to start Load Kernel Modules.
Aug 13 14:59:31 XXXyyy systemd[1]: Unit systemd-modules-load.service entered failed state.

Use journalctl to view the PID’s logs:

[root@XXXyyy Trusty]# journalctl -b _PID=21364
-- Logs begin at Thu 2014-07-31 16:07:13 CST, end at Wed 2014-08-13 15:00:02 CST. --
Aug 13 14:59:31 XXXyyy systemd-modules-load[21364]: Failed to find module 'vboxdrv vboxnetflt vboxnetadp'
[root@XXXyyy Trusty]# systemctl status dkms.service 
● dkms.service - Dynamic Kernel Modules System
   Loaded: loaded (/usr/lib/systemd/system/dkms.service; disabled)
   Active: inactive (dead)

So the problem is quite clear: Failed to find module, and dkms service is not enabled.

Solving Problem

First enable the dkms.service via:

# systemctl enable dkms.service
Created symlink from /etc/systemd/system/multi-user.target.wants/dkms.service to /usr/lib/systemd/system/dkms.service.

Install vboxhost-hook, this will add the hook to compile the virtualbox host modules:

# yaourt -S vboxhost-hook

Add vboxhost into the /etc/mkinitcpio.conf:

HOOKS="base udev autodetect modconf block filesystems keyboard fsck vboxhost"

Now recompile the initramfs via:

mkinitcpio -p linux

dkms should also be installed:

pacman -S linux-headers virtualbox-host-dkms viftualbox-guest-dkms
dkms install vboxhost/4.3.14
dkms install vboxguest/4.3.14

Finally I found the reason:

# cat /etc/modules-load.d/virtualbox.conf
# Load virtualbox related modules at startup
vboxdrv
vboxnetflt
vboxnetadp

But previously I let them in one line!!!!!!!!!!!!!OMG…….
Reboot and examine the result via lsmod | grep vbox.

Puppet On ArchLinux

Aug 11, 2014
Technology

Installation

Install via;

sudo pacman -S puppet

Configurate this machine into server mode.

Install new Virtual Machine

Install a new ubuntu14.04 using qemu, and install puppet in it.
Generate the configuration file for mirror.list of Ubuntu.

Finally use the vdi file in the Ubuntu.

Install puppet in Ubuntu14.04:
http://linuxconfig.org/puppet-installation-on-linux-ubuntu-14-04-trusty-tahr

Make Ubuntu use a fixed IP.

$ cat  /etc/network/interface
# s file describes the network interfaces available on your system
# and how to activate them. For more information, see interfaces(5).

# The loopback network interface
auto lo
iface lo inet loopback
auto eth0
	iface eth0 inet static
	address 10.0.0.88
	netmask 255.255.255.0
	gateway 10.0.0.1

And Copy the virtual disk, and change the UUID of the disk:

$ VBoxManage internalcommands sethduuid ./Ubuntu.vdi 
UUID changed to: d1xxxxxxxxxxxxxxxxxxxxxxxxxx

Be sure to change the ip address to 10.0.0.89.

Now we have 2 machines.

No password enter for ssh login:

$ cat ~/.ssh/id_rsa.pub| ssh Trusty@10.0.0.88 'cat>>~/.ssh/authorized_keys'
$ cat ~/.ssh/id_rsa.pub| ssh Trusty@10.0.0.89 'cat>>~/.ssh/authorized_keys'

Server and Client

Install Server side in 10.0.0.88:

sudo apt-get install puppetmaster

In 10.0.0.88, edit /etc/hosts:

10.0.0.89	client

While in 10.0.0.89, edit /etc/hosts:

10.0.0.88	puppet

In client(10.0.0.89), start the service of puppet:

$ sudo service puppet start
 * Starting puppet agent                                                                                
puppet not configured to start, please edit /etc/default/puppet to enable
                                                                                                 [ OK ]

In server(10.0.0.88), start the service of puppet master:
Add following lines in to /etc/puppet/puppet.conf:

dns_alt_names = puppet, master.local, puppet.terokarvinen.com

Then remove all of the generated ssl :

rm -rf /var/lib/puppet/ssl

Now restart the puppetmaster via:

# service puppetmaster restart

Change the hostname of 10.0.0.88 to Ubuntu88, 10.0.0.89 to Ubuntu89, and then restart the computer.
Now change the Ubuntu88’s configuration Add following lines in 10.0.0.88(Server): In /etc/puppet/puppet.conf, [master] heading:

dns_alt_names = puppet, master.local, puppet.terokarvinen.com

On 10.0.0.89(Client), change the following line in /etc/default/puppet:

START=yes

Then in /etc/puppet/puppet.conf, add following:

[agent]
server = puppet

Restart the puppet service.

Now on server, use following command to list the cert and add signed cert.

Trusty@Ubuntu88:~$ sudo puppet cert --list
sudo: unable to resolve host Ubuntu88
Warning: Setting templatedir is deprecated. See http://links.puppetlabs.com/env-settings-deprecations
   (at /usr/lib/ruby/vendor_ruby/puppet/settings.rb:1095:in `block in issue_deprecations')
  "ubuntu89" (SHA256) xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Trusty@Ubuntu88:~$ sudo puppet cert --sign ubuntu89
sudo: unable to resolve host Ubuntu88
Warning: Setting templatedir is deprecated. See http://links.puppetlabs.com/env-settings-deprecations
   (at /usr/lib/ruby/vendor_ruby/puppet/settings.rb:1095:in `block in issue_deprecations')
Notice: Signed certificate request for ubuntu89
Notice: Removing file Puppet::SSL::CertificateRequest ubuntu89 at '/var/lib/puppet/ssl/ca/requests/ubuntu89.pem'

Create the Site Manifest and a Module

Go to /etc/puppet, run following command:

Trusty@Ubuntu88:/etc/puppet$ sudo mkdir -p manifests/ modules/helloworld/manifests

Edit following file:

Trusty@Ubuntu88:/etc/puppet$ cat manifests/site.pp 
include helloworld

Create the file:

Trusty@Ubuntu88:/etc/puppet$ sudo cat modules/helloworld/manifests/init.pp
class helloworld {
        file { '/tmp/helloFromMaster':
                content => "See you at http://terokarvinen.com/tag/puppet\n"
        }
}

And Now in client, restart the puppet service:

Trusty@Ubuntu89:~$ sudo service puppet restart
sudo: unable to resolve host Ubuntu89
[sudo] password for Trusty: 
no talloc stackframe at ../source3/param/loadparm.c:4864, leaking memory
 * Restarting puppet agent                                               [ OK ] 
Trusty@Ubuntu89:~$ cat /tmp/helloFromMaster 
See you at http://terokarvinen.com/tag/puppet

Now the basic configuration is OK.

很惭愧，就做了一点微小的工作