ESX 5 : how to use API over http instead of https

By default , SDK port is tcp 443 ( https://vcenter IP/sdk )
Here’s an explanation of how to enable both http and https access.
When evaluating SDK , it is useful to use http instead of https to debug.

1.
access to the vCenter server

2.

go to <Drive name>:\Documents and Settings\All Users\Application Data\VMware\VMware VirtualCenter” directory

3.
edit proxy.xml file.

from

<e id="9">      <_type>vim.ProxyService.LocalServiceSpec</_type>      <accessMode>httpsWithRedirect</accessMode>      <port>8085</port>      <serverNamespace>/sdk</serverNamespace>    </e>

to

<e id="9">      <_type>vim.ProxyService.LocalServiceSpec</_type>      <accessMode>httpAndHttps</accessMode>      <port>8085</port>      <serverNamespace>/sdk</serverNamespace>    </e>

4.
restart vCenter service

5. access to http://vCenter IP/sdk

over http

root@ubuntu1204-vm1:~# python Python 2.7.3 (default, Aug  1 2012, 05:14:39) [GCC 4.6.3] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> from pysphere import VIServer >>> server = VIServer() >>> server.connect("http://1.1.1.10/sdk","username",".password.") >>> server.get_server_type() 'VMware vCenter Server' >>> server.get_api_version() '5.1' >>> server.disconnect

over https

>>> server.connect("http://1.1.1.10/sdk","username",".password.") >>> server.get_server_type() 'VMware vCenter Server' >>> server.disconnect

pysphere : vSphere SDK for python

http://code.google.com/p/pysphere/

[ install pysphere on Ubuntu 12.04 LTS ]

# cat /etc/lsb-release DISTRIB_ID=Ubuntu DISTRIB_RELEASE=12.04 DISTRIB_CODENAME=precise DISTRIB_DESCRIPTION="Ubuntu 12.04.2 LTS" # uname -ri 3.2.0-37-virtual x86_64

the installation is very easy , just type

# easy_install -U pysphere

[ sample usage ]

connect to the ESXi not vCenter

# python Python 2.7.3 (default, Aug  1 2012, 05:14:39) [GCC 4.6.3] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> from pysphere import VIServer >>> server = VIServer() >>> server.connect("192.168.0.1","user",".Password.")

get the ESXi info

>>> print server.get_server_type() VMware ESXi >>> print server.get_api_version() 4.1

>>> vmlist = server.get_registered_vms() >>> print vmlist[0] [datastore1] CentOS6-1-32bit/CentOS6-1-32bit.vmx

retrieve VMs

>>> server.get_vm_by_name("CentOS6-1-32bit") Traceback (most recent call last):  File "<stdin>", line 1, in <module>  File "/usr/local/lib/python2.7/dist-packages/pysphere-0.1.7-py2.7.egg/pysphere/vi_server.py", line 323, in get_vm_by_name    FaultTypes.OBJECT_NOT_FOUND) pysphere.resources.vi_exception.VIException: [Object Not Found]: Could not find a VM named 'CentOS6-1-32bit'

if you see the above error , connect to the ESXi again.

>>> server.connect("192.168.0.1","user",".Password.") >>> server.get_vm_by_path("[datastore1] CentOS6-1-32bit/CentOS6-1-32bit.vmx") <pysphere.vi_virtual_machine.VIVirtualMachine instance at 0x3cffd88> >>> vm1 = server.get_vm_by_path("[datastore1] CentOS6-1-32bit/CentOS6-1-32bit.vm x")

>>> vmlist[1] '[datastore1] CentOS6-1-64bit/CentOS6-1-64bit.vmx' >>> vm2 = server.get_vm_by_name("CentOS6-1-64bit")

check the status of the VM

>>> print vm1.get_status(basic_status=True) POWERED OFF >>> print vm1.is_powering_on() False

>>> vm1.get_properties() {'files': {0: {'type': 'diskDescriptor', 'name': '[datastore1] CentOS6-1-32bit/CentOS6-1-32bit.vmdk', 'key': 0, 'size': 499}, 1: {'type': 'diskExtent', 'name': '[datastore1] CentOS6-1-32bit/CentOS6-1-32bit-flat.vmdk', 'key': 1, 'size': 10737418240}, 2: {'type': 'log', 'name': '[datastore1] CentOS6-1-32bit/vmware-1.log' <snip> >>> print vm1.get_property('guest_id') rhel6Guest

power on the VM

>>> vm1.get_status() 'POWERED OFF' >>> vm1.power_on()

>>> vm1.get_status() 'POWERED ON'

suspend / shutdown the VM

>>> vm1.get_status() 'POWERED ON' >>> vm1.suspend() >>> vm1.get_status() 'SUSPENDED' >>> vm1.power_on() >>> vm1.get_status() 'POWERED ON' >>> vm1.power_off() >>> vm1.get_status() 'POWERED OFF'

disconnect

>>> server.disconnect()

How to have Solaris 10 x86 VM running within ESXi recognize vNICs

I’ve installed Solaris 10 on ESXi as a virtual machine.

# uname –a SunOS unknown 5.10 Generic_147441-01 i86pc i386 i86pc

After finishing the installation , I added two additional vNICs to the VM.
However Solaris VM does not seem to recognize vNICs I added to the VM.

The VM has totally three vNICs ( driver is e1000 )
e1000g0 is the vNIC I added when installing Solaris.

# ifconfig  e1000g0 e1000g0: flags=1004843<UP,BROADCAST,RUNNING,MULTICAST,DHCP,IPv4> mtu 1500 index 2        inet 192.168.11.215 netmask ffffff00 broadcast 192.168.11.255        ether aa:aa

There are no two vNICs I added to the VM after the installation

bash-3.2# ifconfig  e1000g1 ifconfig: status: SIOCGLIFFLAGS: e1000g1: no such interface bash-3.2# ifconfig  e1000g2 ifconfig: status: SIOCGLIFFLAGS: e1000g2: no such interface

dmesg

bash-3.2# dmesg | grep e1000g | grep register Dec 26 22:37:06 unknown mac: [ID 469746 kern.info] NOTICE: e1000g0 registered Dec 26 22:37:37 unknown mac: [ID 469746 kern.info] NOTICE: e1000g1 registered Dec 26 22:37:37 unknown mac: [ID 469746 kern.info] NOTICE: e1000g2 registered Dec 26 22:37:59 unknown mac: [ID 736570 kern.info] NOTICE: e1000g1 unregistered Dec 26 22:37:59 unknown mac: [ID 736570 kern.info] NOTICE: e1000g2 unregistered

check “/etc/path_to_inst” file

bash-3.2# egrep e1000 /etc/path_to_inst "/pci@0,0/pci15ad,790@11/pci15ad,750@0" 0 "e1000g" "/pci@0,0/pci15ad,790@11/pci15ad,750@2" 1 "e1000g" "/pci@0,0/pci15ad,790@11/pci15ad,750@3" 2 "e1000g"

Seen from the logs , the VM seems to recognize vNICs at the hardware level.

enable e1000g1 and e1000g2 with ifconfig.

bash-3.2# ifconfig e1000g1 plumb bash-3.2# ifconfig e1000g2 plumb

The VM has recognized all vNICs

bash-3.2# ifconfig e1000g1 e1000g1: flags=1000842<BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 3        inet 0.0.0.0 netmask 0        ether zz:zz bash-3.2# ifconfig e1000g2 e1000g2: flags=1000842<BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 4        inet 0.0.0.0 netmask 0        ether zz:zz

Seen from the man of “ifconfig” , it describes :

plumb         Open the device associated with the  physical  interface         name  and  set  up  the streams needed for IP to use the         device. When used with a logical  interface  name,  this         command  is  used  to  create  a  specific named logical         interface. An interface must be separately  plumbed  for         use  by IPv4 and IPv6. The address_family parameter con-         trols whether the ifconfig command applies  to  IPv4  or         IPv6.         Before an interface has been plumbed, the interface will         not show up in the output of the ifconfig -a command.

ESXi : install CentOS automatically with kickstart

[ prepare kickstart file ]

I made a kickstart file using system-config-kickstart.

install system-config-kickstart via yum.

# yum install –y system-config-kickstart.noarch

make a kickstart file with system-config-kickstart

# system-config-kickstart

GUI will appear like this:

configure installation settings step by step with system-config-kickstart.

Here’s an example ks file of CentOS5.6 which I created with system-config-kickstart.

# cat centos5-6.ks.cfg #platform=x86, AMD64, or Intel EM64T # System authorization information auth  --useshadow  --enablemd5 # System bootloader configuration bootloader --append="console=ttyS0,115200n8 console=tty0" --location=mbr # Clear the Master Boot Record zerombr # Partition clearing information clearpart --all --initlabel # Use text mode install text # Firewall configuration firewall --disabled # Run the Setup Agent on first boot firstboot --disable # System keyboard keyboard jp106 # System language lang ja_JP # Installation logging level logging --level=info # Use CDROM installation media cdrom # Network information network --bootproto=static --device=eth0 --gateway=192.168.11.254 --ip=192.168.11.160 --nameserver=192.168.11.15 --netmask=255.255.255.0 --onboot=on # Reboot after installation reboot #Root password rootpw --iscrypted $1$3QEuw1wV$VUKWhr1WZ1SAUXNVFv7Jk. # SELinux configuration selinux --disabled # System timezone timezone --isUtc Asia/Tokyo # Install OS instead of upgrade install # X Window System configuration information xconfig  --defaultdesktop=GNOME --depth=8 --resolution=640x480 # Disk partitioning information part / --bytes-per-inode=4096 --fstype="ext3" --size=5000 part swap --bytes-per-inode=4096 --fstype="swap" --size=512 %packages @development-libs @development-tools @editors

[ make an ISO image including a ks file ]

mount original CentOS ISO.

# mount -t iso9660 -o loop CentOS-5.6-x86_64-bin-DVD-1of2.iso /media/cdrom/

copy ISO files to an another directory.( /root/CentOS.5.6_Custom )

# cd /media/cdrom/ # mkdir /root/CentOS5.6_Custom # pwd /media/cdrom # tar cf - . | (cd /root/CentOS5.6_Custom; tar xfp -)

copy ks file ( name a ks file you made “ks.cfg” )

# cp /root/centos5-6.ks.cfg /root/CentOS5.6_Custom/ks.cfg

build a customized ISO.

# pwd /root/CentOS5.6_Custom # mkisofs -o /root/CentOS5.6_x86_64_custom.iso -b isolinux/isolinux.bin -c isolinux/boot.cat -no-emul-boot -boot-load-size 4 -boot-info-table -J -R -V "CetnOS 5.6 Custom ISO" # file /root/CentOS5.6_x86_64_custom.iso /root/CentOS5.6_x86_64_custom.iso: # ISO 9660 CD-ROM filesystem data 'CDROM                          ' (bootable)

confirm you can mount customized ISO and confirm there’s ks.cfg into that ISO.

# mount -t iso9660 -o loop CentOS5.6_x86_64_custom.iso /root/aaa

# head -5 /root/aaa/ks.cfg #platform=x86, AMD64, or Intel EM64T # System authorization information auth  --useshadow  --enablemd5 # System bootloader configuration bootloader --append="console=ttyS0,115200n8 console=tty0" --location=mbr

[ install CentOS as a VM using customized ISO on ESXi ]

start a VM from CentOS5.6_x86_64_custom.iso and then specify “linux ks=cdrom:/ks.cfg” as below.

installation has been processed automatically.

With this ks file , after finishing the installation , rebooting system automatically.

I could install CentOS automatically :)