Once you have added your hosts to vSphere it’s time to connect all the hosts to a centralized networking configuration so we can move the guests between hosts for maintenance and DRS. These are the basics- for more detailed information see the vSphere vMotion Networking Requirements documentation from VMware

Step 1: Create a Distributed Switch for vMotion Traffic

  1. Navigate to Networking
  2. Click the Actions Dropdown
  3. Select Distributed Switch
  4. Chose New Distributed Switch

Step 2: Configure the Distributed Switch

Name the Distributed Switch something that will let you know what it is and what it does. [“ds – for distributed switch and vMotion for the service]

Click NEXT

Select the version that will support the oldest version of ESXi in the environment.

Click NEXT

  1. Chose the number of Physical Uplinks [1 for vMotion is plenty]
  2. Enable Network I/O Control (why)
  3. Select Create s default port group
  4. Name Name the Port [pg-XXXXX]

Click NEXT

Review your settings

Click FINISH

Step 3: Edit the Properties of the Distributed Switch

  • Right-Click the new DS
  • Select Settings
  • Click Edit Settings

From the Distributed Switch – Edit Settings window-

  • Select the Advanced tab
  • Increase the MTU as high as the physical network will permit. [My USB NIC devices are capped at 4000]
  • Click OK

[ Read more about Jumbo Frames ]

Step 4: Edit the Properties of the Port Group

  • Right-Click the new PG
  • Click Edit Settings

Under General change the Port Binding to Ephemeral – no binding [This will help later when we install NSX-T]

Click OK

Step 5: Connect the Hosts to the Distributed Switch

From Networking, Right-Click on the vMotion DS and click Add and Manage Hosts…

Select the Add hosts radio button and click NEXT

Click New hosts…

Select the hosts to be added to this DS [generally all the hosts in that cluster]

Click OK

On the next screen click NEXT

Step 6: Assign Uplinks

Select the available Physical Adapter you want to use on each host to use for Uplink(s) for this DS.

Click the Assign uplink button

Select the available Uplink you want to assign to this Physical Adapter

If all your host are of the same configuration, you can click the Apply this uplink assignment to the rest of the hosts and it will automatically configure the other hosts connected to this DS similarly.

If you selected Apply this uplink assignment to the rest of the hosts you will see that all hosts have been configured to connect the same adapter to the DS using the same uplink

Verify your configuration and click NEXT

Click NEXT on the Manage VMkernel adapters screen [we will configure those later]

Click NEXT on the Migrate VM Networking screen [there should be no networking to migrate for vMotion]

Click FINISH

Verify all the Hosts are connected to the DS by selecting the DC and clicking the Hosts tab.

All desired hosts should be present.

Step 7: Add VMkernal Adapters

From Networking, Right-Click on the vMotion PG and click Add VMkernel Adapters…

On the Select member Hosts window, click Attached hosts… and select all the hosts attached to the PG.

Click OK

Verify all the hosts and click NEXT

On the Configure VMkernel adapter window, select vMotion from the TCP/IP stack drop down.

Click NEXT

NOTE: vMotion should use non-routable IP space for isolation.

Enter the IP information for the vMotion interface on each host. There is an auto fill feature to make this process faster in large clusters.

Set the Gateway Configuration Type to Do not configure

Click NEXT

Verify the information is correct

Click FINISH

Select the vMotion PG and click on the Ports tab.

You should see each host connected to the PG and the status should be UP

From Hosts and Clusters, the Networking / VMkernel adapters for each host should show the same vmk{#} with a corresponding IP and vMotion enabled.

Step 8: Test VMK Connectivity Between Hosts

  • Enable SSH on he hosts

vmkping -I vmk{#} -d -s {MTU-28} -S vmotion {vmk ipaddress of other host}

VMKPING RESULTS FOR HOST 1 TO HOSTS 2 & 3

[root@ESX-01:~] vmkping -I vmk1 -d -s 3972 -S vmotion 172.16.0.102
PING 172.16.0.102 (172.16.0.102): 3972 data bytes
3980 bytes from 172.16.0.102: icmp_seq=0 ttl=64 time=0.737 ms
3980 bytes from 172.16.0.102: icmp_seq=1 ttl=64 time=0.649 ms
3980 bytes from 172.16.0.102: icmp_seq=2 ttl=64 time=0.670 ms

--- 172.16.0.102 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max = 0.649/0.685/0.737 ms

[root@ESX-01:~] vmkping -I vmk1 -d -s 3972 -S vmotion 172.16.0.103
PING 172.16.0.103 (172.16.0.103): 3972 data bytes
3980 bytes from 172.16.0.103: icmp_seq=0 ttl=64 time=1.184 ms
3980 bytes from 172.16.0.103: icmp_seq=1 ttl=64 time=0.872 ms
3980 bytes from 172.16.0.103: icmp_seq=2 ttl=64 time=0.699 ms

--- 172.16.0.103 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max = 0.699/0.918/1.184 ms

VMKPING RESULTS FOR HOST 2 TO HOSTS 1 & 3

[root@ESX-02:~] vmkping -I vmk1 -d -s 3972 -S vmotion 172.16.0.101
PING 172.16.0.101 (172.16.0.101): 3972 data bytes
3980 bytes from 172.16.0.101: icmp_seq=0 ttl=64 time=0.654 ms
3980 bytes from 172.16.0.101: icmp_seq=1 ttl=64 time=0.614 ms
3980 bytes from 172.16.0.101: icmp_seq=2 ttl=64 time=0.666 ms

--- 172.16.0.101 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max = 0.614/0.645/0.666 ms

[root@ESX-02:~] vmkping -I vmk1 -d -s 3972 -S vmotion 172.16.0.103
PING 172.16.0.103 (172.16.0.103): 3972 data bytes
3980 bytes from 172.16.0.103: icmp_seq=0 ttl=64 time=1.052 ms
3980 bytes from 172.16.0.103: icmp_seq=1 ttl=64 time=0.810 ms
3980 bytes from 172.16.0.103: icmp_seq=2 ttl=64 time=0.699 ms

--- 172.16.0.103 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max = 0.699/0.854/1.052 ms
  • Continue this process for each of the hosts.

Categories:

BasicsESXiMiscellaneousvSphere

Tags:

AppleESXMacusbVMw