G6 platform and Foundation

Badge +1
Hello guys,

I wanted to ask that we are planning to create a cluster using 1065-G6 servers.

Since the G6 platform has two Copper 10GE ports, I was wondering how are we going to do the foundation and discovery of the nodes when you don't have a 10GbE copper switch. Do the ports support a 1 G speed?

In addition we have two 10G SFP+ ports in the servers as well.

3 replies

Userlevel 3
Badge +6
Hi @Muhammad Usman Ayub

as for foundation , you can create a cluster using 10GbE Copper even if the switch which NX Server connected to is only support for 1GbE.

the default setting of NX is as follows
bond: all physical ports are used , and active - Backup is adoped as default .
so , when creating cluster using Copper , you should plug out 10G SFP+ Port from server if connected .
Badge +1
thanks for the reply.

As you said, We will use the 10G copper port for foundation however for production network we have to use 10G SFP+ ports in active-backup bond. What will have to be done to migrate from copper to 10G SFP+ for 10G connectivity between the servers and for VM traffic as well.
Userlevel 3
Badge +6
after creating cluster and then stop the cluster.

1st , connect to 10G SFP from 10G copper
2nd . remove 10G copper cable form 1065
3rd . check whether ping is working
4th . set up bond for 10G SFP+

to know current bond status.
login to AHV as root
root@ahv# ovs-appctl bond/show bond0
---- bond0 ----
bond_mode: active-backup
bond may use recirculation: no, Recirc-ID : -1
bond-hash-basis: 0
updelay: 0 ms
downdelay: 0 ms
lacp_status: off
active slave mac: 0c:c4:7a:48:b2:68(eth0)

slave eth0: enabled
active slave
may_enable: true

slave eth1: disabled
may_enable: false

to know uplink interfaces
login to CVM as nutanix
nutanix@cvm$ manage_ovs --bridge_name bridge show_uplinks
Bridge: br0
Bond: br0-up
bond_mode: active-backup
interfaces: eth3 eth2 eth1 eth0 => ""default bond interfaces""
lacp: off
lacp-fallback: false
lacp_speed: slow

if you want to know which number of eth is used as copper
login to AHV and ethtool comand which allows you to know what it is.

to know bridge name
login to CVM as nutanix
nutanix@cvm$ manage_ovs show_bridges

if you want to modified interfaces of bond
login to CVM as nutanix
suppose , eth2 and 3 is 10G copper and bridge name is br0 , bond name is br0-up
nutanix@cvm$ allssh manage_ovs --bridge_name br0 --bond_name br0-up --interfaces eth3,eth2 update_uplinks

5th . ping check among cvms and hypervisor
6th . cluster start