I have an active 6 node cluster with 2x 10G fibre NICs in a simple active-backup bond0 (see config below), and want to change the bond to LACP for VMs, External NFS storage and cluster migration. I would then use the copper 1GB NICs for management and corosync.
I need to use DHCP everywhere, with no static IP Addresses.
The intention is to try and make these changes with zero downtime, migrating VMs off one node, reconfiguring that node, then doing the rest of the nodes incrementally.
Current /etc/network/interfaces
After setting up the LAG on the switch for one of the nodes, I've tried the following config with vbr1 selected on some trial VMs:
If vmbr1 is also dhcp, how do I assign it a separate hostname? If I leave it as above, there will be 2 IP Addresses assigned to the hostname.
Do I just leave it as "iface vmbr1 inet manual" with no address set? - I've tried this and it looks to work, but am unsure its correct.
I've searched this forum and found similar posts, but most examples seem to use static addresses.
Also, when I have a solution for the cluster network, as I migrate running VMs onto this server, can I change them from vbr0 to vbr1 without having to reboot them, or will that depend on what OS they are running and need manual network restart from within each VM?
I need to use DHCP everywhere, with no static IP Addresses.
The intention is to try and make these changes with zero downtime, migrating VMs off one node, reconfiguring that node, then doing the rest of the nodes incrementally.
Current /etc/network/interfaces
Code:
auto lo
iface lo inet loopback
auto ens2f0
iface ens2f0 inet manual
auto ens2f1
iface ens2f1 inet manual
iface eno1 inet manual
iface eno2 inet manual
auto bond0
iface bond0 inet manual
bond-slaves ens2f0 ens2f1
bond-miimon 100
bond-mode active-backup
bond-primary ens2f0
bond-minion 100
bond-primary-reselect always
auto vmbr0
iface vmbr0 inet dhcp
bridge-ports bond0
bridge-stp off
bridge-fd 0
hwaddress 2c:44:fd:97:da:78After setting up the LAG on the switch for one of the nodes, I've tried the following config with vbr1 selected on some trial VMs:
Code:
auto lo
iface lo inet loopback
auto ens2f0
iface ens2f0 inet manual
auto ens2f1
iface ens2f1 inet manual
auto eno1
iface eno1 inet manual
auto eno2
iface eno2 inet manual
auto bond0
iface bond0 inet manual
bond-slaves eno1 eno2
bond-mode active-backup
bond-primary eno1
bond-minion 100
bond-primary-reselect always
auto vmbr0
iface vmbr0 inet dhcp
bridge-ports bond0
bridge-stp off
bridge-fd 0
hwaddress 2c:44:fd:97:da:78
auto bond1
iface bond1 inet manual
bond-mode 802.3ad
bond-miimon 100
bond-lacp-rate fast
bond-xmit-hash-policy layer2
bond-slaves ens2f0 ens2f1
auto vmbr1
iface vmbr1 inet dhcp
bridge-ports bond1
bridge-stp off
bridge-fd 0
hwaddress 2c:44:fd:97:da:7cIf vmbr1 is also dhcp, how do I assign it a separate hostname? If I leave it as above, there will be 2 IP Addresses assigned to the hostname.
Do I just leave it as "iface vmbr1 inet manual" with no address set? - I've tried this and it looks to work, but am unsure its correct.
I've searched this forum and found similar posts, but most examples seem to use static addresses.
Also, when I have a solution for the cluster network, as I migrate running VMs onto this server, can I change them from vbr0 to vbr1 without having to reboot them, or will that depend on what OS they are running and need manual network restart from within each VM?