Skip to main content

High availability planner

Enter the addresses of the HA interface and of any additional LAN interface or virtual IP, and the planner gives you the ns-ha-config commands to set up the cluster. See Setup and management for how the cluster works.

Configure

HA interface

One per cluster, this is where VRRP runs. Three addresses in the same network: both nodes and the virtual IP that follows the active one.

Additional LAN interfaces

Optional. DMZ, guest or other non-WAN networks of the cluster, with the same three-address logic.

Extra virtual IPs

Optional. One more virtual IP on an interface already in the cluster. Not needed on WANs: their aliases are replicated automatically.

Commands

Run them on the primary node, from the console or over SSH.

# 1. Check the requirements, from the primary node
# first: static IPs set on both nodes, cables connected on every cluster interface
ns-ha-config check-primary-node lan
# the root password of the secondary node is asked and not shown
ns-ha-config check-backup-node 192.168.100.239 lan
# 2. Initialize the cluster, again from the primary node
ns-ha-config init-primary-node 192.168.100.238 192.168.100.239 192.168.100.240/24 lan
ns-ha-config init-backup-node lan
# 3. Sync the secondary node right away
/usr/libexec/ns-rsync.sh
# 4. Check the result
# Last Sync Status must be Successful
ns-ha-config status
After the initialization
  • Wait at least 10 minutes before testing a failover: the configuration syncs at that interval.
  • Configure the WAN interfaces only on the primary node, from the web interface: they are replicated automatically.
  • Clients must use the virtual IP as gateway and DNS, never a single node's address.