Ping Nodes

A ping node is a pseudo-member of a failover. It is located outside the failover and serves to answer ICMP requests from failover members (this means the ping node cannot be one of the nodes). It is there to make sure the outside connection for the failover members remains online.

For instance, if the ICMP request from node_a to the ping node fails, a similar request is performed by remote node_b. If this one succeeds, resources are moved from node_a to node_b.

Automatic failover will only take place if all ping nodes become unavailable for one of the nodes (e.g. the server does not take action when at least one ping node is reachable). The ping node needs only to be able to receive and reply to ICMP echo requests.

add new ping node

Pressing this button will expand the window and show an empty area to enter an IP address of a ping node. Pressing the button for the second time will add the IP address to the ping node list.

When a new IP address of a ping node is entered, the status for the ping nodes is automatically being verified on both nodes. As a result, the status is shown for both the local node and for the remote node.

trash bin icon ("delete" button)

Expands the window and shows a key to remove the particular ping node from the list.

cancel

Is used to hide the expanded view.

Note

You can indicate up to 10 ping nodes but only one ping node can be entered at once. 

Important: Note that when no ping nodes are available then failover resources will not be served to clients (they will be exported on both nodes). Once at least one ping node is reachable then resources will be available again.

Important: For data safety it’s highly recommended to set ping nodes (at least 2) on interfaces used by rings.