User manual

21/1551-ASP 113 01 Uen N2 2014-01-28 79

OLUTION ARCHITECTURE

other one will continue to serve the operations and the Tele-

phony Servers will be available on the functioning interface.

Ethernet bonding is only supported in the Telephony Server

Link Failover in the Media Gateway MGU

A proprietary active/backup (link failover) network redundancy

technique is used with the MGU in MX-ONE 4.1 SP1 and later

releases.

This means that the MGU's two Ethernet ports are each

connected to a different switch, where one interface is active

and the other interface is in passive mode. The two physical

interfaces share the same IP and MAC addresses (for both

media and signaling). If the active interface fails, the other one

will become active and take over the traffic (both signalling

and media) and the MGU will continue to operate normally.

Although the mechanism used by MGU differs from that of

Telephony Server, they provide similar functionality.

LSU-E and EMG Gateways

Older Media gateways with only one Ethernet port (EMG or

LSU-E) can be connected via a local switch. By using a local

switch, which would be connected to 2 different switches on

the same network, most capabilities in the network would be

restored if there were network faults. The impact on system

behavior depends on the time it takes for the network to

recover. Depending on type of network disturbance with a

Media Gateway Classic with LSU-E configuration, the

capacity for gateway calls could be reduced. A local switch

must be used to secure control of the hardware when a LAN

switch fails.

• Server redundancy

Server redundancy is achieved by adding one or more standby

servers to the network with the ability to take over any failing Tele-

phony Server in the cluster. At failover, the standby server will take

over the identity of the failing server and the control of the media

gateways in the failing server. In case of a server failure ongoing

direct media calls will remain connected but TDM and gateway calls

will be disconnected. All new traffic will be redirected to the standby

server.

In a distributed system connected over limited bandwidth (WAN),

each remote domain must have its own standby server.

A prerequisite for server redundancy is that network redundancy is

implemented to avoid any single point of failure.

• HLR redundancy