
The administrator must make a judgment about the traffic being spread across the aggregated
physical interfaces and choose one of the following criteria for the distribution:
•
DestinationMAC
•
SourceIP
•
DestinationIP
•
SourcePort
•
DestinationPort
•
IP and Ports (the default)
Choosing the Distribution Method
The algorithm that spreads the traffic between the aggregated interfaces uses hashing with the
chosen distribution method as the input. The best distribution method is therefore the one
which varies the most. For example, if the source of traffic is a number of internal clients being
NATed to the Internet via an ISP, the best choice for the distribution method is most likely
SourcePort
since this will be chosen randomly as each connection is opened by a client.
An alternative in the above scenario could be
SourceIP
but only if there is a sufficiently large
number of clients. With just a few clients,
SourceIP
might end up with only one of the aggregated
interfaces being used.
If aggregation is being done for a protected web server receiving external requests from remote
clients over the public Internet, the
DestinationIP
would not be suitable since all connections
would have the server's address. Instead, the more variable
SourceIP
would be a better choice for
the distribution method.
The hashing process to choose the physical Ethernet interface to use takes place each time a new
connection is opened. This means that all packets for a given connection will be sent on the
same physical interface. The chosen interface for the connection would then only subsequently
change if the chosen mode was dynamic and the connection fails.
The Default IP and Ports Distribution Method
The default distribution method is
IP and Ports
and this takes into account both the source and
destination IP address as well as the source and destination port number. It is designed to be a
general catch-all solution where the traffic type is known to be variable or where the
administrator is uncertain which of the more specific distribution is suitable.
Physical Switch Connections
The physical cable links between the firewall and the external switch can be made either before
or after creating the
LinkAggregation
object and activating the changed configuration.
NetDefendOS will try to send data on the aggregated interfaces as soon as the configuration
changes become active.
However, it is recommended that the physical cabling is in place before the
LinkAggregation
object is activated and saved. This will provide the behavior which is expected from the feature
and is particularly relevant if negotiated aggregation (LACP) is used.
Setup with High Availability
When using link aggregation with HA, the connections from the Ethernet ports on each firewall
in the HA cluster can connect to the same or different switches. However, if using the same
switch, the switch must be configured so that the connections from each firewall are kept
separate by creating two link aggregation groups in the switch.
Chapter 3: Fundamentals
194
Summary of Contents for NetDefendOS
Page 30: ...Figure 1 3 Packet Flow Schematic Part III Chapter 1 NetDefendOS Overview 30 ...
Page 32: ...Chapter 1 NetDefendOS Overview 32 ...
Page 144: ...Chapter 2 Management and Maintenance 144 ...
Page 284: ...Chapter 3 Fundamentals 284 ...
Page 392: ...Chapter 4 Routing 392 ...
Page 419: ... Host 2001 DB8 1 MAC 00 90 12 13 14 15 5 Click OK Chapter 5 DHCP Services 419 ...
Page 420: ...Chapter 5 DHCP Services 420 ...
Page 573: ...Chapter 6 Security Mechanisms 573 ...
Page 607: ...Chapter 7 Address Translation 607 ...
Page 666: ...Chapter 8 User Authentication 666 ...
Page 775: ...Chapter 9 VPN 775 ...
Page 819: ...Chapter 10 Traffic Management 819 ...
Page 842: ...Chapter 11 High Availability 842 ...
Page 866: ...Default Enabled Chapter 13 Advanced Settings 866 ...
Page 879: ...Chapter 13 Advanced Settings 879 ...