Nonstop Forwarding - Dell PowerConnect PC7024 Technical White Paper

Stacking Dell PowerConnect 7000 Series Switches

Nonstop Forwarding

A switch can be described in terms of three semi-independent functions called the forwarding plane,
the control plane, and the management plane:
•
Forwarding Plane — The set of hardware components that forward data packets
without intervention from a control CPU, sometimes called the Data Plane. The
forwarding plane is implemented in hardware.
• Control Plane — The firmware layer that manages system and hardware configuration
and runs the network control protocols in order to set system configuration and state.
The control plane determines how the forwarding plane should forward packets,
deciding which data packets are allowed to be forwarded and where they should go.
The control plane is implemented in application firmware running on the management
unit.
•
Management Plane — A set of interfaces that enable the network administrator to
configure the networking device. The management plane is implemented in application
firmware running on the management unit.
Nonstop forwarding (NSF) allows the forwarding plane of stacked units to continue to forward packets
while the control and management planes restart as a result of a power failure, hardware failure, or
firmware fault on the management unit. A nonstop forwarding failover can also be manually initiated
using the initiate failover command. Traffic flows that enter and exit the stack through physical
ports on a unit other than the management continue with virtually no interruption when the
management unit fails. To prepare the standby management unit in case of a failover, applications on
the management unit periodically checkpoint state information to the standby unit. Changes to the
running configuration are automatically copied to the standby unit. The MAC address for each switch in
the stack stays the same across a nonstop forwarding failover, so that neighbors do not have to relearn
them.
For NSF to be effective, adjacent networking devices must not reroute traffic around the restarting
device. Three techniques are used to prevent traffic from being rerouted:
1. A protocol may distribute a part of its control plane to stack units so that the protocol can give
the appearance that it is still functional during the restart. When NSF is enabled on a switch,
various protocols and configurations such as Spanning Tree and Link Access groups
automatically use this technique.
2. A protocol may enlist the cooperation of its neighbors through a technique known as graceful
restart. Graceful restart functionality can be enabled for the OSPF and OSPFv3 protocols so
that the stack can continue to forward packets using the same IPv4 and IPv6 routes while the
standby unit takes over management responsibility. For example, when OSPF executes a
graceful restart, it informs its neighbors that the OSPF control plane is restarting, but that it
will be back shortly. Helpful neighbors continue to advertise to the rest of the network that
they have full adjacencies with the restarting router, avoiding announcement of a topology
change and the network activity that would result (i.e., flooding of LSAs, SPF runs). Helpful
neighbors continue to forward packets through the restarting router. The restarting router
relearns the network topology from its helpful neighbors.
3. A protocol may simply restart after the failover if neighbors react slowly enough that they will
not normally detect the outage. The IP multicast routing protocols are a good example of this
behavior in that the PIM stack restarts before its neighbors detect its absence and drop any
adjacencies.
Page 22