802.1Ag; Ethernet Cfm - Dell S4820T Configuration Manual

802.1ag

Ethernet operations, administration, and maintenance (OAM) are a set of tools used to install, monitor,
troubleshoot, and manage Ethernet infrastructure deployments. Ethernet OAM consists of three main
areas:
• Service layer OAM — IEEE 802.1ag connectivity fault management (CFM)
• Link layer OAM — IEEE 802.3ah OAM
• Ethernet local management Interface — (MEF-16 E-LMI)

Ethernet CFM

Ethernet CFM is an end-to-end per-service-instance Ethernet OAM scheme which enables: proactive
connectivity monitoring, fault verification, and fault isolation.
The service-instance with regard to OAM for Metro/Carrier Ethernet is a virtual local area network (VLAN).
This service is sold to an end-customer by a network service provider. Typically the service provider
contracts with multiple network operators to provide end-to-end service between customers. For end-
to-end service between customer switches, connectivity must be present across the service provider
through multiple network operators.
Layer 2 Ethernet networks usually cannot be managed with IP tools such as internet control message
protocol (ICMP) Ping and IP Traceroute. Traditional IP tools often fail because:
• there are complex interactions between various Layer 2 and Layer 3 protocols such as spanning tree
protocol (STP), link aggregation group (LAG), virtual router redundancy protocol (VRRP), and
electronic commerce messaging protocol (ECMP) configurations.
• ping and traceroute are not designed to verify data connectivity in the network and within each node
in the network (such as in the switching fabric and hardware forwarding tables).
• when networks are built from different operational domains, access controls impose restrictions that
cannot be overcome at the IP level, resulting in poor fault visibility. There is a need for hierarchical
domains that can be monitored and maintained independently by each provider or operator.
• routing protocols choose a subset of the total network topology for forwarding, making it hard to
detect faults in links and nodes that are not included in the active routing topology. This is made more
complex when using some form of traffic engineering (TE) based routing.
• network and element discovery and cataloging is not clearly defined using IP troubleshooting tools.
There is a need for Layer 2 equivalents to manage and troubleshoot native Layer 2 Ethernet networks.
With these tools, you can identify, isolate, and repair faults quickly and easily, which reduces operational
cost of running the network. OAM also increases availability and reduces mean time to recovery, which
allows for tighter service level agreements, resulting in increased revenue for the service provider.
86
5
802.1ag