Web Management Guide-R05

Chapter 13 | Basic Administration Protocols
Connectivity Fault Management
– 479 –
Maintenance domains are designed to provide a transparent method of
verifying and resolving connectivity problems for end-to-end connections. By
default, these connections run between the domain service access points
(DSAPs) within each MA defined for a domain, and are manually configured
(see "Configuring Maintenance End Points").
In contrast, MIPs are interconnection points that make up all possible paths
between the DSAPs within an MA. MIPs are automatically generated by the
CFM protocol when the MIP Creation Type is set to “Default” or “Explicit,” and
the MIP creation state machine is invoked (as defined in IEEE 802.1ag). The
default option allows MIPs to be created for all interconnection points within an
MA, regardless of the domain’s level in the maintenance hierarchy (e.g.,
customer, provider, or operator). While the explicit option only generates MIPs
within an MA if its associated domain is not at the bottom of the maintenance
hierarchy. This option is used to hide the structure of network at the lowest
domain level.
The diagnostic functions provided by CFM can be used to detect connectivity
failures between any pair of MEPs in an MA. Using MIPs allows these failures to
be isolated to smaller segments of the network.
Allowing the CFM to generate MIPs exposes more of the network structure to
users at higher domain levels, but can speed up the process of fault detection
and recovery. This trade-off should be carefully considered when designing a
CFM maintenance structure.
Also note that while MEPs are active agents which can initiate consistency
check messages (CCMs), transmit loop back or link trace messages, and
maintain the local CCM database, MIPs, on the other hand, are passive agents
which can only validate received CFM messages, and respond to loop back and
link trace messages.
The MIP creation method defined for an MA (see "Configuring
CFM Maintenance Associations") takes precedence over the method defined
on the CFM Domain List.
Configuring Fault Notification
◆ A fault alarm can generate an SNMP notification. It is issued when the MEP fault
notification generator state machine detects that the configured time period
(MEP Fault Notify Alarm Time) has passed with one or more defects indicated,
and fault alarms are enabled at or above the specified priority level (MEP Fault
Notify Lowest Priority). The state machine transmits no further fault alarms until
it is reset by the passage of a configured time period (MEP Fault Notify Reset
Time) without a defect indication. The normal procedure upon receiving a fault
alarm is to inspect the reporting MEP’s managed objects using an appropriate
SNMP software tool, diagnose the fault, correct it, re-examine the MEP’s
managed objects to see whether the MEP fault notification generator state
machine has been reset, and repeat those steps until the fault is resolved.
◆ Only the highest priority defect currently detected is reported in the fault
alarm.