Stateful Line Module Configuration Scenarios

The line module high availability functionality is based on the stateful SRP switchover design architecture, with the implementation extended to two pairs of line modules to function as the primary and secondary module. This section describes the behavior of different system functions when line module high availability is configured on the router.

High Availability Configured and Enabled on the Line Module

If line module high availability is configured and the state is active, when a fault occurs on the primary line module, the primary line module performs a warm switchover to a secondary module. After the switchover, the secondary module starts operating as the new primary module. The previously configured primary module, after it becomes operational, takes over the role of the secondary module.

When line module high availability is enabled on a router, the secondary module takes over the role of the primary module, which causes normal system services and subscriber data traffic to continue without interruption after a switchover. The main processor in an SRP module on E120 and E320 routers is referred to as the SRP module and the main processor in a line module is called the interface controller (IC).

High Availability Configured and Disabled on the Line Module

If line module high availability is configured and the state is not active, when a fault occurs on the primary line module, the primary line module performs a cold switchover to a secondary module. After the switchover, the secondary module starts operating as the new primary module. The previously configured primary module, after it becomes operational, takes over the role of the secondary module. Subscribers are disconnected and need to log in again to establish their connections again. This behavior is similar to the functionality experienced during the line module redundancy operation.

High Availability Configured and the Switchover State Is Active or Disabled

Any failure on the secondary line module or a restart of the module causes high availability to move to the disabled state. The show redundancy line-card command displays the HA status on all line modules in the system. When a line module transitions from one state to another, log messages are seen on the SRP console and SNMP traps, if enabled, are sent.

All CLI commands that cause the line module to cold restart behave in the same manner, except for the method adopted to trigger the switchover action. For example, the reload slot and slot disable commands that reload a primary line module, causing the secondary module to take over as the primary. However, the slot erase command clears the configurations on the line module that is fitted in that slot. If you specify the slot erase command to delete the configuration of the module in the selected slot before you install a different type of module on slots that contain line modules that are members of a high availability pair, an error message states that you must deactivate high availability feature for the applicable line modules before erasing or replacing the slot configuration. You need to use the no mode high-availability slot command to disable high availability for the slots in which those line modules reside.

If you enter the reload slot command, after booting up, line modules start operating in HA mode. With the slot disable command, HA is disabled until you reenable the slot.

Rebooting of the System When Line Module High Availability Is Configured

The line modules undergo a cold start, when the router is rebooted, and the secondary line module is held in a state in which it is not online. The primary line module reaches the online state. If the primary line module fails to come up online, within the specified timeout value (of less than 8.5 minutes), the secondary line module takes over as the primary module and HA remains in the disabled state.

Stateful SRP Switchover

During a stateful SRP switchover, a window of time occurs when the communication between interface controllers (IC) is disrupted owing to the switchover to new SRP module, which requires the Ethernet switch to relearn the MAC addresses and the interchassis communication (ICC) sessions to be reestablished. The system infrastructure ensures this task of relearning of details is transparent to the applications. Any notification sent by the applications on the IC-IC communication is either buffered until the communication is reestablished. Otherwise, the Ethernet switch on the standby SRP module that has become active learns the MAC addresses in standby mode without interrupting the IC-IC communication.

Line Module Redundancy

Line module redundancy and line module high availability are mutually exclusive features. You cannot configure the line modules in a redundancy group to operate in HA mode. Because both the mechanisms are mutually exclusive, if a module is a member of a redundancy group, warm switchover is not supported. Similarly, if line modules are configured in a high availability pair, they cannot be members of a redundancy group and the spare module does not take over as the primary.

Unified ISSU

Unified ISSU can continue, if the configured secondary line module takes over as the primary line module. The secondary line module is disabled during the upgrade phase of the unified ISSU operation and cold boots after the unified ISSU operation is complete. The disabled line module during unified ISSU is cold booted after the unified ISSU operation is complete. Only the primary line module can participate in a unified ISSU operation. You cannot perform stateful line module switchover when unified ISSU operation is in progress.

Simultaneous Stateful Line Module Switchover and Stateful SRP Switchover

If you configure both stateful SRP switchover and line module high availability on a router, a window of time can occur during which multiple stateful switchover operations, such as stateful line module switchover and stateful SRP switchover, can be performed. The period during which simultaneous switchover of SRP and line modules occurs is called a double fault window. The behavior of system operations after a double fault window happens is unexpected and undefined.

The following scenarios can occur during a double fault window:

Related Documentation