Netgear XCM8810 - 8800 SERIES 10-SLOT CHASSIS SWITCH ユーザーズマニュアル

ページ / 968
68
   
|   
Chapter 3.  Managing the Switch  
NETGEAR 8800 User Manual 
Dynamic Checkpointing
After an application transfers its saved state to the backup node, dynamic checkpointing 
requires that any new configuration information or state changes that occur on the primary be 
immediately relayed to the backup. This ensures that the backup has the most up-to-date 
and accurate information.
Viewing Checkpoint Statistics
To view and check the status of one or more processes being copied from the primary to the 
backup node, use the following command:
show checkpoint-data {<process>}
This command is also helpful in debugging synchronization problems that occur at run time.
This command displays, in percentages, the amount of copying completed by each process 
and the traffic statistics between the process on both the primary and the backup nodes.
Viewing Node Status
XCM8800 allows you to view node statistical information. Each node in a NETGEAR 8800 
installed in your system is self-sufficient and runs the management applications. By reviewing 
this output, you can see the general health of the system along with other node parameters. 
To view node status, use the following command:
show node {detail}
 lists the node status collected by the switch.
Table 11.  Node States 
Node State
Description
BACKUP
In the backup state, this node becomes the primary node if the primary fails or enters the DOWN 
state. The backup node also receives the checkpoint state data from the primary.
DOWN
In the down state, the node is not available to participate in leader election. The node enters this 
state during any user action, other than a failure, that makes the node unavailable for 
management. Examples of user actions are:
• Upgrading the software
• Rebooting the system using the 
reboot
 command
• Initiating an MSM/MM failover using the 
run msm-failover
 command
• Synchronizing the MSM/MM software and configuration in non-volatile storage using the 
synchronize
 command
FAIL
In the fail state, the node has failed and needs to be restarted or repaired. The node reaches this 
state if the system has a hardware or software failure.
INIT
In the initial state, the node is being initialized. A node stays in this state when it is coming up and 
remains in this state until it has been fully initialized. Being fully initialized means that all of the 
hardware has been initialized correctly and there are no diagnostic faults.