Configuring Link Monitoring on the Cluster Interfaces

Important - In a Cluster, you must configure all the Cluster Members in the same way.

Description

This procedure configures the Cluster Member to monitor only the physical link on the cluster interfaces (instead of monitoring the Cluster Control Protocol (CCP) packets):

  • If a link disappears on the configured interface, the Cluster Member changes the interface's state to DOWN.

    This causes the Cluster Member to change its state to DOWN.

  • If a link appears again on the configured interface, the Cluster Member changes the interface's state back to UP.

    This causes the Cluster Member to change its state back to ACTIVE or STANDBY.

See Viewing Cluster State.

Procedure

Step

Instructions

1

Connect to the command line on the Cluster Member.

2

Log in to the Expert mode.

3

See if the $FWDIR/conf/cpha_link_monitoring.conf file already exists:

stat $FWDIR/conf/cpha_link_monitoring.conf

4

If the $FWDIR/conf/cpha_link_monitoring.conf file already exists, create a backup copy:

cp -v $FWDIR/conf/cpha_link_monitoring.conf{,_BKP}

If the $FWDIR/conf/cpha_link_monitoring.conf file does not exist, create it:

touch $FWDIR/conf/cpha_link_monitoring.conf

5

Edit the $FWDIR/conf/cpha_link_monitoring.conf file:

vi $FWDIR/conf/cpha_link_monitoring.conf

6

  • To monitor the link only on specific interfaces:

    Enter the names of the applicable interfaces - each name on a new separate line.

    Example:

    eth2

    eth4

  • To monitor the link on all interfaces:

    Enter only this word:

    all

7

Save the changes in the file and exit the editor.

8

Reboot the Cluster Member.

Important - This can cause a failover.

Best Practices:

    1. Perform the configuration steps on all Cluster Members

    2. Reboot all the Standby Cluster Members

    3. Initiate a manual failover on the Active Cluster Member

    4. Reboot the former Active Cluster Member

    1. Perform the configuration steps on all Cluster Members

    2. Reboot all the non-Pivot Cluster Members

    3. Initiate a manual failover on the Pivot Cluster Member

    4. Reboot the former Pivot Cluster Member

    1. Perform the configuration steps on all Cluster Members

    2. Reboot all Cluster Members except one

    3. Initiate a manual failover on the remaining Cluster Member

    4. Reboot the remaining Cluster Member

Note - See Initiating Manual Cluster Failover.