set cluster configuration high-availability

This command is supported only on these:

  • ElasticXL Cluster

  • Maestro Security Group

  • Scalable Chassis Security Group

Note - You must run this command in Gaia gClish.

Description

Configures cluster settings - High Availability values.

Related Commands

Syntax

set cluster configuration high-availability<ESC><ESC>

set cluster configuration high-availability

      bond <Name of Bond> min_slaves <1-8>

      down_on_first_join {0 | 1}

      factors

            member <0-1000>

            pnote pingable_hosts <0-1000>

            port bond <0-1000>

            port mgmt <0-1000>

            port other <0-1000>

            port standard <0-1000>

            sensor cmm <0-1000>

            sensor fans <0-1000>

            sensor power_supplies <0-1000>

            sensor ssm <0-1000>

      failover <0-1000>

      freeze_interval {0 | <1-1000>}

      mgmt-monitoring {on | off}

      mode {0 | 1 | 2 | 3}

      port <Name of Port> priority {1 | 2}

      vs site_priority "<ID of Primary Site> <ID of Secondary Site>"

Explanation for Grade Factors

In a Dual Site environment, the Site with the higher grade becomes the Active Site and handles traffic.

The grade for each component is calculated based on this formula:

(Unit Weight) x (Number of Components in the State "UP")

To see the weight of each component, run this command in Gaia gClish and refer to the column "Weight" (see asg stat):

asg stat -v

For additional information, see:

Parameters

Parameter

Description

bond <Name of Bond> min_slaves <1-8>

Configures the minimum number of subordinate interfaces in the "UP" state in the specified bond interface.

If the number of subordinate interfaces in the "UP" state is lower than the configured threshold, then the cluster state of this bond interface changes to "DOWN".

down_on_first_join {0 | 1}

Configures whether a Security Group Member must remain in the Administratively Down state when it joins the Security Group.

  • 0 - Does not enable the administratively DOWN state for the first join.

  • 1 - Enables the administratively DOWN state for the first join.

See:

factors member <0-1000>

Configures the Grade factor for the component "Security Group Member".

factors pnote pingable_hosts <0-1000>

Configures the Grade factor for the component "Critical Device 'Pingable Hosts'".

factors port bond <0-1000>

Configures the Grade factor for the component "Bond Interfaces".

factors port mgmt <0-1000>

Configures the Grade factor for the component "Management Ports".

factors port other <0-1000>

Configures the High Grade factor for the component "Ports".

factors port standard <0-1000>

Configures the Standard Grade factor for the component "Ports".

factors sensor cmm <0-1000>

Configures the Grade factor for the component "CMMs" (only on Scalable Chassis).

factors sensor fans <0-1000>

Configures the Grade factor for the component "Fans" (only on Scalable Chassis).

factors sensor power_supplies <0-1000>

Configures the Grade factor for the component "Power Supply Units" (only on Scalable Chassis).

factors sensor ssm <0-1000>

Configures the Grade factor for the component "SSMs" (only on Scalable Chassis).

This factor applies to all SSMs.

failover <0-1000>

Configures the minimum quality grade differential that causes a failover.

freeze_interval {0 | <1-1000>}

A Standby Site cannot failover a second time until the specified failover freeze interval expires.

If the Standby Site grade changes to a value greater than the minimum quality grade gap for a failover, the Standby Site fails over and becomes a new Active Site.

The failover does not start until the freeze interval expires. This confirms that the Standby Site quality grade is stable, before it becomes a new Active Site.

The default failover freeze interval is:

  • 30 seconds for the "Active Up" mode ("set cluster configuration high-availability mode 0").

  • 150 seconds for the "Primary Up" mode ("set cluster configuration high-availability mode 1").

  • 150 seconds for the "VSLS" mode ("set cluster configuration high-availability mode 3").

Valid freeze time values:

  • Value 0 - Disables this threshold.

  • Values 1-1000 - Specifies the freeze time in seconds.

    If this Security Group is configured to work in the "VSLS" mode or in the "Primary Up" mode, then you must enter the value that is 5 times smaller that the required freeze time.

    Example: If the freeze time must be 250 seconds, you must enter the value 50.

mgmt-monitoring {on | off}

Configures whether to monitoring the Management Ports:

  • on - Enables the monitoring of the Management Ports.

    • The monitored management ports are included in the Standby Site grade mechanism, according to the predefined factors.

    • The output of the "asg stat -v" command shows the Management ports.

    • The "show interfaces" command in Gaia gClish shows the link state of management interfaces based on this feature mechanism.

  • off - Disables the monitoring of the Management Ports (this is the default).

mode {0 | 1 | 2 | 3}

Configures the Dual Site mode:

  • 0 - No primary Site ("Active Up" Mode)

  • 1 - "Primary Up" Mode

  • 2 - "Active Active" Mode

  • 3 - Legacy VSX Virtual System Load Sharing (VSLS) Mode

port <Name of Port> priority {1 | 2}

Configures the Priority Level for the specified port:

  • 1 - Standard.

  • 2 - Other.

Use the "set cluster configuration high-availability port ... priority ..." command together with the "set cluster configuration high-availability factors port ..." command:

  1. Set the port grade as standard or high.

    For example, to set the standard grade at 50, run:

    set cluster configuration high-availability factors port standard 50

  2. Set the port to high grade or standard grade.

    For example, to assign the standard port grade to eth1-01, run:

    set cluster configuration high-availability port eth1-01 priority 1

vs site_priority "<ID of Primary Site> <ID of Secondary Site>"

Configures the priority for each Site for the current Legacy VSX Virtual System.

You must run this command in the context of the applicable Virtual System.

Example - Configuration of Site 2 to be the Primary and Site 1 to be the Secondary for the Virtual System 4:

  1. set virtual-system 4

  2. set cluster high-availability vs site_priority "2 1"