Versions Compared

Old Version 1

changes.mady.by.user Peter Lafler

Saved on

compared with

New Version Current

changes.mady.by.user Brian Nordland

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

The QCST_CRG_CANCEL_FAILOVER policy can prevent some automatic failovers from taking place.

Each QCST_CRG_CANCEL_FAILOVER policy is uniquely defined by the cluster resource group name specified in the domain name field, and the failover scope specified in the SCOPE keyword in the policy qualifier field, and an EVENT value . An EVENT value for the policy value field specifies which events that should not cause an automatic failover.

To add a QCST_CRG_CANCEL_FAILOVER policy to a cluster resource group (CRG) supply:

  • The name of the CRG.

  • The policy qualifier value for the SCOPE policy qualifier. These values are:

Scope

Description

*SITE

Indicates that the policy will apply only to failovers where the new primary node is in the same site as the original primary node.

*CRSSITE

Indicates that the policy will apply only to failovers where the new primary node is in a different site than the current primary node.

*CRG

The policy applies to all failovers regardless of where the current and new primary nodes are located.

  • The policy value field requires a value for the Event. The Event contains a list of event types the QCST_CRG_CANCEL_FAILOVER should cancel. At least one value needs to be specified. Values for EVENT are:

Event

Description

*CLUFAIL

The failover was caused by one of the following cluster events:

  • a CHGCLURCY(*END) ending the RGM or CCTL jobs.

  • a critical job end

  • a node in the cluster crashed.

*CRGFAIL

The failover was caused by one of the following cluster events:

  • a CHGCLURCY(*END) ending jobs other than the RGM or CCTL jobs.

  • a CRG job end

  • a node in the cluster crashed.

*ENDCLUNOD

The failover was caused by the ENDCLUNOD command.

*ENDSYS

The failover was caused by an ENDSYS or an ENDSBS(*ALL) action on the primary cluster node.

*ENDTCP

The failover was caused by invoking the ENDTCP command on the primary cluster node.

*PWRDWNSYS

The failover was caused by the PWRDWNSYS command.

*ALL

All actions listed in the above values that cause failovers.

Example 1 - Disabling automatic failovers for specific events

For example, a QCST_CRG_CANCEL_FAILOVER policy set for a CRG named MYCRG created:

...

Indicates that if an ENDCLUNOD, ENDSYS, ENDTCP, or PWRDWNSYS event occurs a failover will not be attempted for the CRG MYCRG.

Example 2 - Disabling all listed automatic failovers for a CRG

For example, a QCST_CRG_CANCEL_FAILOVER policy set for a CRG named MYCRG created:

Code Block
ADDHAPCY PCY(QCST_CRG_CANCEL_FAILOVER) PCYDMN(MYCRG) QUAL('SCOPE(*CRG)')
VALUE('EVENT(*ALL)')

Indicates that if any of the listed failover events occurs a failover will not be attempted for the CRG MYCRG.

Example 3 - Disabling all cross site failovers for a CRG

For example, a QCST_CRG_CANCEL_FAILOVER policy set for a CRG named MYCRG created:

Code Block
ADDHAPCY PCY(QCST_CRG_CANCEL_FAILOVER) PCYDMN(MYCRG) QUAL('SCOPE(*CRSSITE)')
VALUE('EVENT(*ALL)')

Indicates that if any of the listed failover events occurs and the first active backup node is a node in a different site, a failover will not be attempted for the CRG MYCRG.

This policy is useful to enable automatic failovers within a site (LUN-Level Switching), but disabling failovers that would cause a Metro Mirror or Global Mirror failover to occur.