Building Disaster Recovery Serviceguard Solutions Using Metrocluster with Continuous Access XP P9000 for Linux B.12.00.00

5 Understanding failover/failback scenarios
Metrocluster package by default fails to start if the data is not current or if it is not able to determine
the status of the device group. In such situations, the user has options to start a package either by
setting the value of Metrocluster package AUTO parameters to “1” or by using FORCEFLAG. To
use FORCEFLAG option, you must create a FORCEFLAG file in the package directory
(<package_directory>/FORCEFLAG). Using FORCEFLAG, the package can be started without
making any changes to the Metrocluster package configuration. Before starting the package
forcefully using FORCEFLAG, ensure that the data is consistent and safe to start. Also, run the
cmdrprev command to preview the data replication storage failover.
When the Continuous Access links fail , the status of the journal device group changes from PAIR
to PSUE state if the Mode 449 is set to “OFF” while Mode 449 setting to “ON” retains the PAIR
state. Mode 449 is a system mode setting in an array. The PAIR state is retained as long as the
journal volumes have enough space . If the journal volumes reach full capacity, the pair state is
switched to PFUS and the data written to the data volume is tracked in bitmap.
Failover/failback scenarios in a Metrocluster package
This section describes the package start up behaviors in various failure scenarios depending on
AUTO parameters or FORCEFLAG file and fence level. Table 6 describes the list of failover/failback
scenarios.
Table 6 Failover/failback scenarios
AUTO parameters or FORCEFLAG setMetrocluster Behavior (By default)Fence LevelFailover/Failback Scenarios
The behavior is not affected by the
AUTO parameter setting or by the
presence of the FORCEFLAG file.
Metrocluster package starts and
swaps the replication role by issuing
horctakeover.
AnyFailover to recovery site
due to an application
failure on all nodes in the
primary site.
The behavior is not affected by the
AUTO parameter setting or by the
presence of the FORCEFLAG file.
Metrocluster package starts and
swaps the replication role by issuing
horctakeover.
AnyFailback to primary site
Metrocluster package checks for
AUTO_NONCURDATA setting and
Metrocluster package fails to start
as AUTO_NONCURDATA is set to
“0” and FORCEFLAG is not present.
data or never
or async
(using side
file)
Failover to recovery site
due to failure of all nodes
in the primary site. FORCEFLAG presence. If
AUTO_NONCURDATA is set to “1”
or FORCEFLAG is present,
Metrocluster package starts and
issues horctakeover which results in
swap takeover.
NOTE: In case of fence=never,
horctakeover results in SVOL takeover
and the data does not replicate back
to the primary site. Metrocluster
automatically issue pairresync
-swaps to replicate data back to the
primary site.
The behavior is not affected by the
AUTO parameter setting or by the
presence of the FORCEFLAG file.
Metrocluster package starts and
swaps the replication role by issuing
horctakeover.
async (using
journal)
The behavior is not affected by the
AUTO parameter setting or by the
presence of the FORCEFLAG file.
Metrocluster package starts and
swaps the replication role by issuing
horctakeover.
AnyFailback to primary site
after the recovery of
primary site nodes and the
data is being replicated
from the recovery site to the
primary site.
36 Understanding failover/failback scenarios