Takeover Timeout (for third data center)

When a package is being failed over to the third data center (SVOL of the Continuous Access-Journal device group), the Metrocluster toolkit script issues takeover command on the SVOL. If the journal group pair is flushing the journal data from its PVOL to SVOL and takeover timeout occurs, the following situations would happen:

1.The device group pair state remains in PVOL-PAIR/SVOL-PAIR

2.The journal data continues transferring to the SVOL

In this case, you must wait for the completion of the journal data flushing and the Continuous Access-Journal pair state to be:

•Hot-standby site: PVOL-PAIR or PVOL-PSUS(E)

•Third site: SVOL-PSUS(SSWS) or SVOL-PSUE(SSWS)

Either 1) use FORCEFLAG to startup the package on third site or 2) fix the problem (if any of Continuous Access links was failed) and resume the data replication with the following procedures:

1.Split the Continuous Access-Journal device group pair completely (pairsplit -g <dg>-S>

2.Re-create a Continuous Access-Journal pair from it’s original PVOL as source. (use the paircreate command)

NOTE: You can specify “none” as the copy mode for initial copy operations. If the none mode is selected, full copy operations are not performed. The user is responsible for using the none mode only when the user is sure that data in the primary data volume is exactly the same as data in the secondary data volume.

Continuous Access-Journal Device Group PVOL-PAIR with SVOL-PSUS(SSWS) State

PVOL-PAIR with SVOL-PSUS(SSWS)is an intermediate state. It could happen, but is unlikely to be seen. The state PVOL-PAIR/SVOL-SSWS is an invalid state for P9000 or XP Continuous Access Journal. In this state, if you issue a pairresync or takeover command, it would fail. It is necessary to wait for the PVOL to become PSUE or PSUS.

Additional Reading

The following documents contain additional useful information:

•Managing Serviceguard Nineteenth Edition (5900-1492)

•Understanding and Designing Serviceguard Disaster Tolerant Architectures (B7660-90018)

•Designing Disaster Tolerant HA Clusters Using Metrocluster and Continentalclusters (5900–1484)

•Use the following URL to access HP’s High Availability web page: http://www.hp.com/go/hpux-ha-monitoring-docs

•Use the following URL for access to a wide variety of HP-UX documentation: http://www.hp.com/

To learn more about P9000 or HP StorageWorks XP disk arrays, contact your local HP storage representative.

474 Designing a Three Data Center Disaster Recovery Solution