Timing considerations for Windows Clustering (MSFC)

HP 3PAR Cluster Extension gives priority to storage system operations over cluster software operations. If HP 3PAR Cluster Extension is invoked during a virtual volume synchronization/full synchronization operation, or is gathering information about the remote storage system, it waits until the requested status information is reported. This ensures the priority of data integrity over cluster software failover behavior. However, this behavior can cause a failure of HP 3PAR Cluster Extension resources in some cases:

When HP 3PAR Cluster Extension communicates with the HP 3PAR storage system, online operations could time out depending on the settings of the command time out configuration.

When a virtual volume resynchronization/ full synchronization is in progress, the software waits until the Remote Copy volume group is in a synced state if the ApplicationStartup resource property is set to RESYNCWAIT. In case of resynchronization, the delta between the source and destination virtual volume could be large enough for the resynchronization process to exceed the resource PendingTimeout value.

The values of the ResyncWaitTimeout properties must be less than the configured resource

PendingTimeout value.

In general, because the failover environment is dispersed into two (or more) data centers, the failover time cannot be expected to be the same as that in a single data center with a single shared disk device. Therefore, the following values of the HP 3PAR Cluster Extension resource and the service or application using that resource, must be adjusted, based on failover tests performed to verify the proper configuration setup: FailoverPeriod, RestartPeriod, PendingTimeout, LookAlive, and IsAlive.

In addition, FailoverPeriod value of the service or application must be higher than the RestartPeriod value of the resource and both must be higher than the resource's PendingTimeout value. Microsoft Cluster provides two properties to adjust state change recognition or resolution:

IsAlive

LookAlive

NOTE: For Windows Server 2008/2008 R2/2012, the IsAlive and LookAlive properties are replaced by the Thorough Resource Health Check Interval and Basic Resource Health Check Interval properties.

Bouncing service or application

HP 3PAR Cluster Extension will alternate (start and fail) between local nodes if the ApplicationStartup property has been set to FASTFAILBACK and no remote system is available until the service or application restart limit has been reached.

Creating array password file

HP 3PAR Cluster Extension uses HP 3PAR CLI commands to manage the HP 3PAR storage system objects. HP 3PAR CLI can be installed on the client machine to manage the HP 3PAR storage systems remotely. Each CLI session requires a user name and an encrypted password for successful command execution. Alternatively, a onetime password file can be generated for subsequent use. Generated password file is specific to the HP 3PAR storage system. However, it can be copied to other systems for use. HP 3PAR Cluster extension requires this password file for its operations. For each array managed by HP 3PAR Cluster Extension, appropriate password files must be generated and configured during the HP 3PAR Cluster Extension configuration.

After you log into the HP 3PAR command shell (using the valid username and password credentials), use following command to generate the password file:

setpassword -u <user name> -saveonly -file <file name>

Timing considerations for Windows Clustering (MSFC) 41