There are some restrictions for using the DB2® high availability instance configuration utility (db2haicu).
db2haicu does not support the configuration of mount resources which are based off of Logical Volume Manager (LVM) on any platform other than AIX.
You cannot perform the following tasks using db2haicu:
You cannot configure automatic client reroute using db2haicu.
db2haicu is not supported in a DB2 pureScale® environment. Use the db2cluster command instead to configure clustered environments.
Consider the following db2haicu usage notes when planning your cluster configuration and administration activities:
Even though db2haicu performs some administration tasks that normally require root authority, db2haicu runs with the privileges of the database manager instance owner. db2haicu initialization, which is performed by a root user, enables db2haicu to carry out the required configuration changes despite having only instance owner privileges.
When you create a new cluster domain, db2haicu does not verify that the name you specify for the new cluster domain is valid. For example, db2haicu does not confirm that the name is a valid length, or contains valid characters, or that is not the same name as an existing cluster domain.
db2haicu does not verify or validate information that a user specifies and that is passed to a cluster manager. Because db2haicu cannot be aware of all cluster manager restrictions with respect to cluster object names, for example, db2haicu passes text to the cluster manager without validating it for things like valid characters, or length.
When you run db2haicu with the -delete parameter, db2haicu deletes the resource groups that are associated with the current database manager instance immediately, without confirming whether those resource groups are locked.
If a cluster operation you attempt to perform using db2haicu times out, db2haicu does not return an error to you. When a cluster operation times out, you do not know that the operation timed out unless you review diagnostic logs after you make the db2haicu call; or unless a subsequent cluster action fails, and while you are investigating that subsequent failure, you determine that the original cluster operation timed out.
If you attempt to change the failover policy for a given database instance to active-passive, there is one condition under which that configuration operation fails, but for which db2haicu does not return an error to you. If you specify a machine that is currently offline to be the active machine, db2haicu does not make that machine the active machine, but db2haicu does not return an error that indicates that the change did not succeed.
For a shared disk configuration, db2haicu does not support a nested mount configuration because DB2 does not enforce the disk mount order.
mkequ <name> IBM.NetworkInterface:<eth0>:<node0>,...,<ethN>:<nodeN>
The following is a list of recommendations for configuration your cluster, and your database manager instances when you are using db2haicu.
dev/vpatha1 /db/svtpdb/NODE0010 ext3 noauto 0 0