HP Storage Management for Oracle 5.0 Release Notes, HP-UX 11i v3, First Edition, May 2008

• Problem
When an SFCFS file system is not mounted, it is possible that the fsck operation of the
snapshot will take longer than the usual time to complete, depending on the I/O activity at
the time of the snapshot.
Workaround
When taking a snapshot of an SFCFS file system, ensure that at least one of the volumes
defined in the command line is mounted on the cluster master.
• Problem
Mount operation can cause inconsistencies in snapshots.
Inconsistencies can arise in point-in-time copies if any of the following snapshot operations
are performed on a volume when a file system in the volume is being mounted:vxassist
snapshot vxplex snapshotvxsnap make vxsnap refreshvxsnap restore.
Workaround
There is no workaround for this issue.
• Problem
Adding cache volumes to volume sets causes data corruption. Do not add cache volumes
used by space-optimizes instant snapshots to volume sets.
Workaround
There is no workaround for this issue.
• Problem
VxVM Flashsnap operations on volume sets with mounted VxFS file systems can cause loss
of changes.
The operation to create snapshots on volume sets experiences a known problem when a
VxFS file system is created on top of it and if the filesystem is mounted while creating the
snapshot. The snapshot volume that is created may lose recent changes in the file system of
the parent volume set just before the snapshot operation was performed.
Workaround
When taking a snapshot of an SFCFS file system, ensure that at least one of the volumes
defined in the command line is mounted on the cluster master.
• Problem
When using Oracle with ODM, Oracle encounters an I/O error. One of the following error
messages will be displayed:
ORA-12801: error signaled in parallel query server P005
ORA-01115: IO error reading block from file 5 (block # 64128)
ORA-17500: ODM err:ODM ERROR V-41-4-1-264-14 Bad address
ORA-01115: IO error reading block from file 5 (block # 64128)
ORA-17500: ODM err:ODM ERROR V-41-4-1-264-14
Workaround
There is no workaround for this issue. A fix for this issue is planned for a future patch. If
you need a fix for this issue prior to the availability of the patch, contact the HP support
personnel.
• Problem
If an Oracle 9i database is upgraded to Oracle 10g, and at a later time the database is rolled
back to the Oracle 9i environment, then the recovery procedure for Oracle 9i will log the
following errors in the alert log file:
16 HP Storage Management for Oracle (SMO) 5.0 Release Notes