*******************************************************************************
* *
* This is an update to an existing patch... *
* *
* Online links can be found at *
* http://ftp.support.compaq.com/patches/public/unix/v5.1/earlyreleasepatches/oracle/oracle9i.README
*******************************************************************************

TITLE: Tru64 UNIX/Cluster ORACLE9I Operation Fixes for V5.1
 
New Kit Date: 26-JUL-2001
Modification Date: Not Applicable
Modification Type: The kit replaces the Oracle 9I kit that was
                    previously released on July 11, 2001.

PRODUCT: Oracle 9I

OP/SYS: Tru64 UNIX/Cluster V5.1

SOURCE: Compaq Computer Corporation

This directory contains files needed for proper operation of Oracle 9i
with Tru64 and TruCluster 5.1 with patch kit 3 (BL17).

For each of the selected products, COMPAQ recommends the following
be downloaded from this directory and installed on the cluster.

  Oracle9i.tar (Checksum: 40190 8420)

Which contains:

 T64V5117-C0038308-11129-ER-20010627.tar (Checksum: 15662 6850)

 TCV5117-C0009702-11280-ER-20010723.tar (Checksum: 17878 1560)

This patch contains corrections required for proper operation of Oracle 9i
with Tru64 UNIX/TruCluster 5.1. The problems corrected include:

   - Processes hanging when using Cluster File System/Direct I/O feature.

   - Improper handling of direct I/O to an AdvFS fileset if a clone fileset
     was already in use, potentially resulting in an inconsistent backup.

   - Using ls -l, the Cluster File System file attribute could be seen
     inconsistently from the server and client members. For example a
     file's mode could be seen differently from the server and the client.

   - A file opened for Direct I/O on the Cluster File System server may
     inappropriately be opened in non-direct I/O mode by a client.

   - Oracle processes hanging due to shutting down one cluster member.

   - Performance improvements for Oracle 9i with RDG.

   - A kernel problem which could cause a cluster system to panic with
     the panic string "acquire_rm_spinlock called at ipl."

   - A cluster problem which could cause one member of a cluster to panic
     with the panic string "ics_mct: Error from establish_RM_notification_channel"
     and others to panic with the string "ics_mct: Error from
     register_RM_notification_callback."

   - A problem with the Cluster File System which could cause a cluster
     system to panic with the panic string "kernel memory fault" in the
     routine mc_bcopy().

   - A problem with Cluster File System which could cause a cluster
     member to panic with the panic string "uiomove: mode." This problem
     could cause Oracle multi-instance data bases to crash with the
     message similar to the following:

     "ORA-27063: skgfospo: number of bytes read/written is incorrect"

   - A problem with Cluster File System which could case a cluster member
     to panic with the panic string "cfsdb_assert" in cfs_reclaim().

   - Three problems in the RDG subsystem that could cause a cluster member to
     panic with the panic string "vl_unwire: page is not wired"

---

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]